Skip to main content

Full text of "American journal of philology"

See other formats


This  is  a  digital  copy  of  a  book  that  was  preserved  for  generations  on  library  shelves  before  it  was  carefully  scanned  by  Google  as  part  of  a  project 
to  make  the  world's  books  discoverable  online. 

It  has  survived  long  enough  for  the  copyright  to  expire  and  the  book  to  enter  the  public  domain.  A  public  domain  book  is  one  that  was  never  subject 
to  copyright  or  whose  legal  copyright  term  has  expired.  Whether  a  book  is  in  the  public  domain  may  vary  country  to  country.  Public  domain  books 
are  our  gateways  to  the  past,  representing  a  wealth  of  history,  culture  and  knowledge  that's  often  difficult  to  discover. 

Marks,  notations  and  other  marginalia  present  in  the  original  volume  will  appear  in  this  file  -  a  reminder  of  this  book's  long  journey  from  the 
publisher  to  a  library  and  finally  to  you. 

Usage  guidelines 

Google  is  proud  to  partner  with  libraries  to  digitize  public  domain  materials  and  make  them  widely  accessible.  Public  domain  books  belong  to  the 
public  and  we  are  merely  their  custodians.  Nevertheless,  this  work  is  expensive,  so  in  order  to  keep  providing  this  resource,  we  have  taken  steps  to 
prevent  abuse  by  commercial  parties,  including  placing  technical  restrictions  on  automated  querying. 

We  also  ask  that  you: 

+  Make  non-commercial  use  of  the  files  We  designed  Google  Book  Search  for  use  by  individuals,  and  we  request  that  you  use  these  files  for 
personal,  non-commercial  purposes. 

+  Refrain  from  automated  querying  Do  not  send  automated  queries  of  any  sort  to  Google's  system:  If  you  are  conducting  research  on  machine 
translation,  optical  character  recognition  or  other  areas  where  access  to  a  large  amount  of  text  is  helpful,  please  contact  us.  We  encourage  the 
use  of  public  domain  materials  for  these  purposes  and  may  be  able  to  help. 

+  Maintain  attribution  The  Google  "watermark"  you  see  on  each  file  is  essential  for  informing  people  about  this  project  and  helping  them  find 
additional  materials  through  Google  Book  Search.  Please  do  not  remove  it. 

+  Keep  it  legal  Whatever  your  use,  remember  that  you  are  responsible  for  ensuring  that  what  you  are  doing  is  legal.  Do  not  assume  that  just 
because  we  believe  a  book  is  in  the  public  domain  for  users  in  the  United  States,  that  the  work  is  also  in  the  public  domain  for  users  in  other 
countries.  Whether  a  book  is  still  in  copyright  varies  from  country  to  country,  and  we  can't  offer  guidance  on  whether  any  specific  use  of 
any  specific  book  is  allowed.  Please  do  not  assume  that  a  book's  appearance  in  Google  Book  Search  means  it  can  be  used  in  any  manner 
anywhere  in  the  world.  Copyright  infringement  liability  can  be  quite  severe. 

About  Google  Book  Search 

Google's  mission  is  to  organize  the  world's  information  and  to  make  it  universally  accessible  and  useful.  Google  Book  Search  helps  readers 
discover  the  world's  books  while  helping  authors  and  publishers  reach  new  audiences.  You  can  search  through  the  full  text  of  this  book  on  the  web 

at  http  :  //books  .  google  .  com/| 


Digitized  by  VjOOQ IC 


.   h 


Digitized  by  VjOOQ IC 


THE 


AMERICAN 


Journal  of  Philology 


EDITED  BY 

BASIL  L.  GILDERSLEEVE 

Prof*stor  of  Greek  in  ifu  yohnt  Hopkint  University 


&  Co. 


Digitized  by  VjOOQ IC 


Digitized  by  VjOOQ IC 


THE 


AMERICAN 


Journal  of  Philology 


EDITED  BY 

BASIL  L.  GILDERSLEEVE 

Prof*ttor  of  Grttk  in  tkt  yohnt  Hophint  Univtrtiiy 


VOL.  XVIII 


BALTIMORE:  THE  EDITOR 

Nxvr  York  and  London:  Macmillan  &  Co. 

Lripsic:   F.  a.  Brockhaus 

1897 


Digitized  by  VjOOQ IC 


^it  ftithnmatb  Company 

BALTIMORB.  MD..  U.S.A. 


Digitized  by  VjOOQ IC 


Digitized  by  VjOOQ IC 


iv  CONTENTS, 

No.  70. 

I. — The  Subjunctive  in  Independent  Sentences  in  Plautus.     I.    By  E. 

P.  Morris, 133 

II.— Textual  Notes  and  Queries  on  Plautus.    By  Edwin  W.  Fay,  .  168 

III. — Superstitions  and  Popular  Beliefs  in  Greek  Comedy.     By  Ernst 

RiESS, 189 

IV.— On  the  Definition  of  Some  Rhetorical  Terms.     By  V.  J.  Emery,     .  206 

Reviews  AND  Book  Notices  : 214 

Schanz*s  Beitrage  zur  historischen  Syntax  der  griechischen  Sprache. 

Band  III,  Heft  3  u.  4.  Dyroff*s  Geschichte  des  Pronomen  Reflexivum. 

— Rambeau  and  Passy's  Chrestomathie  frangaise. — Karsten's  Journal 

of  Germanic  Philology. 

Reports: 228 

Revue  de  Philologie. — Englische  Studien. — Rheinisches  Museum  ftir 
Philologie. 

Brief  Mention, 242 

Necrology, 247 

Recent  Publications, 248 

Books  Received, 253 

No.  71. 
I. — The  Ethics  and  Amenities  of  Greek  Historiography.    By  B.  Perrin,  255 

II. — The  Subjunctive  in  Independent  Sentences  in  Plautus.     II.     By 

E.  P.  Morris 275 

III. — Caecilius  of  Calacte.    By  W.  Rhys  Roberts,  .302 

IV. — Are  the  Letters  of  Horace  Satires?    By  G.  L.  Hendrickson,  .  313 

V. — Notes  on  Horace.    By  Charles  Knapp, 325 

Note:  339 

On  Lucian's  Nigrinus.    By  Emily  James  Smith. 
Reviews  and  Book  Notices  : 342 

Jusserand's  A  Literary  History  of  the  English  People. — Hillebrandt's 
Ritual-Literatur. — Kaibers  Sophokles  Elektra. 

Reports: 357 

Romania. — Hermes. 

Brief  Mention, 368 

Necrology, 371 

Recent  Publications, 376 

Books  Received 381 


Digitized  by  VjOOQ IC 


CONTENTS. 
No.  72. 


Digitized  by  VjOOQ IC 


Digitized  by  VjOOQ IC 


Digitized  by  VjOOQ IC 


2  AMERICAN  JOURNAL    OF  PHILOLOGY, 

any  wide  acceptance.    Some  other  young  men,  not  so  thoroughly 
grounded  in  the  literary  tradition  as  their  elders,  returned  from 
Greece ^ith  more  or  less  enthusiasm  for  the  revolutionary  doctrine 
and  for  its  author.    In  Hermes  for  1886  (pp.  597  ft)  appeared  an 
article  by  Wilamowitz,  *Die  Biihne  des  Aischylos.'    This  was 
based  in  part  on  the  new  view,  and  made  it  widely  known  in 
fragmentary  form,  though  the  article  is  now  seen  to  have  con- 
tained some  hay  and  stubble  along  with  better  material.    Then 
it  was  announced  that  our  revolutionary  architect  was  to  publish 
a  book  on  the  subject  soon.    Years  passed  and  the  book  did  not 
appear ;  some  were  inclined  to  scoff  and  others  to  grieve.    Mean- 
time on  other  grounds  Dorpfeld's  reputation  was  growing.     His 
part  in  the  Olympia  publications  and  a  steady  stream  of  papers 
in  the  Mittheilungen  attested  his  power.    Schliemann  secured  his 
help  in  excavating  Tiryns,  and  later  Troy;  in  1887  Dorpfeld 
succeeded  Petersen  as  First  Secretary  of  the  Institute  in  Athens, 
and  the  stream  of  publications  never  ceased.    It  was  more  and 
more  widely  recognized  that  a  mind  of  exceptional  force,  training 
and  candor  was  at  work  on  the  scattered  remnants  of  Greek 
architecture  of  every  sort,  discovering  order  in  supposed  chaos, 
revealing  significance  in  little  facts  till  then  unnoticed,  clarifying 
and  enlarging  our  knowledge  of  Hellenic  and  prehistoric  building, 
and  vastly  improving  the  young  art  of  excavation.    Further, 
even  before  he  became  the  head  of  the  Institute,  he  adopted  the 
plan  of  explaining  on  the  spot  accessible  architectural  remains  to 
'stipendiaries'  of  the  Institute  and  others.    This  plan  developed 
into  three  separate  courses,  identical  in  character  and  differing 
only  in  the  location  of  the  monuments  and  the  means  of  reaching 
them.    The  first  was  a  weekly  peripatetic  lecture  two  or  three 
hours  in  length  during  the  winter,  in  Athens  and  vicinity.    The 
second  was  a  journey  to  Mykenai,  Tiryns,  Epidauros,  and  across 
Arkadia  to  Olympia:  this  has  been  extended  and  enriched  as 
excavation  has  increased  the  material  for  study  and  as  facilities 
for  travel  have  improved.    Third  comes  a  trip  by  sea  for  similar 
examination  of  some  of  the  islands  and  places  on  the  coast  inac- 
cessible otherwise  for  large  parties.    The  severe  charm  of  those 
lectures  no  one  can  forget  who  has  been  so  fortunate  as  to  hear 
them;  and  during  these  years  a  goodly  number  of  scholars, 
younger  and  older,  have  carried  back  from  them  to  Europe  and 
America  more  just  impressions  of  Dorpfeld's  mastery  of  the  entire 
field  of  Greek  architecture,  as  well  as  a  more  accturate  knowledge 


Digitized  by  VjOOQ IC 


DORPFELD  AND  THE  GREEK  THEATRE.  3 

of  existing  theatres  and  a  clearer  perception  of  the  cogency  of 
arguments  drawn  from  the  character,  position,  shape  and  method  of 
working  of  walls,  pavement,  and  the  very  workmen's  chips.  Also, 
with  the  generosity  of  a  mind  too  rich  to  fear  that  some  one  else 
may  by  previous  publication  rob  him  of  personal  glory,  Dorpfeld 
has  allowed  and  freely  assisted  others  to  publish  his  results,  con- 
tenting himself  with  oral  discussion  in  his  lectures,  and  with  occa- 
sional reviews,  in  the  Berliner  Philologische  Wochenschrift,  of 
publications  in  which  his  doctrine  was  either  defended  or  attacked. 
Gradually  the  circle  of  adherents  has  enlarged;  especially  in 
Germany  and  America,  these  have  examined  anew  in  the  light  of 
his  views  every  scrap  of  the  literary  tradition  and  every  fragment 
of  archaeological  evidence  other  than  architectural.  More  theatres 
have  been  uncovered  and  those  already  known  have  been  more 
fully  investigated,  by  Greek,  French,  American  and  English 
scholars.  Sometimes  the  aim  was  to  support,  sometimes  to  over- 
throw the  new  views,  but  always  the  result  was  more  light.  Every 
conceivable  argument  in  defence  of  orthodoxy  has  been  urged, 
and  so  Dorpfeld  has  gained  the  opportunity  to  consider  and  meet 
all  difficulties  that  could  be  raised.  On  the  whole  then — seeing 
the  author  has  lived — it  is  perhaps  just  as  well  that  the  long- 
promised  *Theaterbuch'  has  been  delayed  till  now.  *  The  greater 
part  of  the  investigation  was  substantially  finished,'  we  are  told  in 
the  preface,  *in  the  years  1884-8.'  If  the  delay  has  in  part 
deprived  the  book  of  the  charm  of  novelty,  it  has  made  possible 
broader  and  deeper  foundations  and  far  greater  completeness 
and  permanence  of  superstructure.  It  finds  even  the  stoutest 
defenders  of  the  old  tradition,  among  our  English  brethren  for 
example,  already  abandoning  the  main  fortress  and  withdrawing 
to  one  or  two  small  redoubts.  No  one  any  longer  believes  in  a 
high  stage  for  the  fifth  and  fourth  centuries :  the  utmost  claim 
now  goes  no  farther  than  to  assume  as  probable  a  very  low 
temporary  stage,  wholly  without  evidence,  for  the  Hellenic,  and 
a  high  Vitruvian  stage  for  the  Hellenistic  period.  That,  I  say,  is 
abandoning  the  main  position.  The  very  last  redoubt  will 
eventually  be  carried  by  the  book  before  us. 

The  radical  difference  between  this  and  all  previous  books  on 
the  Greek  theatre  has  already  been  indicated.  Dorpfeld  him- 
self approached  the  problem  from  the  archaeological  instead  of 
the  literary  side,  and  his  book  does  the  same.  As  an  architect 
he  examines  the  Greek  theatres  existing  and  adequately  exca- 


Digitized  by  VjOOQ IC 


4  AMERICAN  JOURNAL    OF  PHILOLOGY. 

vated ;  his  aim  is  to  interpret  the  extant  remains  and  from  them 
to  reconstruct  the  original  form  and  trace  its  history,  appealing  to 
the  literature  and  inscriptions  for  such  help  as  they  can  give. 
History  of  the  drama,  and  its  relation  to  other  literature,  to 
religion  and  to  the  state,  he  leaves  to  others,  except  so  far  as 
these  relations  throw  light  on  his  special  topic,  the  theatre  itself. 
Now  it  is  clear  that  if  we  wish  to  know  what  the  Greek  theatre 
was,  existing  theatres  are  our  best  witnesses.  Nearly  sixty  years 
ago  A.  Schonbom  fully  recognized  this,  and  made  his  journey  to 
Asia  Minor  in  1841-2  mainly  in  hope  of  there  finding  theatres  in 
good  preservation.  But  the  best  he  found  was  that  of  Aspendos. 
With  the  best  possible  use  of  this,  he  was  forced,  as  were  his 
successors,  to  make  Vitruvius  and  the  grammarians  the  real 
starting-point,  appealing  to  the  extant  plays  for  confirmation  or 
occasional  rectification.  The  Athenian  theatre  was  not  brought 
to  light  till  1862,  and  then  but  partially ;  not  till  Dorpfeld  took  it  in 
hand  twenty  years  later  did  the  QKi\vr\  begin  to  become  intelligible ; 
one  important  feature  even  he  did  not  discover  till  1895.  The 
smaller  Peiraeus  theatre  near  the  harbor  of  Zea  was  excavated  in 
1880,  that  of  Epidauros  in  1881;  and  these  three  were  the 
earliest  Greek  theatres  to  become  adequately  known.  Even 
Albert  Muller  in  1886,  and  Haigh  in  1889,  could  go  but  a  little 
way  beyond  their  predecessors  along  this  road,  and  for  that  little 
were  indebted  to  Dorpfeld,  and  to  Dorpfeld's  pupil  Kawerau,  in 
Baumeister's  Denkmdler.  My  object  is  merely  to  note  how 
recently  it  has  become  possible  to  start  from  anything  but  literary 
tradition  in  forming  a  notion  of  the  Greek  theatre.  It  was 
inevitable  that  the  system  built  on  the  old  foundation  should 
differ  from  that  of  Dorpfeld,  built  on  the  new,  and  that  those 
trained  in  the  old  system  should  oppose  the  new  vigorously.  For 
even  now,  if  one  has  not  with  his  own  eyes  seen  those  fragmentary 
ruins  assume  form  and  meaning  under  Dorpfeld's  interpretation, 
it  is  difficult  to  realize  how  full  is  the  story  they  tell  and  how  plain 
is  their  language,  now  that  this  has  once  been  deciphered.  The 
controversy  over  the  Megalopolis  theatre  is  a  good  illustration. 
It  was  really  not  to  be  wondered  at  that  the  English  excavators 
in  1890  and  1891  misread  the  signs  before  them;  yet  on  nearly 
every  point  they  now  read  them  as  Dorpfeld  does,  and  the  visitor 
who  has  learned  from  him  the  elements  of  that  language  may 
easily  do  them  injustice  in  the  confidence  of  his  own  recent 
knowledge.    This  different  starting-point  of  the  entire  discussion, 


Digitized  by  VjOOQ IC 


DORPFELD  AND  THE  GREEK  THEATRE. 


Digitized  by  VjOOQ IC 


6  AMERICAN  JOURNAL    OF  PHILOLOGY. 

mass  of  apparent  confusion  as  the  precinct  of  Dionysos,  one  can 
best  appreciate  who  has  spent  hours  on  the  spot  endeavoring  to 
make  things  out  from  the  plans  and  descriptions  of  others. 

One  peculiarity  that  immediately  strikes  the  eye  is  the  absence 
of  footnotes.  Of  course  the  footnote  has  its  uses.  In  it  one  can 
give  references,  and  an  occasional  aside  or  a  polemic  observation 
that  really  requires  to  be  uttered,  without  interrupting  the  course 
of  demonstration  or  the  peaceful  flow  of  exposition.  But  the 
difficulty  that  many  feel  in  properly  integrating  their  disjointed 
thoughts,  in  properly  relating  the  straggling  observation  and  the 
uncontrollable  polemic  impulse  to  an  organic  whole — this  diffi- 
culty, joined  with  the  tendency  towards  pedantry  to  which  we 
professing  scholars  are  all  liable,  has  led  to  a  monstrous  develop- 
ment of  the  parasitic  footnote  growth.  It  is  a  pleasure,  therefore, 
to  read  a  book  whose  authors  do  not  feel  obliged  to  prove  their 
breadth  of  reading  by  giving  up  a  large  portion  of  the  page  to 
unincorporated  and  unincorporable  matter.  Consequently,  when 
the  eye  does  catch  a  footnote,  one  immediately  finds  it  interesting. 
Now,  it  is  a  frank  acknowledgment  that  in  some  minor  point  the 
authors  are  not  in  agreement  (pp.  lo  and  148)  ;  again  (p.  8),  it  is 
the  announcement  of  a  paper  presented  at  the  Institute  two  years 
ago,  though  not  yet  printed,  in  which  Dorpfeld  has  proved  that 
Kollytos,  Alopeke  and  Kynosarges  lay  across  the  Ilissos  south  of 
the  Acropolis.  As  regards  polemic,  the  preface  puts  the  case 
admirably:  "Wir  glaubten  eine  ins  Einzelne  gehende  Polemik 
vermeiden  zu  sollen  gegen  Behauptungen,  denen  wir  die  Grund- 
lagen  durch  die  von  uns  dargelegten  neuen  Thatsachen  entzogen 
haben."  In  point  of  literary  style  too  the  chapters  by  Dorpfeld 
are  notable.  Those  qualities  which  his  American  audiences  lately 
admired  in  his  speaking  appear  in  more  perfect  form  in  whatever 
he  writes.  No  one  could  more  religiously  avoid  ornament;  there 
is  nothing  Dorpfeld  more  strenuously  disclaims  in  private  than 
literary  skill.  Using  always  the  plainest  language,  he  makes  it 
his  sole  aim  to  present  his  subject  free  from  all  entanglements, 
with  all  attainable  simplicity.  The  result  is  a  German  style 
whose  lucidity  could  not  be  surpassed  in  French;  his  quiet 
description  of  the  old  stones  and  their  meaning  glows  with  a  calm 
but  unflagging  enthusiasm  that  imparts  itself  to  the  reader. 
Would  that  philologists  and  all  other  men  of  science  would  profit 
by  the  example.  The  first  step  towards  a  like  achievement  is 
an  equal  mastery  of  one's  subject ;  the  second  is  a  like  singleness 
and  sincerity  of  purpose. 


Digitized  by  VjOOQ IC 


DORPFELD  AND  THE  GREEK  THEATRE.  J 

In  the  description  of  the  Dionysiac  precinct  the  first  twenty-five 
pages  or  so  have  a  delightfully  familiar  sound  to  one  who  has  had 
the  privilege  of  listening  to  him  under  the  shelter  of  the  ancient 
rock-fortress  on  those  golden  winter  afternoons ;  in  the  succeeding 
part  many  details  are  added  that  those  lectures,  though  extending 
to  eight  or  nine  hours,  did  not  allow  time  for.  It  would  be 
impossible  to  summarize  all  this  here  with  any  useful  result ;  a 
few  significant  points  only  will  be  noticed.  It  is  a  part  of  Dorp- 
feld's  method  that  he  begins  with  other  portions  of  the  precinct — 
the  boundary  wall,  the  altar,  the  newer  and  the  older  temple  of 
Dionysos;  in  this  way  the  theatre  takes  its  proper  place  as  one 
part  only  of  the  enclosure.  A  circumstance  passed  over  Hghtly 
in  the  book,  after  Dorpfeld's  manner,  illustrates  again  the  recent 
date  and  the  main  source  of  our  ability  to  read  the  language  of 
architectural  remains.  Some  of  the  most  important  walls,  includ- 
ing the  foundations  of  the  older  temple,  now  recognized  by  their 
material  and  their  relation  to  the  stoa  as  of  the  sixth  century 
B.  C,  were  taken  by  the  original  excavators  to  be  of  Byzantine 
date  and  were  partially  removed.  Dorpfeld  regards  the  art  of 
excavation  as  in  its  in&ncy  still,  and  believes  that  much  which  is 
in  like  manner  now  thrown  aside  as  of  no  value  will,  twenty-five 
or  fifty  years  hence,  be  recognized  as  full  of  significance.  In 
spite,  however,  of  ancient  destruction  and  modern  demolition,  the 
Dionysiac  theatre  still  exhibits  remnants  fi'om  every  age  of  the  Attic 
drama.  This  fact  alone  would  make  it  appropriately  the  centre 
of  such  a  discussion  as  this.  Five  periods  in  the  structure's 
history  are  clearly  distinguishable,  as  follows : 

1.  To  the  sixth  and  fifth  centuries  B.  C.  belong  fragments  of  the 
wall  that  supported  the  old  circular  orchestra;  this  is  all  that 
survives  from  the  age  of  Aischylos  and  Sophokles. 

2.  To  the  fourth  century,  to  the  administration  of  Lykurgos, 
belong  the  existing  seats  and  the  earliest  permanent  otki^m?,  with 
the  stoa  at  its  back  facing  the  temple  and  altar. 

3.  The  Hellenistic  period  shows  a  new  form  of  o-ki^v^,  with  first 
a  wooden  and  then  a  permanent  stone  npoaKfivtov. 

4.  In  the  imperial  age,  about  the  time  of  Nero,  appears  the  first 
raised  stage,  of  the  Roman  type. 

5.  To  late  Roman  times,  the  third  or  fourth  century  A.  D., 
belongs  the  stage  of  Phaidros,  whose  inscription  still  stands  on 
its  front. 

Each  of  these  periods  in  the  life  of  the  structure  is  presented  to 
us  in  full  description,  and  all  but  the  first  and  last  in  more  or  less 


Digitized  by  VjOOQ IC 


8  AMERICAN  JOURNAL    OF  PHILOLOGY. 

complete  restoration.  It  is  surprising  how  fully  the  material  is 
supplied,  by  drawing  as  well  as  by  description,  that  enables  one 
to  test  for  himself  the  3teps  of  the  interpretation  and  restoration. 
Much,  of  course,  would  be  unintelligible  without  the  testimony  of 
other  theatres,  in  which  now  one,  now  another  feature  is  preserved 
that  has  disappeared  from  this. 

These  other  theatres  are  therefore  presented  in  '  Abschnitt'  IL 
And  here,  I  think,  many  will  agree  with  me  in  wishing  that  in 
several  cases  the  description  had  been  made  fuller.  On  every 
one  Dorpfeld  could  throw  still  more  light  by  reason  of  his 
thorough  acquaintance  with  all  the  rest  But  some  limit  had  to 
be  set  to  the  size  of  the  book,  and  we  may  be  sure  that  no 
essential  matter  bearing  on  general  theatre  construction  is  passed 
over.  I  will  restrict  myself  to  noting  some  of  the  more  important 
contributions  made  by  these  theatres. 

The  theatre  in  Peiraeus  corresponds  in  form  to  the  third  period 
of  the  Athenian  structure,  from  which  it  is  closely  copied.  It  is 
the  form  of  theatre  which  Vitruvius  describes  as  the  Greek  theatre 
of  his  time,  and  shows  particularly  well,  without  a  trace  of  Roman 
rebuilding,  the  shape  of  the  permanent  stone  npwncfiviop,  which 
Vitruvius  called  a  stage.  The  little  theatre  at  Oropos  is  particu- 
larly interesting  as  one  in  which,  wooden  UpuL  were  retained  as 
seats  to  the  latest  period.  But  above  all,  more  of  aiaiprf  and 
^pwriaiviov  are  here  preserved  than  elsewhere ;  so  that  considerable 
space  is  given  to  the  elucidation  of  important  details  that  are  here 
assured,  and  so  enable  us  to  understand  indications  elsewhere  that 
would  otherwise  be  inexplicable.  At  Thorikos  we  have  a  speci- 
men of  a  little  provincial  theatre  of  a  country  deme,  its  irregular 
shape  conforming  to  the  shape  of  the  hill,  with  no  permanent 
triofiffi)  even  the  orchestra  is  rudely  elliptical  instead  of  round. 
This,  and  the  theatres  of  Eretria  and  Sikyon,  were  excavated  by 
the  American  School,  and  are  together  no  mean  contribution. 
Plate  XII  gives  an  excellent  general  view  of  the  Eretrian  theatre, 
and  incidentally  illustrates  how  unfounded  is  the  assertion  that 
actors  and  chorus,  if  on  the  same  level  in  the  orchestra,  could  not 
be  readily  distinguished.  It  is  highly  probable  that  the  oldest 
aiaivTi  here,  whose  polygonal  lower  course  is  well  preserved,  ante- 
dates the  first  permanent  orn^r^  at  Athens.  This  fact  is  explained 
by  the  respective  locations.  The  position  of  the  sixth-century 
orchestra  at  Athens  was  such  that  a  permanent  axijy?  there  would 
have  blocked  up  the  front  of  the  temple  of  Dionysos ;  no  stone 


Digitized  by  VjOOQ IC 


DORPFELD  AND  THE  GREEK  THEATRE.  9 

fTKi\vi\^  therefore,  was  possible  until  they  resolved,  in  the  fourth 
century,  to  remove  the  orchestra  farther  up  the  hill  and  so  make 
room  for  the  o-nym;  of  Lykurgos.  At  Eretria  there  was  room 
enough  to  permit  the  innovation  as  soon  as  it  came  to  appear 
desirable — not  improbably  as  early  as  400.  At  Eretria  also 
appears  another  peculiarity  of  much  significance  in  theatre  con- 
struction— the  lowering  of  the  orchestra.  The  location  is  not  on 
the  slope  of  a  hill,  but  on  the  level  ground  a  little  way  from  the 
acropolis.  To  obtain  the  support  for  permanent  stone  seats, 
therefore,  a  large  mass  of  earth  or  stone  had  to  be  built  up  on  all 
but  the  vfo^  side  of  the  orchestra.  Practical  as  Greeks  usually , 
were,  the  Eretrians,  in  remodelling  their  theatre,  left  the  old  oKi^vi^ 
at  its  original  height,  lowered  by  about  3.20  m.  the  level  of  the 
orchestra,  constructed  sloping  ndpodot,  at  the  sides,  and  used  the 
earth  obtained  by  this  excavation  for  constructing  the  needed 
basis  for  seats.  The  new  orchestra  was  put  a  little  forward  of  the 
old,  the  new  aKrjprf  added  to  the  front  of  the  old,  still  left  standing ; 
and  thus  was  obtained  a  theatre  at  once  commodious  and  inex- 
pensive. The  slight  elevation  above  sea  level  fairly  compelled  a 
method  of  drainage  that  was  a  distinct  improvement  on  that  at 
Athens — a  method,  therefore,  generally  adopted  from  that  time,  as, 
for  instance,  at  Epidauros.  Here  too  is  a  perfect  and  indubitable 
example  of  the  stairs  of  Charon-*an  underground  passage  leading 
from  behind  the  irpoaKqvtov  to  the  centre  of  the  orchestra,  with  a 
flight  of  stone  steps  at  either  end  so  cut  from  a  single  block  as  to 
imitate  closely  the  earlier  wooden  flight.  The  surface  of  the 
orchestra  at  Athens  and  elsewhere  was  usually  of  trodden  earth 
merely,  until  a  late  period.  At  Eretria  and  at  Delos  the  earth 
was  covered  with  a  sort  of  hard  plaster.  At  Sikyon  an  interesting 
feature  is  the  stone  foundation  of  the  wooden  irpoo-nivioy.  This  is 
here  sufficiently  preserved  to  give  some  notion  of  how  the  wooden 
npwriajviov  was  constructed,  and  to  show  that  the  stone  rrpwnaiviov 
partially  preserved  in  many  places  was  modelled  on  its  wooden 
predecessor.  Another  significant  fact  must  not  be  omitted.  Most 
of  the  walls  of  the  o-n^  are  at  Sikyon  cut  from  the  solid  rock,  and 
can  never  have  undergone  alteration;  in  particular,  ramps  on 
either  side  leading  to  the  top  of  the  wpoaiaivtoif  are  so  cut  from  the 
rock.  This  disposes  completely  of  the  assumption  put  forward 
in  their  extremity  by  defenders  of  a  Greek  stage,  that  the  trpwrKt/viow, 
being  such  a  stage,  was  originally  four  or  five  feet  high,  which 
height  was  later  increased  to  eight  to  ten  feet.    These  ramps  cut 


Digitized  by  VjOOQ IC 


lO  AMERICAN  JOURNAL   OF  PHILOLOGY, 

from  the  living  rock,  and  leading  to  the  top  of  the  wpoaKriviov,  the 
pretended  stage,  were  never  lower.  The  Epidaurian  theatre, 
most  beautiful  and  best  preserved  of  all,  is  so  well  known  by  plans 
and  photographs  that  we  need  not  dwell  on  it  The  plan  here 
published  corrects  some  errors  to  which  Dorpfeld  calls  attention 
in  his  own  previous  plan,  the  source  of  all  intervening  publica- 
tions, which  repeat  the  errors.  There  is  no  better  place  for  the 
traveler  to  disprove  by  experiment  the  oft-repeated  assertion  that 
masks  and  a  'raising  and  conventionalizing  of  the  tones  of  the 
voice'  were  made  more  necessary  in  the  ancient  drama  than  in 
the  modern  by  the  distance  between  actors  and  audience.  As  to 
the  date,  it  appears  certain  that  this  theatre  is  later  than  that 
of  Lykurgos:  the  stone  irpo<rKri»io¥  may  belong  to  the  original 
structure  or  it  may  be  part  of  an  alteration.  For  the  theatre  at 
Megalopolis  the  plans  published  by  the  English  architect,  Schultz, 
are  lauded  by  Dorpfeld  as  '*  vorziiglich  und  in  ihrer  Ausfiihrlichkeit 
musterhaft."  From  Dorpfeld's  account,  which  is  longer  and  more 
polemic  than  usual  by  reason  of  Ernest  Gardner's  attitude  on 
certain  points,  I  extract  two  items.  First,  the  irpoaicrivlop  is  here 
no  less  than  seven  meters  from  the  structure  behind  it.  Now,  the 
old  theory  explains  the  otherwise  universal  narrowness  of  the 
wpotriofviov  roof  by  its  height,  which  would  unpleasantly  conceal 
from  the  seats  of  honor  in  front  any  actor  standing  more  than  five 
feet  back  from  the  edge.  A  stage  ten  or  twelve  feet  high  and 
over  twenty  feet  deep  is  wholly  inexplicable  on  this  old  theory. 
On  the  new  theory,  however,  this  extraordinary  depth  of  irpoaicffviow 
is  explained  very  simply.  The  orchestra  had  the  extraordinary 
diameter  of  about  twenty-eight  meters.  When  the  nrydkii  n-dXcr 
had  become  a  fteyaXi;  tprjfua,  such  an  enormous  orchestra  was 
absurd.  To  bring  the  actors  nearer  to  the  shrunken  audience, 
the  irpoo-jc^Mov,  the  background  .of  the  action,  was  brought  forward 
until  the  remaining  orchestral  space  had  about  the  same  breadth 
that  was  found  desirable  at  Athens  and  elsewhere.  Secondly,  we 
find  here,  from  the  earlier  period,  pretty  clear  evidence  of  a  scaena 
ductilis,  or  movable  background,  which  was  drawn  across  before 
the  great  columns  of  the  portico  when  plays  were  to  be  presented. 
The  base  of  this  scaena  duciilis  was  on  the  same  level  with  the 
orchestra.  The  actors  surely  did  not  stand  on  the  top  of  it. 
What  clearer  evidence  could  be  asked  that  the  actors,  standing  in 
front  of  it,  were  on  the  same  level  with  the  chorus  ?  Passing  over 
to  Delos  we  find  a  <r«7in/  of  simple  and  highly  instructive  form. 


Digitized  by  VjOOQ IC 


DORPFELD  AND  THE  GREEK  THEATRE,  II 

It  is  merely  a  rectangular  building  with  a  colonnade  running 
around  all  four  sides.  If  the  colonnade  on  the  side  towards  the 
audience  was  a  stage,  what  was  it  on  the  other  three?  The  new 
theory  makes  all  four  sides  alike  colonnade :  in  front  of  it  on  one 
side  the  action  took  place.  Movable  n-tVfuccr  or  painted  wooden 
panels  closing  the  space  between  the  columns,  with  doors,  made 
this  side  an  ordinary  npoaialvioif,  which  in  other  theatres  of  the 
Vitruvian  type  is,  in  &ct,  as  here,  simply  a  kind  of  closed  portico. 
Statue  bases  before  it,  on  the  orchestra  level,  quite  exclude  the 
possibility  of  any  raised  stage  there.  At  Pergamon  a  notable 
circumstance  is  that  temple  and  theatre  were  located  in  relation 
to  each  other  as  the  old  temple  and  orchestra  at  Athens :  a  per- 
manent iTKfivrf  would  have  blocked  up  the  approach  to  the  temple 
front.  Accordingly,  throughout  the  Greek  period  of  its  history 
this  theatre  had  only  a  temporary  aKip^  erected  for  each  festival. 
Large  stone  bases  level  with  the  ground  contain  carefully  cut 
holes  to  receive  the  posts  of  the  movable  wooden  structure,  which 
we  may  presume  was  stored,  from  one  festival  to  another,  in  a 
form  ready  to  be  set  up  when  needed.  In  Roman  times  a  per- 
manent Roman  stage  and  fixed  aiaiv^  were  erected.  And  it  may 
be  noted  here  that  in  no  Greek  theatre  in  Greece  proper,  so  far 
as  is  yet  known,  with  the  one  exception  of  that  in  Athens,  was  a 
Roman  stage  ever  built. 

Into  the  detailed  consideration  of 'Abschnitt'  III  on  the  textof 
Vitruvius  I  will  not  enter,  being  neither  Latinist  nor  architect. 
Reisch's  chapter  also  on  the  Greek  theatre  according  to  the 
extant  plays  there  is  the  less  reason  for  me  to  report  upon, 
because  readers  of  this  Journal  are  familiar  with  the  discussion 
from  this  side.^  In  all  his  chapters  Reisch  has  a  task  in  one 
sense  easier,  in  another  sense  more  difficult,  than  Dorpfeld. 
Easier,  because  precisely  this  field  has  been  so  thoroughly 
worked  over  in  recent  years  that  not  much  remained  to  do 
beyond  sifting  and  arranging.  His  task  was  more  difficult,  how- 
ever, in  that  so  little  remained  to  glean,  and  in  the  nature  of  the 
case  so  much  must  inevitably  remain  uncertain  so  far  as  evidence 
from  that  side  goes.  The  dramas  from  the  oldest  to  the  youngest 
present  an  abundance  of  proof  that  no  high  stage  existed ;  but 
agreement  is  unattainable  on  the  question  how  much  can  be 

^See  Pickard,  XIV  6S-89,  198-215,  273-304.    Historically  interesting  is 
Allinson*s  reyiew  of  Hoepken,  V  252,  published  in  1884,  when  it  required 
i  courage  to  say  eren  a  word  for  the  new  yiew.— B.  L.  G. 


Digitized  by  VjOOQ IC 


12  AMERICAN  JOURNAL    OF  PHILOLOGY. 

inferred  as  to  the  details  of  theatrical  machinery  and  the  like 
from  the  language  of  characters  in  a  play.  One  fact,  however, 
comes  out  clearly  from  the  mass  of  particulars.  The  dramas  of 
the  fifth  century  contain  nothing  inconsistent,  they  contain  much 
in  positive  agreement,  with  the  doctrine  that  the  earliest  stone 
amyvai  preserved  are  substantially  the  same  in  form  as  the  tem- 
porary o-m^yoi  of  the  fifth  century,  before  which  were  acted  the 
plays  of  the  great  dramatists. 

The  conservatives  long  found  great  comfort  in  the  vases  from 
Magna  Graecia  on  which  are  represented  scenes  from  comedy 
enacted  on  a  raised  stage.  Following  Heydemann  and  others, 
Reisch  classifies  and  examines  all  the  known  examples  of  this 
type,  and  brings  out  these  facts.  Their  date  is  between  300  and 
200  B.  C;  all  were  made  in  southern  Italy;  they  present  scenes 
from  popular  farces  of  that  region,  which  were  usually  played  on 
a  temporary  stage  of  moderate  height ;  these  farces  and  this  stage 
had  no  direct  influence  on  Greek  theatre  construction,  but  may 
well  have  been  one  of  the  influences  that  led  the  Romans  to  prefer 
a  raised  stage. 

Reisch's  chapter  on  the  Greek  names  for  parts  of  the  theatre 
seems  to  me  of  considerable  value.  By  numerous  examples  the 
entire  range  of  usage  is  illustrated  for  each  of  the  familiar  terms, 
especially  those  which  have  figured  so  largely  in  the  argument 
against  the  new  doctrine.  In  this  manner  more  than  one  rooted 
prejudice  of  the  conservatives  is  left  dangling  without  support, 
while  the  method  of  accomplishing  this  operation  is  gentle  and 
impersonal.  It  is  true  that  much  of  the  argument  based  on  this 
materia]  would  be  inconclusive  if  it  stood  alone.  Many  of  these 
passages,  considered  by  themselves,  might  be  interpreted  other- 
wise. But  it  is  false  method  to  consider  any  one  of  them  by 
itself;  and  when  taken  in  the  mass,  it  is  found  that  Dorpfeld's 
interpretation  of  the  theatres  introduces  a  clarifying  and  organizing 
principle  that  was  lacking  before :  what  looked  like  inexplicable 
contradiction  or  confusion  thus  reveals  itself  as  a  natural  succes- 
sion. For  example,  that  opxif^pa  should  be  used  in  Roman  times 
in  the  sense  of  stage  becomes  entirely  natural  when  it  is  seen  that 
this  portion  of  the  6pxn<rrpa  always  was  the  ordinary  place  of  the 
actors,  who  in  the  Roman  theatre  still  occupied  the  same  position 
and  the  same  level,  while  the  remainder  of  the  ipxi^pf^  was 
lowered.  The  history  of  orniym;  and  its  progeny  is  very  instructive 
in  a  similar  way.    The  famous  oKpifiat  passage  from  Plato  (Sym. 


Digitized  by  VjOOQ IC 


DORPFELD  AND  THE  GREEK  THEATRE.  1 3 

194  b)  is  shown  to  furnish  not  the  sh'ghtest  ground  for  argument 
in  favor  of  a  stage  in  the  theatre.  Indeed,  the  same  explanation 
was  eleven  years  ago  accepted  from  Hug  and  Rohde  by  A.  MuUer 
in  his  '  Buhnenaltertiimer '  (p.  365). 

It  is,  however,  in  the  history  of  the  development  of  the  theatre 
as  a  building  that  Dorpfeld  rightly  finds  the  most  striking  and 
finally  decisive  proof  of  the  correctness  of  his  interpretation.  He 
shows  us  the  ancient  theatre  changing  by  slow  and  natural  steps 
from  the  simplest  to  the  latest  form ;  there  are  no  sudden  leaps 
and  no  breaks  in  the  development.  To  maintain  the  old  view 
now  is  to  assume  a  series  of  such  breaks  and  leaps  in  an  art  which, 
by  the  very  nature  of  architecture,  is  exceedingly  conservative, 
even  unreasonably  tenacious  of  old  forms  because  they  are  familiar, 
after  they  have  become  meaningless  and  perhaps  a  hindrance. 
The  whole  history  of  architecture  illustrates  this  quality  in  it ;  and 
he  who  to-day  in  planning  a  house  cuts  loose  from  tradition  and 
builds  with  too  great  originality  illustrates  anew  the  wisdom  of 
such  conservatism.  It  would  contradict  one  of  the  best  established 
characteristics  of  Greek  art  in  general  to  suppose  any  such  sudden 
leaps  in  theatre  construction.  Whoever  is  interested  in  the  Attic 
drama  from  any  point  of  view  should  read  the  last  chapter  on  the 
'  Entwickelungsgeschichte/  if  no  other  part  of  the  work ;  and  I 
cannot  more  fidy  close  this  notice  than  by  giving  a  rapid  summary 
of  this  story  as  Dorpfeld  reveals  it  to  us. 

Five  periods  are  readily  distinguishable.  For  the  early  choral 
songs  and  dances  in  honor  of  the  gods,  and  especially  of  Dionysos 
at  his  festival,  a  level  space  was  required.  When  the  dance  and 
the  participants  in  it  learned  to  follow  a  more  complicated  law, 
the  opxiarpa  became  a  round  floor  of  trodden  earth,  sufficiently 
marked  by  a  visible  boundary  to  delimit  the  space  for  participants 
and  spectators.  At  Athens  in  the  sixth  century  there  were  two 
such  dancing  places:  one,  already  ancient,  somewhere  by  the 
western  slope  of  the  Areopagus,  near  the  ancient  precinct  of 
Dionysos  ip  Xt/iMur,  the  Aijyaiov;  the  other  on  the  southeastern 
slope  of  the  Acropolis,  in  the  precinct  of  Dionysos  Eleuthereus. 
The  latter  developed  into  the  theatre  which  we  know,  and  the 
former  was  eventually  neglected.  The  lookers-on  at  these  festivals 
found  places  as  they  could  on  the  side-hill.  As  the  festival  increased 
in  attractiveness  and  duration,  something  more  in  the  way  of 
seats  became  needful,  not  only  for  priests  and  officials,  but  for  the 
onlooking  throng.    Such  seats,  needed  but  once  or  twice  a  year. 


Digitized  by  VjOOQ IC 


14  AMERICAN  JOURNAL    OF  PHILOLOGY. 

were  naturally  of  wood  and  temporary.  But  on  the  lower  side  of 
the  dpxnarpa  such  seats  were  hardly  practicable,  and  it  was  necessary 
that  a  part  of  the  circle  be  left  free  for  approach,  both  for  spectators 
and'for  the  chorus.  Ramps  on  either  side,  to  east  and  west,  were 
demanded  by  the  nature  of  the  ground.  An  altar  occupied  the 
centre  of  the  circle,  on  which  offering  was  made  at  the  opening  of 
the  festival.  The  introduction  of  an  actor,  carrying  on  dialogue 
with  the  chorus,  of  itself  changed  nothing.  The  common  dress- 
ing-room for  such  slight  change  of  costume  as  the  piece  required 
might  be  anywhere  conveniently  in  the  neighborhood.  The 
central  altar,  of  the  usual  form,  had  a  lower  part  on  which  stood 
the  priest  in  offering  sacrifice,  and  on  the  same  spot  stood  the 
flute-player  who  accompanied  the  dance  and  song.  On  the  same 
platform,  a  mere  step  or  two  above  the  ground,  an  actor  might, 
in  addressing  the  chorus,  also  take  his  stand  for  a  longer  speech. 
The  scene  on  Easter  Tuesday  at  Megara  or  elsewhere  in  Greece 
to-day  offers  many  analogies  to  the  scene  at  such  an  ancient 
festival. 

In  the  second  period,  the  fifth  century,  the  age  of  full  develop- 
ment of  the  drama,  the  6pxn<rrpa  remained  the  same.  The  seats 
for  spectators  were  enlarged,  and  probably  firmer  foundations  of 
earth  and  stone  were  found  needful  to  support  the  benches  for 
such  throngs.  The  addition  of  a  second  actor  involved  no  change. 
All  action  went  on  in  the  opxnirrptu  It  was  a  better  place  for  the 
action  than  our  own  theatres  now  offer.  But  Aischylos,  in  the 
first  half  of  the  fifth  century,  as  a  part  of  his  endeavor  to  enlarge 
the  range  of  myth  available  for  dramatic  treatment,  introduced 
the  affiym^.  Hitherto  the  scene  of  every  play  was  distinctly  a  sacred 
precinct.  Temple  and  altar  were  visibly  there,  and  sacrifice  had 
just  been  offered.  Even  in  the  Suppliants,  where  a  larger  altar, 
a  ffom>/3Mfua,  had  been  erected  beside  the  orchestra,  the  action  is 
still  in  a  sacred  precinct,  if  not  distinctly  that  of  Dionysos,  and 
has  all  the  more  of  reality  and  effectiveness  for  that  fact.  In  the 
Persians  the  tomb  of  Darius  has  taken  the  place  of  the  Koun^iua^ 
and  the  spectators  are  now,  by  the  presence  of  this  tomb,  trans- 
ported in  imagination  to  Persia.  In  Prometheus  Bound  the  mass 
of  rock  on  which  the  victim  is  bound  locates  the  action  in  some 
wild  region,  which  the  opening  lines  define  more  clearly.  There 
is  scarcely  any  limit  to  the  range  of  story  possible,  now  that  the 
axi^  is  available.  So  in  the  Agamemnon  we  have  the  palace  of 
the  king ;  temple,  fortress,  grove,  camp,  may  all  be  represented, 


Digitized  by  VjOOQ IC 


DORP F ELD  AND  THE  GREEK  THEATRE.  1$ 

that  the  action  may  take  place  before  them.  It  was  a  convenience 
also  that  the  o-mjin^  could  serve  as  dressing-room,  though  that  was 
not  its  main  purpose.  It  was  a  matter  of  course  that  any  such 
structure  was  temporary.  The  spot  remains  after  all  the  precinct 
of  Dion]^sos.  During  the  rest  of  the  year  it  belongs  to  the  god 
exclusively.  Not  until  the  theatre,  with  what  was  necessary  for 
the  production  of  plays,  had  come  to  be  definitely  adopted  as  a 
natural  part  of  the  sacred  precinct  could  the  religious  sentiment 
allow  one  portion  of  that  precinct  to  be  cut  off  from  the  other  by 
a  permanent  building  stretching  from  side  to  side.  For  a  time 
each  separate  play,  or  each  trilogy,  might  require  a  different 
fTKi\vi\^  built  up  for  a  few  hours'  use  and  removed  at  the  end  of  the 
play,  or  at  evening.  Probably  by  450,  however,  it  had  been 
found  a  convenience  to  build  the  main  orjo^yi^  somewhat  more 
solidly,  to  stand  throughout  the  festival:  changes  in  the  front 
aloiie  would  suffice  for  the  successive  plays.  Somewhat  larger 
use  was  made  of  machinery  of  various  sorts.  For  such  a  structure 
projecting  wings,  napain^via,  were  desirable,  to  support  and  bound 
the  decoration  between  them  that  constituted  the  visible  back- 
ground. A  passage  is  left  on  either  side  between  the  ends  of 
the  tiers  of  seats  and  the  vapaaiaivia ;  here  the  audience  enter, 
and  chorus  and  actors  use  the  same  entrances  when  the  action 
demands  it,  as  is  usually  the  case  for  the  chorus  and  often  for  actors. 
Doors  are  made  in  the  <rKriprj  front  as  may  be  needed.  A  single 
story  was  usually  enough  for  such  a  cno^M? ;  when  needed,  a  lighter 
upper  story  could  be  added  as  easily  as  the  front  was  changed. 
The  entire  structure  was  still  of  wood  and  light  material,  and  was 
all  removed  at  the  close  of  the  festival. 

Not  till  the  fourth  century,  after  the  great  period  of  the  drama 
was  past,  and  the  art  of  acting  more  independently  developed, 
was  the  o-ki^i^  built  of  stone.  At  Athens  the  change  was  made 
under  Lykurgos  in  the  third  quarter  of  the  fourth  century.  The 
opxnirrpa,  removed  several  meters  up  the  hill  to  make  room  for 
the  stone  0-1071^,  is  still  circular,  enlarged  on  the  side  towards  the 
(rKfi9Ti  by  the  space  bounded  by  that  structure.  The  spectators' 
seats  were  built  of  fine  limestone  and  marble,  supported  in  part 
on  great  masses  of  earth  and  masonry,  in  part  on  the  solid  rock 
hewn  out  to  receive  them.  Of  the  trKfjp^,  that  which  before  had 
been  made  semi-permanent,  probably  preserved  and  set  up  anew 
for  each  festival  to  remain  till  its  close,  was  now  built  of  stone — a 
long  hall  or  series  of  connected  rooms,  with  projecting  wings, 


Digitized  by  VjOOQ IC 


l6  AMERICAN  JOURNAL    OF  PHILOLOGY. 

between  which  the  temporary  decoration  needed  for  each  piece 
was  set  up.  ira^<rjci7ria  and  front  of  the  ataivr^  proper  were  adorned 
with  columns ;  for  the  theatre  has  now  become  a  regular  place  of 
assembly  on  other  than  festal  days,  and  when  the  -Rpwrtaiviiw  was 
removed  it  was  necessary  that  the  building  should  have  some 
architectural  character.  The  irKi\vi\  is  still  of  one  story,  but  a 
second  or  even  a  third  could  be  added  temporarily  if  any  piece 
required  it.  With  the  growing  prevalence  of  pieces  in  which  a 
temple,  palace,  or  ordinary  house  was  called  for  as  background, 
the  npoiTKfiifiop  takes  more  and  more  the  form  of  a  row  of  columns, 
or  rather  pilasters,  with  movable  painted  panels  between  them  ; 
this  arrangement  itself  permitted  considerable  variety  of  make-up 
for  the  individual  play,  and  if  an  older  play  requiring  something 
different  was  presented,  the  proper  form  of  ir/MMric^Hoir  could  be  put 
in.  A  space  of  several  feet  between  the  front  vpovitripiop  wall  and 
aaiwfi  was  convenient  for  the  actors,  and  on  the  flat  roof  of  the 
irpotriafpiov  gods  appeared,  or,  e.  g.,  the  watchman  in  the  Agamem- 
non. The  floor  level  in  wpotriofvtov  and  aai^fl  is  about  the  same  as 
that  of  the  Ip^mpa^  and  there  is  not  the  slightest  reason  for 
supposing  any  change  in  the  relative  position  of  actors  and  chorus. 

In  the  Hellenistic  theatre  the  one  essential  change  is  merely 
that  the  npwriaipwv  itself  is  now  made  of  permanent  stone,  and 
doubtless  in  connection  therewith  a  second  story  of  stone  is  added 
to  the  aiaivfi.  The  vpoindpaov  has  the  form  which  had  more  and 
more  become  the  typical  one  for  the  wooden  frpocno^Mor— the  form 
which  suited  the  great  majority  of  plays  now  presented.  It  was 
a  closed  portico,  of  pilasters  with  movable  mpaMt  or  panels  between 
them ;  by  varying  their  position,  number  and  decoration,  a  single 
palace  or  three  houses,  or  a  temple,  could  be  conventionally  repre- 
sented. If  required,  on  rare  occasions  a  more  elaborate  special 
decoration  could  still  be  placed  before  the  vpovtaiwiw.  The  height 
of  its  flat  roof  is  three  to  four  meters — the  height  which  Vitruvius 
gives  for  it.  The  wapaaKffvia  have  now  become  of  little  or  no 
importance,  and  in  many  theatres  disappear,  or  at  least  no  longer 
project  in  front  of  the  wpoaK^piop. 

There  remains,  as  the  final  type,  the  Roman  theatre,  which 
Vitruvius  describes  as  very  different  from  the  Greek,  because  he 
was  unable  to  trace  the  course  of  development  of  one  from  the 
other.  Probably  none  of  his  contemporaries  could  have  done 
better ;  the  historical  method  in  the  study  and  presentation  of  such 
matters  was  unknown.    The  typical  course  of  this  development, 


Digitized  by  VjOOQ IC 


DORPF^LD  AND  THE  GREEK  THEATRE,  1 7 

illustrated  abundantly  in  existing  examples,  is  as  follows.  Inas- 
much as  the  dance  of  the  chorus,  so  far  as  the  chorus  still  existed, 
had  disappeared,  the  wide  space  of  the  orchestra  needed  for  their 
evolutions  was  no  longer  wanted.  For  their  songs  and  their  part 
in  the  action,  that  portion  of  the  opxnfrrpa  near  the  npoaicrfviov,  where 
the  actors  ordinarily  stood,  was  quite  sufficient.  This  was  a  space 
say  sixty  by  twenty  feet.  Also  we  can  trace  in  many  ways  the 
great  increase  in  the  number  of  dignitaries  and  personages  to 
whom  the  community  desired  to  give  seats  of  honor.  By  lower- 
ing a  little  over  one  half  the  opx^i^p^*  the  half  farthest  from  the 
9x171^,  space  could  be  got  for  more  seats  of  honor  without  at  all 
interfering  with  the  remaining  seats ;  the  rest  of  the  6pxfi<rrpaj  left 
at  its  original  level,  was  large  enough  for  actors  and  chorus 
together.  On  some  occasions,  too,  with  the  growing  taste  for 
gladiatorial  combats  and  the  like,  this  lowered  portion  could  be 
used  for  the  combatants  without  danger  to  the  spectators.  Exactly 
this  change  was  made,  e.  g.,  in  Aspendos.  Thereby  was  produced 
the  appearance  of  a  low,  broad  and  long  stage,  such  as  Vitruvius 
describes.  Certain  minor  changes  were  necessarily  brought  with 
this.  The  ndpodot  still  leads  as  of  old  to  the  part  of  the  6pxn<rrpa 
left  at  its  original  level.  The  enlarged  <rKfjv^  and  the  seats  are 
united  by  a  mass  of  masonry.  It  is  too  awkward  for  the  entire 
I  audience  to  enter  the  old  napobos  and  pass  over  wb'at  is  now  a 

/  stage,  down  steps  into  the  lowered  portion,  and  the  :i  mount  to 

their  seats.  Therefore  new  entrances  are  made — kot^  ndpodoi — to 
that  lowered  portion.  The  new  entrance  becomes  a  vaulted 
passage  under  the  end  of  the  upper  tiers  of  seats,  and  cuts  off  the 
end  of  the  lower  tiers ;  several  steps  are  necessary  for  descending 
from  the  outer  level,  that  of  the  old  6pxi(rrpa  and  new  stage,  to  the 
lowered  space,  the  Kovltrrpa.  But  not  only  does  the  stage  remain 
on  the  same  level  as  before,  in  relation  to  seats  and  the  ground 
without,  even  the  irpoaKffvtoy  before  the  old  aiaiv^  remains  also  in 
the  old  place,  back  of  the  new  stage,  back  of  actors  and  chorus. 
Changes  in  style  of  decoration  follow  the  changing  taste ;  but  in 
essentials  tl^ese  are  all  the  changes  made.  True,  the  lowering  of 
part  of  the  opxnarpa  was  not  the  only  way  of  producing  the  new 
type  of  theatre.  At  Athens  a  barrier  was  erected  before  the  seats, 
and  a  stage  raised ;  elsewhere  the  lowest  rows  of  seats  were  cut 
away,  and  a  stage  raised ;  but  the  net  result  was  the  same.  In  the 
Herodes  theatre  at  Athens  one  may  see  a  striking  illustration  of 
this.     Here  is  a  theatre  piu-ely  Roman,  of  the  second  century 


4 


Digitized  by  VjOOQ IC 


1 8  AMERICAN  JOURNAL    OF  PHILOLOGY. 

A.  D.  Yet  so  conservatively  is  the  type  preserved  that  arose 
from  lowering  the  opxntrrpa,  that  even  here  the  stage  is  on  the 
level  of  adjoining  rooms  and  the  outside  ground,  while  the  xorv 
ndpodoi  leading  into  the  Kovlarpa  make  a  descent  of  nine  steps. 

The  universal  acceptance  of  Dorpfeld's  view  of  the  Greek 
theatre  is  merely  a  question  of  time.  Some  will  doubtless  hold 
out  for  many  years  yet ;  but  the  great  mass  of  scholars,  I  believe, 
will  find  that  the  case  is  sufficiently  proved  by  the  architectural 
evidence  presented  in  this  important  book.  In  the  words  of  the 
preface,  "  Es  war  Zeit,  die  Welt  der  Antike  von  jenem  wunder- 
lichen  Zerrbilde  zu  befreien,  das  uns  als  'griechisches  Theater' 
gelaufig  geworden  war." 

Yals  UmvBMiTT,  yam.  1897.  THOMAS  DWIGHT  GOODELL. 


Digitized  by  VjOOQ IC 


II.— THE  QUESTION  OF  LANGUAGE-STANDARD  IN 
MODERN  GREECE. 

The  appearance  of  Dr.  Thumb's  *  Handbuch  der  neugriech- 
ischen  Volkssprache '  marks  a  significant  advance  in  the  question 
of  the  Modern  Greek  language-standard.  It  does  this  by  virtue 
of  being  the  first  attempt  to  state  with  scientific  method  and  pre- 
cision the  facts  of  the  colloquial  language. 

The  book  is  neither  a  beginner's  hand-book  nor  an  historical 
grammar.  It  is  not  an  historical  grammar,  because  it  does  not 
consistently  attempt  to  present  the  material  of  the  language  with 
which  it  deals  according  to  its  genetic  relations.  Its  method  of 
arrangement  and  presentation  is  primarily  that  of  descriptive 
grammar,  but  it  differs  from  the  ordinary  descriptive  grammar,  in 
that  its  classifications  are  invariably  made  in  the  consciousness  of 
the  real  historical  relations,  and  that  incidentally  an  abundance  of 
historical  explanation  is  afforded.  Most  of  our  common  hand- 
books of  Modern  Greek  have  been  pseudo-historical  in  character. 
They  have  simply  pared,  patched,  and  re-vamped  the  paradigms 
and  rules  of  the  Ancient  Greek.  They  were  neither  descriptive 
nor  historical,  for  they  undertook  to  set  the  language  forth  neither 
in  terms  of  itself  nor  in  terms  of  the  old.  They  did  not  even 
start  with  the  existing  language.  They  used  chips  of  it  to  stop 
gaps  with.  The  result  was  the  presentation  to  the  world  of  a 
linguistic  monstrosity,  looking  as  fit  in  a  grammar  as  the  beasts  ' 
of  the  Apocalypse  in  a  hand-book  of  zoology. 

The  reputation  which  the  Modern  Greek  standard  literary  idiom 
has  acquired  for  artificiality  is  in  considerable  measure  due  to  the 
manuals  which  have  undertaken  to  represent  it.  It  is  of  course 
unquestioned  that  both  the  standard  literary  idiom  and  the 
standard  folk-speech  are  abundantly  mixed  with  materials  firom 
the  older  language.  But  mixture  does  not  make  monstrosity. 
The  grammatical  'great  bad'  consists  in  putting  together  things 
that  do  not  belong  together  and  do  not  occur  together  in  the  life 
of  a  living  speech.  A  book  which  is  to  describe  grammatically 
the  Modem  Greek  speech  must  take  as  its  basis  ^ther  the  modern 
literary  idiom  or  the  modern  folk-idiom,  and  not  Ancient  Greek; 


Digitized  by  VjOOQ IC 


20  AMERICAN  JOURNAL    OF  PHILOLOGY. 

and  if  it  takes,  e.  g.,  the  literary  idiom  as  its  basis,  it  may  cite  the 
folk-speech  forms,  but  must  cite  them  as  such. 

Self-consistency  will  probably  prove  more  difficult  of  attain- 
ment in  describing  the  literary  standard,  for  the  reason  that  the 
language  of  the  educated  still  remains  strongly  individualized. 
The  wide-spread  acquaintance  with  the  ancient  vocabulary  and 
morphology  among  persons  of  this  class  gives  an  abnormally  free 
rein  to  individual  freedom  in  borrowing.  Perhaps  the  most  suc- 
cessful attempt  to  describe  with  consistency  the  literary  idiom,  in 
the  form  of  a  lesson-book,  is  that  of  Captain  Eugene  Rizo-Ran- 
gab6  in  his  recently  published  *  Practical  Method  in  Modem 
Greek'  (Boston,  1896).  He  selects  skilfully  the  essential  things, 
and  arranges  the  material  conveniently  for  the  learner's  purpose. 
He  may  be  trusted  to  have  excluded  everything,  whether  form 
or  expression,  which  would  appear  out  of  place  either  as  written 
in  a  letter  or  as  spoken  in  the  formal  conversation  of  the  drawing- 
room  or  in  formal  oratory.  It  may  be  questioned  only  whether 
his  purism  has  not  gone  too  far.  The  forms  which  he  occasion- 
ally adds  in  parentheses  or  foot-notes  and  marks  '  vulgar,'  are, 
though  an  educated  man  would  generally  avoid  writing  them,  the 
almost  universally  accepted  forms  in  the  current  speech  of  the 
caf€s  and  shops.  Nevertheless  the  book  is  self-consistent  in  its 
attempt  to  represent  the  higher  literary  language  and  the  polite 
language  of  the  highest  social  circles.  The  *  Modern  Greek 
Mastery,'  by  Thomas  L.  Stedman  (New  York,  1896)  is  also  a 
good  book,  though  it  covers  a  somewhat  wider  range  and  makes 
more  concessions  to  the  every-day  speech.  It  represents  with 
fair  consistency  the  literary  language  as  it  appears,  for  instance,  in 
the  daily  newspapers. 

As  an  introduction  to  the  standard  folk-speech,  the  common 
spoken  language  of  the  great  masses  of  the  people,  no  book  can 
vie  with  the  little  manual  of  Wied, '  Praktisches  Lehrbuch  der 
neugriechischen  Volkssprache '  (English  translation  by  Mrs. 
Gardner).  This  has  been  found  by  me  in  experience  to  repre- 
sent very  accurately  the  actual  facts  of  the  common  spoken  idiom. 
Educated  Greeks  of  the  upper  class,  to  whom  purism  is  a  part  of 
patriotism,  regard  the  book  as  an  abomination.  *'  That  is  the  way 
our  cooks  speak,"  I  heard  a  lady  say,  in  intended  denunciation. 
One  who  knows  Ancient  Greek  should  begin  the  study  of 
Modem  Greek  with  the  folk-speech,  for  this  added  to  Ancient 
Greek  gives  ready  access  to  all  types  of  the  written  and  spoken 


Digitized  by  VjOOQ IC 


LANGUAGE-STANDARD  IN  MODERN  GREECE.  21 

language  alike.  Inasmuch  as  those  Greeks  who  speak  the  literary 
idiom  usually  speak  also  French  or  German,  and  perhaps  English, 
and  seem  to  prefer  the  foreign  language,  there  is  little  practical 
value  in  learning  the  literary  form  simply  for  the  purpose  of 
speaking  it.  One  who  does  not  know  Ancient  Greek  probably 
will  do  best  to  learn  the  literary  idiom  first,  as  this  enables  him 
to  read. 

The '  Praktische  Grammatik  der  neugriechischen  Schriftsprache 
und  Umgangssprache '  von  J.  K.  Mitsodakis  (Berlin,  1891)  at- 
tempts with  fair  success  to  describe  both  idioms  and  distinguish 
between  them  by  means  of  parallel  exercises.  The  book  shows, 
however,  the  lack  of  a  firm  and  clear  idea  of  what  the  spoken 
language  in  its  norm  and  standards  really  is.  The  facts  of  this 
language  had  never  been  collected  and  sifted,  so  that  perspective 
in  its  material  was  possible.^  It  was  to  supply  this  lack  that 
Thumb's  book  was  written. 

This  *  Handbuch  der  neugriech.  Volkssprache '  is  not  a  begin- 
ner's book.  Its  material  is  arranged  systematically  and  not 
paedagogically.  It  has  no  exercises  and  vocabularies.  The  texts 
at  the  end  of  the  book  are  essentially  its  sources,  and  are  appended 
rather  for  verification  than  primarily  as  reading  exercises.  It  is 
essentially  a  first  attempt  to  collect,  sift,  and  assort  the  facts  of  the 
standard  language  of  common  spoken  intercourse.  It  is  by  no 
means,  however,  an  accepted  proposition,  especially  among  the 
archaizers,  that  such  a  standard  exists.  In  the  eyes  of  those  who 
insist  that  the  weal  of  modern  Greece  is  to  be  found  only  in 
clinging  to  its  great  past,  in  reviving  the  forms,  if  not  the  spirit, 
of  its  literature,  its  architecture,  its  manners,  and  its  language, 
there  can  be  nothing  so  odiously  heretical  as  this.  They  claim 
that  aside  from  the  literary  standard  there  is  nothing  that  can  be 
called  a  standard — ^that  over  against  it  stands  only  the  multiplicity 
of  the  local  dialects  or  patois.  Thumb,  however,  asserts  his  belief 
that  there  is  such  a  thing  as  a  popular  Greek  icoiv^,  and  his  book 
is  at  once  the  product  and  vindication  of  his  faith. 

The  only  fault  we  have  to  find  with  him  in  this  matter  is  that 
he  states  his  faith  too  timidly  and  sets  forth  his  case  too  weakly. 

^  The  most  successful  attempt  to  discriminate  fully  and  practically  between 
the  purely  colloquial,  the  standard  literary  and  the  learned-archaistic  mate- 
rials of  the  vocabulary  has  been  made  by  Jannaris  (A.  N.)  in  his  *  Concise 
Dictionary  of  the  English  and  Greek  Languages,  as  actually  written  and 
spoken.'    1895. 


Digitized  by  VjOOQ IC 


22  AMERICAN  JOURNAL    OF  PHILOLOGY. 

While  it  must  clearly  be  admitted  that  this  colloquial  koci^  is  by 
no  means  yet  a  completely  defined  and  well  digested  standard,  it 
is  unmistakably  well  advanced  in  the  process  of  crystallization, 
and  will  respond  more  or  less  satisfactorily  to  all  the  character- 
istic tests  of  a  standard  language.  Thus,  (i)  The  dialects  are 
shrinking  back  into  the  more  isolated  districts,  such  as  the 
islands,  and  the  country  districts  of  Epirus  and  Asia  Minor. 

(2)  While  many  of  the  popular  songs  and  popular  stories  are 
published  in  a  form  strongly  colored  with  dialectal  characteristics, 
their  dialectal  character  is  distinctly  felt,  and  is  recognized  and 
enjoyed  as  such.  The  great  mass  of  the  folk-songs  of  general 
currency  are  cast  in  the  common  idiom,  and  are  dependent  for 
such  currency  either  upon  having  originally  had  that  form  or 
upon  having  with  time  assumed  it. 

(3)  Certain  departments  of  literary  composition  have  come  to 
recognize  the  colloquial  idiom  as  their  proper  vehicle ;  thus  the 
comedy,  the  satire  of  the  comic  journals  (notably  the  admirable 
'Ptf^off),  the  humorous  anecdote  of  the  daily  paper,  especially 
when  cast  in  the  dialogue  form,  comic  poetry,  and  lyric  poetry 
(prevailingly).  The  very  fact  of  the  continual  use  of  this  idiom 
for  publication  in  journals  which  are  current  throughout  the  most 
of  the  Greek-speaking  territory  guarantees  a  regard  for  a  univer- 
sally intelligible  medium,  and  is  furthermore  working  steadily 
toward  the  more  precise  establishment  and  definition  of  a  norm. 
There  is  no  test  which  more  certainly  vindicates  the  character  of 
an  idiom  as  '  standard  *  than  just  this ;  when  a  given  language, 
even  if  originally  no  more  than  a  local  dialect,  asserts  for  itself 
noHofuU  recognition  as  the  proper  vehicle  for  a  certain  form  of 
intercourse  written  or  spoken,  or  for  a  certain  type  of  literature, 
then  it  has  ceased  to  be  a  dialect  and  has  become  a  standard. 
This  is  not  a  matter  merely  of  correspondence  to  a  set  and 
arbitrary  definition;  it  is  so  in  the  nature  of  the  case.  A  language 
or  dialect  which  has  for  any  purpose  come  into  inter-tribal  or 
national  use  has  passed  out  of  the  state  of  nature  and  become  a 
tool  of  formal,  organized  civilization — a  '  Kultur'  fact 

(4)  It  is  not  alone  as  a  printed  language  that  the  colloquial 
idiom  is  assuming  the  character  of  a  standard.  There  exists  also 
the  substratum,  at  least,  of  a  standard  spoken  language.  The 
language  which  one  can  use,  and  which  one  will  hear  in  the 
transaction  of  business  and  in  the  carrying  on  of  ordinary  inter- 
course, is,  in  respect  to  the  ordinary  phraseology  and  the  corn- 


Digitized  by  VjOOQ  IC 


LANGUAGE^STANDARD  IN  MODERN  GREECE,  23 

moaly  used  words,  essentially  one  throughout  nearly  the  entire 
extent  of  free  and  *  enslaved  *  Greece.  One  is  not  a  litde  surprised 
to  find,  for  example,  on  landing  in  Samothrace,  an  apparently 
isolated  island  under  Turkish  sway,  that  the  common  language 
of  the  people,  at  least  those  with  whom  a  stranger  comes  most 
naturally  in  contact,  the  officials,  the  priests,  the  inn-keepers,  and 
the  agogiates,  is  such  that  Athenian  and  Samothracian  converse 
freely  and  without  the  consciousness  of  any  dialectal  barrier 
between  them.  This  language  of  general  intercourse  is  not  the 
formal  literary  language,  that  of  the  newspapers  for  instance,  but 
the  colloquial  idiom  as  one  hears  it  in  the  ordinary  conversation 
of  the  shops  in  Athens.  If  one  met  the  village  schoolmaster  he 
would  undoubtedly  seek  to  delectate  the  philological  enthusiasm 
of  his  hearer  and  the  pedantic  vanity  of  himself  by  resurrecting 
an  antique  form  or  two,  but  this  would  scarcely  survive  the 
formalities  of  the  first  interview. 

An  educated  Greek  in  conversing  with  a  foreigner  who  has 
not  acquired  full  control  of  the  language,  especially  if  the  relation 
be  that  of  teacher  and  pupil,  is  likely  to  give  the  impression  that 
the  colloquial  language  is  in  a  niost  confused  condition  so  far  as 
any  standard  of  usage  is  concerned.  This  is  largely  due,  as  my 
own  experience  has  shown,  to  the  artificial  consciousness  of  his 
language  awakened  in  the  Greek  by  the  necessities  of  teaching. 
Under  the  influence  of  this  reflective  consciousness  he  reverts  to 
the  fuller  or  more  dignified  forms  of  the  higher  literary  idiom, 
with  which  as  a  written  and  to  a  greater  or  less  extent  as  a  spoken 
language  he  is  acquainted.  Unless  he  has  by  continued  experi- 
ence in  teaching  acquired  the  habit  of  consciously  facing  his  own 
language,  he  is  likely,  when  called  upon  to  commend  to  his 
trustful  disciple  as  a  cold  grammatical  fact  something  hitherto 
only  known  to  him  warm,  to  repudiate  a  plain  usage  of  his 
common  speech  in  deference  to  what  seems  the  better  social 
presentability  of  the  literary  standard. 

There  is,  however,  though  the  Greeks  may  be  slow  to  confess 
it  to  themselves,  a  standard  spoken  language,  not  yet  indeed  a 
fixed  and  unified  norm,  but  a  common  and  universally  current 
form  of  language  which  has  found  a  literary  expression,  is  well 
established  in  the  essentials  of  phonology,  inflexion  and  syntax, 
and  is  perfecdy  capable  of  being  summarized  and  stated  in 
grammatical  form.  Beside  it  there  exist:  (i)  The  archaizing 
language,  used  in  learned  works,  forma    announcements,  etc.; 


Digitized  by  VjOOQ IC 


24  AMERICAN  JOURNAL    OF  PHILOLOGY. 

(2)  the  literary  standard,  used  in  newspapers,  general  litera- 
ture, letters,  speeches  and  addresses,  school-books,  laws,  etc.; 

(3)  the  folk-dialects,  found  in  the  more  isolated  districts,  as  in 
the  islands  and  in  Turkey.  The  colloquial  language  is  steadily 
enriching  itself  from  the  literary  idiom,  and  the  tendency  of  its 
development  is  unmistakably  toward  that  idiom.  The  language 
which  is  to  become  ultimately  the  standard  national  language  of 
Greece  is  evidently  to  be  built  upon  this  basis,  and  not  upon  the 
basis  of  the  present  literary  language.  It  will  be  built  upon  this 
basis  with  free  use  of  materials  from  the  literary  language,  so  that 
the  resultant  will  be  the  product  of  a  series  of  progressive  com- 
promises. This  is  the  linguistic  forecast  for  Greece,  but  it  is  a 
prognostic  based  upon  laws  which  experience  has  shown  unerr- 
ingly rule  in  the  evolution  of  national  standards  of  speech. 
When  the  various  local  forms  of  a  folk-speech  show  a  tendency 
to  solidify  in  a  sub-standard,  no  matter  how  strong  the  crust  of 
the  literary  standard  above  tending  to  hold  the  folk-speech  down, 
it  sooner  or  later  bursts  through  and  overspreads  the  first  crust, 
gradually  in  turn  becoming  a  crust  itself. 

The  crust  with  which  the  literary  standard  has  thus  far  suc- 
ceeded in  repressing  the  colloquial  speech  owes  its  strength  to  a 
variety  of  causes.  The  Attic  kolvti  which  took  shape  and  held 
sway  between  300  B.  C.  and  600  A.  D.  is  the  parent  speech  of  the 
modern  popular  dialects  as  spoken  languages,  out  of  which  by 
consolidation  and  compromise,  as  well  as  by  gradual  acceptance 
of  the  influence  of  the  modern  literary  language,  the  modern 
colloquial  standard  has  been  developed.  The  modern  literary 
standard  is  also  a  continuation  of  the  Attic  jcoii^,  but  by  way  of 
the  written  language.  The  continuity  of  the  written  language 
through  the  middle  ages  (Byzantine)  and  the  influence  of  the 
church  tended  to  shore  up  the  literary  idiom  and  hold  it  aloof 
from  the  declining  levels  of  the  living  popular  speech.  The 
re-birth  of  Greek  nationality  in  the  early  decades  of  this  century 
made  a  sudden  demand  for  a  national  language.  There  existed 
with  which  to  meet  this  demand,  on  the  one  hand  the  half- 
formed  Church-Byzantine  standard,  on  the  other  a  mass  of 
formless  patois.  Upon  the  former  was  constructed  through  the 
naturally  concurrent  choice  of  historians,  publicists,  pamphleteers, 
journalists  and  public  speakers,  the  modern  literary  standard.  It 
adapted  itself  to  the  need  of  the  times  by  free  use  of  ancient 
materials.    The  Philhellenic  spirit  which  gave  impulse  to  the  war 


Digitized  by  VjOOQ IC 


LANGUAGE-STANDARD  IN  MODERN  GREECE,  2$ 

of  the  liberation,  the  overshadowing  greatness  of  the  ancient 
literature,  the  patriotic  adherence  to  the  church  as  the  one  insti- 
tution binding  the  people  to  their  past,  the  ambition  to  find  a 
raison  {Tilre  for  the  nation  in  its  re-embodiment  of  Old  Greece, — 
all  combined  to  make  and  keep  linguistic  toryism  an  essential 
part  of  national  loyalty.  Unsoundness  on  the  language  question 
is  generally  viewed  in  the  higher  social  circles,  and  especially  in 
academic  circles,  as  both  a  heresy  and  a  disloyalty.  It  has  been 
made  an  issue  in  the  '  educational  politics '  of  the  country,  and 
men  tainted  with  a  suspicion  of  this  heresy  are  likely  to  find 
themselves  debarred  from  positions  in  the  university  and  in  the 
public  schools.  The  curious  sensitiveness  of  the  Greeks  on  the 
subject  of  the  ancient  pronunciation  is  merely  symptomatic. 

This  linguistic  orthodoxy  which  has  created  and  by  main  force 
kept  in  use  for  two  generations  the  modern  literary  standard, 
artificial  and  bizarre  as  it  is,  has  served  nevertheless  a  useful  pur- 
pose. The  language  itself  has  in  the  first  place  furnished  during 
an  emergency  the  means  of  communication.  It  has  furthermore, 
and  what  is  more  important,  served  as  a  temporary  terrace 
between  the  plane  of  the  old  icoii^  and  that  of  the  new, — or  rather 
it  has  been  a  staging  about  a  structure  in  building,  helping  to  for- 
ward into  their  proper  and  well-determined  place  the  builders' 
materials.  The  stern  orthodoxy  which  has  enforced  the  standard, 
has  spent,  it  is  true,  all  its  labor  upon  strengthening  the  staging, 
and  has  even  sought  to  convince  us  that  the  staging  was  the 
house,  but  it  has  ensured  slower  and  more  careful  construction, — 
it  has  indeed  made  building  possible. 

Ithaca,  Nov.  as.  1896.  BeNJ.  IdE  WhEELER. 


Digitized  by  VjOOQ IC 


111.— ET/AM  IN  PLAUTUS  AND  TERENCE. 

It  is  not  uncommon  to  find  the  temporal  etiam  familiar  in 
Plautus  and  Terence  explained  as  =  odAuc,  non  eiiam  as  =  mm- 
dum;  but  from  any  other  point  of  view  than  the  narrow  one  of 
the  Ciceronian  this  is,  of  course,  an  inversion  of  the  facts.  Eiiam 
is  originally,  as  Kiihner  (Lat.  Gr.  II  i66,  A.  i)  truly  says,  a 
temporal  particle ;  it  is  the  particle  of  continuance  in  time,  which, 
through  its  development  into  something  much  wider  and  more 
varied,  lost  its  temporal  rights  in  the  negative  sentence  to  dum^ 
in  the  positive,  less  early  and  less  completely,  to  culkuc — both  of 
them  mere  usurpers.  Its  primitive  signification,  living  to  Cicero 
and  Caesar,  perhaps  an  archaism  for  Vergil  and  Ovid,  is  not  only 
prominent  in  Latin  comedy,  but  is  often  to  be  felt  there  even 
when  another  sense  is  growing  up  beside  and  out  of  it ;  and  while 
the  fact  that  its  rhetorical  value  is  much  more  pronounced  in  the 
Ciceronian  age  depends  in  some  degree  at  least  on  a  difference  of 
literary  sphere,  the  comparative  reluctance  of  Plautus  and  Terence 
to  admit  a  complete  disjunction  of  eiiam  from  the  verb  shows 
clearly  enough  that  in  the  sixth  century  of  the  city  the  adverb  was 
still  in  a  period  of  transition,  still  clung  to  its  original  narrowness 
of  relation.  This  was  ignored  by  Hand,  who  paid  no  regard  to 
the  historical  point  of  view  and  drew  his  examples  indifferently 
from  all  periods  of  Latin  literature ;  Holtze  merely  copied  Hand 
so  far  as  eiiam  was  concerned ;  and  the  influence  of  these  gram- 
marians is  still  to  be  felt,  not  altogether  happily,  in  Plautine  and 
Terentian  commentaries.^ 

The  following  list  of  examples*  from  the  twenty  fairly  complete 
plays  of  Plautus  (quoted  from  the  edition  of  Loewe,  Goetz  and 

^It  is  not  to  be  supposed  that  any  modem  editor  would  follow  Hand  in 
rendering  Haut.  1057  nil  etiam  audio  ipswn  bj  *'  ich  hOre  auf  ihn  wirkiich 
nicht" ;  but  the  spectre  of  the  vis  affirmoHva  is  not  jet  wholly  banished  to  its 
proper  place. 

'  The  only  similar  list  known  to  me  is  the  incomplete  one  for  Plautus  in 
Ramsay's  edition  of  the  Mostellaria,  Excursus  III,  which  is  made  useless  for 
purposes  of  interpretation  by  an  insistence  on  the  false  rendering  *  even  now/ 
This  would  be  incorrect  even  granting— what  I  doubt — the  correctness  of  the 
derivation  from  et  iam.  Neither  iam  nor  eiiam  denotes  a  single  point  of  time 
or  suffers  us  to  feel  action  as  stationary :  both  include  a  point  of  departure  and 
a  point  of  arrival  and  mark  the  flow  of  action. 


Digitized  by  VjOOQ IC 


ETIAM  IN  PLAUTUS  AND  TERENCE, 


27 


Schoell)  and  the  six  of  Terence  (quoted  from  Dziatzko)  is  meant 
to  serve  as  a  contribution  to  the  history  of  the  particle  and  the 
interpretation  of  our  texts.  The  three  distinct  values  of  etiam 
I  have  called,  with  some  departure  from  the  usual  terminology, 
the  temporal  ('still,'  with  negative  'yet'),  the  add i tory  (* also') 
and  the  intensive  ('even*);  and  I  have  proceeded  throughout 
on  the  assumption,  which  is,  I  think,  justified  by  the  passages 
cited,  that  etiam  has,  at  this  stage  of  its  existence,  a  certain 
mobility,  a  sympathetic  quality ;  that  it  attracts  and  is  attracted 
by  words  or  forms  of  kindred  meaning,  which  often  serve  to 
determine  its  wavering  signification. 

I.  The  temporal  etiam  is  found  with  the  following  verbs  or 
verbal  phrases^  est  True.  886,  Poen.  315,  Aul.  507,  Cure.  172, 
Cas.  306,  Ad.  358;  est?  Haut.  742;  opus  est  Cas.  502,  Pseud. 
735;  de  integro  est  Ph.  174;  sunt  True.  174.  probably  Trin. 
1039  eae  miserae  etiam  ad  parietem  sunt  fixae,  cf.  Andr.  282-3 ; 
posse  Stich.  617;  habeo  Mil.  640;  pergo?  Poen.  1224;  pergin? 
Cure.  196.  With  expressions  of  uncertainty :  Merc.  896  metuis  ?, 
Trin.  572  consulis?,  Capt.  892  dubium  habebis?.  Ph.  774 
dubiumst?  (cf.  Trin.  594),  Haut.  188  incertumst,  Hec.  614  incer- 
tus  sum  (cf.  True.  785).  With  sto  and  its  derivatives  and  other 
verbs  denoting  position  or  delay :  stas  ?  Cas.  749,  Hec.  430  (cf. 
Eun.  286),  asto?  Merc.  129,  astas?  Most.  522  (on  Men.  697,  Cas. 
728  cf.  inf.  IX  S),  adstante  Amph.  747,  restas?  Most.  851,  restat, 
restare  (in  metaphorical  sense)  Ad.  190.  444,  restitas?  Eun.  668, 
retentas  ?  Rud.  877,  maneo  Ad.  279,  manes  ?  Men.  422,  mane  Men. 
177,  maneam  (fut.)  Trin.  1136  (c£  Aul.  805,  Cas.  606),  cesso? 
Merc.  129,  cessant  Haut.  175.  So  of  position:  Asin.  923  cubat; 
of  physical  action :  Asin.  327  anhelitum  ducere ;  of  physical  per- 
ception: Asin.  109,  True.  331  audin?;  of  physical  condition: 
Stich.  574  valet?;  of  mental  condition:  Capt.  137  beat;  of  a  con- 
dition of  life:  Pseud.  610  servis,  Asin.  871  in  senatu  dare  operam. 

Spengel  on  Andr.  116  cites  Haut.  175,  haud  quaquam  etiam 
cessant,  as  an  example  of  etiam  with  negative  =  nandvm^  which 
would  yield  no  sense;  Hand's  (Turs.  H  573)  *no  longer*  is 
impossible.    The  negative  here  modifies  in  the  first  place  only 

^  This  list  is  not  exhaustive ;  other  instances  of  the  temporal  character  are 
dealt  with  later  in  different  connections.  Here  and  there  I  have  called  atten- 
tion  to  explanations  opposed  to  or  agreeing  with  my  own ;  in  general  I  have 
necessarily  refrained  from  discussion. 


Digitized  by  VjOOQ IC 


28  AMERICAN  JOURNAL    OF  PHILOLOGY, 

the  modal  adverb,  and  this  combination  modifies  the  complex 
eiiam  cessanty  which  words  reveal  to  the  audience  the  substance 
of  the  fear  expressed  by  Clinias  and  com  batted  by  Clitipho: 
"  they  are  by  no  means  (as  you  imagine)  still  delaying.*'  Brix' 
on  Capt.  892,  dubium  habebis  etiam  sancte  quom  iurem  tibi, 
translates  'selbst  wenn.'  But  the  instances  cited  above  of  the 
temporal  adverb  with  expressions  of  uncertainty  go  far  towards 
enforcing  a  similar  interpretation  here  (and  so  Lindsay  renders 
by  *  still ') ;  moreover,  the  combination  eiiatn  quom  =  *  even  when ' 
occurs  nowhere  else  in  Plautus,  except  in  Capt.  256/  a  line  which 
Brix  follows  Biicheler  in  regarding  as  spurious. 

II.  The  notion  of  continuance  sometimes  passes  into  that  of 
repetition,  and  so  we  have  'again'  as  an  alternative  rendering 
for  Mil.  1418  verberetur  etiam,  1424  verberone  etiam  an  iam 
mittis?  (cf.  1339),  Amph.  369  at  mentiris  etiam;  in  this  last 
passage  the  sense  is  to  be  determined  from  366-7  compositis 
mendaciis  advenbti.  The  rendering  'again'  is  inevitable  in 
Most.  474  circumspice  etiam,  Merc.  1013  etiam  vide,  where  the 
same  imperative  has  just  been  used,  and  can  hardly  be  avoided  in 
Merc.  324,  Hec.  841  vide  etiam ;  AuL  326  fur,  etiam  fur  trifurcifer 
shows  the  same  force  in  an  elliptical  construction,  etiam  denoting 
*  I  repeat '  (cf.  Mil.  1373). 

III.  The  additory  value  of  eiiam  is  an  inference  from  the 
temporal,  and  passages  are  not  lacking  to  illustrate  the  develop- 
ment of  the  inference.  Brix  on  Capt.  prol.  53,  etiamst  paucis 
quod  vos  monitos  voluerim,  translates  the  adverb  by  '  noch  aus- 
serdem,'  which  would  correspond  to  the  Plautine  eiiam  ht super  \ 
but  the  similar  phrase,  Merc.  569  prius  etiamst  quod  te  facere  ego 
aequom  censeo,  shows  Plautus  emphasizing  the  temporal  side. 
Eiiam  =  eiiam  prius  (cf.  inf.  VII)  is  clear  enough  in  Men.  431 
iam  sequar  ted:  hunc  volo  etiam  conloqui.  Pseud.  11 58  mane: 
iam  redeo  ad  te  . .  •  hunc  advocare  etiam  volo;  the  contrast  with 
the  future  fixes  the  anticipatory  sense  of  eiiam.  In  Bac.  1161, 
Most.  118  the  English  idiom  admits  only  the  additory  idea;  that 
this  is  for  the  Latin  only  a  connotation  appears  from  the  contrast 
in  the  first  passage  with  iam,  in  the  second  with  a  perfect.    Here 

^  Quom  etiam,  Capt.  355,  is  not  analogous ;  see  under  VI ;  and  even  if  we 
read  cum  with  the  MSS  in  Rud.  112^,  etiam  belongs  there  to  the  preceding 
clause. 


Digitized  by  VjOOQ IC 


ETIAM  IN  PLAUTUS  AND  TERENCE,  29 

the  particle  is  used  with  verbs  of  desire ;  elsewhere  with  a  future 
tense  or  a  tense  or  phrase  implying  futurity.  It  is  purely  temporal 
in  Poen.  188,  Rud.  441  (cf.  inf.  XI);  usually  it  has  an  additory 
connotation:  Eun.  717  ludificabere,  Ph.  547  instigemus,  Pers.  669 
dedimus  dabimusque  etiam  (the  notion  of  time  is  held  fast  by  the 
contrasting  tenses,  that  of  addition  is  brought  out  by  the  identity 
of  action).  The  same  close  connection  with  dabo  (the  adverb 
suggesting  the  undefined  object)  appears  in  Ph.  877  immo  etiam 
dabo  quo  magis  credas,  Hec.  869  immo  etiam  qui  hoc  occultari 
fadlius  credas  dabo ;  the  verb  is  omitted  in  Capt.  290  quin  etiam 
ut  magis  noscas,  Men.  1018  em  tibi  etiam ;  similar  is  the  connec- 
tion with  dicam:  Pseud.  522  vin  etiam  dicam  quod  vos  magis 
miremini  (cf.  324).  The  additory  sense,  which  penetrates  by 
implication  into  such  phrases  without  expelling  the  temporal,  is 
reinforced  when  a  word  denoting  or  implying  addition  stands 
beside  etiam:  porro^  Cure.  453,  Bac.  273.  274;  ultro  Bac  567, 
Rud.  484;  insuper  Cas.  441,  Trin.  1025,  Eun.  1014,  Ad.  246; 
unus  Poen.  403.  491-2,  Andr.  940,  Haut.  895,  Eun.  1084,  Ph. 
831 ;  alius  Men.  922,  Stich.  449-50,  Pseud.  370  (cf.  524-5);  alter 
Bac.  692.  954.  971.  A  simitar  and  stronger  effect  is  produced  by 
a  verb  denoting  addition  or  accession:  addo  Poen.  385,  Epid. 
473,  Rud.  1007,  Merc.  435';  accedo  Merc.  24,  Pers.  669,  Andr. 
215;  adscribo  Pers.  69,  Bac  745;  accudo  Merc  432;  accerso 
Bac.  424;  and  in  elliptical  phrases,  where  the  verbal  notion  is 
suggested  by  eHam,  Stich.  427,  Bac.  546,  Andr.  300,  Eun.  1081 
(cfl  inf.  X).  Some  of  these  clauses  contain  also  a  word  of  number ; 
and  in  some  (as  Pers.  69  atque  etiam  hoc  in  ea  lege  adscribier) 
the  additory  force  is  further  brought  out  by  a  position  which  seems 
to  throw  the  stress  of  the  adverb  on  the  pronoun,  while  it  still 
belongs  logically  to  the  verb. 

IV.  Etiam  is  used  with  words  of  degree  partly  in  a  temporal, 
partly  in  a  derived  intensive  sense.  In  Men.  158  etiam  concede 
hue,  AuK  55  etiamne?  (sc.  abscedam)  the  adverb  of  time  is 
transferred  to  a  relation  in  space.  For  Asin.  40  etiamne?  (sc. 
exscreem  ?)  the  answer,  usque  a  penitis  faucibus,  gives  the  notion 

^  In  Asin.  875  the  rhythm  separates  etiam  from  porro  (cf.  inf.  V  a,  i).  Some 
of  the  examples  with  ultro  and  insuper  are  cited  under  VIII ;  the  tense  decides 
whether  etiam  retains  or  not  its  temporal  quality. 

'So  MSS  and  Ritschl  (with  the  easy  addition  of  me) ;  on  Bothe*s  etiam  nufu, 
adopted  by  Schoell,  cf.  inf.  VII. 


Digitized  by  VjOOQ IC 


30  AMERICAN  JOURNAL    OF  PHILOLOGY. 

of  extent  or  degree,  which  is  continued  by  the  following  question, 
etiam  amplius  ?  The  employment  of  the  particle  with  compara- 
tives is  by  no  means  so  frequent  or  so  fixed  as  at  a  later  period ; 
both  dramatists  connect  it  with  amplius^  Plautus  also  with  plus^ 
magis.prius,  but  the  connection  is  given  by  rhythm  and  context, 
not  by  formally  established  usage.  For  etiam  prizes  cf  inf.  VII ; 
the  only  example  which  belongs  here  is  Bac.  221  atque  eo  fortasse 
iam  opust. — immo  etiam  prius,  which  shows  an  unusual  intensifi- 
cation of  the  comparative  notion  and  consequent  subordination  of 
etiam.  The  temporal  force  of  etiam  is  felt  with  amplius  where 
the  latter  is  an  adverb  and  the  tense  is  present  or  future:  Asin.  41, 
Men.  791,  Ad.  468 ;  it  is  lost  when  the  tense  is  perfect :  Haut.  132, 
Eun.  143;  or  when  amplius  is  a  substantive:  [Trin.  249],  Capt. 
777,  Rud.  961.  With  the  substantive //«j  :  Amph.  610,  Pseud. 
1329,  Rud.  504,  the  adverb  has  a  purely  intensive  value ;  so  with 
magis,  Capt  150,  but  in  Pseud.  324  the  temporal  notion  is  held 
fast  by  the  future  tense.  With  parum,  Amph.  374,  True.  898, 
Mil.  1 142,  the  temporal  sense  is  always  felt,  being  retained  even 
with  the  perfect  in  the  third  passage  by  the  negative  value  of 
this  adverb ;  with  adaucta,  Haut.  435,  etiam  is  of  course  purely 
intensive. 

V  a.  The  examples  so  far  cited  exhibit  etiam  as  either  retaining, 
wholly  or  in  part,  its  temporal  character,  or  as  losing  it  by  close 
association  with  classes  of  words  to  which  it  was  originally  attached 
in  virtue  of  that  character.  It  resigns  all  connection  with  the  idea 
of  time  and  assumes  a  purely  additory  signification  in  the  follow- 
ing cases : 

First,  where  it  defines  a  contrast  between  two  difierent  actions 
in  like  time.  Here  it  is  still  formally  a  verbal  adjunct,  but  its 
relation  to  the  verb  is  not  an  intimate  one,  manifested  in  the 
sympathy  of  the  tense;  the  tenses  are  fi^ee,  and  the  adverb  bears 
upon  the  whole  clause,  not  especially  upon  its  verbal  element, 
for  which  reason  it  frequently  stands  at  the  beginning  of  the 
clause.  Secondly,  where  it  contrasts  two  objects.  Here  the  verb 
is  the  same  for  both  clauses,  and  etiam  has  become  a  nominal 
adjunct. 

i)  Most.  978  quadraginta  etiam  dedit  huic?,  Mil.  1 147-8  quin 
etiam  .  . .  omnia  dat  dono,  Merc.  1002  quin  loris  caedite  etiam, 
Aul.  304  etiamne  obturat  inferiorem  gutturem,  452.  465,  Asin. 
276  [482],  Cas.  367-8  perperam  iamdudum  hercle  fabulor. — pol  tu 


Digitized  by  VjOOQ IC 


ETIAM  IN  PLAUTUS  AND  TERENCE,  3 1 

quidem,  atque  edam  facis,  Capt.  561  at  etiam  .  . .  aibat,  Asin.  875 
(where  eiiam  adds  the  reproach  contained  in  this  line  to  the  pre- 
ceding,/^tt^  expresses  the  transition  from  conrupius  to  canrum- 
///),  Cist.  775,  Rud.  1275  etiamne  salutem  earn,  1277,  Men.  69i» 
Trin.  942.  943,  Pseud.  1075  atque  etiam  habeto  mulierem  dono 
tibi,  Andr.  368  etiam  puerum  inde  abiens  convent  Chremi,  Haut. 
999,  Ad.  209. 

2)  Most  513  etiam  tu  fuge,  Mil.  1206  etiam  me?,  Stich.  709, 
Aul.  561. 641  ostende  hue  manus  . . .  ostende  etiam  tertiam,  Pers. 
783,  Amph.  760,  Asin.  184,  Cas.314.  612.  991,  Cist.  522,  Rud. 
1 124.  1275  anne  etiam  patrem?,  Pseud.  I95\  1223  atque  etiam 
mihi  aliae  viginti  minae,  True  248,  Ph.  238  etiamne  id  lex  coegit  ?, 
940,  Ad.  532  etiam  noctu,  664  atque  etiam  inliberaliter.  Under 
thb  head  may  be  classed  the  cases  where  eitam  qualifies  a  clause 
or  verbal  noun  used  as  object:  Poen.  281,  Most.  272,  Pseud.  1178 
etiamne  facere  solitus  es  ?,  Rud.  402.  Sometimes  the  corresponding 
clause  is  only  implied :  Merc.  751,  Poen.  251,  True.  248,  Hec.  507 ; 
or  the  clauses  cover  each  other  in  sense,  not  in  terms,  so  that  for 
a  repetition  of  the  same  verb  we  have  two  verbs  expressing  essen- 
tially the  same  notion:  Merc.  437  etiam  mens  adnutat  (cf.  435 
iubet).  Men.  939  qui  mihi  etiam  me  iunctis  quadrigis  minitatu's 
prostemere  (c£  935,  which  recalls  851  f.),  Pseud.  628,  Rud.  201. 
382,  Mil  811,  Hec.  221. 

i.  Under  these  two  divisions  belong  the  infrequent  instances  of 
S€d  {verum)  eiiam :  preceded  by  nan  modo  Most.  390.  994,  Ad. 
387 ;  by  nan'-quidem  Most.  1 1 12 ;  without  any  preceding  particle, 
Poen.  1386. 

c.  Occasionally  the  two  divisions  are  not  sharply  distinguished. 
There  is  a  curious  transference  of  emphasis  in  Bac.  417  (cf.  Hec. 
543),  Rud.  1270,  where  the  adverb  belongs  logically  to  the  clause 
but  is  so  placed  as  to  throw  its  whole  stress  on  the  noun  or  pro- 
noun. In  Men.  944  it  qualifies  the  repeated  governing  verb  scto\ 
in  Amph.  91.  902,  Aul.  99,  Merc.  728,  Most.  422,  Haut.  865  (cf. 
inf.  VIII).  980,  Eun.  660  the  adverb  introduces  the  clause,  but 
by  position  emphasizes  especially  a  nominal  element  This  is,  I 
think,  also  the  case  in  the  much-discussed  passage  Men.  1039  f.; 
the  manuscript  reading  yields  a  perfectly  good  sense  if  we  under- 
stand etiam  hie  as  giving  a  definite  contrast  to  the  indefinite  aliii 
"  Some  deny  my  identity  and  shut  me  out  of  my  house ;  on  the 
other  hand  this  fellow  said,"  etc.  This  use  of  the  additory 
particle    where  English    requires  an  adversative    phrase  finds 


Digitized  by  VjOOQ IC 


32  AMERICAN  JOURNAL    OF  PHILOLOGY. 

parallels  in  Andr.  849/  Hec.  535,  and  a  more  perfect  one,  as 
showing  also  the  contrast  with  alias^  in  Cic.  Fin.  Ill  19.  63  ut 
enim  in  membris  alia  sunt  tamquam  sibi  nata  . . .  aliqua  etiam 
ceterorum  membrorum  usus  adiuvant  (so  MSS,  Madvig,  Orelli  ; 
Baiter  needlessly  adopts  the  conjecture  of  Marsus,  alia  etiam). 

d.  In  a  number  of  passages,  many  of  which  are  interrogative, 
the  additory  eiiant  is  weakened  to  the  point  of  being  for  us 
untranslatable  except  by  a  vocal  stress  upon  the  modified  word : 
a  weakening  which  arises  from  the  absence  of  any  contrast,  or 
even  any  very  clear  implication  of  a  contrast,  to  the  phrase 
modified.  Not  that  the  contrasting  notion  is  always  completely 
effaced :  thus  in  Trin.  934  it  is  easily  found,  but  not  without  going 
outside  of  the  text.  I  cite  the  passages  in  full,  placing  side  by 
side  those  which  exhibit  the  same  emphasized  verb ;  the  stress  is 
of  course  variously  laid  on  the  clause  or  a  single  element : 

Aul.  307  at  scin  etiam  quomodo?,  Amph.  773  an  etiam  credis 
id?,  Capt.  556  etiam  huic  credis?,  Eun.  loii  at  etiam  primo  calli- 
dum  ac  disertum  credidi  hominem :  Capt.  255  vix  cavet  quom 
etiam  cavet :  Haut.  235  etiam  caves  ?,  Capt.  327  est  etiam  ubi  pro- 
fecto  damnum  facere  praestet  quam  lucrum  :  455  at  etiam  dubitavi 
. . .  diu :  Amph.  814  haeret  haec  res  si  quidem  etiam  mulier 
factast  ex  viro :  Bac.  216  sed  B.  etiam  fortis  tibi  visast  ?,  910  etiam 
me  mones?,  Epid.  524  is  etiam  sese  sapere  memorat :  Most.  377 
quid  illi  reditio  hue  etiam  fuit?,  552  etiam  fatetur  de  hospite?, 
Merc.  202  etiam  rogas?  (cf.  inf.  IX  c),  Men.  1072  ego  hunc  cense- 
bam  ted  esse :  huic  etiam  exhibui  negotium :  Pseud.  1 172  an  etiam 
ille  umquam  expugnavit  ?,  1 177  tune  etiam  cubitare  solitu's  ?,  Cure. 
191  tune  etiam  . . .  odium  me  vocas?,  Poen.  271  tune  audes  etiam 
servos  spemere  ?,  Rud.  982  ausis  etiam  comparare  ?,  Trin.  934  an 
etiam  Arabiast  in  Ponto?  Cas.  prol.  74  quam  liberales  etiam 
probably  belongs  under  this  head. 

VI.  The  intensive  value  of  eiiam,  where  not  coming  from  the 
temporal  (c£  sup.  IV),  is  an  outgrowth  of  the  additory,  from  which, 
strictly  speaking,  it  does  not  differ ;  only  the  varying  tone  of  the 

*  Where  the  stress  rests  of  course  on  hoe  and  the  contrast  is  with  the  pre- 
ceding id:  **Of  that  there  is  now  no  question;  but  do  you  answer  me  as  to 
this."  AHud  tu  responds  would  express  the  contrast  in  a  less  concrete  and 
definite  fashion  ;  and  this  apparently  adversative  use  of  etiam  is  due  to  the  fact 
that  it  can  suggest  alius.  Cf.  Pseud.  370,  Bac.  274.  So  in  Hec,  I.  c,  etiam 
illorum  is  the  definite  substitute  required  for  aliorum. 


Digitized  by  VjOOQ IC 


vw*r  A  mr    tat    tot  a  rrfTTo    a\7t\    tz?  di?  Arr^i? 


Digitized  by  VjOOQ IC 


34  AMERICAN  JOURNAL   OF  PHILOLOGY. 

etiamnum  mali  pendii^  for  which  Schoell  conjectures  tt  iam 
malli\  the  simple  inversion,  nunc  cHam,  would  yield  a  good  sense 
(eh'am  =  'even ')  and  would  save  malipendii  as  a  pendant  to  bani 
cansulas  True.  429,  ctegui  banique facto  Haut.  788. 

prius:  On  Bac.  221  cf.  sup.  IV.  Elsewhere  the  two  words, 
while  having  much  liberty  of  position,  are  essentially  fused  into 
one  temporal  expression;  but  the  context  sometimes  gives  to 
eiiam  an  additory  connotation,  while /rrt^  serves  to  bring  out  the 
anticipatory  side.  In  the  following  passages  eiiam  might  stand 
alone :  Amph.  202,  Cist.  586,  Merc.  389.  569,  Mil.  1401,  Pseud. 
331 ;  in  the  following,  where  a  quam  clause  follows  or  precedes, 
prius  might  stand  alone,  save  in  so  far  as  eiiam  conveys  an 
additory  notion:  Asin.  232.  939,  Cure.  210,  Merc.  169,  Pseud. 
524-5,  Bac.  920-1  (this  last  the  only  instance  where  the  phrase 
is  negatived) ;  cf.  also  Mil.  1339. 

turn :  Pers.  356,  Rud.  846,  Hec.  145. 

etiam  atque  etiam :  Trin.  674,  Eun.  56,  cf.  Aul.  614. 

diu :  Cas.  606,  Haut.  402. 

denuo:  Amph.  394,  Bac.  923.  Here  the  idea  of  repetition 
possible  to  eiiam  is  made  more  distinct  by  the  accompanying 
adverb. 

parumper  Mil.  596;  paulisper  Aul.  805;  modo  Triil.  910 
(with  imperfect ;  the  only  other  example  of  this  tense  is  with 
neqiu  eiiam  Eun.  113). 

dum :  I  cite  the  passages  which  show  the  use  of  eiiam  and 
dum^  together  or  apart,  with  a  negative  word  : 

i)  etiam :  Amph.  248.  733,  Asin.  385.  445.  491,  Bac.  920,  Epid. 
336,  Merc.  386,  Mil.  1400,  Pers.  128.  231.  552.  630,  Pseud.  280, 
567,  Rud.  959,  Stich.  356,  True.  526,  Andr.  116.  503,  Haut.  433. 
1057,  Eun.  113.  360.  710.  1030.  1092,  Ph.  474.  The  negatives 
are :  non^  neque^  haud,  numquam,  nemo,  nihil)  they  precede  the 
temporal  particle  except  in  Pseud.  280  etiam  non  dedit,  Eun,  710 
etiam  non  credes  ? 

2)  etiam  dum  (dum  etiam):  Mil.  992,  Pers.  174,  Pseud.  957, 
True.  321,  Andr.  201.  807,  Haut.  229,  Eun.  570,  Hec.  192.  745. 
The  negative  is  nan  or  neque,  except  True.  321  haud,  Pseud.  957 
nihil. 

3)  dum:  Cure,  57,  Mil.  641.  787,  Pers.  137.  Pseud.  622.  730, 
Rud.  1201.  True.  205,  Andr.  340.  659,  Ph.  147.  445.  492,  Ad.  467. 
The  negative  is  non  or  neque;  in  Pers.  137  Schoell  adopts 
Ritschl's  metrical  correction,  hau  dum  for  nondum. 


Digitized  by  VjOOQ IC 


ETIAM  IN  PLAUTUS  AND  TERENCE,  35 

In  two  instances  the  MSS  of  Plautus  exhibit  eHamdum  without 
a  negative.     In  Rud.  1380-2  they  give: 

cedo  quicum  habeam  iudicem 
ni  dolo  malo  instipulatas  sis  nive  etiamdum  siem 
quinque  et  vigintl  natus  annos : 

annos  naius  and  sive  appear  in  Priscian's  quotation.  Sive  is 
adopted  by  Sonnenschein  and  approved  by  O.  SeyfTert  (Berl. 
Phil.  W.  XVI,  p.  1 291)  on  the  ground  that,  as  Sonnenschein  puts 
it,  "in  spansiones  what  the  challenger  denies  is  introduced  by  «, 
what  he  affirms  by  niJ*  He  adds,  as  a  further  argument  for 
adopting  sive^  that  the  second  clause  "  contains  the  word  eitamdum, 
'as  yet,'  which  is  used  only  where  a  negative  is  expressed  or 
implied."  I  know  of  no  instance  of  such  an  implication  of  the 
negative,  and  it  seems  impossible  because  eitam  (to  which  eiiam- 
dum  is  exactly  equivalent)  has  a  totally  different  meaning  in 
positive  clauses  from  that  which  belongs  to  it  with  a  negative.^ 
The  sponsio^  says  S.,  is  ''always  expressed  from  the  point  of  view 
of  the  person  challenged";  this  is  true  here  in  the  first  clause, 
where  ni  instipulaius  sis  represents  the  denial  of  Griphus,  non 
siipulaius  sum.  The  reading  sive  in  the  second  clause  would 
require  an  affirmation  on  the  part  of  Griphus ;  but  this  could  only 
be  expressed  by  iam  naius  eSj  since  eiiam  (dum)  naius  es  would 
be  contrary  to  the  syntax  of  the  temporal  eiiam  and,  even  if  that 
were  not  the  case,  could  only  mean  *you  are  still  of  that  age,' 
which  would  be  nonsense.  It  is  evident  therefore  that  Labrax 
formulates  this  part  of  the  challenge  from  his  own  point  of  view, 
and  that  we  must  on  the  one  hand  retain  nive,  and  on  the  other 
follow  Acidalius,  Bentley  and  Reiz  in  inserting  a  negative,  which, 
however,  should  stand,  according  to  the  normal  form,  before 
eiiam.* 

^English  'yet'  is  misleading;  alike  by  sense  and  construction  (with  perfect 
tense)  the  positive  of  non  etiam  {Jum)  is  not  etiam  but  iam.  This  appears 
clearly  in  the  question  which  is  akin  to  the  negative ;  cf.  Andr.  806-7  ^^^  suos 
parentis  repperit?  an  nondum  etiam?  *has  she  yet  found  her  parents?  not 
yet,  say  yon?*    In  the  first  clause  fiiam  would  be  impossible. 

'The  corrupt  lines  1 381-4  may,  I  think,  be  thus  emended:  ni  dolo  malo 
instipulatus  sis  nive  hand  dum  etiam  siem  |  quinque  et  viginti  annos  natus. — 
habe  cum  hoc. — at  aliost  opus.  |  nam  ego  ab  isto  auferre  haut  ausim  si  istunc 
condemnavero.  The  last  line  is  spoken  by  Labrax  to  the  audience :  **  When 
I  have  got  sentence  against  that  fellow  (Griphus),  I  shall  not  dare  to  claim 
my  money  from  that  one  (Daemones).*'    At  might  easily  be  dropped  by  haplog- 


Digitized  by  VjOOQ IC 


36  AMERICAN  JOURNAL   OF  PHILOLOGY. 

Pseud.  1028  reads:  metuo  autem  ne  erus  redeat  etiatndutn  a 
foro.  Here  no  negative  can  be  inserted ;  etiamdum  has  the  value 
of  the  afiirmative  etiam,  an  irregularity  to  which  we  have  a  pendant 
in  Epid.  42,  where  etiamnunc  is  used  with  a  negative,  and  a 
parallel  in  Cic.  Att.  13.  3i.  i  quoniam  etiamdum  abes.  As  nunc 
and  dum  are  mere  satellites,  adding  nothing  to  the  sense,  there  is 
no  ground  for  questioning  these  irregularities,  which  show  that 
the  restriction  of  etiamdum  to  the  negative,  of  eiiatpnunc  to  the 
affirmative,  function  of  etiam,  though  usual,  was  not  absolute. 
The  meaning  here  is  that  of  eiiam  =  etiam  prius. 

VIII.  The  additory  etiam  also  lends  itself  to  conjunction  with 
particles  kindred  in  meaning :  ultro :  Amph.  587,  Asin.  440  [Aul. 
530],  True.  112,  Eun.  860,  Ph.  360.  769;  insuper:  Eun.  645  (cf. 
sup.  Ill,  note);  adeo:  Most.  629.  Much  the  most  important 
of  these  combinations  is  that  with  quoque.  The  refusal  (as  of 
Brix  on  Men.  1160)  to  recognize  this  as  a  pleonasm  rests  on  the 
assumption  that  etiam  is  properly  an  intensive  particle ;  so  Brix 
says  "quoque  ist  vergleichend,  etiam  steigernd";  and  Lorenz  on 
Most,  mo,  while  speaking  oi  etiam  quoque  as  a  pleonasm,  incon- 
sistently agrees  with  Brix  in  translating  it  by  "sogar  auch."  But 
this  rendering  is  permissible  in  few  cases,  necessary  in  none ;  and 
in  fact  etiam  and  quoque,  so  far  as  they  coincide  in  range,  corres- 
pond exactly  in  meaning,  both  having  the  additory  sense  'also,* 
from  which  the  intensive  'even'  is  derived  for  both,  though  less 
frequently  for  quoque}  The  real  difference  between  the  two  words 
lies  in  the  fact  that  etiam,  by  reason  of  its  temporal  origin,  belongs 
primarily  to  the  verb,  while  quoque  is  a  nominal  adjunct.'    This 

raphy  before  a/—;  and  the  reduction  of  haut  ausim  to  {h)aut  sim  would  be 
merely  another  form  of  the  same  error.  V^ith  haud  dum  etiam  cf.  Pers.  174 
nondum  eiiam,  Etiam  was  probably  omitted  by  the  copyist  of  the  archetype 
and  inserted  in  such  a  manner  as  to  make  it  seem  intended  as  a  substitute  for 
hand, 

^Prof.  Gildersleeve,  Lat.  Gr.  479,  Remark*  calls  attention  to  the  fact  that 
**  the  difference  between  etiam  and  quoque  is  not  to  be  insisted  on  too  rigidly/' 
and  cite^,  for  quoque  ^^tytviy  Juv.  4.  116.  Plautine  examples  are  Asin.  207, 
Foen.  166.  888,  Pseud.  295. 

'  So  regularly  in  Terence,  the  only  exception  being  Haut.  866,  where,  how- 
ever, desponsam  is  a  nominal  element  of  the  verb.  Out  of  some  fifty  examples 
of  quoque  which  I  have  noticed  in  Plautus,  five  exhibit  it  as  a  verbal  adjunct : 
Pers.  234,  Cist.  35,  Pseud.  367,  Capt.  284,  Bac.  892  (in  Most.  538  text  and  sense 
are  uncertain). 


Digitized  by  VjOOQ IC 


ETJAM  IN  PLAUTUS  AND  TERENCE.  37 

distinction  is  actually  retained  in  Haut.  865-6  postremo  etiam  si 
voles  desponsam  quoque  esse  dicito,  where  etiam  by  its  position 
serves  to  introduce  the  clause  (cf.  Most.  422  quin  etiam  illi  hoc 
dicito).  This  is  the  only  Terentian  example  of  eiiam — guoqtie ; 
in  Plautus  the  fusion  is  complete  and  the  phrase  qualifies  always 
a  noun  or  pronoun:  Amph.  461.  702,  Asin.  502.  567,  Cure.  128, 
Most,  mo,  Pseud.  121.  353.  Quoque  eiiam,  following  the  noun 
or  pronoun,  is  found  in  Pers,  145.  744,  Stich.  258,  Poen.  40,  Merc. 
299-  328,  True.  94.  731.  875,  Amph.  30.  81.  717.  753,  Epid.  234. 
589,  Men.  1 160,  Pseud.  932,  Hec.  543.  762.  In  Amph.  281  the 
words  guoque — eiiam  after  a  pronoun  are  separated  by  edepol,  in 
Trin.  1048  by  the  verb ;  and  in  Hec.  734  ego  pol  quoque  etiam 
timida  sum,  the  adverbs  precede  the  noun  (as  quoque  alone  does, 
Asin.  184). 

IX.  It  has  been  said  that  eiiam  in  questions  loses  wholly  or  to 
a  great  extent  its  proper  signification.  Thus  Ussing  on  Asin.  677 
(670)  etiam  me  delusisti,  says:  '* etiam  saepe  ponitur  in  principio 
interrogationis,  ubi  quaeritur  aut  quod  per  se  vix  credibile  videtur 
.  . .  aut  quod  propter  alterius  cunctationem  fore  non  videtur  . .  . 
Hoc  adhortandi  quandam  vim  habet  nee  multum  abest  a  quin." 
This  view  appears  in  the  notes  of  Sonnenschein  and  Lorenz  on 
Most.  383,  Palmer  on  Amph.  369  and  Brix  on  Trin.  514;  objec- 
tion is  raised  to  it  by  Langen,  Beitr.,  p.  160,  who  says  that  eiiam^ 
in  a  question  retains  its  proper  meaning ;  but  he  admits  as  an 
exception  *'einen  bestimmten  Fall,  in  welchem  das  abgeschwachte 
etiam  lediglich  dazu  dient,  der  Frage  eine  besondere  Niiance  zu 
geben,  wenn  sie  namUch  statt  einer  AufTorderung  dient."  This 
weakening  of  the  adverb  Langen  finds  in  questions  where  "etiam 
dem  Verbum  immer  vorangeht  und,  wenn  die  Fragepartikel  ne 
hinzutritt,  diese  sich  an  etiam,  nicht  an  das  Verbum  anschliesst" 
Morris,  Am.  Jour.  Phil.  XI,  p.  180,  says  that  "of  words  of  restric- 
tion and  definition  . . .  only  num  clearly  assumed  the  function  of 
an  interrogative  particle,  though  etiam  came  very  near  doing  so." 
SeyfTert,  in  a  review  of  Morris's  article.  Burs.  Jahresb.  80  (1894), 
p.  346,  denies  that  in  Bac.  216,  Most.  553,  Pers.  651  eiiam  stands 
"  without  any  proper  meaning,"  and  adds,  "  ob  etiam  iiberhaupt 
in  Fragen  wie  etiam  rogas,  minitaris  seine  Bedeutung  soweit  ver- 
liert,  dass  es  nur  dazu  dient  der  Frage  den  Ton  des  Dringlichen, 
Ungeduldigen  zu  geben,  wie  M.  meint,  scheint  mir  gleichfalls 
zweifelhaft;   ich  wenigstens   fiihle   an  alien  diesen  Stellen  die 


Digitized  by  VjOOQ IC 


38  AMERICAN  JOURt^AL    OF  PHILOLOGY, 

Bedeutung  von  etiam  noch  klar  heraus.  Freilich  liesse  sich  dies 
nur  im  Zusammenhang  einer  eingehenden  Untersuchung  iiber 
den  Gebrauch  von  etiam  bei  Plautus  darlegen." 

Langen's  theory  seems  untenable ;  etiam  and  etiamne  preceding 
the  verb  are  found  in  other  passages  than  the  score  which  he 
cites,  and  we  cannot  rely  on  a  formal  distinction  which  is  not 
universally  valid.  In  general,  SeyfTert  is,  I  think,  right  in  feeling 
that  the  significance  of  etiam  cannot  be  obscured ;  yet  a  few 
passages  support  the  view  of  Ussing  and  Morris  by  exhibiting  an 
etiam  which  has  no  distinct  characteristic  other  than  that  it  intro- 
duces an  emphatic  question.  But  instead  of  extending  this 
apparently  meaningless  use  to  passages  where  a  meaning  may 
easily  be  found,  we  must  rather  seek  to  restrict  it  to  cases  where 
it  is  inevitable,  and  to  find  its  origin,  if  possible,  in  some  significant 
employment  of  the  adverb.  I  have  already  cited  many  interroga- 
tive clauses  in  which  the  value  of  the  particle  seemed  sufficiently 
obvious,  and  have  reserved  for  this  place  certain  questions  which 
can  be  easily  classified  under  a  few  heads. 

<i)  etiam  temporal  with  verbs  of  speaking:  muttis?  Pers.  827, 
Amph.  381  ("nondum  taces?"  Ussing),  muttire  audes?  Men. 
710-1,  loquere?  Merc.  982,  Pers.  848,  clamas?  Amph.  376;  and 
with  omission  of  verb,  etiamne  'opinor'?  Pers.  651  (^'nondum 
certumst?"  SeyfTert,  1.  c.).  In  Epid.  711  etiam  inclamitor  quasi 
servos,  the  suggested  contrast  may  be  given  in  the  form  of  an 
assertion:  iam  non  inclamitandus  sum,  In  the  following  (where 
the  verb  of  speaking  is  specialized  as  a  verb  of  abuse)  the  idea  of 
repetition  enters  in,  reference  being  had  to  previous  utterances  of 
the  same  kind:  etiam  male  loquere?  Pers.  289,  at  etiam  male 
loqui  mi  audes?  Capt  563-4  (this  being  the  second  insulting 
remark  of  Tyndarus,  cf.  551  f.),  at  etiam  maledicis?  Trin.  991  (cf. 
926  ne  male  loquere  apsenti  amico),  etiam  minitare?  True.  621 
(cf.  612  f.). 

b)  etiam  temporal  in  the  elliptical  construction  noticed  under 
II  fin :  cf.  Weise,  Lex.  Plant.,  p.  463,  with  whose  choice  of  examples, 
however,  I  cannot  wholly  agree :  respicis  ?  Pers.  275,  Most.  886, 
dicis?  Pers.  277,  tenes?  413,  vigilas?  Most.  383,  aperis?  937. 
938,  adstas?  Cas.  728,  imus?  977,  abis?  Poen.  431,  despondes? 
Aul.  255,  acceptura  es  ?  Rud.  467. 469,  redditis?  Bac.  1167,  taces? 
Trin.  514.  In  all  these  cases  the  demand  (or  command)  has 
already  been  uttered,  and  etiam  reintroduces  it  with  emphasis : 
"once  more  (I  say),  will  you  — ?  "     In  Pers.  542  etiam  tu  illam 


Digitized  by  VjOOQ IC 


mmti 


ETIAM  IN  PLAUTUS  AND  TERENCE,  39 

destinas?,  Cure.  i88  etiam  dispertimini  ?  it  is  rather  a  suggestion 
than  a  demand  that  has  preceded ;  in  Bac.  670  etiam  quid  respon- 
detis  mihi  ?  there  is  a  sudden  shift  in  the  kind  of  question  used ; 
and  in  Men.  697  mane:  etiam  astas?  etiam  audes  mea  revorti 
gratia  ?  the  particle  follows  so  quickly  on  the  command  that  we 
have  hardly  time  to  feel  its  force.  In  the  following  there  is  no 
backward  reference  and  the  meaning  is  quite  effaced :  amoves  ? 
Asin.  714,  taces?  Pers.  152,  Cure.  41,  Ad.  550,  abis?  Ph.  542.  It 
is  only  to  these  last  instances  that  Ussing's  formula  ''etiam  inter- 
rogantis"  can  properly  be  applied ;  and  these  can  easily  be  under- 
stood as  representing  the  degeneracy  of  an  idiom  which  in  its 
original  value  seems  not  to  have  survived  Plautus.  For  the  two 
Terentian  examples  show  no  trace  of  repetition ;  and  we  may 
conclude  that,  while  etiam  was,  as  Morris  says,  on  its  way  to 
become  a  mere  particle  of  interrogation,  it  was  diverted  from 
this  tendency  by  the  disappearance  of  the  usage  from  which  the 
tendency  was  derived. 

c)  etiam  additory  or  intensive  in  questions  (occasionally  also  in 
affirmative  clauses)  which  imply  that  the  utterance,  or  act,  is  an 
aggravation  of  some  wrong  done,  an  adding  of  insult  to  injury. 
Thus,  in  affirmative  clauses :  Amph.  586-7  qui  quoniam  erus  quod 
imperavit  neglexisti  persequi  |  nunc  venis  etiam  ultro  inrisum 
dominum;  Poen.  1280  f.  si  ego  minam  n'on  ultus  fuero  probe 
quam  lenoni  dedi  ...  |  is  etiam  me  ad  prandium  ad  se  abduxit 
ignavissumus,  |  ipse  abiit  foras;  Rud.  325  f.  data  verbo  ero  sunt : 
leno  abit  scelestus  exulatum.  |  in  navem  ascendit,  mulieres  avexit 
...  I  is  hue  erum  etiam  ad  prandium  vocavit,  sceleris  semen.  In 
interrogative  clauses,  expressing  anger  at  denial,  mockery,  evasion 
or  threats:  negas?  Mere.  763,  inrides?  Most.  1132,  derides? 
Men.  499,  rides?  Eun.  10x7,  ut  etiam  inrideat?  Ph.  669,  delusisti? 
Asin.  677,  exordire  argutias?  Bac.  127,  minitare  Bac.  785-6 
(differentiated  by  the  context  from  True.  621) ;  so  of  a  question 
which  is  regarded  as  evasion  or  denial:  rogas?  Amph.  571.  1025, 
Andr.  762,  rogitas?  Amph.  1029,  Aul.  424.  437.  633,  Cas.  997. 
But  for  Merc.  202  etiam  rogas?  cf.  sup.  V;  here  no  contrast  is 
implied,  and  etiam  therefore  merely  gives  emphasis  to  the  verb. 

X.  The  ins  affirmativa  of  etiam  is  to  be  found  in  its  use  as  a 
particle  of  response  = 'yes';  but  this  use  is  very  rare,  and  as 
a  rule  etiam  in  the  response  has  some  meaning  of  its  own.  In 
Poen.  406,  atque  audin? — etiam,  the  temporal  sense  is  evident. 


Digitized  by  VjOOQ IC 


40  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Amph.  544  numquid  vis  ? — etiam :  ut  actutum  advents :  Bac.  757 
numquid  aliud? — hoc  atque  etiam:  Hec.  811  nil  aliud  dicam? 
etiam ;  show  the  half-temporal,  half-additory  value  noted  under 
III ;  indeed,  from  the  elliptical  phrases  there  cited  these  differ 
only  by  standing  after  a  question.  The  same  notion  is  underlying 
in  Most.  1000  numquid  processit . .  •  novi  ? — etiam  :  but  the  use 
of  the  perfect  tense  puts  the  verb  out  of  sympathy  with  the 
particle,  and  thus  reduces  the  latter  to  a  mere  affirmative. 

XI.  This  is  the  case  also  in  Baa  214-5  sed  nilne  hue  attulistis 
inde  auri  domum  ? — ^immo  etiam.  The  fusion  of  eHam  with  immo 
to  a  phrase  of  affirmation  is  therefore  here  only  apparent ;  and 
elsewhere,  in  Plautus  as  in  Terence,  the  two  words  have  no 
relation  to  each  other.  The  following  instances  of  their  occur- 
rence in  close  proximity  have  already  been  cited :  under  I,  Poen. 
315  ;  III,  Cure.  453,  Ph.  877,  Hec.  869;  IV,  Rud.  961,  Pseud. 
324,  Bac.  221;  VI,  Epid.  518;  VII,  Mil.  1401,  Rud.  1123.  In 
Poen.  188  immo  etiam  ubi  expolivero  magis  hoc  tum  demum 
dices,  Rud.  441  immo  etiam  tibi,  mea  voluptas,  quae  voles  faciam 
omnia  (both  referred  to  under  III)  the  sympathetic  tense  decides, 
as  so  often,  the  value  of  the  adverb;  the  singularity  of  both 
passages  lies  in  the  fact  that  they  alone  exhibit  the  purely  tem- 
poral notion  with  a  future,  all  other  examples  having  additory 
connotation,  unless  eiiam  is  reinforced  by  another  adverb  of  time, 
as  in  Haut.  402  diu  etiam  duras  dabit,  or  by  a  sympathetic  signifi- 
cation of  the  verb,  as  in  Trin.  11 36  maneam  (these  two  elements 
combine  in  Aul.  805,  Cas.  606).  Poen.  188  is  also  noticeable 
because  the  more  exact  definition  of  time,  ubi — demum,  following 
on  etiam  detaches  the  latter  to  some  extent  from  the  main  verb 
and  makes  it  almost  equivalent  to  mane. 

Andr.  655  reads:  immo  etiam,  quo  tu  minus  scis  aerumnas 
meas,  hae  nuptiae  non  adparabantur  mihi.  Dziatzko's  change  of 
guo  to  guom  shows  that  he  understands  immo  eiiam  in  the  sense 
of  Spengel's  explanation,  "ironisch  bestatigcnd."  But  it  is  only 
immo  that  confirms,  and  the  relative  pronoun  is  necessary  to 
eiiam,  as  may  be  seen  by  a  comparison  with  Ph.  877  immo  etiam 
dabo  quo  magis  credas,  Hec.  869  immo  etiam  qui  hoc  occultari 
facilius  credas  dabo ;  the  ellipsis  is  like  that  in  Capt  290  immo 
etiam  ut  magis  noscas,  where  also  etiam  is  independent  of  the 
principal  verb,  uHhir,  as  here  it  is  independent  of  adparabantur. 
These  three  passages  differ  firom  the  one  under  discussion  only 


Digitized  by  VjOOQ IC 


ETIAM  IN  PLAUTUS  AND  TERENCE,  4I 

by  the  circumstance  that  in  them  the  subordinate  clause  gives  the 
purpose  for  which  the  statement  is  made,  while  here  it  merely 
defines  the  character  of  the  statement.  Pamphilus  has  several 
times  declared  to  his  friend  that  he  does  not  understand  the  cir- 
cumstances: 647  ialsu's,  649  nescis,  652  baud  istuc  dicas  si  cog- 
noris;  Charinus  insists:  scio  e.  q.  s.  653-4;  "I  understand  very 
well ;  you  have  had  a  quarrel  with  your  father,  he  is  angry  with 
you  and  could  not  get  you  to  marry  to-day."  "Nay,"  answers 
Pamphilus, "  I  will  tell  you  another  thing  by  (ignorance  of)  which 
you  fail  to  understand  my  troubles ;  no  one  was  asking  me  to 
marry." 

Andr.  673  Davus  says:  nisi  id  putas,  quia  primo  processit 
parum,  |  non  posse  iam  ad  salutem  convorti  hoc  malum.  His 
master  replies :  immo  etiam :  nam  satis  credo,  si  advigilaveris,  |  ex 
unis  geminas  mihi  conficies  nuptias.  He  means:  immo  etiam 
posse  puto ;  the  temporal  sense  is  clearly  given  by  the  preceding 
nan  tarn,  which  is  the  negative  of  the  temporal  efiam,  as  non  etiam 
is  the  negative  of  tarn ;  and  cf.  Stich.  617  posse  edepol  tibi  opinor 
etiam  uni  locum  condi  bonum.  In  Andr.  708-9  Dziatzko  and 
Spengel  put  a  stop  after  eitam ;  but  etiam  volOy  or  dicitOy  implies 
an  object:  "I  wish  (say)  something  else":  so  in  Most.  1000,  Bac. 
215  an  etiam  aliquid  was  implied  by  the  preceding  quid  and  nil* 
The  punctuation  of  Umpfenbach  is  therefore  to  be  preferred : 
verum  vis  dicam? — immo  etiam  |  narrationis  mi  incipit  initium. 
The  force  is  temporal:  "nay,  he  is  still  (etiam  prius  quam  abit) 
beginning  a  speech  to  me."^ 

XII.  In  Mil.  1014  immo  etiam  might  be  treated  as  temporal  if 
we  restore  the  metre  and  furnish  the  temporal  contrast  by  reading 
sed  iam  non  celas ;  but  Studemund's  reading  of  A  makes  such 
conjecture  needless.  For  Trin.  708  [Pseud.  566]  I  can  find  no 
explanation ;  in  Rud.  783  etiam  may  have  the  value  considered 
under  X  ^ ;  in  Cure.  612  it  cannot  easily  be  defended.  Cist.  518 
seems  hopelessly  corrupt ;  but  in  any  attempt  to  restore  it  the 
reading  of  A  deserves  more  consideration  than  it  seems  to  have 
received :  anne  etiam  quid  consultura  es  ?  "  are  you  going  to  form 

^Narratio  in  the  technical  sense ;  the  jest  is  Menander's,  not  Terence's ;  for 
stock  oratorical  forms  were  not  familiar  to  the  lattefs  audiences,  and  it  is  the 
Attic  orators  who  constantly  preface  their  St^ytfot^  with  an  assurance  of  truth- 
fulness :  awavra  inideiiu  •  .  .  Xiyuv  rahid^^  says  the  Lysianic  Euphiletus ; 
similar  examples  abound. 


Digitized  by  VjOOQ IC 


42  AMERICAN  JOURNAL    OF  PHILOLOGY. 

any  other,  any  new  decision?"  Poen.  570  may  perhaps  be 
referred  to  VI  df;  but  this  interpolation  shows  an  un-Plautine 
lack  of  emphasis  in  the  particle.  If  we  accept  Lachmann's 
emendation  for  Asin.  499,  it  is  to  be  referred  to  VI  c^  the  closest 
parallel  being  Amph.  91 ;  but  the  passage  may  be  regarded  as 
still  doubtful.  But  Bac.  321  etiam  dimidium  censes?  sc.  eum 
attulisse,  though  unique,  need  cause  no  difficulty  when  we 
remember  that  the  additory  or  intensive  etiam  has  ne — quidem 
for  its  negative.  "Has  he  brought  even  the  half,  do  you  think?" 
asks  Nicobulus ;  in  a  less  hopeful  mood  he  would  have  exclaimed : 
ne  dimidium  quidem  attulit?^ 

VAifDMBiLT  umvwsiTY.  Wm.  HAMILTON  Kirk. 

^  At  the  time  of  writing  this  article  I  had  not  seen  Birt's  discussion  of  the 
etymology  of  etiam^  Rh.  M.  51  (1896),  p.  70  f.  It  was  gratifying  to  find  Birt 
arguing  with  force  and,  as  it  seems  to  me,  with  cogency  for  the  rejection  of  the 
ordinary  derivation  and  the  acceptance  of  that  from  eti  (=  fri)  -f*  ioM,  This 
naturally  leads  him  to  conclude  that  the  original  sense  was  temporal,  and  he 
notices  this  fact  briefly,  p.  107-8,  quoting  some  passages  in  illustration,  chiefly 
from  Plautus. 


Digitized  by  VjOOQ IC 


IV.— THE  ORIGIN  6F  LATIN  BAUD  AND  GREEK  oy; 

AND  THE  EXTENSIONS  OF  THE  ORIGINALLY 

UNEXTENDED  FORM.^ 

A.— The  Origin  of  Lat.  Aaud  Gk.  ov  *not.' 
§1.  Iniroductton. 

Lat  hand  and  Gk.  ov  'not'  have  long  been  the  subject  of 
discussion,  but  it  will  hardly  be  contended  that  the  question  of 
their  origin  has  yet  received  a  satisfactory  answer.  Under  these 
circumstances  I  would  venture  to  offer  a  new  explanation  in  the 
following  pages. 

§2.  The  three  farms  Aau,  Aaud,  kauL  The  evidence  of  (a)  /»- 
scrtpiians^  (^)  MSS  and  Libri,  {c)  The  Ancient  Grammarians. 
The  form  *au  established  as  the  earlier  Latin  form. 

One  of  the  chief  difficulties  lies  in  the  Latin  form.  Hence  we 
shall  do  well  to  examine  this  word  first,  to  see  what  its  earlier 
form  was  in  Latin.  * 

The  three  forms  hau,  haud,  haul. 

We  have  apparently  three  forms  to  deal  with — namely,  hau^ 
hand,  haut,  Ritschl,  Prol.  ad  Plant.  Trin.,  pp.  xcix-cii  (1848), 
writes :  "  Corruptelis  autem  etiam  haut  scriptura  non  raro  prodi- 
tur,  velut  cum  pro  eo  aui  positum  est,  e.  g.  Trinummus,  vv.  362, 
721.  Sed  novum  est  quod  duabus  haui  et  haud  formis  tertia  hau 
accessit,  suscepta  a  me  ex  Ambrosiano  vv.  233  {hau  liquet)  y  462 
{hau  bonumst),  in  eodemque  codice  aliis  in  fabulis  tam  saepe 
exstans,  ut  de  calami  lapsu  cogitari  nequeat." 

(a)  Inscriptional  Evidence. 

The  usual  form  of  the  word  on  inscriptions  is  haud^  e.  g.  C.  I.  L. 
I  1306  quaniam  haud  licitum,  but  we  find  hau  in  one  inscription, 

'  The  present  is  the  paper  to  which  an  advance-reference  has  already  been 
made  in  the  Essay  on  the  'Establishment  and  Extension  of  the  Law  of 
Thurneysen  and  Havet/  Part  II  (Amer.  Joum.  of  Philology,  vol.  XVII,  part  2, 
July,  X896).  §6,  p.  180,  n.  3,  and  §8.  p.  189,  n.  i.  and  p.  193. 


Digitized  by  VjOOQ IC 


44  AMERICAN  JOURNAL   OF  PHILOLOGY. 

C.  I.  L.  I  1007  (=  Orelli  4848  =  Gruterus,  p.  769)  heic  est  sepul- 
crum  hau  pulcrum  pulcrai  feminae}  hau  does  not  occur  again 
in  C.  I.  L.  I,  nor  does  it  occur  at  all  in  the  indices  to  C.  I.  L.  II-V 
(incL),  VII-X  (ind.),  XII,  XIV.  The  form  haui  is  found  in  C.  I.  L. 
II  562  ^haut  liciium^  (an  inscription  'aevi  Antoniani'  probably), 
and  XII  ti499  ^kaut  dispar*  ('ex  titulis  Christianorum ')f  but  is 
not  found  in  the  indices  toX.  I.  L.  I,'  III-V  (incL),  VII-X  (inch), 
XIV. 

{b)  Evidence  of  MSS  and  Libri. 

Otto  Ribbeck,  Scaenicae  Romanorum  Poesis  Fragmenta,  vol.  I 
(Tragicorum  Fragm.),  187 1,  gives  the  following  readings : — 

hatiqucLquaniy  Att  6x8. 

haud  in  Ennius  330';  Pacuvius  325,  426';  inc.  inc.  fab.  30.' 

haut  in  Livius  35;  Enn.  340;  Attius  108,  115,  193,  330,  360, 
466;  Fabul.  Praetext.  31. 

Otto  Ribbeck,  op.  cit.,  vol.  II  (Comicorum  Fragm.),  1873: — 

hau,  Liv.  3  {haui  codices) ;  Naev.  16  {hau  om.  libri) ;  Titin.  181 
(The  MS  has  room  for  three  letters  only) ;  Afran.  58  {aui  libri) ; 
Pub.  Syri  Sententiae  461  (=  694)  (reading  very  doubtful). 

?iaudy  Fabul.  Palliat.  inc.  inc.  47^;  Afranius  12^;  Sententiae 
Turicenses  693  (=  850). 

hauiy  Naevius  60,  112;  Caecilius  i8r;  Turpilius  9,  10;  Fab. 
Pall.  inc.  inc.  74;  Titinius  127,  166;  Afran.  7,  51. 

L.  Miiller,  Q.  Enni  Carminum  Reliquiae  (1885),  reads  as 
follows : — 

haudy  Annales  278,  389,  578 ;  Fabulae  127,  424  (=  Ribbeck, 
voL  I,  Enn.  340  haui,  v.  supra). 

haudquaquam,  Ann.  293. 

haut,  Ann.  499. 

In  connexion  with  Ribbeck's  '  velut  cum  pro  eo  (s.  c.  hauf)  aui 

*C.  I.  L.  I  1007  (=also  F.  D.  Allen,  Remnants  of  Early  Latin.  No.  138, 
where  it  is  Included  among  '  Epitaphs  dating  from  about  the  Gracchan  period 
on')  and  I  1306  are  among  the  "  Inscriptiones  a  bello  Hannibalico  ad  C. 
Caesaris  mortem"  (see  Mommsen  in  C.  I.  L.  I,  pp.  5,  43),  and  they  are,  I 
believe,  the  only  instances  of  the  negative  in  question  to  be  found  in  C.  I.  L.  I. 

^haut  does  not  occur  in  C.  I.  L.  I  1306,  as  might  perhaps  be  at  first  sight 
inferred  from  Stolz,  Lat.  Gr.',  §69,  p.  316.  This  particular  inscription  (quoted 
above  in  the  text)  shews  *  haud  licitum!  with  which  contrast  hata  lieitum  in 
C.  I.  L.  II  562  (also  quoted  in  the  text  above). 

*  Wrongly  placed  under  haui  in  Rtbbeck's  index  to  vol.  I. 

*  Wrongly  placed  under  haut  in  Ribbeck's  index  to  vol.  II. 


Digitized  by  VjOOQ IC 


LATIN  HAUD  AND  GREEK  OT.  45 

positum  est'  we  may  note  that  in  the  following  of  the  above- 
mentioned  passages  the  reading  out  is  supported  either  by  Libri 
or  by  at  least  one  MS : 

Ribbeck,  op.  cit.,  vol.  I,  Enn.  340;  Pac.  426;  Att.  108,  115, 

I93»  330- 

Ribbeck,  op.  cit,  vol.  II,  Naev.  60,  112;  Turpil.  9,  10;  Pall, 
inc.  inc.  74;  Titin.  166;  Afran.  7,  12,  51,  58. 

Miiller,  op.  cit.,  Enn.  Ann.  499 ;  and  Fab.  424  (=  Ribbeck,  op. 
cit.,  I,  Enn.  340). 

Ritschl,  in  his  edition  of  Plautus,  Trin.  1848,  reads  as  follows 
(I  add  critical  notes  in  brackets) : 

hau  in  lines  233  (Jiau  A,  ut  videtur.  hand  reliqui),  462  (Jiau  A, 
haud  reliqui).* 

haui  in  lines  60  {haui  A,  haud  reliqui,  et  sic  constanter  nisi  ubi 
contrarium  testabimur) ;  62  {haui  A.  R.,  haud  reliqui);  90  {aut 
H.) ;  1 15  {Jiaut  B,  haud  reliqui) ;  362  (jxui  A,  haud  reliqui) ;  445 
[haud  BCDE,  hau  Camerarius,  au  Palmerius  Spicil.,  p.  859)  ;  584 
[haud  dare  Pius,  haddare  B,  addere  CDEZ);  625  [haui  B, 
haud  reliqui) ;  721  [haud  Dousa  iv.  24,  Scaliger.  aut  libri) ;  835 ; 
II57- 

[c)  The  opinions  of  the  Ancieni  Grammarians. 

Marius  Victorinus  (flor.  about  360  A.  D.),  according  to  the 
reading  of  Ritschl,  Proleg.  ad  Plant.  Trinum.,  page  c,  writes : 

^*Hau  adverbium  est  negandi  et  significat  idem  quod  apud 
Graecos  ov:  sed  ab  antiquis  cum  adspiratione,  ut  alia  quoque 
verba,  dictum  est  et  adiecta  d  littera,  quam  plerisque  verbis 
adiiciebant.  d  tamen  litteram  conservat,  si  sequens  verbum 
incipiat  a  vocali  ut  haud  aliter  muros  et  haud  equidem.  at  cum 
verbum  a  consonanti  incipit,  d  perdit  ut  hau  duiiam  et  hau  multa 
et  hauplacitura  refer '^'^ 

An  alternative  reading  given  by  Ritschl  (1.  c.)  runs  as  follows : 

*  To  these  statistics  we  may  add  that  Georges,  Lex.  der  Lat.  Wortf.,  gives 
hau  in  Plant.  Bacch.  506;  Men.  927;  Most.  434,  720,  919;  Pers.  11,  23,  500; 
Poen.,  Prol.  94;  Pseud.  215.  Nipperdey,  Ritter  read  hau  in  Tacitus,  Ann. 
(e.  g.)  II  36  and  VI  43  (49).— P.  S.  Reference  may  profiUbly  be  made  also  to 
Friedr.  Neue.  Formenl.  d.  Lat.  Spr.  II'  664  sqq. 

'  The  reading  "  at  cum  verbum  a  consonanti  incipit,  d  perdit  ut  haut  dudum 
et  haut  muUum  et  hatU  placitura  refert^  et  inducit  /"  cannot  possibly  stand. 
Keil's  reading  is,  in  the  main,  identical  with  that  of  Ritschl,  and  runs  thus : 
"  at  cum  sequens  verbum  a  consonanti  incipit,  d  perdit,  ut  hau  dudum  et  hau 
mulhtm  et  *  hau  placitura  refer*  [et  inducit  /]." 


Digitized  by  VjOOQ IC 


46  AMERICAN  JOURNAL  PF  PHILOLOGY. 

^'Haud  adverbium  est  negandi  et  signiiicat  idem  quod  apud 
Graecos  ov,  et  fuit  au :  sed  ab  antiquis  etc."  ^ 

Flavius  Caper  (flor.  before  the  end  of  the  4th  century  A.  D.)f 
according  to  the  reading  of  Keil,  Gramm.  Lat.,  vol.  VII,  p.  96, 
1.  4,  writes : — 

''*hau  dolo^  per  d  recte  scribitur,  etenim  d  inter  duas  vocal es 
esse  debet,    quod  si  consonans  sequitur,  d  addi  non  debet,  ut 

Charisius  (flor.  some  time  between  the  middle  of  the  4th  and 
end  of  the  5th  centuries  A.  D.),  Institut  Gramm.,  bk.  I,  §xv  ad 
fin.,  after  a  brief  discussion  of  the  particle  sed^  continues: — 

'^Haud  similiter  d  littera  terminatur:  du  enim  Graeca  vox'  d 
littera  termina[ri  apud  antiquos]  coepit  quibus  mos  erat  d  litt[eram 
omnibus]  paene  vocibus  vocali  littera  finitis  adiungere,  ut  quo  ted 
hoc  noctis  [dicam  pro\ficisci  foras,  Sed  et  per  /  scribi  sonus 
vocis  admittit." 

Charisius  thus  gives  authority  also  for  the  form  haid. 

Resulis  of  the  foregoing  investigation. 

(i)  The  three  forms  haUy  haudy  haut  are  established  by  the 
united  evidence  of  Inscriptions,  MSS  and  Grammarians. 

(ii)  The  earlier  form  of  the  Latin  word  under  discussion  seems 
beyond  all  reasonable  doubt  to  have  been  *au. 

Thus  we  have  before  us  Lat.  "^au  'not'  :  Gk.  ov  'not';  and  the 
problem  is  to  find  the  connexion  between  them. 

The  explanation  of  the  h  and  d:  t ot  hau  hand :  haut^  which 
have  been  shewn  above  to  be  non-original  extensions  of  the 
earlier  Latin  form  *auy  and  the  examination  of  the  extensions  of 
Gk.  oly  viz.  ov-K  ol'x  o^-ic/  ov'Xh  will  be  deferred  to  the  latter  part 
of  the  present  essay  (§8,  pp.  61  sqq.),  where  the  various  Latin 

^  This,  with  the  omission  of  Ritschl's  '  et  fait  au^  is  the  reading  given  in 
Keil's  Grammatici  Latini,  vol.  VI,  p.  15, 11.  21,  22. 

'  Keil  gives  the  following  critical  note :  **  haud  oh  per  d  recte  scriHtur  M 
(=  Codex  Montepessulanus  306);  Mattd  doh  sic  recte  scriHtur  C  (=  Codex 
Bernensis  338) ;  hauddoto  sic  cUioqid  recte  scriHtur  B  (=  Codex  Bemensis  330) ; 
rectius  erat  hau  dohper  unum  d  recte  scriHtur,  nisi  potius  Aaud  aliter  scriptum 
erat,  quod  ex  Vergilii  versa  Aen.  Villi  65  Marius  Victorinus  p.  15,  24 
adscripsit/'  Lindsay,  The  Latin  Language,  ch.  II,  §136,  p.  122,  suggests  a 
new  reading :  haud  uolo. 

'Ritschl  (1.  c,  p.  ci)  prefers  to  read  "Aan  enim,  graeca  vox  ov,  </ littera  etc." 
Keil,  Gramm.  Lat.,  vol.  I,  p.  112, 1.  8,  and  Lindsay,  The  Latin  Language,  ch. 
X,  §18,  read  **ov  enim  Graeca  vox  </ littera,"  etc. 


Digitized  by  VjOOQ IC 


LATIN  HAUD  AND  GREEK  OY.  47 

and  Greek  extensions  of  the  forms  *au  :  o^,  together  with  some 
kindred  forms  in  the  same  or  in  some  other  Idg.  languages,  will 
be  dealt  with  m  detail. 

§3.  Older  expianaUans  of  hand :  ov  examined. 

Before  I  venture  to  put  forward  my  own  views  on  the  vocalism 
of  Gk.  o^  and  Lat  {K)au{d')y  it  may  be  well  to  examine  one  or 
two  of  the  older  etymologies  or  explanations  offered. 

Corssen,  Ausspr.  Vocalism.  und  Beton.  d.  Lat.  Spr.',  vol.  I 
(1868),  p.  205,  regards  the  au  of  {JC)au{dy  as  the  'Pronominal 
Particle'  au^  which  in  au-fero  au-fugio  has  the  meaning  'away, 
apart/  and  occurs  also,  according  to  Corssen,  1.  c,  in  Lat.  au-iem^ 
Osc.  aV'H,  Umbr.  o-ie^  Lat.  au-i  (see  id.  ib.,  p.  157).  This  au,  he 
says  further,  corresponds  to  Skr.  dva,  which  properly  means 
'down,  downwards,'  but  which,  he  adds  (on  the  authority  of 
Benfey,  Chrestom.  Gloss.,  p.  32  £),  in  composition  often  contains 
the  pure  negative  meaning  '  -less,  un-,  not'  On  the  ground,  too, 
that  Pott  (Etymol.  Forsch.,  part  II,  1836,  pp.  64,  134)  identified 
Gk.  o^jc  with  Skr.  avd-k,  Corssen  (1.  c.)  identifies  Greek  6{f  with 
the  same  Skr.  dva,*  which,  according  to  him,  has  the  form  au  in 
Lat.  h-au'd. 

Thus  Corssen  identifies  the  vocalism  of  Lat.  (Ji)au{d'),  au{ferd)y 
au{fugio\  au(Jtem)y  au[i\  Umbr.  a{te),  Osc.  av(tt),  Skr.  dva,  Gk. 
oi.  Leaving  Lat.  h^au-d  and  Gk.  ov  for  the  moment  out  of  the 
question,  it  may  be  remarked  that  only  on  one  condition — viz, 
that  Skr.  dva  and  the  au-  of  Lat.  au-fero  aufugio  ret)resent  an 
Idg.  ay,  (a  view  which  I  believe  to  be  right,  v.  infra,  p.  50,  n.  3,) 
— can  we  regard  as  correct  Corssen^s  identification  of  these  with 
Lat.  auitem),  tf»(0>  Umbr.  o{ie'),  Osc.  avift),  which  must  represent 
Idg.  *au  (see  Brugmann,  Grundriss,  I,  §§96,  97 ;  Lindsay,  The 
Lat.  Lang.,  ch.  X,  §§4,  p.  599,  5,  p.  601).  If,  therefore,  Skr.  dva, 
Lat.  au{Jerd)  au(Jugid)  au{ien£)  au(f),  Umbr.  o{te'),  Osc.  av{tt) 
are  all  to  be  identified  together  as  representing  Idg.  *au,  they 
must  all  be  separated  from  Gk.  ov  'not'  (which  cannot  represent 
Idg.  *a«f),  and  probably,  therefore,  also  from  Lat  *a«  'not.' 
Thus  Corssen's  identification  of  Gk.  ov  with  Skr.  dva  [based  on 

^  For  Corssen's  yiew  on  the  h  and  d  of  h-au-d  see  below,  §7  a  and  h,  pp.  55, 

59- 

'Bopp  (as  I  conclude  from  Brngmann,  Gr.  Gr.',  p.  236)  was  the  first  to 
identify  Gk.  ov  vith  Skr.  dva. 


Digitized  by 


Googte 


48  AMERICAN  JOURl^AL    OF  PHILOLOGY. 

Pott's  incorrect  identification  of  Gk.  oW  with  Skr.  avdk  (on  the 
latter  syllable  of  which  v.  infra,  §8,  p.  63)]  cannot  possibly  stand. 

Osthoff,  in  Hiibschmann's  Das  Idg.  Vocalsystem  (1885),  pp. 
190,  191,  regards  **Lat.  (Ji)aud\  Gk.  ov  (from  *oW)"  as  shewing 
different  ablaut-grades  of  the  same  root.  For  the  ablaut  '*Gk.  ov  : 
Lat.  a«"  Osthoff,  1.  c,  compares  Gk.  tAaxa  (Att.  Zvra  from  *aaro)  : 
Lat.  auris,  from  Idg.  if  aus-  'to  draw,  gather,  take,  obtain'  [Gk. 
a{^»  Lat.  haurio  (from  ^aus-id,  Osthoff,  Zur  Geschichte  des  Per- 
fects, p.  486)  O.Norse  ausa]^  the  'ear'  thus  meaning  'the  grasping 
organ' — middle  grade  of  Gk.  oZ^  in  Lesb.  nap-ava  'cheek'  and 
weak  grade  in  Avest.  uh-* 

Osthoff's  derivation  of  Gk.  ov  from  *ovd  seems  unlikely.  It  is 
much  more  probable  that  ov  was  the  earliest  Greek  form,  and  that 
the  forms  such  as  oid-€is  shewing  d  came  in  later  (cf.  Brugmann, 
Gr.  II,  §31  *ad  fin.).  With  respect  to  Osthoff's  view  that  Lat. 
hand  Gk.  ov  belong  to  the  Jf-series,  I  would  not  deny  that  d 
appears  beside  d  in  the  strong  grade  of  this  series,'  but  I  would 
raise  the  objection  that  there  is  (so  to  speak)  no  Indo-Germanic 
'peg'  on  which  to  hang  Lat.  Aaud  Gk.  ov,  thus  referred  to  the 
df-series. 

Victor  Henry,  in  M6m.  d.  1.  Soc.  d.  Ling.,  vol.  VI,  part  5  (1889), 
pp.  378  sqq.,  seeks  {umuccess/uify,  I  think)  to  justify  Bopp's  and 
Corssen's  above-mentioned  identification  of  Skr.  dva  with  Gk.  o£. 
He  observes  that  at  first  sight  the  disparity  o(  meaning  between 
dva  and  ov  is  difficult,  but  remarks  (on  p.  378)  that  it  is  possible 
to  see  in  5kr.  the  commencement  of  the  proceeding,  by  which 

^  It  should  be  mentioned  that  Havet,  in  M^m.  de  la  Soc.  d.  Ling,  de  Paris, 
vol.  VI,  part  I  (1885),  p.  18,  and  King  and  Cookson*  Sounds  and  Inflexions  in 
Greek  and  Latin,  chh.  V«  p.  86,  IX,  p.  187,  regard  Idg.  ^offs-  as  the  origin  of 
Lat.  auris  Gk.  ov; .  But  these  scholars  seem  certainly  to  be  mistaken  in  their 
view.  Cf.  Osthoff,  Perf.,  pp.  486  sqq.;  HObschmann,  Das  Idg.  Vocals.,  p.  159 ; 
Lindsay,  The  Latin  Language,  ch.  IV,  §31,  and  the  present  writer  in  his 
Essay  on  the  '  Establishment  and  Extension  of  the  Law  of  Thurneysen  and 
Havet/  Part  II  (Amer.  Joum.  of  Philology,  vol.  XVII,  part  2,  July,  1896).  §8, 
p.  194,  n.  3. 

'Till  recently  'understanding'  was  the  only  known  meaning  of  Avest.  uH- 
(cf.  Armen.  ui  'understanding/  regarded  as  a  borrowed  word  by  HQbschmann, 
Arm.  Stud.  I,  p.  47),  but  the  meaning  'ear*  has  been  established  by  the  new 
fragments  of  the  *  Nirangistan,*  vid.  Darmesteter,  Le  Zend-Avesta,  fragment 
vi,  verse  26,  p.  95  htfoHbya  uiibya  *  with  the  two  ears.' 

'  Cf.  the  above-mentioned  Essay  on  the  *  Establishment  and  Extension  of 
the  Law  of  Thurneysen  and  Havet,'  Part  I  (A.  J.  P..  vol.  XVI,  part  4,  Dec. 
1895).  §3  (p.  447.  note  x). 


Digitized  by  VjOOQ IC 


LATIN  MAUD  AND  GREEK  OT.  49 

dva^  ''qui  a  parfois  en  Sanscrit  le  sens  inversif,"  has  been  capable 
of  transformation  into  a  negative  particle.  But  to  pass  on  to 
what  is  perhaps  more  important,  his  attempted  explanation  of  the 
form  of  the  words : — he  suggests  that  dva  may  have  come  from  an 
Idg.  ^a^O'  [adding  that  Latin  is  not  against  this,  seeing  that  the 
shortened  form  au-  (au-fero^  etc.),  in  virtue  of  the  Law  of  Thum- 
eysen  and  Havet,  may  represent  Idg.  *<?^  just  as  well  as  Idg.  *a|^]. 
The  Old  Irish  preposition  6  ila  (sicO,  which  comes  from  Idg.  *^, 
would  correspond  to  Skr.  dva^  if  this  latter  be  rightly  derived 
from  Idg.  *^|f(^).'  According  to  Brugmann's  law  that  Idg.  d  in 
an  open  syllable  becomes  a  in  Skr.,  Victor  Henry  would  have 
expected  Idg.  *Guo  to  become  Skr.  *dvd,  and  not  Skr.  dva^ 
adding,  however,  that  this  rule  is  not  absolute,  being  violated 
notably  in  the  particles,  e.  g.  Skr.  dpa  :  Gk.  dn6 ;  he  further 
explains  dva  by  supposing  the  co-existence  of  two  original  forms, 
a  full  one  *w>  (or  *^>ud  or  *^>?^/),  and  a  shortened  one  *du  (with 
which  he  compares  Lat.  a«-),  the  former  yielding  Skr.  *dvd  (or 
*dvd),  the  latter  yielding  Skr.  *dv,  and  contamination  of  the  two 
producing  dvd.    So  far  as  Ionic  Attic  alone  is  concerned,  Victor 

^tia  (as  accentuated  by  Victor  Henry),  which  should  strictly  represent  lia 
(for  the  *  accent/  which  is  written  in  Gaelic,  is  really  no  accent  at  all,  but  only 
a  mark  of  quantity),  may  be  criticised  as  not  being  quite  an  exact  way  of 
representing  the  true  sound  of  the  word.  Some  diphthongs  in  Gaelic  may  be 
(i)  short,  (2)  long  with  respect  to  the  first  vowel,  (3)  long  with  respect  to  the 
second,  but  there  are  also  (4)  a  few  others,  which  are  always  long\  to  this 
latter  class  ua  belongs.  Diphthongs  of  this  latter  class  are  never  'accen- 
tuated' ;  thus  ila,  as  written  by  V.  Henry,  is  both  right  and  wrong — ^wrong 
with  respect  to  the  notation,  right  with  respect  to  the  quantity  (ua  being  a 
long  diphthong). 

The  form  o  belongs  to  Scots  (as  well  as  Irish)  Gaelic,  e.  g.  o  umtdgh  *  from 
prayer,'  St.  Luke  xxii.  45 ;  c  Ghalile  *"  from  Galilee,'  id.  xxiii.  5.  (This  o  of 
Scots  Gaelic  is  of  course  long,  as  in  Irish  Gaelic,  but  Scots  Gaelic  very 
rarely  makes  use  of  the  '  accent.')  The  form  ua,  however,  is,  so  far  as  I  know, 
peculiar  to  Irish,  and  even  there  is,  I  believe,  retained  only  in  the  preposi- 
tional pronouns  which  are  formed  with  this  preposition,  e.  g.  uaim  (:=ua-\-pte) 
'from  me,' fAzi/(=fAs4-/!K)  'from  thee'  [all  the  simple  prepositions  in  Irish 
being  thus  compounded  with  the  personal  pronouns,  an  odd  feature  common 
to  Gaelic  and  the  totally  unrelated  languages,  Hebrew  (e.  g.  Immanu-el 
'with-us  God')  and,  I  am  told,  Hungarian].  Oi  ua  \  can  find  no  trace  in 
Scots  Gaelic.  Curiously  enough,  in  O'Donovan's  Irish  Dictionary,  the  only 
form  given  of  the  pre[>osition  in  question  is  ua^  but  the  form  d  (=s  o)  is  now 
almost  invariably  written,  even  in  printing  old  texts  which  shew  ua. 

^See  infra,  §4.  p.  52,  n.  i. 


Digitized  by  VjOOQ IC 


50  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Henry  observes  that  ov  might  come  from  the  full  form  *of  ©,*  but 
that  as  this  would  not  suit  Lesb.,  Boeot.  and  Doric,  in  which  the 
said  *o/o  would  have  become  *«,  he  prefers  to  derive  ov  from  the 
shortened  form  *hf  (comparing  irdp  =  ita^y  av  =  dra,  etc.)i  which 
satisfies  all  the  phonetic  needs,  the  /  before  consonants  forming  a 
diphthong  with  the  preceding  vowel,  so  that,  e.  g.,  *6f  (jm/xi  (which, 
according  to  V.  Henry,  =  Skr.  dva  bhdtni')  became  od  <tniiu. 

On  Victor  Henry's  theory  the  following  remarks  may  be  made. 
He  is  surely  wrong  in  supposing  that  Brugmann's  law,  viz.  that 
"Idg.  d  in  open  syllables  became  a  in  the  Prim.  Aryan  period" 
(Brugmann,  Gr.  I,  §78),  is  intended  to  include  final  syllables,  for 
Brugmann  himself,  in  Gr.  HI,  §409,  regards  Skr.  sd  (:  Gk.  6)  sd-s 
as  the  Skr.  representative  of  Idg.  *sd  *sd'S,  and  in  Gr.  IV,  §1047, 
he  exemplifies  Idg.  "^-sd,  an  Idg.  personal  ending  of  the  2  sing, 
middle,  by  Avest.  bara-vha  (:  Gk.  ^pfo  xfUpov  ^-(^>co  i'<t>ipov) ; 
thus  we  should  have  expected  Victor  Henry's  postulated  Idg, 
*^^  to  yield  Skr.  *dvdj  and  not  (as  Victor  Henry  thinks)  Skr. 
*dvd.  To  return  to  the  main  question : — his  suggested  explanation, 
that  Skr.  dva  arose  by  the  contamination  of  two  original  Idg. 
forms  *(^d  (or  '^i>^d  or  *^)  and  *du,  does  not  seem  at  all 
satisfactory,  and  consequently  I  would  reject  the  view  that  Skr. 
dva  is  the  outcome  of  an  Idg.  form  containing  *^-.  In  the  next 
place,  although  I  believe  firmly  in  the  truth  of  the  Law  of 
Thumeysen  and  Havet,'  I  yet  think  that  the  au-  of  Lat.  au-fero 
au-fugio  is  not  an  instance  illustrative  of  this  Law.  Much  rather, 
Skr.  dva  and  the  an-  in  these  two  Latin  verbs  (compare  together 
Lat.  aU'/eroxSkx.  ava-bkr-  *aufero'),  together  with  O.CSl.  «- 
Pruss.  aU'  'ofT,  away'  (e.  g.  O.CSl.  u-myii  *to  wash  ofT,*  u-daii 
'to  give  away,  iwhwvoi  Bvyaripa,*  Pruss.  au-mu'Sna-n  'ablution'), 
should  all  be  regarded  as  representatives  of  the  Idg.  preposition 
*at4(€).  Cf.  Brugmann,  Gr.  I,  §100;  Stolz,  Lat.  Gr.'  (1890),  §50, 
p.  293,  and  in  the  Historische  Grammatik  der  Lateinischen 
Sprache,  vol.  I,  part  I  (1894),  hs^f  P*  154;  Lindsay,  The  Latin 
Language  (1894),  ch.  IX,  §12,  p.  576.'    From  this  Idg.  prepo- 

^  It  is  curious  to  compare  the  representation  in  Cyprian,  viz.  o-vo  (y.  Cauer, 
Delect.  Inscrr.  Graec.  474i  line  3). 

'See  the  above-mentioned  Essay  on  the  *  Establishment  and  Extension  of 
the  Law  of  Thumeysen  and  Havet*  (Amer.  Joum.  of  Philology,  vol.  XVI, 
part  4,  Dec.  1895.  and  vol.  XVII,  part  2,  July.  1896). 

»M.  Br^al,  in  Mem.  de  la  Soc.  de  Ling.,  vol.  V,  part  3  (1883),  pp.  197.  198, 
maintains  that  the  au-  oiau^fero  au-fugio  is  the  preposition  ab  or  d  (cf.  abstuii^ 
abhtum).    According  to  M.  Br^al,  *'  Vd  a  subi  la  diphthongaison  en  du  comme 


Digitized  by  VjOOQ IC 


LATIN  BAUD  AND  GREEK  OT.  5I 

sition  *aii{e)^  Gk.  ov  and  (in  my  opinion)  Lat  h-au-d  must  be 
entirely  separated. 

§4.  Reference  of  Lai.  *au  Gk,  ov  io  a  common  ground-form, 
namely,  Idg.  iauiosylkMc  *&u,  from  the  Idg.  nf  eu-  Ho  fail,  be 
deficient,  be  wanting^ 

The  views  hitherto  advanced  on  the  etymology  of  Gk.  ov  and 
Lat.  h'OU'd  have  thus  been  briefly  criticised.  None  of  them 
seems  at  all  satisfactory.  I  would  therefore  venture  to  suggest 
the  following  view,  which  has,  at  least,  the  advantage  of  attaching 
Gk.  ov  and  Lat.  h-au-d  to  an  established  Idg.  root  with  a  well- 
defined  meaning.  According  to  my  view,  Gk.  06  and  Lat 
{K)au{d)  are  identical  in  origin,  their  Idg.  ground-form  being 
tautosyllabic  *Su.  This  Idg.  ^D^  became  regularly  in  Greek  o^, 
in  Latin  first  *ou  and  then  at  a  later  date,  some  time  in  the  3d 
century  B.  C.,^  ^atu    Granting,  then,  that  Gk.  oCr  and  Lat.  ^au 

ZdrerUia  est  devenu  Laurentia"  He  relies  mainly  on  the  authority  of  Cicero, 
Orat.  XLVII,  §158:  "Qnid,  si  etiam  'abfugit*  turpe  visum  est  et  'abfer^  nol- 
uemnt,  {^aufugit'*  et)  *aufer'  maluerunt?  quae  praepositio  praeter  haec  duo 
verba  nuUo  alio  in  verbo  reperietur."  He  might  much  more  appropriately 
(from  his  own  point  of  view)  have  cited  Quintil.  I  5,  §69 :  "  Frequenter  autem 
praepositiones  quoque  compositio  ista  corrumpit :  inde  abstulit,  aufugit^  a$msit, 
cum  praepositio  sit  ad  sola."  Such  a  view,  however,  can  hardly  be  regarded 
as  correct ;  cf.  the  authorities  cited  in  the  text,  especially  Lindsay,  1.  c,  where, 
speaking  of  the  au-  in  au-fero  and  au-fugio^  he  writes:  "It  has  not  been 
produced  from  t^  by  any  phonetic  process,  but  represents  a  different  I.-£ur. 
preposition  *aw{/)  (O.Ind.  dzMi,  Pruss.  au-^  e.g.  O.Ind.  ava-bkr-  *  au-fero'), 
which  was  brought  into  requisition  in  these  compounds  before  an  initial/  to 
avoid  confusion  with  the  compounds  of  ad^  e.  g.  affero*^  With  this  explana- 
tion of  Lindsay,  I  would  agree  entirely,  save  in  one  point : — Surely  the  au- 
of  au'ferv  au-fugio  was  not  "  brought  into  requisition "  by  Latin  "  to  avoid 
confusion  with  the  compounds  of  a/,  e.  g.  affero"  but  rather  is  a  relic  pre- 
served from  Idg.  times  (cf.  Skr.  ava-bhr-^  quoted  above),  long  before  the  Idg. 
bh  of  the  two  Idg.  roots  bher-  and  bhe^^-  had  become  Latin  /  (/ero  /ugio). 
It  would  be  more  true,  therefore,  to  say  that  the  au-  of  au-fero  au-fugio  was 
maintained  in  Latin  "  to  avoid  confusion  with  the  compounds  of  a/,"  a  con- 
fusion which  would  have  ensued  if  au  had  been  exchanged  for  ab  (e.  g.  aufero 
exchanged  for  ab-fero,  whence,  of  course,  affero,  which  would  have  been 
ambiguous). 

^  See  the  above-mentioned  Essay  on  the  *  Establishment  and  Extension  of 
the  Law  of  Thumeysen  and  Havet,'  Part  I  (A.  J.  P.,  vol.  XVI,  part  4),  §3  ad 
fin.  (pp.  456  sq.),  Part  II  (A.  J.  P.,  vol.  XVII,  part  2),  §8  ad  fin.  (p.  195)  and 

§9  (pp.  195  sq.). 


Digitized  by  VjOOQ IC 


52  AMERICAN  JOURNAL    OF  PHILOLOGY. 

come  from  Idg.  tautosyllabic  *e>^,  we  have  next  to  ask :  What 
can  this  Idg.  "^ti^  mean  ?  So  far  as  I  am  aware,  no  one  has  yet 
suggested  any  Idg.  root  with  a  known  and  definite  meaning,  with 
which  to  connect  this  Idg.  "^du ;  but  there  exists  one  Idg.  root, 
the  meaning  of  which  is  well  established,  and  from  which  there 
are  numerous  derivatives,  a  root  which  exactly  suits  the  needs 
required  of  it  here,  both  as  to  form  and  meaning:  the  Idg.  V^' 
*to  fail,  be  deficient,  be  wanting.'  To  this  Idg.  V^-»  therefore,  I 
would  refer,  as  preserving  the  strong  grade  d  of  the  /-series,  Idg. 
tautosyllabic  *^>{^,  the  common  ground-form  of  Gk.  ov  and  Lat. 

From  this  derivation  we  can  easily  trace  the  development  of 
meaning.  In  Gk.  o\i  and  Lat.  h-au-d  the  idea  of  'want'  or 
'deficiency'  has  produced  the  purely  negative  meaning.  In  Old 
Irish  6ua  'away  from,'  if  connected  herewith  (as  is  quite  possible 
from  the  phonetic  point  of  view^,  the  meaning  has  further 
developed  through  the  idea  of  'absence'  or  'separation'  implied 
in  the  primitive  root' 

§5.  Other  derivatwes  of  the  Idg.  nj  e^-. 

The  meaning  of  this  Idg.  V^*  *^o  fail,  to  be  wanting,'  which 
is  thus  given  by  Brugmann,  Gr.  II,  §66,  p.  141  (Engl,  ed.),  §67, 
p.  153  (E.  E.),  and  Osthoff  in  Morph.  Untersuch.  IV,  p.  370,  is 
well  established  from  the  following  derivatives,  most  of  which  will 
be  found  in  Brugmann,  Gr.  I,  §63,  II,  §66,  p.  141  (E.  E.))  §67,  p. 
153.  §95»  P-  286,  III,  §175,  p.  25  (E.  E.),  IV,  §574,  and  in  Pick, 
Vergl.  Worterb.  d.  Idg.  Spr.*,  part  I,  p.  123,  s.  v.  "i/a  'mangeln, 

^  O.Ir.  ^  ua  can  represent  either  Idg.  tautosyllabic  *i^  or  *du  (v.  Brugmann, 
Gr.  I,  §§66,  82),  but  not  Idg.  tautosyllabic  *du  (v.  Brugmann,  Gr.  I,  §98). 

'  If  my  derivation  of  Gk.  ot>  Lat.  (h)au{d)  from  Idg.  tautosyllabic  *du  from 
Idg.  y^-'to  fail,  to  be  deficient,  to  be  wanting' — ^a  derivation  to  which  no 
exception  can  possibly  be  taken  from  the  point  of  view  of  the  meaning — is 
correct,  it  will  hardly  be  denied  that  this  is  an  example  very  strongly 
supporting  the  views  set  forth  in  the  above-mentioned  Essay  on  the  '  Establish- 
ment and  Extension  of  the  Law  of  Thumeysen  and  Havet.'  For  according 
to  no  system  of  ablaut  can  an  original  (i.  e.  Idg.)  d  appear  in  the  Idg.  /-series,  to 
which  the  Idg.  ye^-  (:  Gk.  ev-vi-c  *  bereft,  mulcted,'  etc.)  undoubtedly  belongs. 
See  HQbschmann,  Idg.  Vocalsystem  (1885) ;  Brugmann,  Gr.  I  (1886),  §309  and 
§§311-14;  Bartholomae  in  Bezzenberger's  BeitrSlge,  vol.  XVII  (1891),  pp.91 
sqq.  Cf.  also  P.  Giles,  Short  Manual  of  Comparative  Philology  (1895),  §§258- 
265,  and  the  Note  following  §265.  pp.  186-94. 


Digitized  by  VjOOQ IC 


LATIN  HAUD  AND  GREEK  OT.  53 

fehlen,' "  part  II,  p.  179,  s.  v.  "«  'mangeln,' "  part  III,  p.  542,  s.  v. 
''vd  'mangeln'": 

Avestic  root  fi^  'to  want,'  Uyamna  pple.  mid.  'wanting,  failing,' 
Una  'empty'  also  subst.  f.  'want'  Sanskrit  nnd-^  'defective, 
deficient  in  something,'  unay  'to  leave  (a  wish)  unfulfilled  (find-\^ 
based  on  Und-}  Armenian  unain  'empty.'*  Greek  c5-w-ff  'bereft, 
mulcted.'  Latin  vdnus,  vdcare, vdcuus}  Gothic*  v-an-s  'wanting, 
absent,  lacking,'  ^  v-an  n. '  want.'  O.H.G.  w-an '  wanting,  lacking.' 
English  wan-  'lacking,  without.'^ 

To  the  derivatives  just  given  may  be  added,  I.  Greek  ap-tv 
'without'  and  II.  (i)  Skr.  vd  Gk.  *-f€  Lat  -vi  'or.'  (2)  Skr.  va 
'as,  like,'  Gk.  *-f€  'as,  like  as,'  Lat.  *-z/i?  'as,  like  as,'  for  a  full 
discussion  of  which  see  the  paper  on  "Some  Sanskrit,  Greek  and 
Latm  Derivatives  of  the  Idg.  V^-  'to  fail,  to  be  deficient,  to  be 
wanting,'"  published  recently  in  Bezzenberger's  Beitrage,  vol. 
XXII,  3/4  (1896),  pp.  189-202. 

^Cf.  also,  in  addition  to  the  authorities  cited  in  the  text,  Grassmann, 
W6rterb.  zum  RV.,  s.  v.  (und),  col.  272. 

*and  is  apparently  not  used  independently  in  the  RV.  (vid.  Grassmann),  bat 
it  is  found  in  the  compound  dn-una  'not  defective,  perfect'  [cf.  also  dnuna- 
varcas  *  possessing  perfect  glory  (vdrcasy  Grassmann,  WOrterb.  zum  RV.,  col. 
61  and  272].  Una  is  found  independent  in  classical  Sanskrit,  e.  g.  (Raghu- 
▼anSa)  Onam  na  scUtvehf  adhiko  hahadhe  *  a  strong  one  amongst  animals  has  not 
hurt  a  weak  (or  inferior)  one.'  Una  is  frequently  used  to  form  phrases  of 
subtraction  in  the  numerals,  e.  g.  19  ikdnaviHatii  (from  thM-ana-ifpati}  *  a  score 
wanting  one')  and  unavtsatil,  57  tryundidHi}  (see  Whitney,  Skr.  Gr.,  §§477a, 
478b;  Brugmann,  Gr.  Ill,  §175,  p.  25,  E.  E.). 

'Cf.  also  Grassmann,  op.  cit.,  s.  y.  (und)^  col.  272. 

^Cf.  also  Httbschmann,  Arm.  Stud.  I  (1883),  pp.  47.  62. 

^On  these  Latin  words  cf.  also  the  above-mentioned  Essay  on  the  '  Establish- 
ment and  Extension  of  the  Law  of  Thumeysen  and  Havet,'  Part  II  (A.  J.  P., 
vol.  XVII,  part  2),  §6  (pp.  178  sqq.). 

'  With  the  above-mentioned  use  of  Skr.  Una^  to  form  phrases  of  subtraction 
in  the  numerals,  we  may  compare  the  similar  phenomenon  in  Gothic,  e.  g. 
2  Cor.  II,  24  fidvdr  Hguns  dinamma  vanans  'forty  save  one'  (cited  by  Brug- 
mann, Gr.  Ill,  §175.  P*  35,  E.  E.). 

'  Cf.  also  Osthoff  in  Morph.  Unters.  IV,  p.  375. 

'  E.  g.  wofUon  from  M.E.  watu  from  A.S.  Tvan  *  lacking,'  and  iowen  =  A.S. 
ttfgen,  past  pple.  of  tedn  'to  draw,  to  educate'  (v.  Skeat,  Concise  Etym.  Diet, 
of  the  Engl.  Lang.,  1887),  'lacking  or  without  education.'  Cf.  also  the  follow- 
ing, cited  by  R.  C.  Trench  (English  Past  and  Present,  Lect.  Ill,  p.  112,  and 
note  on  pp.  112,  113):  wanhope  'despair,*  wanthHft  'extravagance,'  wanluck 
'  misfortune,'  wanlust '  languor,'  Tvanwit  *  folly,'  wangrace  '  wickedness,'  wanfrust 
(Chaucer)  *  distrust.' 


Digitized  by  VjOOQ IC 


54  AMERICAN  JOURNAL   OF  PHILOLOGY. 

§6.  The  ablaut  "ou  :  cw"  in  Greek. 

The  ablaut  ov  :  rv  in  Greek  is  not  common ;  few  examples  are 
known ;  but  it  must  have  existed  originally  in  Prim.  Greek,  none 
the  less ;  and  an  isolated  word  like  ov  is  just  such  as  might  have 
been  expected  to  keep  its  original  Wocalismus/  untouched  by 
external  influences.  The  following^  are  examples  of  the  ablaut 
ov  :  cv  in  Greek :  , 

oirovdi;  :  cnrcvdw,' 
wXov-TO-ff  :  ij pl'e^-^ 

\oZ<r<ro¥^ :  Xcvo-o-w*  from  *Xev<c-i«  ij  leuq-  'lucere/^ 

u\rfkov6'fUv  f  :  root  iXtvd'  (fut.  Af  vo-o/MU,  cXcvoreoy).' 

In  0o6t :  Oidf  n\6ot :  irXco),  x^^^  -  X^^t  ^^^^  ^^^  sonantal  element 
became  at  an  early  period  consonantal.^^ 

*  I  omit  the  late  word  /xiOaiog  •  reddish/  which  Schleicher,  Comp.  67,  derived 
from  *i)Mu)c,  and  placed  beside  kpMu^  but  wrongly  so,  pohaiog  being  really  a 
borrowed  word,  Lat.  russus  russeus.    Cf.  G.  Meyer,  Gr.  Gr.',  §9,  p.  9. 

'Cf.  Habschmann,  Das  Idg.  Vocalsystem  (1885),  §165,  p.  116;  Brugmann, 
Gr.  I.  §80,  p.  72,  Engl.  ed. 

'Cf.  Brugmann,  Gr.  Gr.',  §70,  p.  96. 

*Cf.  Httbschmann,  Das  Idg.  Vocals.,  p.  116,  and  especially  J.  Schmidt, 
*  Asstmilationen  benachbarter  einander  nicht  berUhrender  Vocale,*  in  Kuhn's 
Zeitschr.  XXXII,  p.  325,  who  there  observes  that  only  one  0  in  anSTjcn^  can 
be  occasioned  by  ablaut.  A  comparison  of  andikai^  with  nk'kEvBoq  proves,  as 
Schmidt  rightly  says,  that  the  ablaut  syllable  is  that  containing  the  diphthong. 
Schmidt  considers  the  original  flexion  to  have  been  *a/ceAev^  nom.,  aiuikobBov 
gen.,  and  levelling  to  have  thence  ensued  in  the  historical  time. 

^hivaaw  *  the  pith  of  the  fir-tree,*  first  attested  by  Theophrastus,  H.  P.  3. 9.  7. 

•Cf.  G.  Meyer,  Gr.  Gr.«,  §9,  p.  9. 

^The  evof  AevK^  *  white'  cannot  be  original,  but  has  come  in  for  older  ov 
(*Aov/c<Jf ) ;  cf.  G.  Meyer.  Gr.  Gr.«,  §9,  p.  9 ;  Prellwitz,  Etym.  W6rterb.  d.  Griech. 
Spr.,  p.  185,  s.  V.  Xovffow;  and  Bartholomae  in  Bezz.  Beitr.  XVII,  p.  99. 

*  In  the  Idg.  perfect  the  ^grade  prevailed  in  the  2.  3.  sing,  indie,  act.  {oUjOa 
ol6e  J^fftid')  and,  according  to  the  view  of  most  linguists,  in  the  r.  sing,  as  well 
[oIcJo,  *Ke^ovya  (in  Fick,  Vergl.  WOrterb.*,  p.  89,  read  Hhlbouga  for  ^bMbauga, 
cf.  id.  ib.,  p.  xxxv),  from  J^bhiftq-  or  ^bhef^g-  'to  bend,  decline'],  although 
some  think  that  the  /-grade  prevailed  in  the  r.  sing.  (e.  g.  Tzit^euya).  On  this 
question  cf.  Brugmann,  Gr.  IV,  §843. 

*Cf.  HQbschmann,  Das  Idg.  Vocals.,  p.  116;  Brugmann,  Gr.  I,  §77.  p.  68 
(Engl.  ed.).  g8o.  p.  72  (Engl,  ed.),  IV.  §856;  id.  Gr.  Gr.«,  §9,  p.  26;  and  G. 
Meyer,  Gr.  Gr.«,  §552,  p.  484. 

»Cf.  G.  Meyer,  Gr.  Gr.«,  §9,  p.  9. 


Digitized  by  VjOOQ IC 


LATIN  MAUD  AND  GREEK  OT.  55 

B. — The  Extensions  of  Lat.  *««  and  of  Gk.  ov  'not.' 

Having  thus  examined  thoroughly  the  vocalism  of  Gk.  ov  and 
Lat.  *au  'not/  and  having  referred  them  to  the  common  Idg. 
ground-form  *di^,  representing  the  strong-grade  d  of  the  /-series, 
from  the  Idg.  kj €%-  'to  fail,  be  deficient,  be  wanting,'  we  will  now 
proceed  to  discuss  the  extensions  of  Lat.  *a«,  viz.  h-au^  A-au-d, 
A-au-i,  and  also  of  Gk.  ov,  viz.  ov-x«.  ohx,  ov-kI,  ov-x,  together  with 
some  other  kindred  forms  in  the  same  or  in  other  Idg.  languages. 

§7.  Laitn  h-au  h-au-d  h-au-i: — (a)  ike  h  of  h-au  A-au-d  A-au-t; 
(*)  iAedofh-aU'd;  (jc)  tAe  t  of  A-au-U 

(a)  TAe  A  of  A-au  A-au-d  A-au-L 

The  explanation  of  this  A-  is  no  easy  matter.  Various  expla- 
nations have  been  offered ;  but  none  seems  convincing : — 

Corssen,  Ausspr.  Vocalism.  und  Beton.  d.  Lat  Spr.',  vol.  I 
(1868),  p.  205,  regards  Aaud  as  a  compound  word,  consisting  of 
three  parts:  A-,  au,  d.  His  explanation  of  the  au-  as  the 
'Pronominal  Particle'  meaning  'away,  apart,'  considered  above 
(§3,  pp.  47  sqq.),  did  not  appear  satisfactory.  For  his  explanation 
of  the  d,  see  below  (p.  59).  The  A-,  according  to  him,  is  the 
remnant  of  the  demonstrative  pronominal  stem  Ao-  of  Ai-c  Aae-c 
Ao-Cy  which  appears  blunted  to  A-  in  A-or-nu-s  from  ^Ao-jpr-nU'S 
'this  year's.'*  Thus,  according  to  his  theory,  A-au  must  = 
literally  'this  away/  and  would  be  a  formation  similar  to  Skr.  so 
(contracted  from  Skr.  sd  u)  Old  Pers.  Aauv  Gk.  o$[-ro(]  from  Idg. 
*so  (demonstrative  pronoun)  +«  (the  particle  of  place,  meaning 
'here'  or  'there').'  Thus,  while  Idg.  *sou  strictly  means  'this 
man  here,'  Lat.  ^Ao-au,  according  to  Corssen's  explanation,  would 
mean  just  the  opposite,  'this  away,  this  not  here.'  Although 
Corssen's  etymology  of  the  au  did  not  seem  satisfactory,  this 
need  not  really  affect  our  view  concerning  his  explanation  of  the 
A',  which,  if  suitable  to  the  au  as  derived  by  Corssen,  might  be 
equally  suitable  to  the  au  as  derived  in  the  present  paper  (v. 
supra,  §4,  p.  51).  That  Corssen's  explanation  of  the  A-  is  possible 
may  be  admitted ;  but  at  the  same  time  it  can  hardly  be  considered 
satisfactory. 

^  This  Latin  slem  A^-  is  generally  extended  by  'Ce  {kic  =  ^ho-uee)^  but  ho-die 
ho-rsum  and  Faltsc.  hei  he-sz* heic  hie'  (Schneider,  Dial.  Ital.  Exx.  Select.,  to]. 
I,  part  2,  p.  106,  Nos.  ao,  3ia,  32)  are  instances  of  the  nnextended  root  (▼.  Stolz, 
Lat.  Gr.«.  §90,  p.  347). 

''Cf.  Bmgmann,  Gr.  I,  §603  ad  fin.,  II,  §4,  p.  9,  £.  E.,  Ill,  $414,  p.  337,  E.  E. 


Digitized  by  VjOOQ IC 


56  AMERICAN  JOURNAL   OF  PHILOLOGY. 

Osthoff,  Zur  Geschichte  des  Perfects  (1884),  Excurs  III,  on 
*  Indog.  sa^'y  aus-,  eus'  im  griechischen/  has  on  pp.  491,  492 
worked  out  an  ingenious  theory  to  explain  (i)  the  incorrect 
presence  of  h  in  A-aurid  (for  *auS'idt  Osthoff,  ib.,  p.  486)  :  Gk. 
aC»t  Old  Norse  aus-a  'to  draw,  obtain/  A-umerus :  the  correct 
umertiSy  A-erus  :  the  correct  erusy  A-dldre  A-dlUus :  the  correct 
alum  dlarCy  and  (2)  the  incorrect  absence  of  A  in  anser :  Skr. 
Af^sd'  Gk.  -xfivy  arena  :  the  correct  but  less  frequent  Aarena  (=a 
Sab.  fasina\  olus  :  the  correct  Aolus.  Osthoff,  it  is  true,  makes 
here  no  mention  of  Aaud,  but  inasmuch  as  this  theory  of  Osthoff 
is  cited  in  Hiibschmann,  Das  Idg.  Vocalsystem  (1885),  p.  191,  in 
explanation  of  the  A-  oiA-aud,  we  must  briefly  review  the  theory 
to  see  whether  it  can  be  applied  to  A-aud.  Osthoff  explains  the 
incorrect  presence  and  absence  of  the  A  as  the  outcome  of 
sentence-doublets.  He  suggests  that,  e.  g.,  Lat  ex  Aarena^  in 
Aarena  came  to  be  pronounced  ec  s-arendy  i  n-arend  (cf.  French 
I'd  Z'am^=iUs  Aommes),  the  spiritus  asper  having  disappeared 
after  the  new  consonant  thus  prefixed.  But  elsewhere,  he  says, 
and  especially  at  the  beginning  of  a  sentence,  the  spiritus  asper  • 
will  have  maintained  itself  unweakened,  so  that  in  such  a  position 
only  Aarena  *Aanser  Aolus  were  pronounced.  But  inasmuch  as 
in  phrases  like  ec  s-arend,  i  n-arend  the  A,  though  no  longer 
pronounced,  continued  still  to  be  written,  confusion  arose,  followed 
by  the  not  unfrequent  wrong  insertion  of  A ;  the  pronunciation 
was  not  altered  when  the  A  was  wrongly  written  in  cases  like 
exAdldre  tnhdldre,  just  as  in  Aumero. 

This  theory  may  or  may  not  give  the  correct  explanation  of 
the  words  discussed,  but  it  is  difficult  to  consider  it  applicable  to 
the  negative  A-aud  'not,'  which  can  but  rarely  have  been  preceded 
by  ex  or  in. 

Bugge,  Beitrage  zur  Erforschung  der  Etruskischen  Sprache,  in 
Bezz.  Beitr.  X  (1886),  pp.  75,  76,  attempts  to  prove  that  the  A  of 
the  Latin  demonstrative  stem  Ao-  was  merely  'vorgeschoben.^ 
Latin,  he  says,  does  not  know  an  inflected  demonstrative  stem  o-^ 
but  rather  Ao-  :  Aoc  (from  *Aod'ce')y  Aunc  and  so  forth.  But,  he 
continues,  no  other  Idg.  language  shews  an  inflected  stem  Ao-  or 
£^Ao'y  and  it  is  well  known  that  Latin  sometimes  shews  a  'vorge- 
schoben  A'  (e.  g.  A-auriOy  where  the  related  Etruscan  words 
Ausmana  Auzmaire  likewise  shew  A),  hence  he  regards  the  A  of 
the  Latin  demonstrative  stem  Ao-  as  'vorgeschoben,'  and  iden- 
tifies formally  *Aain  (contained  in  Lat  Aunc)  with  Etrusc  am,  an. 


Digitized  by  VjOOQ IC 


LATIN  HAUD  AND  GREEK  OT.  57 

Bu|^ge  thinks  that  the  older  unaspirated  forms  *om  *am  ^od  were 
aspirated  because  they  were  felt  to  be  too  unsubstantial,  when  the 
pronoun  was  in  an  accented  position.  He  accordingly  felt  himself 
able  to  identify  Lat  hd\rdie\  with  the  synonymous  Skr.  a\jdyd\. 
In  his  discussion  of  Lat.  haud  itself,  he  appears  to  call  it  'a  word 
of  pronominal  origin'  and  says  that  "if  the  h  of  hand  is  of  the 
same  origin  as  the  h  of  haurio,  hic^  hodic^  then  haiid^  as  has  long 
been  conjectured,  may  be  related  to  Gk.  ov." 

It  is  difficult  to  see  whether  Bugge  means  that  the  h  of  h-aiid 
is  the  'vorgeschoben  A'  of  the  demonstrative  stem  h-o-  (according 
to  his  own  explanation  of  the  latter),  or  that  it  is  a  '  vorgeschoben 
A,'  as  in  h-auriOj  quite  unconnected  with  the  demonstrative  stem 
ho-.  He  apparently  means  the  former,  in  which  case  his  expla- 
nation is  much  the  same  as  that  of  Corssen,  given  above,  and  no 
more  satisfactory.  Moreover,  Bugge  is,  I  think,  quite  wrong  in 
regarding  the  h-  of  the  demonstrative  stem  ho-  as  a  '  vorgeschoben 
k'J  ^  Much  rather  does  the  Lat.  ho-  represent  an  Idg.  stem  ^gho- 
(v.  infra,  §8,  p.  62). 

Lindsay,  The  Latin  Language  (1894),  ch.  X,  §18,  suggests  that 
the  initial  h-  of  haud  must  have  been  used  as  a  distinguishing 
mark  to  differentiate  the  word  from  aui,  comparing  M.  Valerius 
Probus  (temp.  Nero  and  the  Flavian  Caesars),  Inst.  Art.,  in  Keil's 
Gramm.  Lat.,  vol.  IV,  p.  145,  1.  9:  "*a«/,'  si  sine  adspiratione 
scribatur  et  in  /  litteram  exeat,  erit  coniunctio;  si  vero  *Aaiid* 
cum  adspiratione  scribatur  et  in  ^litteram  exeat,  erit  adverbium." 
We  might  compare  also  Cledonius  (flor.  5th  century  A.  D.)  in 
Keil,  op.  cit.,  vol.  V,  p.  74, 1.  28 :  "  aui  si  sine  A  aspiratione  et  in 
/  exit,  coniunctio  est;  si  vero  cum  A  aspiratione  et  in  d  exit, 
adverbium  " ;  and  Cassiodorus  (circ.  500  A.  D.),  De  Orthographia, 
in  Keil,  op.  cit.,  vol.  VII,  p.  158, 1.  20:  "Haud,  quando  adverbium 
est  negandi,  d  littera  cerminatur  et  adspiratur  in  capite ;  quando 
autem  coniunctio  disiunctiva  est,  per  /  litteram  sine  adspiratione 

'  It  is  quite  trne  that  we  find  unaspirated  forms  on  inscriptions  fairly  often, 
e.  g.  ic  in  C.  I.  L.  Ill  809;  IV  1321  ;  V  6400;  VIII  5257,  5501,  8297,  9344, 
9638. 9768 ;  X  7123, 7172. 7763.  «^  XII  S70.  oc  V  4488 ;  VIII  9192 ;  X  1541. 
uc  (=  oc)  XII  t2i47.  ««^  V  tl642 ;  IX  306  (bis) ;  XII  2584.  uius  V  +1741 ; 
VIII  9200;  X  2184.  4410.  «'/  I  1297.  aduc  V  t6244.  But  these  unaspirated 
forms  (which  do  not  occur  on  any  early  inscription)  are  not  the  survival  of  an 
original  unaspirated  demonstrative  stem  ;  they  are  due  merely  to  the  incorrect 
'  dropping  of  ^/  as  is  proved  by  the  fact  that  the  aspirated,  not  the  unaspirated, 
forms  occur  in  the  earliest  inscriptions,  e.  g.  hofu  and  Aic  in  C.  I.  L.  I  32  (of 
353  B.  C.  circ.)- 


Digitized  by  VjOOQ IC 


S8  AMERICAN  JOURNAL    OF  PHILOLOGY. 

scribitur."  Albinus  Magister,  in  Keil,  op.  cit.,  vol.  VII,  p.  303, 
1.  3,  uses  identically  the  same  words  as  Cassiodorus,  only  differing 
from  them  by  the  insertion  of  the  word  *  auV  before  *  per  /  litteram.' 

We  can  scarcely  accept  Lindsay's  explanation  of  the  h-  of 
h-aud.  The  grammarians  are  merely  stating  linguistic  facts,  and 
not  advancing  theories.  Had  the  usual  form  of  the  word  been 
hauiy  the  aspiration  might  in  that  case  have  been  used,  as  Lindsay 
suggests,  to  distinguish  h-aui  'not'  from  aut  'or.'  But  this  was 
not  the  case,  hatd  is  very  rare  on  inscriptions,  and  quite  late,  the 
only  two  inscriptions,  on  which  I  have  been  able  to  discover  it, 
being  respectively  *aevi  Antoniani'  and  *ex  titulis  Christianorum ' 
(v.  supra,  §2,  p.  43).  The  aspiration  is  already  on  the  earlier 
and  more  usual  hand  and  on  the  still  earlier  form  hau  (v.  supra, 
§2).  Hence  we  must  find  an  explanation  of  the  h-  which  will 
suit  the  earlier  forms  hau,  haud,  irrespective  of  the  later  haui. 

The  only  ancient  grammarian,  who  did  not  confine  himself 
(like  those  just  quoted)  to  the  statement  of  the  fact  that  ''aid 
'or'  is  a  conjunction,  hand  'not'  an  adverb"  or  the  like,  is  Marius 
Victorinus,  whose  remark  (quoted  above,  §2,  p.  45)  is :  ''Hau  ad- 
verbium  est  negandi  et  significat  idem  quod  apud  Graecos  ov  [et 
fuit  au\ :  sed  ab  antiquis  cum  adspiratione,  ut  alia  quoque  verba, 
dictum  est."  Marius  Victorinus  would  thus  seem  to  regard  the 
k'  of  h-aud  as  a  mere  '  vorgeschoben  A.' 

Instances  of  h  'vorgeschoben'  are  numerous  on  inscriptions, 
e.  g.: 

hcLegregius  in  C.  I.  L.  V  ti709 ;  homnium  XIV  3323 ;  hetema 
V  ti72o;  hegit  V  7647;  hardo  IX  5577,  X  477;  haliquit  (= 
aliquid)  XII  915;  Hillyricus  V  3620;  hac  III  5839,  IX  5961, 
X  7995;  his  XIV  497;  heius  III  3917,  VIII  3520;  hiKOic  XII 
915;  have  IV  1983,  2148. 

Compare  also  Catullus'  poem  (No.  84)  on  Arrius,  whom  he 
represents  as  speaking  hinsidias  and  Hionios  instead  of  insidias 
and  Ionics.  To  say  nothing  of  inscriptions,  this  poem  alone  is 
sufficient  to  prove  the  fluctuation  of  A  in  Latin  as  early  as  the  first 
century  B.  C;  while  a  century  later  Quintilian  (I  6,  21)  laughs  at 
those,  as  affected,  who  greet  one  another  with  ave  instead  oih-avi 
on  account  of  the  derivation  from  avere  (c£  Blass,  Ausspr.  d. 
Griech.',  §25,  and  Corssen,  Ausspr.  Voc.  und  Beton.  d.  Lat. 
Spr.*,  vol.  I,  p.  104).* 

^  Both  forms  are  found  on  inscriptions :  have  (v.  snpra  in  text) ;  ave  in  C.  I.  L. 
XIV  1473,  aveu  IV  2071. 


Digitized  by  VjOOQ IC 


LATIN  HAUD  AND  GREEK  OT.  59 

For  the  converse  we  may  compare  iCy  etc.  (quoted  above,  p.  57, 
n.  i),  and  also  arrespex^  in  C.  I.  L.  I  1216;  erceiscunda  I  205, 
part  2,  55  (48  B.  C);  Iriius  I  625  (42  B.  C  circa),  IX  3771; 
OraHa*  1 924 ;  osHa  (for  hosiia)  I  819  ;  Osti(Jius)  1 1 170 ;  Ypsaeus 
I  467  bis  (57  B.  C.  circa);  also  numerous  instances  of  the  common 
verb  habeo  docked  of  its  h,  e.g.  abes  V  1712;  abis  (^=  abes) 
VIII  9277;  abeas  XII  915;  abiai  IV  538;  abeto  IV  2013;  abere 
V  4488,  X  1365,  4539,  XIV  3323;  abebat  IX  2893;  «*««"^  V  914, 
1707,  XII  230;  abuis(i)e  IV  3121. 

This  extreme  fluctuation  in  the  matter  of  the  aspirate,  which 
became  quite  frequent  in  the  second  half  of  the  second  century 
A.  D.  (v.  Blass,  1.  c;  Corssen,  op.  cit.,  vol.  I,  p.  no;  Seelmann, 
Ausspr.  d.  Lat.,  p.  265  f.),  may  very  well  have  arisen  early.  The 
h'  of  h-au  h-aud  will,  I  think,  be  best  explained  as  a  'vorge- 
schoben  h*  which  (as  in  h-aurio,  v.  supra),  when  once  prefixed, 
was  always  retained.  Nor  is  it  unlikely  that  the  prefixing  of  this 
(etymologically  incorrect)  h  was  partly  due  to  a  desire  for 
increased  emphasis  on  a  word  which,  as  directly  negativing  the 
sentence  in  which  it  stood,  or  the  word  with  which  it  was 
connected,  was  one  of  emphatic  importance.' 

(*)  The  d  of  h-au'd. 

Corssen,  Ausspr.  Voc.  und  Beton.  d.  Lat.  Spr.',  vol.  I  (1868), 
p.  205,  regards  the  d  of  hau-d  as  the  remains  of  the  same  -de 
which  he  sees  preserved  in  quam-de,  un-dey  in-de,  ex-in-de^ 
de-in-de^  pro-in-de,  and  which  he  translates  *even,  precisely.' 
He  holds  that,  whereas  in  ex-in,  de-in,  pro-in  the  -de  has  again 
disappeared,  the  ending  e  having  first  fallen  off,  and  the  (thus 
final)  d  disappearing  thereafter,  in  haud  the  influence  of  the 
preceding  vowel  preserved  the  d  after  the  disappearance  of  the 
final  e.  Corssen's  explanation  of  h-au  has  been  given  above 
(§3i  P-  47»  §7.  P«  55)-    The  whole  word  h-au-d,  therefore,  accord- 

^I  quote  this  word  chiefly  in  order  to  point  out  that  C.  I.  L.  I  1 216  is  an 
incorrect  reference  for  arrespex. 

*  Oraiia  for  Horatia  beside  Praenestine  Foratia  [Schneider,  Dial.  Ital.  Exx. 
Select.,  No.  200  (=:  Gamurrini,  add.  1881,  No.  2354)]. 

'  It  is  true  that  (so  far  as  I  have  been  able  to  examine  the  question)  *  vorge- 
schoben  A*  is  at  least  uncommon  on  inscriptions  prior  to  the  Hannibalic  war 
(no  instance  is  found  in  C.  I.  L.  I  1-195,  which  consists  of  *  Inscriptiones 
▼etustissimae  Bello  Hannibalico  quae  videntur  antiquiores'),  but  this  does  not 
in  the  least  affect  my  explanation  of  the  h-  in  h-au  h-aud^  for  neither  does  any 
example  of  hau  haud  occur  prior  to  the  Hannibalic  war  [cf.  supra,  §2  (a),  p.  43]. 


Digitized  by  VjOOQ IC 


60  AMERICAN  JOURNAL    OF  PHILOLOGY. 

ing  to  Corssen's  theory,  should  strictly  mean  'this  away  pre- 
cisely': not  a  very  convincing  explanation. 

Osthoff  in  Hiibschmann,  Das  Idg.  Vocals.  (1885),  p.  191, 
deriving  Gk.  ov  from  earlier  *ovd,  would  apparently  regard  the  d 
of  AaU'daiS  pre- Latin;  but  his  derivation  of  Gk.  ov,  on  which  his 
explanation  of  the  d  of  Aau-d  rests,  would  appear  to  be  incorrect 
(see  supra,  §3,  p.  48). 

Of  the  ancient  grammarians,  Marius  Victorinus  (quoted  above, 
§2,  p.  45)  remarks :  "  adiecta  (est)  d  littera,  quam  plerisque  verbis 
(antiqui)  adiciebant";  and  Charisius  (quoted  above,  §2,  p.  46), 
after  a  brief  discussion  of  the  particle  sed,  continues :  *'I/aud  simi- 
liter d  littera  terminatur :  au  enim  , . .  d  littera  termina[ri  apud 
antiquos]  coepit  quibus  mos  erat  d  litt[eram  omnibus]  paene 
vocibus  vocali  littera  finitis  adiungere,  ut  quo  ted  hoc  noctis  \dicam 
pro']Jicisci  forasJ*  The  words  of  these  grammarians  can  hardly 
be  termed  scientific,  but  I  think  that  their  explanation  of  the  d  of 
haU'd  lies  nearest  to  the  truth.  It  seems  most  probable  that  the 
negative  particle  hau  became  hatid  on  analogy  of  the  adversative 
particle  sed,  aided  probably  by  the  extension  of  the  -dy  the  abla- 
tival  termination  of  the  ^-stem-nouns  and  of  the  pronouns,  far 
beyond  its  original  limits^  (cf.  Brugmann,  Gr.  Ill,  §240,  pp.  133 
sq.,  §243»  PP-  139  sqq.,  §442»  part  2,  §444). 

(c)  The  i  of  h-au't. 

The  /  of  haui  affords  no  difficulty;  we  find  d  and  /  very 
frequently  interchanged  on  inscriptions,  e.  g.: 

iioxdx 

apui  C.  I.  L.  I  206  (15, 34, 120) ;  818 ;  II  1963  (2,  20)';  1964  (I 
13,  17,  19;  IV  35);  VIII  619,  2634,4238;  IX  259,  339;  X  3334 
(10);  XIV  474  iguaier),  1597, 1661. 

at  (for  ad^  I  1252;  III  633  (I  i,  7,  13.  15);  IV  1880,  2013;  V 

1469,  3408,  8003;  VII  1310;  VIII 284,  1557;  IX  2893, 3314; 

X  787,  3147.  6565 ;  XII  5961  (5  b) ;  XIV  78,  380,  527. 

aihuc  VIII  9624. 

aliut  II 1964  (II  68 ;  IV  6) ;  V  532  (I  23),  1 102 ;  VIII  212  (36) ; 
X  4787,  4842  (11) ;  XIV  586,  1828  a. 

^V875(6);  VIII  2728, 4055;  IX 136;  X  2780, 7852  (3);  XIV 
2112  [(I  12)  (a.  /j<5)],  2795  [(13)  (a.  140)'],  3679  [a.  127]. 

>  We  actually  find  advorsus  tad  in  C.  I.  L.  I  Z96,  35. 
^apudC,  I.  L.  II 1963  (3, 19). 


Digitized  by  VjOOQ IC 


LATIN  HAUD  AND  GREEK  OT.  6l 

quit  IV  1547  a ;  V  3415 ;  VIII  212  (49),  2532  B,  a,  6 ;  IX  3161, 
5860;  X  761 ;  XII  915;  XIV  1874,  3956. 

aliquit  IX  5860,  {haliquit^  XII  915. 

quot  (for  qiiod)  II 144,  462,  813, 1120 ;  IV  i860 ;  V  2090,  3221 ; 
VIII  212  (37),  2728;  IX  2164  add.,  2475;  XII  729,  4326;  XIV 

I357>  1731,  3014.  3435- 

set  III  847  (al.);  IV  1516,  2400;  V  5049  (5);  VIII  403,  434, 
I557>  95191  10570;  IX  1164,  3337;  X  2496,  3334  (19),  5429  (12), 
7024  (I  6) ;  XII  743,  t575o;  XIV  166,  480,  914. 

rffor/: 

quod  (for  quof)  1 1016;  II  1964  (I  2, 10;  II  59) ;  IX  2827  (23). 

quodquod  annis  V  7450. 

quodannis  V  4410,  4448. 

W VIII  4770;  edYill  ♦427,  141 1. 

adqtu  (for  aique^)  VIII  828,  1027,  1179  bis,  2530;  IX  1588, 
1685,  2974;  XII  894,  2228,  3619;  XIV  126,  1826,  2046,  2919  bis. 

Also  not  unfrequently  we  find  such  interchange  in  the  termi- 
nations of  the  verb,  e.  g.  asied,  siedy  feced  in  the  old  Dvenos 
inscription,*  beside  mit{i)at  in  the  same  inscription ;  cf.  9\sofecid 
in  C.  I.  L.  I  54.  V  1870,  VIII  3028,  XIV  41 12;  diced  IV  1700; 
rogtid  IV  2388 ;  liquid  V  7570. 

To  this  great  fluctuation  between  /  and  d  (aided  very  possibly 
by  the  analogy  of  the  form  aui  *  or')  we  ought,  I  think,  certainly 
to  ascribe  the  form  haul  beside  haud, 

§8.  Gk.  ow-x*  ol'x  ov'u  ov'K : — (a)  the  ori^n  of  these  extensions 
[foprether  with  an  examination  of  Skr.  ht  -hi,  AvesU  zi  zi,  Lot. 
hie)  ;  {b)  the  accent  of  Gk.  w-xl  ov-kI  beside  that  of  ?-xt  voi-x*. 

(a)  The  origin  of  the  extensions  -x*  'x  -kI  -k. 
(l)  ov-x*  ov-ict. 
There  is,  I  think,  no  doubt  whatever  that  ot-xl  and  ov-m  must 
be  entirely  separated  from  one  another.    Roscher  indeed,  in 

^The  ultimate  deriYation  of  Lat.  atque  from  earlier  Lat.  ad-{-qtte  (cf.  P. 
Giles,  Short  Manual  of  Comparative  Philology,  §244,  p.  180)  does  not  of  course 
affect  our  point,  adque  in  the  examples  cited  is  not  the  old  form  itself  (a^-^ 
qtu)  preserved,  but  the  new  form  (aique)  altered.  The  form  atque  may  be 
found  in  C.  I.  L.  33  ('end  of  6th  century  U.  C./  Ritschl)  and  thrice  in  C.  I.  L. 
196  (*  supposed  to  be  . . .  568  A.  U.  C./  Roby). 

*  Given  in  Zvetaieff,  Inscrr.  Ital.  Inf.  Dial  (1886),  p.  80. 


Digitized  by  VjOOQ IC 


62  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Curtius  Stud.  Ill  (1870),  p.  144,  endeavoured  to  prove  '*dass  in 
alien  vier  Worten  (i.  e.  ^x^  ^^^-X^  ^X^  ix')  "X*  ^^^  urspriinglichem 
-Ki  durch  Aspiration  entstanden  ist,  dass  wir  also  von  der  Tenuis 
ausgehen  miissen,  wenn  wir  diese  Formen  erklaren  woUen";  but 
such  a  suggestion,  as  Osthoff  has  clearly  shewn  in  Morph. 
Unters.  IV  (1881),  pp.  239  ft,  "fiihrt  lautgesetzlich  nicht  zum 
befriedigenden  ziele." 

ov'xL — The  'Xi  of  ov'Xh  as  also  that  of  val-xh  is  almost  certainly 
to  be  identified  with  Skr.  hi  *for.'  This  view,  first  advanced  by 
Pott  (Wurzelworterb.  I  i,  567),  and  accepted  by  OsthoflT  (1.  c.)i 
Victor  Henry  (in  M6m.  de  la  Soc.  de  Ling.,  vol.  VI,  part  5,  1889, 
p.  379)  and  Per  Persson  (in  Idg.  Forsch.,  vol.  II,  1893,  p.  247), 
seems  beyond  all  reasonable  doubt  to  be  correct. 

The  common  Idg.  ground-form  of  this  Gk.  -x/  Skr.  kt,  beside 
which  latter  we  find  also  Skr.  -ki,^  is  *gh%  as  given  by  Osthoff, 
1.  c,  i.  e.  *gM.  This  is  proved  to  be  correct  by  the  Avestic  zl, 
beside  which  we  find  also  Avest.  zl  'for."  Skr.  A/ and  Gk.  -xh 
taken  by  themselves,  might  quite  regularly  be  derived  from  either 
Idg.  *g/il  (cf.  Brugmann,  Gr.  I,  §§386,  405)  or  Idg.  *^Ai  (cf. 
Brugmann,  Gr.  I,  §§425,  445,  454).  but  if  (as  seems  undoubtedly 
right)  they  are  to  be  identified  with  Avest.  zi,  whose  z  can  only 
represent  Idg.  *^A  (cf.  Brugmann,  Gr.  I,  §405,  and  also  §452  on 
Idg.  flA  as  represented  in  Iranian),  it  follows  that  *2^l  must  be 
set  up  as  the  original  Idg.  form  (cf.  also  infra,  p.  65,  note  2,  on 
OldPers.  *fl?/*forO. 

The  meaning  of  this  Idg.  *gXi  (Gk.  -x*  Skr.  At  Avest.  zt)  may 
be  fixed  with  tolerable  certainty  by  a  comparison  of  the  Latin 
demonstrative  pronominal  stem  *kO',  in  Lat.  ktc  (from  earlier 
Lat.  *Ao't'Ce,  the  -ce  of  which  comes  from  the  Idg.  demonstr. 
pronom.  stem  ^ko-),  etc.,  which  can  come  quite  regularly  from 

^Beside  Skr.  naA{  (from  nd-]^M)  we  find  nahi  before  nti  in  Rigveda  167,  9; 
314,  4;  623,  13;  also  Pr&tis.  443.  483  [vid.  Grassmann,  W5rterb.  zum  Rigv. 
(1873),  s.  V.  nah^'].  Is  the  f  of  this  Skr.  -hi  to  be  compared  to  the  I  of  Avestic 
If,  or  is  it  due  merely  to  metrical  reasons  ?  We  find  also  naht  nti  in  RV.  i.  80, 
15 ;  vi.  27,  3 ;  naht  nA  in  RV.  viii.  3I,  7  (v.  Osthoff  in  Morph.  Untersuch.,  vol. 
IV.  p.  240). 

'The  I  of  Avestic  «l  is  curious;  beside  it  we  find  once  sffor,'  Yt.  xiv.  12 
(v.  Justi,  Handb.  d.  Zendspr.,  125  b),  also  -«f  in ya-Mi  •  if,  whether'  from  *yad'Mi 
seen  in  yatUca  *and  if,'  Yt.  xxiv.  47  ("lies  yitica"!  says  Justi,  op.  cit.,  s.  v. 
yasifa), yf'Mi  * \{,  whether'  for  regular  ya-Mi  [v.  Osthoff  in  Morph.  Unters.,  vol. 
IV  (1881),  p.  240]. 


Digitized  by  VjOOQ IC 


LATIN  HAUD  AND  GREEK  OT.  63 

Idg*  *ghO'}  The  relation  between  Idg.  ^gho-  and  Idg.  ^ghi 
would  be  the  same  as  that  between  Idg.  *kO'  and  Idg.  *ku 
(Brugmann,  Gr.  Ill,  §409,  p.  329),  and  that  between  Idg.  *qo- 
and  Idg.  *qu  (Brugmann,  Gr.  Ill,  §411,  p.  333).  Thus  Idg. 
"^gho-  (:  Lat.  stem  ho-^  *ghu  (Gk.  -X4  Skr.  ht  Avest.  zt)  would  be 
a  demonstrative  pronominal  stem,  identical  in  meaning  with  Idg. 
*kO'  *kt'  *this.'  The  development  of  the  meaning  *for'  in  Skr. 
and  Avest  is  easily  traced:  Idg.  *^Ai  would  originally  mean 
'this'  (the  meaning  perhaps  still  discernible  in  the  -xi  of  Gk. 
ov'xOt  ^"^  might  h^ve  been  used  in  answers  to  a  question, 
as  we  often  say  'Just  this,*  'Just  so,'  'Why,  this,'  'Why, 
just  this,'  before  proceeding  with  our  answer  to  the  question. 
From  the  use  in  such  phrases,  it  might  well  have  developed 
gradually  into  a  pure  conjunction  'for.'  It  may  be  observed  in 
this  connexion  that  the  Skr.  Vedic  negative  nahi,  besides  its 
usual  meaning  'for  not,'  shews  sometimes  the  simple  meaning 
'not'  or  'indeed  not'  (v.  Grassmann,  Worterb.  zum  RV.,  s.  h.  v., 
and  Delbriick,  Vedische  Chrestomathie,  p.  84,  s.  h.  v.),  in  which 
cases  it  comes  very  close  to  Gk.  ov-^t- 

ou-jti. — The  -lei  of  oO-#c/,  quite  distinct  from  the  -xl  of  ov-x* 
(just  discussed),  is  the  neuter  singular  of  the  demonstrative 
pronominal  stem  *^i-  'this'  (cf.  Osthoff,  1.  c;  Brugmann,  Gr.  Gr.*, 
§95.  P-  131.  and  Gr.  Ill,  §182,  p.  49,  E.  E.,  §331,  p.  330.  E.  E.), 
so  that  ov'Ki  meant  originally  'not  this,'  and  is  thus  identical  in 
meaning  with  ov-x^  'not  this'  (v.  supra). 

(2)    OVXi  OVK. 

None  of  the  theories  yet  advanced  in  explanation  of  Gk.  olx, 
ovK  seems  at  all  satisfactory. 

We  may  at  once  dispose  of  Pott's  above-mentioned  (§3,  p.  47) 
identification  of  Gk.  ovk  with  Skr.  dvdk.  Skr.  dvdk  is  of  course, 
strictly,  the  nom.  ace.  sing.  neut.  of  Skr.  dvdc  'turned  downwards,' 
which  is  compounded  of  Skr.  dva  'off,  down'  and  -ar-  'bent  in  a 
certain  direction,  turned '  (from  Idg.  ^-Pg-,  seen  in  Gk.  irodawo'tf 
\j9i\..prop-inqu-o-s)\  cf.  Whitney,  Skr.  Gr.,  §§407, 409 ;  Brugmann, 
Gr.  I,  §228,  p.  195. 

Osthoff,  1.  c,  explains  o^x  :  ovx  in  the  following  way : — he  holds 
that  in  cases  of  apostrophe  of  the  -t  before  a  following  aspirated 

'  Bnxgmann,  Gr.  Ill,  §409,  Rem.  i,  pp.  330,  331,  was  doubtful  how  to  derive 
the  Latin  stem  ho-. 


Digitized  by  VjOOQ IC 


64  AMERICAN  JOURNAL    OF  PHILOLOGY, 

vowel,  ovict  and  fAiyy  fell  together  in  oO^' ;  so  that  e.  g.  o{»x  ort,  o&x 
SnTOfxai  can  belong  to  both  o{>jci  and  olxl ;  before  &t  he  thinks  there 
probably  existed  both  oI-k-  and  ov-x'  (ov-K-m  and  *ov-x-m)i  com- 
paring the  Etym.  Magn.  368,  30  tt<^ccXc  yap  Xryco^ai  o^x'  ^pxo/Aai. 
Then  at  a  later  date,  according  to  his  theory,  ^olx'tn  and  *ovx* 
HpxofJMi  gave  way,  the  use  of  ovx,  beside  ovx  from  oind,  becoming 
confined  to  the  position  before  spiritus  asper  on  the  analogy  of 
other  cases,  such  as  a<^'  beside  air',  kqB*  beside  icary  aiS'  beside  avr\ 
so  that  thenceforth  abx  appeared  only  as  a  phonetic  (or  graphic) 
modification  of  ovx. 

Victor  Henry,  in  M6m.  de  la  Soc.  de  Ling.,  vol.  VI,  part  5, 
1889,  pp.  379  sq.,  without  taking  oIkI  into  consideration  at  all, 
regards  ovx  ^"^  ^*^  ^  ^^^  derivable  primarily  from  ovx(-  His 
explanation  is  that  "  before  an  initial  vowel  the  I  is  elided,  e.  g. 
ohx  c£«.  stnd  if  the  following  consonant  was  an  aspirate,  then  the 
X  had  to  lose  its  aspiration,  whence  ovc  cx^,  ovc  ^X^oi^ ;  whereupon,'' 
according  to  his  theory,  "the  relation  of  ovk  «x»  ^^  ^^X  *^t  ^^^ 
others  similar,  gave  rise  to  the  belief  that  the  x  of  this  last  combi- 
nation was  due  to  the  rough  breathing  of  f^cA,  so  that  the  *deas- 
piration'  has  gained  more  and  more,  whence  oitK  con,  ovit  oXcaXe." 

But  surely  there  is  at  hand  a  much  simpler  explanation  than 
either  of  these  two  latter.  If,  as  is  most  probable,  oyi-yl,  and  ov-« 
both  had  originally  the  same  meaning  ^not  this'  (v.  supra,  pp. 
62,  63),  then  both  would  be  used  indiscriminately.  This  being 
so,  it  was  but  natural  that  in  cases  of  the  elision  of  the  [  before  a 
following  aspirated  word  the  form  ovx(0  would  be  chosen,  while 
on  the  other  hand  before  a  following  non-aspirated  word  the  form 
used  would  be  ovje(i). 

(J>)  The  accent  of  oh'Xi  oI-kI  beside  that  of  J-xi  i^at-x*. 

Assuming  the  correctness  of  the  view  that  "-xi  =  Skr.  Ai  = 
Idg.  accented  "^"^hi^^  it  appears  that  the  words  were  originally 
accented  thus :  *o^  x^»  *5  X*»  *»'«*  X'»  ^'^'^  'hen  when  composition  of 
the  two  members  took  place,  the  already  existing  accent  on  ^  and 
val  ousted  the  competing  accent  on  *xt,  while  in  ohxl  the  accent  of 
"^X}  had  no  rival,  and  consequently  remained. 

In  the  same  way  ovu  was  originally  accented  "^ov  kc,  and  then 
when  the  two  words  became  one,  the  accent  remained  on  the  -xt 
in  ovci  for  the  same  reason  that  it  remained  on  the  -x^'  i°  ovxt.' 

^Osthoff,  in  Morph.  Untersuch..  vol.  IV  (1881),  p.  244,  thinks  it  possible 
that  the  prim.  Gk.  form  of  Gk.  ovkI  was  *o£>-«£  (cf,  troWd-Kt),  and  that  then 
♦o^-Kt  later  became  <w/ct  on  analogy  of  ov^:*. 


Digitized  by  VjOOQ IC 


LATIN  MAUD  AND  GREEK  OT.  65 

§9.  Armenian  zi. 

Arm.  zi^  *for'  ought,  I  think,  certainly  to  be  identified  with 
Skr.  ht  and  (more  rarely)  -^f ,  Avest.  zl  and  (more  rarely)  zl^ 
Old  Pers.  *di*  'for/  from  Idg.  *ghl  (*ghi). 

Hiibschmann  (Arm.  Stud.  I,  p.  79)  and  Brugmann  (Gr.  I,  §410) 
hold  that  Idg.  gA,  initially  and  after  n,  r,  was  represented  in 
Armenian  by  j\  and  only  after  vowels  by  z ;  e.  g.  jaune-m  '  I 

But  why  should  Gk.  -kl  (from  Idg.  *>b'-)  be  regarded  as  originally  accentless  ? 
It  is  tme  that  we  cannot  adduce  any  evidence  from  Sanskrit  to  settle  the 
question  either  way,  as  there  seems  to  be  no  Skr.  representative  of  Idg.  *^'- 
(Brugmann,  at  any  rate,  gives  none  in  the  Grundr.  Ill,  §409,  pp.  329.  330, 
E.  E.).  Nor  do  the  representatives  of  Idg.  *ki-  in  other  Idg.  languages — e.  g. 
Lith.  sti'S  O.C.Sl.  si  O.Ir.  ee  *  this  (masc.),'  Goth.  kU(-a)  Ags.  hit  O.Norse  hit{() 
*this  (neut.)* — prove  anything  for  the  accent.  Nevertheless,  my  explanation 
of  the  accent  of  ovx/,  given  above  in  the  text,  seems  to  me  preferable  to  that 
of  Osthoff. 

In  itoXka-ia  (which  seems  manifestly  later  than  ovxt)  we  see  the  same 
process  as  we  saw  above  in  ^-;t<  '^^*-X^*  where  ij  vol  having  already  an  accent 
of  their  own,  ousted  the  competing  accent  already  existing  on  -x^.  Thus 
V0XX6  and  */d,  when  combined,  produced  *iTo?Xaid^  whence  iroXXdtu  (whence 
many  an  analogical  formation,  e.  g.  ir^^iardiu,  o^^diu,  daadiu^  roaadiu^  etc.). 

In  ov-ici,  on  the  other  hand,  there  was  no  competing  accent  on  the  av  to 
oust  that  on  -ki,  whence  the  combination  produced  ovkL 

^Habschmann,  Arm.  Stud.  I,  p.  12,  and  note  z);  also  Justi,  Handb.  d. 
Zendspr.  (1864),  s.  v.  if. 

'  I  have  written  an  asterisk  against  Old  Pers.*</f  *  for.'  because  Old  Pers.  does 
not  actually  shew  a  word  *di  with  the  meaning  *  for.'  We  find,  indeed,  in  Old 
Persian  a  pronominal  stem  di-  =  *  this  (masc.),'  of  which  the  accus.  sing,  dim 
and  the  accus.  plur.  dt}  appear  [v.  Spiegel,  Die  Altpers.  Keilinschriften  (1881), 
p.  325].  These  Persian  forms  are  enclitic  (v.  Justi,  op.  cit.,  s.  v.  di).  With 
this  Old  Persian  stem  di-  Justi  (1.  c.)  and  Jackson  (Avesta  Reader,  1893,  p.  73) 
identify  Avestic  di-,  which  also  is  an  enclitic  pronoun,  third  person  (v.  Jackson, 
1.  c).  Now  it  is  obvious  that  this  Avestjc  <&'-  cannot  be  identified  with  Avestic 
si  f f,  Skr.  Ai  »M,  Gk.  x'^*  from  Idg.  *f ^ ,  hence  from  these  we  must  separate 
also  Old  Pers.  di-  (if  this  latter  is  identical  with  Avest.  di-).  But  the  Old 
Pers.  form  *di  'for'  may  quite  regularly  be  referred  to  *g^M  [v.  HQbschmann, 
Arm.  Stud.  (1883),  I,  p.  12,  note  i ;  Brugmann,  Grundr.  I,  §405],  hence  it  is 
possible  that  Old  Pers.  *di  *for'  from  Idg.  *fAi  fell  together  with  the 
(etymologically)  quite  different  stem  which  was  represented  in  Avest.  and  Old 
Pers.  by  di-  '  this,'  and  hence  lost  its  own,  apparently  Prim.  Aryan  (Skr.  and 
Avest.),  meaning  'for.'  Indeed,  if  I  am  right  in  assigning  the  meaning  ' this ' 
to  Idg.  *gki  (v.  supra.  §8,  p.  63),  it  is  possible  that  Old  Pers.  *di,  in  contrast  to 
Skr.  Ai  -H  and  Avest.  A  if,  never  reached  the  development  of  meaning  *  for,' 
and  hence  all  the  more  easily  fell  together  with  the  synonymous  (but  etymo* 
logically  different)  form,  Old  Pers.  di*  *  this.' 


Digitized  by  VjOOQ IC 


66  AMERICAN  JOURNAL    OF  PHILOLOGY. 

dedicate,  sacrifice'  from  hlgkey^-  *pour/  lizum  *I  lick'  (:  Skr. 
leh-mt).  If  this  rule  is  correct,  we  can  only  derive  Arm.  zi  from 
Idg.  *gki  by  supposing  that  *ghi  was  already  in  Idg.  so 
closely  attached  to  any  preceding  word  which  ended  in  a  vowel 
that  the  compound  thus  formed  came  to  be  regarded  as  a  word- 
unity,^  and  that  later  in  some  of  the  individual  developments  of 
Idg.  the  compound  split  up  again,  e.  g.  Arm.  zi  Avest.  zl  zl  Skr. 
hi,  contrasted  with  Avest.  _y<?--8l  ye-zl  Mf  Skr.  nahi  nahi  Gk.  owx*, 
where  the  compound  form  has  remained.  Perhaps,  however, 
the  rule,  as  stated  by  Hiibschmann  and  Brugmann  (v.  supra),  is 
too  narrow.* 

Besides  Arm.  zi  *  for '  we  find  also  in  Armenian  zi  =  ori,  zif  = 
ri;  Hiibschmann,  Arm.  Stud.  I,  p.  12,  says  that  if  we  identify 
Arm.  zi  'for'  with  Skr.  hi  Avest.  zl  *for'  we  must  separate  it 
from  zi  =STiy  zif  ==ti;  I  cannot  see  from  what  other  original 
than  *ghi  we  can  derive  these  latter;  the  original  meaning  of 
Idg.  *ghi  appears  to  be  *this'  (v.  supra,  §8,  p.  63),  from  which 
the  transition  to  the  meaning  on  is  easy  enough,  although  the 
transition  to  the  meaning  Wy  is,  I  admit,  not  quite  so  clear.  But, 
after  all,  it  is  much  the  same  difficulty  as  we  have  in  the  transition 

*  In  view  of  the  fact  that  Skr.  At  -At  Avestic  a  gi  Greek  -x^  ^^^  always  (so 
to  speak)  '  postposition-particles/  this  would  not  be  a  very  rash  assumption. 

'Ann.  Mard  'adornment'  I  would  derive  from  Idg.  ^ghr-tu-^  and  identify 
(save  for  the  suffix,  on  which  see  Biugmann,  Gr.  II,  §108,  p.  327)  with  Gk. 
;tap-rd-c  '  delightful,  in  which  one  takes  pleasure,'  ^gher^  which  is  seen  also  in 
Skr.  Adr-yaM  *  takes  pleasure  in'  Gk.  x^P^i  Umbr.  heris  *  vis'  heriest  (fut.) 
'volet/  Osc.  heriiad  'velit.'  Thus  the  relation  between  Arm.  sar-d  from 
*gAr.tu-  and  Gk.  ;^:ap-T6-f  from  *gA^'td'S  is  exactly  the  same  as  that  between  Skr. 
^'tU'}  and  Skr.  r-td-s.  Thus  Arm.  sard  means  primarily  'something  in  which 
one  takes  a  delight,  has  pleasure.'  So  too  Skr.  Airariya  (for  *Aaranya  ?),  Avest. 
zaranya  'gold*  (Pr.  Iran.  *sArranya)  are  also  traceable  to  this  ^gAer-  (cf. 
Whitney,  Skr.  Roots,  p.  203,  s.  v.  Aat),  and  originally  meant  *  something  in 
which  one  takes  pleasure.' 

If  the  derivation  of  Arm.  zi  from  Idg.  *gAi,  and  of  Arm.  zard  from  Idg. 
*gAr.tU'  is  correct,  then  the  rule  concerning  the  Armenian  representation  of 
initial  Idg.  gA  as  stated  by  Hiibschmann  and  Brugmann  is  too  narrow. 

What,  again,  is  the  explanation  of  the  so-called  'prefix'  z  which  so  often 
occurs  in  Armenian,  e.g.  z-erc-ani-m  *I  free  myself*  (HQbschmann,  Arm.  Stud. 
I,  p.  31).  z-ge-uu-m  *  I  dress  myself  x-gest  'dress'  |/w/-  'dress'  (Habschmann, 
op.  cit.,  p.  30),  z-air-ana-l  'to  be  in  a  passion'  beside  air-tl* to  burn'  (Hflbsch- 
mann,  op.  cit,  p.  12)?  Cf.  also  the  accusatives  z-is  'me'  z-A'tz  'thee'  t-m<s 
'us'  Z'j'ez  'you'  (HQbschmann,  op.  cit.,  p.  70);  vide  on  the  subject  HQbsch- 
mann, Arm.  Stud.,  p.  12,  where  he  refers  to  his  Kasuslehre,  p.  317. 


Digitized  by  VjOOQ IC 


LATIN  HAUD  AND  GREEK  OT.  6/ 

of  meaning  between  Gk.  n  (indef.)  :  Gk.  W;  (interrogOi  both  of 
which  come  from  the  same  original  Idg.  form. 

§io.  The  c  of  Armenian  6c  *noC 

Before  concluding  the  present  investigation,  it  will  be  well  to 
add  a  short  remark  on  the  c  of  Arm.  oc  *not,'  which  I  would 
analyse  o-c} 

I  will  not  venture  to  explain  the  difficult  vocalism  of  o-c  (of 
which,  by  the  way,  I  can  find  no  mention  in  Brugmann's  Grund- 
riss).  Arm.  oc  has  long  been  compared  with  Gk.  olKt  and  de 
Lagarde,  Arm.  Stud.,  p.  i8i,  regarded  their  identification  as 
"richtig  Oder  doch  in  hohem  grade  wahrscheinlich."  Hubsch- 
mann,  however,  is  undoubtedly  right  in  holding  (Arm.  Stud.  I, 
p.  13)  that  **oc  'nicht'  =  gr.  ovk  zu  setzen,  ist  aus  lautlichen 
Griinden  im  hochsten  Masse  bedenlich."  If  (as  I  believe  to  be 
the  correct  view)  Gk.  ov  represents  Idg.  *<?^,*  the  o-  of  o-c  certainly 
cannot  be  identified  with  Gk.  ov,  seeing  that  0  is  never  the  Arme- 
nian representative  of  Idg.  *ou  (for  the  true  representation  of 
which  in  Armenian,  see  Hiibschmann  in  Zeitschr.  d.  Deutsch. 
Morgenl.  Ges.  XXXV,  i88i,  p.  172,  and  Arm.  Stud.  I,  1883,  p. 
62;  Brugmann,  Grundr.  I,  1886,  §63;  Bartholomae  in  Bezz. 
Beitr.  XVII,  1891,  p.  99).  Nor  can  Bugge's  attempt  to  derive 
oc  firom  Arm.  *dc  from  Arm.  *auc  bom  Idg.  *au-  (Kuhn's 
Zeitschr.  XXXII,  1893,  p.  30)  be  accepted ;  this  will  be  manifest 
fi'om  the  following  communication,  which  I  have  received  from 
Prof.  Hiibschmann  himself,  concerning  (i)  the  examples  which 
Bugge  cites  in  support  of  his  etymology,  (2)  the  Armenian 
representation  of  Idg.  a^-:  '*Arm.  sosk  heisst  'bios,  allein,  leer' 
und  hat  mit  skr.  stiika-,  etc.,  nichts  zu  ihun.    Ob  arm.  60c 

^  The  first  member  of  Arm.  o-c,  namely  0-,  is  never  (like  Gk.  ov)  used  inde- 
pendently. Bnt  it  is  interesting  to  note  that  the  second  member,  namely  f ,  is 
used  independently,  instead  of  the  full  form  cic,  frequently  with  verbs  and 
sometimes  with  substantives,  e.  g.  Arm.  c-astvac  *  not  god,*  i.  e.  *  idol.'  Cf. 
also  the  Old  Armenian  proper-name  C-unak^  Catholicos.  successor  of  Nerses, 
'not-having/  i.  e.  'poor'  (v.  Hiibschmann,  'Die  altarm.  Personennamen,'  in 
Festgruss  an  Rudolf  von  Roth,  1893).  Curiously  enough,  the  same  phenom- 
enon is  visible  in  Modern  Greek  where  Old  Greek  ov6kv  appears  as  6tv^  the 
tmly  negative  part  of  the  word  having  been  dropped  off.  without  the  meaning 
being  thereby  affected.  Examples  of  the  same  phenomenon  occurring  in 
other  languages  will  be  found  in  Ziemer,  Vergl.  Synt.  der  Idg.  Comparation 
(1884),  p.  186,  and  note  I. 

»Vid.supra,§4,p.  51- 


Digitized  by  VjOOQ IC 


68  AMERICAN  JOURNAL    OF  PHILOLOGY. 

'Flamme'  zu  n-t^avarxtt  gehort  ist  gl^ichfalls  nicht  sicher  und 
beweist  jedenfalls  fiir  arm.  o  =  ursp.  au  nichts.  £s  giebt  bisher 
kein  sicheres  Beispiel  fiir  die  Gleichung  arm.  o  =  ursp.  a»." 

In  discussing  the  c  of  o-c  we  are  on  firmer  ground.  In 
Armenian,  which  does  not  labialise  the  velars,  Idg.  q  (except  after 
a  nasal  or  liquid)  becon>es  regularly  k^  k\  but  c  from  k  before 
original  e-  and  f-vowels  (see  Hiibschmann  in  Zeitschr.  d.  Deutsch. 
Morgenl.  Ges.  XXXV  172  f.,  and  Armen.  Stud.  I,  pp.  66,  67 ; 
Brugmann,  Or.  I,  §455).  Thus  we  hiay  derive  the  c  of  Arm.  o-c 
from  Idg.  *qi  (:  Gk.  n,  cf.  o0-ti)  or  Idg.  *qe  (:  Gk.  re,  cf.  00-re). 
Bugge,  in  Kuhn's  Zeitschr.  XXXII,  p.  31,  comparing  Arm.  ack^ 
beside  Gk.  ^o-orc  from  *6k^€,  thinks  that  the  c  of  Arm.  oc  can 
correspond  to  the  -ici  of  Greek  ov-ict;  but  he  is,  I  think,  at  fault 
herein;  for,  whereas  Arm.  c  cannot  be  the  outcome  of  an  Idg.  k 
(v.  Brugmann,  Gr.  I,  §§380,  408),  the  -m  which  occurs  in  ov-m 
iroXXa-M,  etc,  and  in  the  Greek  numerical  adverbs  such  as  rfrpaxi 
ircyram,  etc.,  comes  from  the  Idg.  demonstrative  stem  *ki'  (v. 
supra,  §8,  p.  63)  and  not  from  the  Idg.  interrogative  and  indefinite 
pronominal  stem  *qi'}  Hence  we  must  separate  the  c  of  Arm. 
O'C  from  Gk.  •«,  and  identify  it  with  Gk.  r*  (from  Idg.  *qi)  or  re 
(from  Idg.  *qe). 

§11.  Classificaiicn  of  the  farms  discussed  in  the  foregoing 
§§8-io. 

The  forms  which  we  have  been  discussing  in  §§8-io  (incl.)  may 
be  classed  accordingly : — 

I. 
Idg.  *^^-*lt.: 
Lat.  -ce  in  ^ho-i-ce  (whence  hie)  from  the  Idg.  demonstrative 

pronominal  stem  ^ko- '  this.' 
Gk.  -M  in  ov-ici,  nom.  ace.  sing.  neut.  of  the  Idg.  demonstrative 
pronominal  stem  *&-  'this";  cf.  Goth.  AiV(-a),  Ags.  A//, 
O.Norse  hii(i)  *this  (neut.)/  and   Lith.  sal-s  O.C.Sl.  si 
O.Ir.  ce  'this  (masc.)." 
Gk.  'K  in  ol'K,  abridged  from  -kI  in  ov-ju.* 

*■  If  Greek  -tu  were  from  Idg.  *^'-,  as  is  assumed  by  Wackemagel  (in  Kuhn's 
Zeitschr.  XXV,  p.  286  f.)  and  J.  Schmidt  (Pluralb.,  p.  352),  all  the  Greek 
dialects  except  Thessalian  must  have  had  -ri  in  place  of  it  (v.  Brugmann,  Gr. 
Ill,  §182.  p.  49  and  note  i,  §409,  p.  330.  Engl.  edit.). 

'Supra,  §8  {a)  (i),  s.  v.  oif-id,  p.  63,  and  (6)  text  and  note,  p.  64. 

*  Supra,  §8  (^),  note,  p.  64.  *  Supra,  §8  (a)  (2)  ad  fin.,  p.  64. 


Digitized  by  VjOOQ IC 


LATIN  HAUD  AND  GREEK  OT. 


69 


11. 

Lat.  hO'  in  ^tV  (from  ^ho-i-ce)  from  the  Idg.  demonstrative 
pronom.  stem  ^gho-  'this.'* 


Skr.  ht  (and  -Aft*^ 
Avest.-arl  (and -aril)* 
Arm.  zi* 
Gk.  -;(i  in  ov-^t' 


Nom.  ace.  sing  neut.  of  the  Idg.  demonstr. 
pronom.  stem  *gki-  'this*  (whence  in 
Skr.  Avest.  and  Arm.  the  meaning 
'for'). 


Gk.  'X  in  ov-x,  abridged  from  -x*  in  ov-x»-* 


III. 


Idg.  *qO'  *qi' : 
Gk.  T€  in  o0-T€  from  the  uninflected  Idg.  *ge  'how'  (indefinite 
'somehow'  and  'as  also'  =  'and')  of  the  Idg.  interrogative 
and  indefinite  pronominal  stem  *gO'y  and  identical  with 
Skr.  and  Avest.  ca,  Lat.  -gue.^ 
Gk.  n  in  oihri,  nom.  ace.  sing.  neut.  of  the  Idg.  interrog.  and 
indefl  pronom.  stem  *gi',  whence  come  also  Skr.  nd-ki-} 
(for  *nd'Ci'f)  'no  one,'  Lat  quidy  etc.* 
Arm.  c  in  o-c :  either  (i)  from  Idg.  *qe  (:  Gk.  re  in  o0-re),  mean- 
ing 'some- how,'  so  that  ^-f  =  ' no-how,  in 
no  wise.'* 
or  (ii)  from  Idg.  *qi  (:  Gk.  t*  in  o0-ti)  'some- 
what,' so  that  O'C  =  'not  at  all.'* 

Lionel  Horton-Smith. 

53  Queen'9  Gardens,  Lancaster  Gate,  London,  W., 
and  St.  John's  College,  Cambridge;  England. 

*  Supra,  §8  (a)  (i),  s.  v.  <w-;r«.  P-  62.  •  Sapra,  §9,  pp.  65,  66. 
'Supra,  §8  {a)  (i),  s.  v.  ou-;t*.  PP-  6x.  62.  and  (b\  p.  64. 

*  Supra,  §8  (a)  (2)  ad  fin.,  p.  64.  ^  Supra,  §10  ad  fin.,  p.  68. 


Digitized  by  VjOOQ IC 


NOTES. 
Latin  -astro-. 

The  material  for  the  study  of  this  suffix  has  been  collected  by 
Franz  Seek,  Das  lateinische  Suffix  aster^  asira^  asirum,  Arch.  I 
390-404 ;  cf.  also  Sittl,  Zum  Suffix  aster,  Arch.  VI  508.  Addi- 
tional examples  are  also  found  in  H.  Stadler's  article,  Lateinische 
Pflanzennamen  im  Dioskorides,  Arch.  X  83-115. 

Previous  attempts  to  explain  the  origin  of  this  suffix  have  been 
unsatisfactory.  A  criticism  of  the  earlier  explanations  is  given 
by  Schnorr  von  Carolsfeld,  Arch.  I  404-7.  His  own  solution  is 
that  the  suffix  -siro-j  whether  coming  from  -d+fro-  (OsthofF)  or 
from  substantives  in  -os+fro-y  e.  g.  flusirMm>*flovostrum  (Cors- 
sen),  was  added  to  nominal  stems,  e.  g.  oiea-strum,  halica-sirum, 
and  from  these  a  suffix  -asiro-  was  propagated.  On  the  other 
hand,  Lindsay,  The  Latin  Language,  pp.  329  f.,  and  Stolz,  Histo- 
rische  Grammatik  der  lateinischen  Sprache,  I  543  ff.,  prefer  to 
follow  Ascoli  in  seeing  here  the  comparative  suffix  -iero-.  Lind- 
say does  not  concern  himself  about  the  'prefixed  -as-.*  Stolz's 
suggestion  is  that,  in  spite  of  the  quantity  of  the  a,^  these  words 
are  to  be  derived  from  formations  in  -dius,  e.  g.  ''^ped%iaster> 
*pediidN(/)rO'^formasier>formdti(je)rOy  etc. 

The  objection  to  all  of  these  explanations  is  that  they  fail  to 
consider  sufficiently  the  lexical  contents  of  the  suffix.  The  suffix 
is  of  pejorative  value  and  expresses  the  resemblance  of  the  deriv- 
ative to  the  primitive  noun,  and  generally  with  a  connotation  of 
contempt — 'a  poor  copy  of.'  This  highly  specialized  meaning 
and  the  restriction  of  the  suffix  to  certain  congeneric  groups  of 
words  suggest  at  once  that  we  have  before  us  a  case  of  adaptation, 

^This  is  one  of  his  objections  to  yon  Carolsfeld^s  explanation.  The 
difficulty,  however,  is  not  serious,  as  there  seems  to  be  no  evidence  for  the 
quantity  of  the  a  except  the  analogy  of  adjectives  in  estus ;  cf.  Marx,  s.  v. 
oleaster. 


Digitized  by  VjOOQ IC 


NOTES,  71 

meaning  by  that  term  ''the  infusion  with  some  grammatical  or 
lexical  value  of  a  formal  element  originally  either  devoid  of  any 
special  functional  value  or  possessed  of  a  value  which  has  faded 
out  so  completely  as  to  make  this  infusion  possible,"  Bloomfield, 
A.  J.  P.  XII  I ;  cf.  also  for  the  general  treatment  and  other 
examples  of  this  factor  in  language,  A.  J.  P.  XVI  409-34. 

That  adaptation  had  been  at  work  in  the  case  of  this  formation 
was  noted  in  the  first  of  the  articles  cited  above,  but  no  attempt 
was  made  to  determine  the  mode  in  which  it  had  operated.  The 
purpose  of  the  present  paper  is  to  offer  a  suggestion  that  bears 
upon  this  point. 

In  considering  this  suffix  we  must  keep  in  mind  that  it  belonged 
especially  to  popular  Latin  and  that  its  use  was  established  before 
the  beginning  of  our  records.  Notice  the  manifestly  secondary 
character  of  the  words  that  occur  in  the  archaic  period :  filiaster^ 
pediiasielltis,  parasiiasier.  Consequently  a  chronological  arrange- 
ment of  the.  words  would  give  not  the  order  of  their  formation, 
but  of  their  emergence  in  literature,  and  accordingly  we  are 
warranted  in  departing  from  it. 

Excepting  the  names  of  plants,  there  is  no  class  of  words 
formed  with  this  suffix  that  does  not  make  the  impression  of 
being  a  secondary  formation.  But  in  the  names  of  plants  this 
suffix  seems  to  be  quite  at  home,  and  has  a  very  definite  value. 
Oleaster  denotes  the  wild  olive  in  opposition  to  olea^  the  culti- 
vated olive.  The  distinction  is  unmistakable ;  cf.  e.  g.  Verg.  G. 
II  182  and  Vulg.  Ep.  ad  Rom.  11,  17  and  24,  where  the  two 
translate  aypuXaw  and  Ama;  cf.  also  Isid.  Orig.  XVII  7.  61 
Oleaster  dictus  quod  sit  foliis  oleae  similibus  sed  latioribus  arbor 
inculta  atque  silvestris  amara  atque  infructuosa.  So  apiasirutn 
denotes  the  wild  apium,  meniasirum  the  wild  menta;  cf.  Pliny, 
N.  H.  XX  144  Mentastrum  silvesMs  menta  est;  cf.  XIX  159;  cf. 

also   Arch.    X    103   fuvSaarpovfJL  =  ^dvoa-fiof   aypiogy    III    3^  G*      ^OT 

pinasier  cf.  Pliny,  XVI  39  Pinaster  nihil  est  aliud  quam  pinus 
silvestris ;  cf.  80.  Lotaster  =  wild  lotus ;  pyriaster  is  glossed  by 
pyrus  agresiis ;  Pliny,  XVI  205  uses  pirns  silvestris.  To  these 
should  be  added  salicastrum.  Pliny,  XXIII  20  uses  it  to  denote 
a  wild  vine  that  grows  on  willows,  but  the  Italian  points  to  the 
more  primitive  meaning  wild-willow. 

There  is  some  slight  evidence  for  a  similar  use  with  names  of 
animals.  Du  Cange  reports  a  gloss  caiulaster  :  lo  cane  sahaticOy 
and  Sitd's  quotation,  1.  c,  from  the  Etymologicum  Gudianum, 


Digitized  by  VjOOQ IC 


72  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Alyatrrptog  ai(  iypiot^  looks  like  an  imitation  of  the  Latin  under  the 
influence,  perhaps,  ofSypios.  But  pu/Iasira  in  Varro,  if  the  emen- 
dation be  accepted,  has  only  diminutive,  and /^r^a^/^r  in  Aldhelm 
only  contemptuous  signification. 

Now,  other  formations  of  this  kind  in  Latin  are  clearly  adapta- 
tive — the  lexical  value  of  the  suffix  can  be  plainly  felt — and  if  this 
suffix  served  originally  to  express  the  relation  between  the  culti- 
vated and  uncultivated  varieties  of  the  same  plant,  I  think  it  may 
easily  be  understood  how  its  meaning  could  be  extended  to 
denote  resemblance  in  general  with  the  connotation  of  inferiority, 
and  how  it  might  then  develop  into  either  a  pejorative  or  a 
diminutive  suffix,  and  sometimes  fade  out  into  a  mere  expression 
of  resemblance  or  approximation.  Thus  *peditasUr  is  a  *  mock 
iooi'Soldxtr,^  pkilosophaster  ^di  mock  philosopher,' /^n/^nta^/^r '  a 
man  that  apes  Antony.'  So  too  it  expresses  the  fictitious  relation- 
ships yf/raj/fr  *a  stepson,' y^fVu/ra  *a  stepdaughter,' etc.;  cf.  the 
German  wiUe  Ehe  =  concubinage. 

The  suffix  comes  back  on  the  class  from  which  it  started,  and 
we  have  siliquastrum^  so  called  from  its  resemblance  to  the 
siliqua:  liliastrum  planta  lilii  similis;  ocymastrum  herba  ocymo 
similis.  Apiaster  or  apiastra,  the  name  of  a  bird  so  called  quia 
apes  comeduniy  seems  at  first  a  strange  formation.  But  it  gets  its 
name  (cf.  Antoniaster)  *  Bee*s-fi-iend*  by  a  sort  of  oxymoron. 
Nothing,  in  reality,  but  another  application  of  the  principle  of  the 
analogy  of  opposites,  which  will  account  for  the  development  of  a 
meliorative  signification  as  far  as  it  occurs. 

Finally  the  suffix  was  added  to  adjectives.  It  is  interesting  to 
note  the  closely  congeneric  nature  of  the  words.  They  are  either 
designations  of  bodily  defects — calvaster,  adcalvasier^  recalvasier^ 
claudasier^  mancasUr,  surdasier—ox  colors  that  lend  themselves 
naturally  to  such  use — canaster,  *gravaster,fulvaster  and  nigel- 
laster\  besides  these  are  only  crudasier,  novellasier  and  *medu 
osier. 

The  next  question  is  the  origin  of  the  suffix  for  the  names  of 
wild  plants.  The  definiteness  of  its  lexical  value  leads  us  to  look 
again  for  the  working  of  adaptation,  and  I  would  suggest  that  the 
suffix  came  from  silvestro-  before  its  passage  into  the  t-declension, 
being  added  first  to  nouns  like  oleay  and  from  these  propagating 
the  suffix  -astro-. 

In  confirmation  of  this  view  it  may  be  noticed  that  the  literary 
expression  corresponding  to  these  popular  formations  is  the  use 


Digitized  by  VjOOQ IC 


-^c*^- 


NOTES.  73 

of  silvesiris ;  thus,  pinaster  :  pinus  sihesiris,  meniasirum  :  menta 
silvestris\  cf.  the  examples  cited  above.  Notice  also  how  in 
English  wild  has  almost  passed  into  a  prefix  for  the  expression  of 
the  same  idea. 

For  the  putting  of  the  whole  lexical  value  of  an  adjective  into 
its  sufHx,  I  know  of  no  good  parallel ;  but  the  process  does  not 
seem  to  me  improbable  nor  essentially  different  from  the  forma- 
tion execuiian  elecirocuiion  *hydrocuii(m. 

In  conclusion  I  would  call  attention  to  the  closely  congeneric 
nature  of  the  group  of  words  to  which  Silvester  belongs :  campes- 
ier,  equester,  Fanesier,  illustris,  nemestrinuSy  paluster^  paludester^ 
pedester,  rurestriSy  semestris  (and  other  compounds),  telluster, 
terrestery  vallestria.  Lanesiris  belongs  to  late  Latin  and  it 
stands  alone.  By  the  side  of  this  group  are  agrestis  and  caelestis^ 
and  it  seems  desirable  to  regard  them,  if  possible,  as  belonging  to 
the  same  formation. 

Brugmann,  II  184,  sees  in  these  words  the  comparative  suf&x 
'terO'  under  the  influence  of  the  analogy  of  -^j-stems,  and  is 
followed  by  Lindsay,  p.  330.  Stolz,  p.  503,  adopts  the  same 
view,  but  considers  eqtiester  for  ^equit-'tero-  and  pedester  for 
*pedit''ter(h  more  likely  starting-points  for  the  development  of 
the  suf&x.  At  the  same  time  he  admits,  p.  420,  the  possibility 
(cf.  Schweizer-Sidler,  KZ.  IV  309 ;  Schulze,  ib.  XXIX  270)  that 
we  have  in  these  words,  as  well  as  in  agrestis  and  caelestis^ 
compounds  with  sta-.  This  view  would  be  supported  by  the 
restriction  of  the  suf&x  in  Latin:  contrast  the  very  different 
range  of  -repo-  with  nouns  in  Greek;  cf.  Otto  Keller,  Zur  lat. 
Sprachgeschichte,  I  150. 

In  neither  case,  however,  does  it  seem  necessary  to  divorce  the 
two  formations.  In  the  former  case  agrestis  might  stand  for 
*agrestriSj  with  dissimilation — in  terrestris  the  conditions  are  not 
precisely  the  same  on  account  of  the  double  r,  while  in  caelestis 
we  also  have  a  liquid  in  the  first  part  of  the  word.  On  the  other 
hand,  if  we  consider  these  words  as  compounds  of  j/a-,  the  differ- 
ence between  -stri-  and  -j/t-  may  possibly  suggest  the  Aryan 
forms :  Avestan  rafaestqm  and  rdpaestdrpm,  Sanskrit  rathe^kdm 
and  saTrye^ltdram;  cf.  Jackson,  §§249,  330;  Bartholomae,  Altir 
Dial.  82 ;  Ar.  Forsch.  I  30. 

Cath.  Umv.  OF  Am..  GeORGE  MeLVILLE  BoLLING*. 

yam,  X897. 


Digitized  by  VjOOQ IC 


74  american  journal  of  philology. 

On  the  Alleged  Confusion  of  Nymph-Names. 

Appendix. 

In  my  paper  on  the  above  subject  (A.  J.  P.  XVII,  pp.  30  sqq.) 
I  referred  to  Virgil,  Eclogues,  X  62,  in  terras  which  may  be  pro- 
ductive of  misconception.  I  said :  '*  What  inference  can  be  drawn 
from  the  words  which  Virgil  puts  into  the  mouth  of  Gallus,  Eel. 
X  62  sq. '  iam  neque  Hamadtyades  rursus  nee  carmina  nobis  |  ipsa 
placent ;  ipsae  rursus  concedite  siluae'  ?  Why  should  Gallus  not 
have  sung  of  tree-nymphs  just  as  well  as  wood-nymphs  ?  "  The 
purport  of  this  remark  was  to  point  out  that  in  this  mention  of 
the  Hamadryads  there  is  a  reference  to  Gallus'  own  poems,  and 
that  consequendy  the  interpretation  of  the  words  involves  as  one 
of  its  factors  the  consideration  of  the  passage  thus  referred  to.  A 
theme  which  runs  through  the  whole  of  the  eclogue  is  the  inability 
of  the  nymphs  of  whom  Gallus  has  sung  to  aid  him  in  his  day  of 
trial.  To  pass  over  Arethusa  (v.  i)  the  puellae  Naides  (9  sqq.) 
are  reproached  for  their  failure  to  help  their  poet.  (It  may  be 
remarked  in  passing  that  the  language  of  9 '  nemora — saltus'  would 
probably  have  been  twisted  into  another  'confusion,'  but  for  the 
circumstance  that  Virgil,  in  the  commentator's  interest,  has  added 
the  necessary  reference  to  water  in  '  Aonie  Aganippe,'  12.)  The 
practice  of  making  learned  allusion  to  the  actual  expressions  of 
brother  poets  is  a  natural  habit  of  Latin  writers.  It  is  hardly 
necessary  to  do  more  than  refer  to  the  well-known  passages,  Ovid 
Am.  3.  9.  58,  Pont.  5.  16.  34  (whence  Grattius  23  has  been 
emended),  Statins  Silu.  i.  2.  255,  Mart.  14.  193.  2,  and  indeed 
Propertius  2.  34.  76,  a  reminiscence  more  or  less  conscious  of 
this  very  passage.  Further  down  (52  sqq.)  Virgil  puts  into  Gallus' 
mouth  the  words  'certum  est  in  siluis  inter  spelaea  ferarum  |  malle 
pati — intereamixtislustraboMaenalaNymphis  |  aut  acres  uenabor 
apros.'  And  it  is  argued  that,  because  in  the  sequel  to  these  lines, 
which  I  have  already  quoted,  Hamadtyades  is  used,  it  is  legiti- 
mate to  equate  Hamadryades  with  Nympkae.  It  might  have  been 
thought  that  by  this  time  the  principle  of  poetic  variation  would 
have  been  better  understood.  From  the  variation  Hamadtyades 
— Nymphae  (which  latter  word  includes  strictly  all  kinds  of 
nymphs,  and  is  of  course  not  limited  to  'wood-nymphs'),  all  that 
can  be  legitimately  inferred  is  that  the  '  Nymphs '  of  v.  55  included 
the  'Hamadryads';  and  of  course  this  of  necessity  neglects  any 
reference  that  there  may  be  to  actual  words  of  Gallus.  I  have  already 


Digitized  by  VjOOQ IC 


NOTES.  75 

(L  c.)  spoken  of  the  impossibility  of  drawing  any  inference  from 
the  phrase  of  Propertius,  'facilis — Hamadryadas/  which  appears 
to  unite  allusions  to  this  passage  (JHamadryas  occurs  nowhere 
else  in  Virgil)  and  to  Eclogue  3.  9  *  faciles  Nymphae.*  There  is 
an  allusion  to  this  latter  passage  in  a  much  later  writer,  Nemesianus 
Cyneget  94  sq. '  tecum  Naiades  faciles,  uiridique  iuuenta  |  puben- 
tes  Dryades,  nymphaeque  unde  amnibus  umor  |  adsint,  et  docilis 
decantet  Oreadas  Echo.'  This  place  has  a  special  interest,  as  it 
distinguishes  the  Naiads  from  the  Dryads  and  from  the  Hydriads 
('unde  amnibus  umor')*  An  earlier  writer  on  hunting  distin- 
guishes the  Naiads  and  the  Dryads,  though  he  does  not  use  the 
latter  name.  The  words  are,  unfortunately,  corrupt,  but  should 
probably  be  read  as  follows:  'adsciuere  tuo  comijtes  sub  numine 
diuae  |  centum,  omnes,  nemorum,  umentes  de  fontibus  omnes  | 
Naides  et  Latii  .  .  .  Faunus  .  .  .  Maenaliusque  puer,'^  ei  q.  s. 
(Grattius,  16  sqq.).  Cf.  Calpurn.,  Eel.  2.  14  'aifuerant  sicca 
Dryades  pede,  Naides  udo,' 

A  summary  of  our  results  may  be  appended.  A  Dryad  was  a 
nymph  of  the  forest  or  woodland,  as  an  Oread  was  a  nymph  of 
the  mountain.  The  term  is  more  general  than  Hamadryad^  which 
means  the  protectress  of  a  particular  tree.  A  Hydriad  is  a  water- 
nymph.  So  is  a  Naiad,  but  with  this  difference  of  use,  that  the 
word  also  denotes  the  protecting  nymph  (or  Hamadryad)  of  a 
tree  growing  in  or  out  of  the  water. 

The  passages  of  classical  authors  which  have  or  might  have 
been  adduced  to  prove  any  confusion  of  the  above  uses  are  either 
corrupt :  Anth.  Pal.  6.  189,  Culex  94,  Prop.  i.  20  (excluding 
those  where,  as  in  Anth.  P.  9. 668,  the  error  is  obvious).  12,  32, 45, 
Isidorus  8. 11. 9]^;  or  irrelevant.  Amongst  the  latter  we  class  (a) 
those  places  where  Nais  is  used  for  a  Naiad-Hamadryad :  Prop. 
2*  32*  37  sqq*,  Ov.  Met.  i.  690  sq..  Fast.  4.  231,  Stat.  Silu.  i.  3. 62, 
to  which  Nemesianus  Cyn.  94  should  (apparently)  be  added; 
{b)  those  places  where  poetical  modes  of  employing  language 
have  been  misunderstood :  Virg.  Eel.  10.  62,  Prop.  2.  34.  76,  Ov. 
Met.  14.  623,  Fast.  2.  155,  Stat.  Ach.  i.  294  sq.,  Silius  Italicus  15. 
769  sq.;  and  Ov.  Met.  14.  556  *  Naides  aequoreae.' 

J.  P.  POSTGATE. 

8 

^  The  MS  has  nomine  diune  |  centem  omnes  nemorumentes  de  f.'  The  dots 
indicate  lacunae  in  the  MS. 


Digitized  by  VjOOQ IC 


76  AMERICAN  JOURNAL    OF  PHILOLOGY. 

V€fA€iv  and  v€fita-3ai* 

Frohberger,  ad  Lys.  XIX  37  and  XXXII  4,  distinguishes 
between  ptftMip  and  vifucrBai  as  used  to  express  the  division  of  an 
inheritance,  and  lays  down  the  rule  that  the  active  is  applied  to  a 
father  (or  guardian)  dividing  a  property  among  heirs,  the  middle 
to  heirs  dividing  with  each  other.  G.  Huettner,  in  Acta  Semi- 
narii  Erlangensis  III,  p.  107,  asserts  that  this  rule  has  its  excep- 
tions, and  that  the  middle  has  the  force  of  the  active  in  Dem* 
XXXVI  8-9  and  Isae.  I  16.  Paley  and  Sandys,  in  Select  Private 
Orations  II,  3d  ed.,  p.  12,  quote  Donaldson,  Gr.  Gr.  p.  450,  to 
the  same  effect  and  refer  to  eleven  passages  in  Demosthenes, 
Isaeus  and  Lysias  for  examples  of  **  vtfitaBai  used  in  the  middle 
voice  generally  (but  not  always)  of  the  heirs."  They  probably 
rely  only  on  Huettner's  two  passages  to  prove  their  point,  as  a 
glance  at  their  other  references  will  show  that  in  them  all  Froh- 
berger's  rule  is  observed. 

Before  undertaking  to  show  that  it  is  observed  also  in  Huett- 
ner's  supposed  exceptions,  I  will  mention  the  two  legal  significa- 
tions of  v€fi€aBai  (which  are  not  to  be  found  in  Liddell  and  Scott) 
and  will  give  references  for  each. 

1.  "To  divide  with  another  or  others."  a,  followed  by  irp6t 
c.  ace.  expressing  the  other  party  to  the  division :  Lys.  XVI  10 ; 
Isae.  VII  5  and  25 ;  Dem.  XL  42  and  52,  XLVII  34.  d,  without 
vp6f:  Lys.  XXXII  4;  Isae.  I  16;  Dem.  XXXVI  8-9,  11,  32; 
XL  14;  XLVII  35. 

2.  **To  take  or  receive  as  one's  portion":  Lys.  XIX  46; 
Dem.  XXXVI  38 ;  XXXIX  6 ;  XL V  76.  The  passive  in  Dem. 
XXXVI  .38  belongs  also  under  this  head. 

Turning  to  the  two  disputed  passages,  we  find  in  Isae.  I  16:  ol 

rovrmf  ffnXoi  .  .  •  rf^low  ptifuurBai  ttjv  ovcrLav  koi  t6  rpirov  fupof  ^fiat  ^X'^"* 
The  usual  construction  of  d{i«  would  require  an  accusative  to  be 
inserted  before  v^ifiaaBai,  if  the  two  verbs  are  not  to  be  understood 
as  referring  to  the  same  subject.  But  from  Thuc.  II  89,  i,  Antiph. 
VI 46  we  see  that  the  subject-accusative  is  not  absolutely  required 
after  a{u»,  and  conclude  that  it  may  be  dropped  where  there  is  no 
danger  of  any  confusion  arising  in  the  hearer's  mind. .  Isaeus,  in 
the  passage  before  us,  did  not  fear  to  be  misunderstood,  first, 
because  rovnap  and  ^fias,  denoting  the  two  parties  to  the  suit, 
readily  suggested  a  subject  for  the  infinitive,  and,  secondly, 
because  no  person  is  ever  said  either  vtfitiv  or  vtfifcrBai  who  has 


Digitized  by  VjOOQ IC 


NOTES,  77 

not  some  legal  right  upon  the  property  in  question.  The  whole 
tenor  of  the  speech,  and  especially  paragraphs  2,  28,  51,  show 
that  the  ^'Xot  of  16  were  only  friendly  mediators,  urging  a  division 
but  claiming  no  right  to  make  one.  The  writer  could  therefore 
safely  give  to  ji^Lovv  the  construction  of  Uiktvov ;  he  was  probably 
led  to  do  so  by  the  fact  that  the  only  possible  subject  for  vtifuicrBai 
was  an  17/iar  which  should  include  both  contestants,  and  he  was 
intending  to  use  fifias  emphatically  of  one  side  alone  with  the 
second  infinitive,  Kx^ip. 

In  Dem.   XXXVI    8-9  we  read:    \oyiC6fA€Poi  irp6s  tavrovs  ol 

tnirpoiroif  ^t,  tl  de^o-ci  Kara  rat  diaBffKag,  ocr*  hp  cZrot  in  KoivZiv  t&p 
XpflfJMT€»p  avaknofft  Tovroig  €^€\6yras  dyriftoiptl  tg  \oiir6.  vtfituff  oid  6tiovp 
Zarai  vtpUp^  ptiiuLo-Boi  ra  tpB*  vnip  rov  iraMs  Hypturap.  xal  ptpoprai  rrjp 
SKkiip  oifcrlap,  •  •  • 

Here  the  middle  is  certainly  used  of  the  guardians,  but  its 
meaning  and  the  nature  of  their  action  are  defined  by  the  words 
wrtp  rov  woMg.  As  the  younger  son,  Pasicles,  was  not  of  age,  the 
operation  denoted  by  the  active  verb  ycficir— a  complete  trans- 
ference of  property  from  guardians  to  heirs — could  not  take 
place.  The  division  here  spoken  of  was  indeed  made  between 
the  elder  and  younger  sons ;  but  as  the  latter  was  legally  incap- 
able of  acting  for  himself,  he  was  represented  by  his  guardians, 
who,  in  their  capacity  of  trustees,  could  properly  be  said  "  to 
receive  a  portion  of  the  inheritance  on  behalf  of  the  child" 
(ptfitcrBai  vtrip  rov  troMg).  A  similar  definition  of  the  action  of  the 
middle  voice  by  vnip  appears  in  Dem.  XXVII  7-8,  where  the 
orator  says  of  his  own  guardians,  €h  y^p  rrip  avfApopiop  vnip  ipoZ 

avprrd^apTO  •  •  •   trvprrifiifirapB^  vnip  ifuw, 

Wm.  Hamilton  Kirk. 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES. 

Harper's  Dictionary  of  Classical  Literature  and  Antiquities,  edited  by 
Harry  Thurston  Peck,  M.A.,  Ph.D.  Illustrated.  New  York, 
Harper  &  Brothers,  1896.     Pp.  xv+  1701. 

The  genesis  of  most  dictionaries  and  books  of  reference  is  an  interesting 
subject  for  investigation,  and  the  portly  and  comely  volume  before  us  is 
no  exception  to  the  rule. 

The  preface  leads  one  to  suppose  that  the  larger  part  of  the  biographical 
and  geographical  material  is  based  upon  Anthon's  revision  of  Smith's 
Classical  Dictionary^  1852.  It  would  have  been  more  accurate  to  refer 
specifically  to  Smith's  Smaller  Classical  Dictionary  (London,  1852)  as  the 
source  of  a  very  large  number  of  the  minor  articles  which  appear  in  this 
work.  An  examination,  for  instance,  of  pages  78,  809  82  of  Harper  will 
show  the  importance  of  this  little  volume  in  the  compilation  of  the  new 
work.  Several  minor  errors  would  have  been  avoided  if  an  independent 
abridgment  of  the  Anthon  revision  had  been  made  oftener. 

The  preface  also  states  that  the  archaeological  portion  of  the  dictionary 
is  based  in  part,  but  only  in  part,  upon  Anthon's  revision  of  Smith's 
Dictionary  of  Greek  and  Roman  Antiquities^  1846.  This  statement  is  quite 
correct,  but  the  dependence  of  the  new  work  upon  that  dictionary  is  slight, 
and  no  serious  injustice  would  have  been  done  if  special  attention  had  not 
been  called  to  it  in  the  preface.  Nor  will  any  one  criticise  the  editor  for 
placing  his  main  reliance  in  archaeology  upon  more  recent  works. 

Certainly  neither  of  the  volumes  revised  by  Anthon  so  long  ago  was 
essential  to  the  production  of  Harper's  Dictionary.  But  this  assertion 
cannot  be  made  of  the  third  edition  of  Smith's  Dictionary  of  Greek  and 
Roman  Antiquities^  London,  1890-^1 ;  nor  of  Seyffert's  Dictionary  of  Class^ 
ical  Antiquities^  &c.,  revised  and  edited,  with  additions,  by  Nettleship  and 
Sandys,  London,  1891.  In  the  absence  of  either  of  these  works  there 
might  have  been  a  Harper's  Dictionary,  but  it  would  not  have  been  this 
dictionary. 

A  few  extracts  are  found  in  this  work  from  the  second  edition  of  Smith's 
Dictionary  of  Greek  and  Roman  Antiquities  (1848),  but  the  drafts  upon  the 
third  edition  have  been  frequent  and  copious.  To  illustrate  the  important 
place  which  it  holds  in  the  production  of  the  Harper,  it  may  be  stated  that 
about  one  hundred  articles  appear  to  have  been  based  upon  it  under  the 
letters  A  and  B,  of  which  a  large  majority  are  substantially  in  the  words  of 
the  original.  In  a  few  cases  the  articles  in  the  third  edition  of  Smith  are 
identical  with  those  in  the  second  edition.  In  such  cases  it  may  be  fair  not 
to  emphasize  the  dependence  of  Harper  upon  the  last  edition. 


Digitized  by  VjOOQ IC 


REVIEIVS  AND  BOOK  NOTICES.  79 

The  topics  are,  however,  treated  in  Smith  on  too  liberal  a  scale  for  the 
scheme  of  this  work,  and  therefore  the  original  articles  have  been  cat 
down,  sometimes  by  condensation,  more  often  by  omissions.  That  is, 
most  of  the  articles  from  this  source  are,  in  the  main,  in  the  original  words, 
but  with  a  less  complete  treatment  of  details.  The  editor  has  usually 
made  his  selections  and  omissions  with  judgment,  and  in  such  a  way  as  to 
bring  over  the  gist  of  the  subject,  though  necessarily  the  abbreviated  matter 
sometimes  makes  the  impression  of  incompleteness  in  comparison  with  the 
original  (e.  g.  Areopagus).  This,  however,  could  not  be  avoided  if  the 
whole  field  of  ancient  knowledge  was  to  be  covered  within  the  compass  of 
a  single  volume. 

Of  the  articles  drawn  from  Smith's  Dictionary  of  Greek  and  Roman 
Antiquities,  third  edition,  may  be  cited  a  few  titles  as  examples  of  the 
prominent  place  which  this  work  holds  in  the  compilation  of  Harper's 
Dictionary:  Abacus^  Achaean  League ^  Acta,  Aes,  Agrariae  Leges ,  Atnbar^ 
valia^  Anulust  Arcus  (arch),  Areopagus,  Athletatt  Aurum^  Balneae,  Bona, 
Breviarium  Alaricianum. 

The  scale  and  method  of  the  Harper  have  often  necessitated  slight 
changes  in  order  to  unite  properly  passages  not  originally  connected.  The 
editor  has  also  exercised  an  independent  judgment  in  modifying  or  reversing 
statements  (cf.  Antefixa),  in  making  many  minor  additions,  and  in  numerous 
verbal  changes,  sometimes  obviously  on  grounds  of  taste,  often  without 
any  obvious  reason.  Such  changes  are  especially  common  in  the  opening 
lines  of  the  articles. 

Sey£Eert's  Dictionary  has  for  its  field  not  only  Antiquities,  but  also 
Mythology,  Religion,  Literature,  Art,  and  the  Biography  of  authors,  artists 
and  philosophers.  Hence  from  its  wider  field  and  from  the  fact  that  the 
scale  of  treatment  of  topics  was  satisfactory,  the  obligation  to  this  volume 
in  respect  to  the  number  of  articles  adopted  from  it  is  even  greater  than 
to  Smith's  Dictionary.  For  example,  more  than  eighty  articles  under  the 
letter  A  are  to  be  credited  to  this  source.  In  most  cases  these  articles  are 
transferred  bodily,  though  with  occasional  verbal  changes.  Seyffert  prints 
Greek  words  in  Roman  letters,  while  Harper  substitutes  the  Greek  form. 
In  Sey£Eert  the  historical  present  tense  is  somewhat  overworked,  especially 
in  mythological  articles,  but  the  editor  of  Harper  prefers  a  past  tense. 
Seyffert  rarely  gives  any  bibliography,  but  this  is  often  supplied  in  Harper, 
and  sometimes  illustrative  references  to  English  literature  are  added. 

Among  the  more  important  articles  due  to  Seyffert  under  the  letter  A  are 
the  following  :  Acropolis,  Aediles,  Anaxagoras,  Aphrodite,  Apollo,  Archi* 
tectura,  Archon,  Ares,  Argonautae,  Aristophanes,  Arma,  Astronomia. 

The  obligation  to  Smith  and  Seyffert  is,  however,  by  no  means  limited 
to  the  earlier  letters  of  the  alphabet,  as  a  chance  examination  of  almost 
any  page  will  show.  For  example,  from  Smith  are  taken  also  the  articles 
Caelaiura  (seven  columns,  slightly  reduced  by  omissions),  Domus  (thirty- 
five  pages  reduced  to  sixteen  and  a  half),  Theatrum  (the  letter-press  some- 
what reduced,  but  some  excellent  cuts  added),  Vcu  (twenty-five  columns 
reduced  to  five  by  omissions,  leaving  the  impression  of  rather  scanty 


Digitized  by  VjOOQ IC 


80  AMERICAN  JOURNAL   OF  PHILOLOGY. 

treatment  for  so  important  a  topic  in  comparison  with  the  original).  From 
Seyffert  have  also  been  taken  the  following  titles,  as  well  as  others  here 
and  there  :  Comoedia^  Education  (the  first  four  columns),  Eisphora^  Fulcra^ 
Heracles,  Musvuum  Opus^  Otiracismus,  Pkihsophia,  Pictura^  Siatuaria  Ars, 
Tra^oedia, 

Occasionally  articles  have  a  composite  origin.  For  example,  Augur^ 
four  columns,  of  which  the  first  half-column  and  the  final  bibliography  are 
from  Smith,  the  intervening  matter  from  Seyffert;  Aquae  Ductus,  three 
and  a  half  columns,  mainly  from  Smith  (but  reduced  from  twenty  columns) 
with  a  half-column  from  Seyffert  inserted;  Musica,  six  columns,  mainly 
from  Seyffert,  but  with  two-thirds  of  a  column  on  Notation  drawn  from 
Smith  ;  Servus,  eight  columns,  of  which  six  from  Seyffert  and  a  half-column 
from  Smith;  Solon,  mainly  from  Smith's  Classical  Dictionary,  but  with 
additions  from  Seyffert,  including  references  to  Aristotle,  Ath.  PoL; 
Boeotarchis,  a  quarter-column  from  the  new  and  a  column  and  a  half  from 
the  old  edition  (1848)  of  Smith's  Dictionary  of  Antiquities.  The  effort  to 
bring  the  articles  up  to  date,  as  shown  above,  under  Solon,  is  also  shown 
by  additions  to  the  article  Tkemistocles  (from  Smith's  Classical  Dictionary), 
It  would  have  been  well  to  make  similar  additions  to  the  article  Archon, 

The  indebtedness  to  Seyffert  is  not  limited  to  articles  which  the  English 
editors  have  simply  translated  from  the  original  Lexikon  der  klassischen 
Altertkumskunde,  but  extends  also  to  original  contributions  or  additions 
made  by  them,  as  in  the  articles  Fulcra^  Musivum  Opus,  Philo,  and  Pictura, 
The  preface  expresses  in  complimentary  terms  a  regret  that  similar  addi- 
tions were  not  made  by  the  English  editors  to  all  the  articles. 

At  this  point  justice  to  the  editor  demands  the  insertion  of  a  paragraph 
from  the  preface :  **  In  drawing  upon  these  and  all  his  other  sources,  the 
Editor  has  allowed  himself  the  very  greatest  freedom.  Whatever  he  has 
taken  he  has  used  in  the  way  best  adapted  to  secure  the  end  he  had  in 
view.  When  material  was,  in  its  original  form,  precisely  suited  to  his  pur- 
pose, he  incorporated  it  without  a  change.  When  change  for  any  reason 
was  desirable,  he  enlarged,  condensed,  modified,  transposed,  or  para- 
phrased according  to  his  conception  of  what  was  most  needed  in  the  given 
case ;  and  as  the  greater  part  of  his  work  was  compilation  rather  than 
original  exposition,  he  wishes  here  to  express  his  very  great  indebtedness 
to  the  many  books  that  have  been  drawn  upon.  No  acknowledgment  can 
be  too  full  or  too  comprehensive ;  and  if  the  completed  work  be  found  of 
service  to  the  student  of  the  classics,  this  result  must  be  very  largely  due 
to  the  original  sources  whence  so  great  a  portion  of  the  Dictionary  is 
derived." 

This  quotation  accurately  describes  the  method  employed.  The  state- 
ment is  frank  and  comprehensive.  There  is  no  concealment.  But  it 
remains  a  fact  that  very  few  people  who  consult  a  work  of  reference 
stop  to  read  the  preface  to  it.  And  as  quotation  marks  are  not  used, 
except  as  they  are  brought  over  from  the  original  article,  and  as  even  the 
special  contributions  secured  from  American  or  foreign  scholars  are  not 
signed,  there  is  no  finger-post  of  any  kind  in  the  body  of  the  work  to  dis- 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES.  8 1 

tingaish  one  class  of  material  from  another,  or  to  prevent  the  reader  from 
getting  a  wholly  wrong  idea  of  the  character  of  the  book.  Most  students 
will  be  misled  in  this  respect  in  spite  of  the  candid  avowal  of  the  preface. 

Bat  even  if  every  one  were  to  read  the  preface  in  fall  and  understand 
thoroughly  the  composite  sources  of  the  work,  the  objections  to  the 
methods  adopted  in  making  it  cannot  be  regarded  as  satisfactorily  met. 
Both  Smith's  Dictionary  and  Seyffert's  Dictionary  are  recent,  both  in 
English,  both  prepared  at  a  large  expense  of  time,  labor  and  money,  and 
both  largely  dependent  upon  the  sales  in  the  American  market  for  their 
pecuniary  return.  It  may  very  well  happen  that  such  return  will  be 
lessened  by  the  new  dictionary,  and  the  question  naturally  arises  whether 
there  was  any  arrangement  with  the  English  publishers  by  which  this  use 
of  their  material  was  authorized.  There  is  no  hint  of  such  authorization 
in  the  preface.  If  it  was  obtained,  the  fact  should  have  been  stated  in  the 
interest  of  both  publishers  and  editor.  In  the  absence  of  such  a  statement 
no  one  can  justly  protest  if  the  reader  assumes  that  the  borrowed  matter 
was  taken  without  authority,  and  if  public  opinion  fails  to  justify  this 
method  of  compilation. 

It  may  be  justly  urged  that  there  is  a  great  mass  of  literary  material 
which  is,  in  substance  at  least,  open  to  the  free  use  of  all  men  without  the 
risk  of  criticism,  and  that  all  contributions  to  human  knowledge  made  by 
scholars  and  scientists  soon  become  merged  in  the  common  stock.  Per- 
haps the  matter  in  the  old  editions  of  Smith's  dictionaries  may  be  con- 
sidered as  belonging  to  that  class.  But  will  any  one  claim  that  six  years 
are  enough  for  the  completion  of  this  communistic  process  ?  It  may,  how- 
ever, be  contended  that  these  works  were  not  copyrighted  in  this  country, 
and  so  were  legally  open  to  republication  here  in  whole  or  in  part.  And 
that  is  true.  Whether  that  fact  changes  the  essential  character  of  the 
transaction  is  an  interesting  question  in  literary  ethics  upon  which  scholars 
are  likely  to  have  strong  opinions  one  way  or  the  other.  Certainly  there 
are  people  who  are  surprised  to  learn  that  the  attractive  literary  wares  of 
foreign  scholars  need  the  protection  of  the  law  to  prevent  their  being 
offered  for  sale  at  a  competing  establishment  in  our  own  country. 

It  is  not  the  purpose  of  this  article  to  present  any  exhaustive  examina- 
tion and  criticism  of  Harper's  Dictionary.  It  is  not  possible  within  these 
limits  to  review  half  a  dozen  important  works  of  reference,  but  that  is 
what  an  adequate  review  of  this  volume  would  mean.  But  a  few  points 
of  some  interest  may  be  noted. 

In  transferring  matter  from  other  works  it  is  inevitable  that  occasionally 
the  errors  and  questionable  statements  of  the  original  article  will  be 
included  as  well  as  its  excellences.  Only  the  most  searching  and  coopera- 
tive editorial  scrutiny  could  avoid  this.  And  it  will  not  be  surprising  if  it 
prove  that  this  volume  is  not  wholly  exempt  from  this  kind  of  criticism. 

In  the  article  Metallum  contributed  by  Professor  Percy  Gardner  to 
Smith's  Dictionary  of  Antiquities  the  following  passage  occurs  (not 
repeated,  however,  in  Harper) :  "  In  one  passage  of  the  Odyssey  (IX  391) 


Digitized  by  VjOOQ IC 


82  AMERICAN  JOURNAL    OF  PHILOLOGY. 

knowledge  is  shown  of  the  process  of  hardening  iron  by  repeated  plunging 
when  hot  in  water."  In  the  article  Aes  from  the  same  competent  hand, 
transferred  with  slight  verbal  changes  from  Smith  to  Harper,  may  also  be 
found  the  following  passage :  *'  The  abundance  of  copper  sufficiently 
accounts  for  its  general  use  among  the  ancients.  We  have  a  remarkable 
result  of  this  fact  in  the  use  of  ;to^'^f  ^nd  x^^^^^^^v  where  working  in 
iron  is  meant  (Hom.,  Od.  IX  391 ;  Aristot.,  Poet.  25)." 

These  two  passages  are  consistent  with  each  other  and  with  the  generally 
accepted  view  that  x^^^^i  was  used  not  merely  in  its  natural  sense  of  a 
coppersmith,  but  also  as  a  metal-worker  in  general  (^.^.  Od.  Ill  432,  where 
it  must  mean  a  goldsmith,  whatever  its  meaning  may  be  at  IX  391)-  ^^^ 
in  the  next  column  of  the  same  article  {Ags)  we  find  that  the  ancients 
**  seem  to  have  understood  the  art  of  hardening  it  [copper]  by  dipping  it  in 
water  and  exposure  to  air.  There  is  a  passage  even  in  Homer  which  is 
supposed  to  allude  to  this  process  (Od.  IX  391)." 

Superficially  at  least  there  seems  to  be  an  inconsistency  between  this 
passage  and  that  which  precedes.  If  x^^-i^C  is  used  in  Od.  IX  391,  **  where 
working  in  iron  is  meant,*'  can  it  also  be  used  in  the  same  passage  to  mean 
a  coppersmith  tempering  copper  tools  by  a  similar  process?  Certainly 
more  clearness  of  statement  was  needed.  The  average  student  who  con- 
sults this  article  will  fail  to  understand  how  the  same  passage  may  refer  to 
the  working  of  both  iron  and  copper,  and  it  would  have  been  well  if  the 
writer  had  cited  authorities.  Still  further,  it  would  have  been  well  if  the 
American  editor  had  found  space  for  the  new  evidence  which  has  converted 
a  supposition  in  Smith  into  a  certainty  in  Harper.  For  while  Smith 
cautiously  states  that  the  ancients  **seem  to  have  understood"  the  art  of 
hardening  copper.  Harper  says  positively  that  **  the  ancients  also  under- 
stood "  the  art. 

The  statement  (from  Smith's  Smaller  Classical  Dictionary)  in  the  article 
upon  Aristidet  that  he  was  recalled  from  exile  after  the  battle  of  Salamis, 
rather  than  before  it,  is,  to  say  the  least,  open  to  question. 

The  statement  (page  126),  taken  from  the  same  source,  that  Aristarchut 
** divided  the  Iliad  and  Odyssey  into  twenty-four  books  each,"  is  incon- 
sistent with  the  more  careful  statement  of  p.  837  that  Zenodotus  "is 
thought  to  have  been  the  first  to  divide  "  them  thus. 

The  statement  on  page  13,  taken  from  Seyffert,  that  the  Acropolis  is 
"abont  200  feet  in  height"  is  seriously  at  variance  with  the  statement  of 
page  149  that  the  hill  is  **  156  meters  high."  In  neither  case  is  it  stated 
whether  the  reckoning  is  from  the  sea-level  or  the  city  below.  The  dis- 
crepancy will  confuse  most  students. 

Mithridates  remains  son-in-law  of  Tigranes,  instead  of  father-in-law 
(p.  1584),  as  in  Smith's  Smaller  Classical  Dictionary  from  which  the  article 
is  taken.  (The  same  error  occurs  in  the  Anthon-Smith  and  even  in  the 
Marindin-Smith  of  1894.) 

Under  the  article  Ostracismus  (from  Seyffert)  it  is  stated  that  this 
measure  *'  was  introduced  at  Athens  in  B.  C.  509."  (Under  Clisthenes^  the 
date  is  B.  C.  508.)    It  would  have  been  more  exact  to  say  that  though 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES.  83 

ostracism  was  legalized  at  Athens  among  the  measures  of  Cleisthenes,  it 
was  first  applied  twenty  years  later  (Aristotle,  Ath.  Pol.  22). 

The  plan  of  Salamis  on  a  fairly  generous  scale  was  hardly  demanded  to 
illustrate  a  five-line  article  in  which  there  is  a  mere  mention  of  the  battle 
of  Salamis  without  any  details.  The  map  adopted,  however,  silently  com- 
mits the  editor  to  the  traditional  view  of  the  position  of  the  hostile  fleets 
within  the  narrow  confines  of  the  strait.  This  view  has  been  hardly  tenable 
since  the  publication  of  Professor  Goodwin's  paper  in  1885  (Vol.  I,  Papers 
of  the  American  School  at  Athens). 

The  value  of  the  Attic  talent  is  almost  as  uncertain  as  that  of  the  Amer- 
ican silver  dollar.  Under  EispAor a  it  is  just  |io8o;  under  Liiurgia  just 
I1180;  under  NumismatUs  " about  |iooo";  under  Talentum  **  about  |xi8o." 
Each  of  these  articles  is  borrowed  from  SeyfFert,  in  which  the  value  of  the 
talent  is  uniformly  given  as  ;^2oo,  or  **  about  ;^200." 

Under  Domus  (p.  537,  b)  may  be  found  the  following  statement :  *<  In  the 
palace  of  Odysseus  the  three  hundred  suitors  of  Penelope  feasted  in  "  the 
fUyapov,  This  sentence  is  taken,  with  the  rest  of  the  article,  from  Smith's 
Dictionary  of  Antiquities.  But  the  Homeric  account  (Od.  XVI  247  seqq.) 
represents  the  number  of  the  suitors  as  one  hundred  and  eight  with  ten 
retainers.    That  would  seem  to  be  enough. 

No  such  book  is  free  from  misprints  and  minor  defects.  A  few  may  be 
noted  :  Artios  for  Athos,  p.  16, 1. 1 ;  Ephesian  History  of  Xenophon  for  Epkf 
sian  History  of  Xenophon,  p.  lox,  ^ ;  the  AcheloQs,  instead  of  the  Alpheius, 
called  the  largest  river  of  the  Peloponnesus,  p.  109,  b\  Cleodctcus  for  Cleo^ 
dcieusy  p.  127,  a\  dpiratva  for  apffratva,  p.  186,  3;  prcufumium  (twice)  for 
praefurnium^  p.  192,  3,  top  ;  Wagen  for  JVdgen^  p.  200,  a;  (Iv.  30)  for  (lib. 
V.  30),  p.  219,  3,  bot;  Pergamun  ior  Pergamum^  p.  1065,^;  Mount  Vesula 
for  Visulus^  p.  1 1 54,  a;  the  omission  of  Argolis  from  the  list  of  districts  in 
the  Peloponnesus,  p.  1x95,  a\  or  for  of^  p.  1223,  3,  1.  14;  [B.  C]  296  for 
196,  p.  1430,  a,  1.  4 ;  ^e<a  for  ^eZa,  p.  1560,  a\  confusion  due  to  misplaced 
commas  in  the  article  Eisphora,  The  revised  spelling  Munichia  is  found 
under  Athenatt  elsewhere  Munychia.  Under  Artemis  is  found  the  cross- 
reference  Elethyia.  The  patient  seeker  will  finally  find  it  under  Ilithyia 
(to  which  one  article  is  devoted  in  the  body  of  the  book  and  another  in  the 
Appendix).  Under  Barathron  there  is  a  very  plausible  cross-reference  to 
Caedes^  a  title  not  to  be  found.  Perhaps  Ceadas  was  intended.  The  error 
by  which  the  Laocoon  (instead  of  Dirce)  group  is  attributed  to  Apollonius 
of  Tralles  had  already  been  corrected  in  the  list  of  corrigenda  in  Seyffert, 
but  it  is  reproduced  in  Harper  (p.  xoi). 

The  illustrations  in  Harper  are  abundant  and  of  a  much  higher  average 
quality  than  those  found  in  Smith  or  Seyffert.  In  many  cases  articles 
borrowed  from  those  sources  are  supplied  with  additional  as  well  as  better 
cuts.  Poor  illustrations  are  exceptional,  but  one  marked  instance  is  the 
cut  of  the  Vatican  Demosthenes.  The  cut  of  Tiryns  is  not  called  for  at 
page  1068,  especially  as  it  is  duplicated  in  its  proper  place,  page  1587.  A 
few  of  the  cuts  seem  to  be  mere  embellishments  of  the  book  rather  than 
illustrations  of  the  subjects.  Such  are  some  of  the  cuts  taken  from  modern 
paintings. 


Digitized  by  VjOOQ IC 


84  AMERICAN  JOURNAL    OF  PHILOLOGY. 

The  special  contributions  to  this  dictionary,  according  to  the  list  in  the 
preface,  number  somewhat  over  fifty  (counting  alphabetical  Abbreviations 
as  one  article).  They  cover  less  than  one  hundred  and  fifty  pages  of  the 
total  seventeen  hundred.  The  special  contributions  of  the  editor  are 
additional  to  this  amount,  but  no  list  of  them  is  given,  and  there  is  no  dis- 
tinction in  the  printing  between  them  and  other  matter,  but  his  influence 
may  be  traced  in  many  articles  known  to  be  mainly  borrowed,  and  often 
the  changes  and  additions  are  distinct  improvements.  Several  of  the 
special  contributions  are  from  the  highest  authority  in  their  special  fields 
and  are  models  of  what  such  articles  should  be.  In  one  or  two  cases  the 
writer  seems  to  have  allowed  himself,  or  to  have  been  allowed,  inadequate 
space  for  the  complete  presentation  of  an  important  topic. 

The  scope  of  Harper's  Dictionary  is  very  comprehensive.  It  aims  to 
include  under  a  single  alphabet  the  subjects  of  Archaeology,  Art,  Biography 
(including  notices  of  classical  scholars  and  philologians  of  the  recent 
centuries).  Geography,  History,  Language,  Literature,  Mythology,  etc.  It 
is  thus  a  cyclopaedia  rather  than  a  dictionary  in  its  scope,  and,  supplying, 
as  it  will,  the  place  of  several  books  of  reference,  it  will  prove  a  great  con- 
venience to  any  one  seeking  information  in  an  accessible  and  compact 
form.  The  student  will  find  in  it  a  wide  and  interesting  range  of  informa- 
tion, attractively  presented,  and  it  will  not  occur  to  him  to  be  fastidious  or 
even  inquisitive  concerning  the  sources  of  that  information.  Thus  the 
book  will  doubtless,  as  the  editor  hopes,  *'  be  found  at  least  to  have  done 
something  to  promote  the  comprehensive,  intelligent,  and  sympathetic 
study  of  classical  antiquity  "  in  this  country. 

WlLUAMS  COLX.BGB.  O.  M.   FERNALD. 


L'Inno  Omerico  a  Demetra.    Con  apparato  critico  scelto  e  un*  introduzione. 
Di  ViTTORio  PUNTONI.    Livomo,  Raffaello  Gittsti,  1 896.    8vo,  viii,  165  pp. 

It  is  unfortunate  that  the  work  of  which  the  Homeric  Hymns  at  present 
stand  most  in  need  is  also  that  which  is  most  difficult  to  perform.  There  is 
still  opportunity  for  something  in  the  way  of  illustration  and  interpretation, 
though  perhaps  not  very  much.  Dissection  is  always  possible;  here  every 
man  can  be  a  law  unto  himself  without  gainsaying,  and  establish  his  originality 
at  a  comparatively  small  expense.  The  real  struggle  comes  in  the  restoration 
of  the  text,  where  leader  after  leader  has  fallen  without  catching  even  a 
glimpse  of  the  Canaan  of  correctness.  Perhaps,  indeed,  we  have  come  to  a 
standstill  until  such  time  as  we  may  be  able  to  enrich  ourselves  here  also  from 
the  spoils  of  Egyptian  tombs. 

The  latest  editor  has  chosen  the  easier  way.  He  has  not  neglected  the  text, 
though  he  has  added  no  improvement  of  his  own,  scarcely  even  a  conjecture 
except  such  as  result  from  and  help  to  support  his  theories.  Illustration  and 
interpretation  he  has  not  attempted,  except  incidentally.  By  far  the  larger 
portion  of  his  book  consists  of  the  Introduction  of  134  pages,  in  which  his 
argument  concerning  the  origin  of  our  present  text  of  the  Hymn  to  Demeter 
is  set  forth  carefully  and  at  length,  though  the  author  admits  that  he  has  left 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES.  85 

mach  to  be  done  in  the  way  of  a  minute  criticism  of  each  separate  verse  of 
the  hymn. 

Signor  Puntoni  believes  in  his  own  theory,  and  has  presented  it  clearly  and 
emphatically.  This  theory  is  briefly  stated  at  pp.  2-3  as  follows :  The  tradi- 
tional text  of  the  Moscow  MS  results  from  the  enlargement  of  a  more  ancient 
hymn,  A,  by  the  insertion  of  a  considerable  number  of  fragments  of  two 
others,  B  and  C,  introduced  and  adapted  more  or  less  successfully  by  two 
revisers.  The  argument  of  A  was  the  rape  of  Persephone  and  the  consequent 
/^<C  of  Demeter ;  the  residence  of  Demeter  in  Eleusis  was  not  included.  B 
resembled  the  hymn  attributed  to  Pamphos  (Paus.  8.  37.  9;  9.  31.  9;  i.  39.  i ; 
I.  38.  3) ;  its  principal  content  was  the  irtvdo^^  not  the  frnviQ^  of  Demeter,  her 
mournful  wanderings  over  the  earth,  and  her  sojourn  in  the  house  of  Celeus ; 
the  foundation  of  her  temple  at  Eleusis,  and  the  establishment  of  the 
mysteries.  C  combined  the  motives  of  A  and  B  in  a  version  probably  not 
widely  different  from  that  followed  in  the  lost  Orphic  KdpTr  'Apiray^, 

Of  these  three  hymns  A  was  the  oldest,  C  the  youngest ;  and  the  difference 
in  age  is  marked  by  an  increasing  use  of  non -Homeric  words  and  forms. 
Puntoni  has  catalogued  these  at  pp.  112-^5,  and  finds  in  the  fragments  he 
assigns  to  each  hymn  the  following  ratios:  in  A,  22  non-Homeric  words 
and  forms  in,  100  verses ;  in  B,  32 ;  in  C,  42  (fractions  omitted).  In  the  55 
transitional  lines  which  he  attributes  to  the  revisers  he  finds  a  sufficiently 
high  proportion  of  40  such  words  and  forms  to  100  verses. 

At  pp.  111-12  Puntoni  gives  a  list  of  the  fragments  attributed  to  each  of  his 
three  hymns,  which  will  enable  any  one  who  cares  to  examine  his  theory  to 
take  a  rapid  survey  of  it,  and  therefore  may  be  repeated : 

A  (the  original  nucleus)  =  1-4+8-17  + 19-20  +  38-46***  +  62(?)-8i  +  87- 

90  +  305-335  +  337-351  +  370-394  +  404-413 + 433*^  +  441  -450  +  459-476 
+  483-484 + 486-489.    Totol,  1 65  verses. 

Fragments  of  B :  5-7,  22-23  +  30-37,  357  +  359-3^9,  82-86,  92-188  (to  iroat) 
+  212-301,  and  probably  478-482  and  490-491  +  (?)494.    Total,  225  verses. 

Fragments  of  C :  24-25  +  27-29, 47-54  +  (?)*58. 191-199+  202-211, 395-399 
+  401-403  +  434-440.    Total ,  48  verses. 

Principal  amplifications:  352-356,  414-432,  451-458. 

The  reasons  for  the  assignment  of  each  of  these  passages  to  A,  B,  C,  or  to 
the  contaminator  himself,  are  fully  and  clearly  stated,  and  in  most  cases  rest 
upon  inconsistencies  or  incoherences  of  more  or  less  weight.  Occasionally 
the  criticisms  are  captious  and  unreal,  as  at  pp.  77-8,  where  it  is  said  that, 
since  Demeter  had  gone  into  the  temple  to  stay  {Ivda  KoBe^oftivrf . . .  fiifive  303 
f.),  the  statement  that  Iris  found  her  in  the  temple  (3x9)  is  out  of  place  (!).  It 
is  but  fair  to  say  that  this  is  the  worst  instance  of  a  perverted  microscopic 
criticism  in  the  entire  introduction. 

A  detailed  examination  of  the  arguments  here  advanced  is,  in  the  space  of  a 
review,  obviously  impossible.  As  it  is  not  customary  for  any  two  investigators 
to  agree  in  matters  of  this  kind,  it  would  be  necessary  either  to  refute  Puntoni^s 
fundamental  doctrine  of  contamination  and  interpolation,  or  else  to  propose  a 
wholly  new  analysis.  It  is  a  matter  of  curiosity,  if  scarcely  of  critical  value, 
to  remember  how  Gottfried  Hermann  took  up  Creuzer's  theory  of  five-line 
stanzas  in  the  Hymn  to  the  Pythian  Apollo,  but  made  an  entirely  new  set, 


Digitized  by  VjOOQ IC 


86  AMERICAN  JOURNAL    OF  PHILOLOGY. 

which  doubtless  pleased  him  better  (Jahrbb.  53.  355-73).  That  sort  of  thing 
can  be  done  indefinitely.  On  an  a  priori  consideration  of  the  matter,  it  seems 
not  at  all  improbable  that  a  hymn  which  was  used  for  ritualistic  purposes 
perhaps  for  centuries  should  have  undergone  extensive  alterations  and  enlarge- 
ments. Not  is  it  at  all  impossible  that  an  interpolator  should  have  chosen  the 
method  of  contamination.  There  are  real  difficulties  and  incoherences  in  the 
text  of  the  Hymn  to  Demeter,  as  we  now  have  it,  such  as  ayytlJkowia  53,  the 
uncertain  and  shifting  rdles  of  Hecate  and  Helios  as  informer,  etc.  Some  of 
these  may  be  accounted  for  by  the  fact  that  we  have  the  hymn  only  in  a  single 
manuscript,  which  Cobet  regarded  as  one  of  the  worst  known,  though  it  has 
latterly  risen  in  esteem.  Others  cannot  be  so  explained.  But  the  theory  that 
the  hymn  can  be  analyzed  into  passages  definitely  referable  to  a  number  of 
different  pre-existing  poems  will  probably  be  received  with  skepticism.  At 
pp.  71-2  Puntoni  himself  admits  the  possibility  that  an  original  poet  might 
have  felt  himself  bound  to  introduce  and  harmonize  varying  versions  of  the 
myth. 

In  his  critical  notes  to  the  text  Puntoni  has  in  one  respect  rendered  a 
valuable  service  to  the  future  editor  of  the  Hymns,  if  not  to  the  general 
scholar.  Though  on  his  title-page  he  has  cautiously  promised  only  an 
"  apparato  critico  scelto,"  he  has  in  reality  recorded  nearly  all  conjectures  and 
hypotheses  advanced  concerning  the  hymn  from  Ruhnken*s  day  down.  Any- 
thing so  complete  would  be  hard  to  find  elsewhere,  and  everything  is  noted 
with  admirable  conciseness  and  lucidity.  Frequent  protests  have  been 
uttered  in  these  days  against  the  perpetuation  of  absurd  or  improbable  conjec- 
tures ;  but,  after  all,  every  new  investigator  is  obliged  at  least  to  read  what 
his  predecessors  have  written,  and  in  the  41  pages  of  Puntoni*s  text  and 
commentary  he  will  find  nearly  everything,  and  save  himself  a  deal  of  note- 
taking.  Puntoni,  however,  is  ignorant  of  Brunck,  Peppmtlller,  and  some  of 
the  recent  English  and  Irish  contributions  inspired  by  the  Goodwin-Allen 
edition. 

— iroA^i  yap  66bv  irp^covatv  dSlrai^ 

rw  ol  fihf  Kwcd  froAAd  fxefia&reCt  ol  di  ft&k*  to6X&, 

^Tuaiv  ;faA«rdv  6i  dafjfievai  ktrnv  eKoarov, 

In  the  last-mentioned  monumental  edition,  which  might  well  have  been 
somewhat  less  monumental  in  size  and  price,  we  have  the  latest  results  of  a 
conservative  textual  criticism.  Puntoni  has  due  reverence  for  this  work,  and 
except  for  a  less  liberal  punctuation,  and  the  absence  of  paragraphing,  differs 
from  it  in  not  many  important  passages.    The  principal  ones  are  as  follows: 

10.    The  editor  keeps,  with  Gemoll,  the  MS  aipof  rdre. 

46.    The  MS  olf  oiuvav  tic  ^  is  retained. 

87.  Puntoni  prints  ^furavaierai,  but  believes  that  fieravaierdetv  was  the 
reading  of  A,  8a-6  belonging  to  B.  Goodwin  reads,  after  Voss,  fieravaierdei, 
Valckenaer  proposed  fiira  vaier&eiv,  and  the  presence  of  82-6  does  not 
necessarily  exclude  the  infinitive. 

115.  irtXvp  Hermann,  Puntoni ;  irtXva^  M ;  TriXvoffoi  Voss,  Goodwin.  iriXvaoai, 
which  has  also  Cobet's  authority,  is  preferable. 

118.    In  the  confusion  regarding  the  daughters  of  Celeus,  Puntoni  keeps  the 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES.  8/ 

137.  Pantoni  does  not,  like  Hermann,  Goodwin,  and  others,  see  a  lacuna 
after  this  verse,  nor  does  it  appear  especially  necessary.  Hermann's  other 
suggestion,  delirvov  cf  rfprivovro,  or  Voss's  (T  hrhvovro,  restores  the  connection, 
and  an  intermediate  action  does  not  seem  called  for.  Puntoni,  however, 
follows  M. 

137-8.  Puntoni  keeps  the  MS  efie  &  aif  olicreipaTe  .  .  .  riuv  irpd^  66ftaff 
Uufioij  which  is  of  course  impossible.    But  what  is  to  be  done  with  it? 

an.  SaitK  kveK£v  is  retained,  and  a  lacuna  marked  after  the  verse.  Voss's 
Mptf^  adopted  by  Goodwin,  is  not  convincing. 

227.  dphlfu'  KxA)  ftiv,  the  reading  of  M,  is  given.  Hermann's  dphJHu  or 
Goodwin's  Opk-^ltowf  would  be  better.  It  is  needless  to  say  that  Puntoni  has 
done  nothing  for  228-9. 

236.    Puntoni  does. not  mark  a  lacuna. 

269.  Puntoni  gives  adavdroic  Ovrrroiai  r'  bvtiap  Koi  x^Pf^  rirvKTai,  Goodwin 
rightly  places  a  mark  of  desperation  at  bveiap.  The  animated  controversy  in 
recent  numbers  of  the  Classical  Review  has  left  matters  where  they  stood.  It 
is  not  clear  how  Puntoni  reads  bvetap, 

328.  TifA&q  ff  ag  Ksv  i?joiTo  fisi'  adavdroiaiv  iXiaOcu,  Puntoni  fails  to  observe 
in  his  critical  note  that  the  reading  of  M  is  ag  k'  ediXotro.  kbv  iXotro  is 
Hermann's. 

344-5.  The  reading  of  M,  i&  kif  arXfymv  |  ipyotc  deuv  ficucdpuv  fUfrieeTO 
PovXy^  is  repeated  with  two  daggers,  as  by  Goodwin.  I  should  almost  venture 
to  introduce  Btlcheler's  ctt  dr^^otc  |  ipyotg  <ola>  6eov  fuucdpuv  furriaaro  Povkif 
into  the  text.  It  gives  exactly  the  sense  which  we  look  for,  with  less  violent 
alterations  than  any  other  conjectures  of  satisfactory  meaning.  The  purport 
of  the  passage  is  clear:  Persephone  was  suffering  (i)  from  longing  for  her 
mother,  (2)  at  the  intolerable  treatment  inflicted  upon  her  by  the  will  of  the 
gods.  V.  345  lacks  either  two  or  three  syllables,  and  they  may  be  supplied 
either  by  the  insertion  of  a  word,  or  by  the  substitution  of  a  longer  word  for  a 
shorter.  The  former  method  commends  itself.  Furthermore,  fpyoTg  ^eav 
suggests  that  the  two  syllables  have  been  lost  here.  A  relative  is  necessary 
to  give  the  meaning  expressed  above ;  and  dia  answers  the  purpose  not  only 
metrically,  but  admirably  in  sense.    The  other  changes  are  not  difficult. 

364.  Puntoni  finds  the  MS  hGdS'  lovaa  appropriate  to  the  position  he  gives 
to  vv.  359-69  in  B. 

428.    oowep  Kpdioov  is  kept.    Goodwin  adopts  the  inreipoxov  of  Voss. 

490-95.  Puntoni  reads  the  sing,  throughout,  though,  as  these  verses  are  in 
his  view  patchwork,  it  is  not  a  matter  of  great  importance. 

The  editor  has  done  well  in  making  the  numbering  of  his  lines  agree  with 
the  actual  number  of  verses.  There  was  no  reason  why  the  Goodwin-Allen 
edition  should  follow  the  erroneous  numbering  of  the  manuscript. 

Puntoni  justly  says  that  Goodwin's  supplement  of  the  lacuna  in  w.  387-99 
is  the  best  that  has  been  offered,  though  it  does  not  entirely  square  with  his 
theories.  Puntoni  himself  is  very  shy  of  attempting  Greek  composition. 
What  he  can  do  in  that  line  is  shown  by  his  attempt  on  p.  77 :  'RvfioXirift  r*  &p* 
afivfiovt  Kal  KeXe^  Ao^;t9  ^^  (•)•  '^his  lack  of  feeling  for  the  caesura  has 
allowed  him  to  print  the  reading  of  M  in  v.  203 :  troAAa  trapaoKCmrovaa  Tpi^ffaro 
ir&rviav  ayvffv^  where  other  editors  rightly  change,  with  Voss,  to  rcapcujKCinrovo' 
irpiffMiTO, 


Digitized  by  VjOOQ IC 


I 


88  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Whatever  the  value  of  his  results,  the  editor  has  worked  at  his  material 
faithfully  and  conscientiously.  His  scrupulousness  has  extended  even  to  the 
proof-reading,  affording  a  striking  contrast  to  the  slovenliness  of  GemoU, 
whose  otherwise  valuable  edition  is  as  full  of  misprints  as  a  country  news- 
paper. Few  typographical  errors  have  been  noticed  in  Puntoni :  h^pa  appears 
in  the  text,  v.  io6,  as  Zi>^pa,  A  wrong  citation,  copied  from  Hermann  without 
correction  on  p.  54,  might  arouse  suspicion ;  but  this  suspicion  is  not  elsewhere 
confirmed. 

Charles  J.  Goodwin. 


Handbooks  on  the  History  of  Religions.  Edited  by  Morris  Jastrow,  Jr. 
Vol.  I.  The  Religions  of  India,  by  Edward  Washburn  Hopkins. 
Boston,  Ginn  &  Co.,  1895. 

In  a  prefatory  note  to  this  first  volume,  Professor  Jastrow  outlines  the  plan 
of  the  series  of  Handbooks  in  which  he  desires  to  present  to  students  of  the 
history  of  religions  the  results  of  the  scholarly  activity  of  recent  years  in  the 
several  departments  concerned.  Each  volume  is  to  include  an  account  of 
the  sources  and  the  method  of  study,  and  a  chapter  on  the  land  and  people, 
**  presenting  those  ethnographical  and  geographical  considerations,  together 
with  a  brief  historical  sketch  of  the  people  in  question,  so  essential  to  an 
understanding  of  intellectual  and  religious  life  ever3rwhere."  The  main 
portion  of  the  work  is  to  present  in  greater  detail  a  description  of  the  beliefs — 
the  pantheon,  the  relation  to  the  gods,  views  of  life  and  death — the  official 
rites  and  popular  customs,  the  religious  literature  and  architecture,  followed 
by  a  general  estimate  of  the  religion,  its  history  and  the  relation  it  bears  to 
others.  In  each  instance  a  full  bibliography,  an  index  and  the  necessary 
maps  and  illustrations  will  be  provided. 

It  is  much  to  be  regretted  that  Professor  Hopkins  has  not  allowed  himself 
the  space  necessary  for  the  generous  treatment  suggested  by  the  editor.  The 
length  of  time  through  which  we  can  trace  the  course  of  religious  ideas  and 
practices  in  India  and  the  manifold  character  of  the  native  developments  give 
a  special  interest  and  importance  to  the  study  of  the  Indian  religions,  the  con- 
sideration of  which  might  well  claim  in  this  series  more  than  a  single  volume. 
Within  these  narrow  limits,  Barth,  with  whose  admirable  sketch  the  present 
work  challenges  comparison,  has  succeeded  only  by  careful  abstinence  from 
the  discussion  of  all  matters  not  of  the  first  importance  and  by  rigorous  exclu- 
sion of  illustrative  citations  from  his  text.  Professor  Hopkins,  however, 
writing  for  "students  ignorant  of  Sanskrit  who  yet  desire  independently  to 
examine  and  to  make  their  own  the  very  words  of  the  Hindu  sages,"  desires 
not  merely  to  summarize  but  **  to  open  up  the  religions  of  India  from  within 
and  in  orderly  succession  to  explain  them  as  they  display  themselves." 

The  order  of  treatment,  accordingly,  follows  what  the  author  conceives  to 
have  been  the  (order  of  development.  The  difficulty  of  the  undertaking  is 
apparent.  •*  For  [none  of  the  native  religious  works  has  one  a  certain  date. 
Nor  is  there  for  any  one  of  the  earlier  compositions  the  certainty  that  it 
belongs,  as  a  whole,  to  any  one  time.  The  Rig  Veda  was  composed  by  suc- 
cessive generations ;  the  Atharvan  represents  different  ages ;  each  Brahmana 


Digitized  by  VjOOQ IC 


appears  to  b< 
have  been  in 
epic  is  the  ¥ 
represent  co] 
each  individi 
case  of  the  ] 
a  quo,  and  ev 
First  with 
assign  the  composition  of  the  hymns  of  the  Rig- Veda  ?  Are  they  the  work  of 
priests,  and  were  they  composed  for  sacrificial  purposes?  Or  were  they  made 
"independently  of  any  ritual,  as  their  own  excuse  for  being"?  Professor 
Hopkins  warns  us  that  **  the  Rig  Veda  is  not  a  homogeneous  whole.  It  is  a 
work  which  successive  generations  have  produced,  and  in  which  are  represented 
different  views  of  local  or  sectarian  origin ;  while  the  hymns  from  a  literary 
point  of  view  are  of  varying  value.  The  latter  is  a  fact  which  has  been 
ignored  frequently,  but  it  is  more  important  than  any  other."  **A  large 
number  of  hymns  are  formal,  conventional  and  mechanical  in  expression," 
and  **  it  may  be  argued  with  plausibility  that  these  were  composed  to  serve 
the  purpose  of  an  established  cult"  ;^  but  in  others  is  found  "  poetry,  not  great 
poetry  perhaps,  but  certainly  not  ground  out  to  order,  as  some  of  the  hymns 
appear  to  have  been."  Mechanical  hymns,  then,  are  late.  **  It  must  not  be 
forgotten  that  the  ritual,  as  it  is  known  in  the  Brahmanas,  without  the  slightest 
doubt,  from  the  point  of  view  of  language,  social  conditions  and  theology, 
represents  an  age  that  is  very  different  to  that  illustrated  by  the  mass  of 
the  hymns.  Such  hymns,  therefore,  and  only  such  as  can  be  proved  to  have  a 
ritualistic  setting,  can  be  referred  to  a  ritualistic  age.  There  is  no  convincing 
reason  why  one  should  not  take  the  fully  justified  view  that  some  of  the  hymns 
represent  a  freer  and  more  natural  (less  priest-bound)  age,  as  they  represent  a 
spirit  freer  and  less  mechanical  than  that  of  other  hymns."  Elsewhere,  how- 
ever, the  existence  of  priestly  families  and  of  a  litany  prepared  for  the  warrior 
class  by  priests  is  recognized  in  the  earliest  period.  And  in  describing  the 
several  gods  of  the  Rig- Veda  the  author  refuses  to  adopt  the  method  suggested 
by  the  distinctions  to  which  he  has  attached  such  importance.  **After  what 
has  been  said  in  the  introductory  chapter  concerning  the  necessity  of  distin- 
guishing between  good  and  bad  poetry,  it  may  be  regarded  as  incumbent  upon 
us  to  seek  to  make  such  a  division  of  the  hymns  as  shall  illustrate  our  words. 
But  we  shall  not  attempt  to  do  this  here,  because  the  distinction  between  late 
mechanical  and  poetic  hymns  is  either  very  evident,  and  it  would  be  super- 
fluous to  burden  the  pages  with  the  trash  contained  in  the  former,  or  the  dis- 
tinction is  one  liable  to  reversion  at  the  hands  of  those  whose  judgment  differs 
from  ours,  for  there  are,  of  course,  some  hymns  that  to  one  may  seem  poetical 
and  to  another  artificial.  Moreover,  we  admit  that  hymns  of  true  feeling 
may  be  composed  late  as  well  as  early,  while  as  to  beauty  of  style  the  chances 

^  Again  we  are  told  "  Indra,  most  honored  with  Soma,  and  Agni,  most  closely  connected 
with  the  execution  of  sacrifice,  not  only  receive  the  most  hymns,  but  these  hymns  are«  for  the 
most  part,  palpably  made  for  ritualistic  purposes  ...  In  every  family  book,  besides  this 
baksheesh  verse,  occur  the  older  purer  hymns  that  have  been  retained  af^er  th€  worship/or 
which  thty  were  composed  had  become  changed  into  a  trite  making  of  phrases."  Is  a  con- 
trast intended  here  also  between  worship  and  ritual? 


Digitized  by  VjOOQ IC 


90  AMERICAN  JOURNAL   OF  PHILOLOGY, 

are  that  the  best  literary  production  will  be  found  among  the  latest  rather 
than  among  the  earliest  hymns.  It  would  indeed  be  admissible,  if  one  had 
any  certainty  in  regard  to  the  age  of  the  different  parts  of  the  Rig  Veda,  simply 
to  divide  the  hymns  into  early,  middle  and  late,  as  they  are  sometimes  divided 
in  philological  works,  but  here  one  rests  on  the  weakest  of  all  supports  for 
*  historical  judgment,  a  linguistic  and  metrical  basis,  when  one  is  ignorant  alike 
of  what  may  have  been  accomplished  by  imitation,  and  of  the  work  of  those 
later  priests  who  remade  the  poems  of  their  ancestors." 

Throughout  the  discussion  it  is  assumed  that  the  linguistic  differences  between 
mantra  and  brahmana  are  wholly  due  to  difference  of  time  and  not  to  the 
varying  usage  of  literary  tradition.  In  the  Brahmanas,  it  is  remarked,  **  religion 
has  apparently  become  a  form,  in  some  regards  it  is  a  farce " ;  "  the  whole 
moral  atmosphere  is  now  surcharged  with  hocus-pocus,  mysticism,  religiosity, 
instead  of  the  cheerful,  real  religion,  which,  however  formal,  is  the  soul  of  the 
Rik " ;  and  this  is  "  the  only  new  literature  which  centuries  have  to  show.'* 
Yet  in  the  same  chapter  there  is  a  passing  recognition  of  the  limitations  of 
liturgical  works.  Again  in  the  exposition  of  the  law-books,  documents  of  a 
period  subsequent  to  the  pantheism  of  the  Upanishads  (itself  later  than 
Brahmanic  formalism),  we  are  told  that  "  there  is  a  reversion  to  Vedic  belief ; 
or  rather  not  a  reversion,  but  here  one  sees  again,  through  the  froth  of  rites 
and  the  murk  of  philosophy,  the  understream  of  faith  that  still  flows  from  the 
old  fount,  if  somewhat  discolored,  and  waters  the  hearts  of  the  people." 

Attention  has  already  been  directed  to  the  author's  inability  to  conceive 
that  priests  interested  in  the  details  of  sacrifice  could  produce  anything  better 
than  **  mechanical "  poetry.  He  also  hesitates  to  credit  them  with  "  a  devo- 
tional spirit  that  gave  voice  to  genuine  feeling.'*  This  personal  lack  of 
sympathy  with  religious  ritual  has  perhaps  contributed  to  Professor  Hopkins* 
success  in  his  exposition  of  early  Buddhism,  altogether  the  most  attractive 
chapter  in  the  book,  but  none  the  less  must  be  our  regret  that  the  whole  body 
of  usages,  the  significance  of  which  is  made  so  evident  in  Oldenberg's  Religion 
des  Veda,  has  been  passed  by  with  the  simple  remark  that  **  the  sacrifice  is 
but  show;  sjrmbolism  without  folk-lore,  only  the  imbecile  imaginings  of  a 
daft  mysticism,  is  the  soul  of  it :  and  its  outer  form  is  a  certain  number  of 
formulae,  mechanical  movements,  oblations  and  slaughterings." 

The  failure  to  co-ordinate  the  evidence  of  the  earlier  texts  has  resulted  in 
a  picture  of  ancient  conditions  inherently  improbable  and  inconsistent  with 
the  subsequent  persistence  of  belief  in  India.  The  theory  of  "  Hindu  influence 
on  the  Aryan  mind"  beginning  in  the  late  Vedic  period  is  suggested,  but 
nowhere  seriously  supported. 

Apart  from  such  considerations  of  method  in  the  earlier  chaptevs,  the  book 
deserves  generous  praise.  The  wide  range  of  the  author's  reading  in  the 
native  literature  and  his  extensive  acquaintance  with  the  work  of  Western 
scholars  are  everywhere  apparent.  From  his  own  more  special  studies  he  has 
contributed  not  a  little.  Particular  mention  must  be  made  of  the  excellent 
chapters  on  the  epic.  Observations,  too,  such  as  are  made  regarding  the 
general  character  and  the  position  in  the  Rig-Veda  of  the  hymns  in  which  the 
worship  of  certain  deities  or  the  expression  of  certain  ideas  especially  appears, 
are  welcome  to  all. 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES.  9 1 

Two  other  features  of  the  book  deserve  mention  here :  the  survey  of  the 

religious  traits  of  the  wild  tribes,  an  interesting  supplement  to  the  discussion 

of  the  Indo-Aryan  religions,  and  the  bibliography,  which  is  well  arranged, 

and  will  call  forth  the  thanks  of  many  students. 

A.  W.  Stratton. 


Vemer  Dahlerup :  Nekrolog  Over  Karl  Vemer.    Arkiv  for  Nordisk  Filologi. 
New  Series,  Vol.  IX.  Part  3. 

Not  a  few  great  talents  have  been  known  as  men  of  one  book.  Karl  Vemer 
is  probably  the  only  scholar  of  distinguished  ability  that  is  generally  known 
as  a  man  of  one  article.  Although  the  name  of  Vemer  is  a  household  word 
among  all  students  of  language,  probably  few  in  this  country  have  any  knowl- 
edge of  the  life  and  character  of  the  man  that  bore  it.  For  this  reason,  if  for 
no  other,  a  brief  account  of  some  of  the  salient  features  in  the  career  of  this 
so  greatly  lamented  Danish  scholar  must  be  of  interest  to  English  readers. 
Additional  interest  is  given  to  the  article  on  which  this  review  is  based  by  the 
fact  that  it  is  written  by  a  Dane,  a  friend  of  Vemer^s,  and  one  in  every  way 
competent  to  judge  of  the  significance  for  linguistic  science  of  Verner's  work. 

Karl  Adolph  Vemer  was  bom  in  Aarhus,  Jutland,  Denmark,  March  7, 1846. 
Even  before  entering  the  University  in  1864  he  had  shown  an  interest  in 
the  study  of  language,  his  attention  having  been  drawn  in  that  direction  by 
reading  the  life  of  his  great  countryman  Rask.  While  at  the  University  he 
devoted  himself  chiefly  to  the  study  of  the  Oriental,  the  Germanic  and  the 
Slavic  languages.  In  the  beginning  of  his  course  he  had  intended  to  pursue 
the  classics,  and  in  spite  of  his  later  investigation  of  the  so-called  modem 
languages,  he  always  retained  a  vivid  interest  in  the  older  tongues. 

After  serving  in  the  army,  Vemer  went  in  December,  1871,  to  Russia,  where 
he  remained  almost  a  year,  learning  to  speak  Russian,  and  pursuing  his  studies 
with  great  zeal.  On  his  retum  his  friends  induced  him  with  difficulty  to  take 
the  master^s  examination,  for  which  he  had  the  utmost  dread.  The  next  year 
and  a  half  were  spent  in  his  native  town,  his  health  not  permitting  him  to 
engage  in  any  regular  occupation.  During  this  enforced  vacation  he  wrote 
his  first  scientific  treatise, '*  Nogle  Raskiana"  (1874).  At  the  same  time  he 
carefully  investigated  accent  in  the  Slavic  languages  and  in  Danish,  and  out- 
lined the  changes  in  Danish  pronunciation  from  Holberg*s  time  to  the  present 
day,  only  the  main  results  of  which  were  published. 

The  account  of  the  development  in  his  mind  of  the  law  that  was  destined  to 
make  him  famous  is  so  interesting  that  nothing  but  a  full  translation  of  it  will 
suffice.  "  According  to  a  verbal  account  of  Vemer*s  (repeated  to  Dahlerup  by 
Hoffory  and  endorsed  by  Vemer),  he  happened  one  morning  on  getting  up  to 
reflect  that  it  was  strange  that  the  Gothic  words /a/ot  and  ^fvpar  had  different 
consonants  after  the  root  vowel.  As  he  was  just  then  engaged  in  studying 
accent,  it  was  natural  for  him  to  seek  the  explanation  in  this  direction.  He 
examined  the  conditions  in  Sanskrit  and  found  there  ^7i/r  and  Mr<F/iir.  He 
had  discovered  the  clue,  which  he  quickly  followed  out."  After  briefly 
explaining  the  law,  Dahlerap  adds :  **  Vemer  finished  his  epoch-making  treatise 


Digitized  by  VjOOQ IC 


92  AMERICAN  JOURNAL    OF  PHILOLOGY. 

in  the  spring  of  1875,  and  sent  it  in  a  letter  dated  May  i  to  Vilhelm  Thomsen. 
Shortly  before  this  he  had  learned  that  a  petition  for  a  travelling  stipendium 
had  been  granted  him.  On  his  way  to  the  continent  he  remained  a  few  weeks 
in  Copenhagen.  Vilhelm  Thomsen  had  immediately  on  receiving  his  letter 
encouraged  Vemer  to  print  his  paper,  which  he  said  would  overturn  many 
accepted  views.  During  his  stay  in  Copenhagen,  Vemer  wrote  his  treatise  in 
German  and  sent  it  to  Adolph  Kuhn."  The  results  of  the  publication  of  the 
Ausnahtne  in  the  following  year  are  too  familiar  to  require  telling  here. 

Surprise  has  often  been  expressed  that  Vemer,  the  most  famous  philologist 
in  Denmark,  should  have  had  so  comparatively  insignificant  a  career,  and  his 
fatherland  has  been  reproached  by  those  unacquainted  with  the  facts  for 
having  neglected  so  brilliant  a  son.  As  a  matter  of  fact,  Vemer  was 
thoroughly  appreciated  at  home,  and  had  it  not  been  for  the  encouragement 
of  his  many  Danish  friends  he  would  probably  have  remained  in  utter 
obscurity.  Strange  as  it  may  sound,  the  discoverer  of  Verner^s  Law  con- 
sidered himself  a  mere  amateur,  a  dilettante  in  Germanic  philology.  When 
in  1876  Wilh.  Scherer  offered  to  obtain  for  him  a  professorship  in  Germany, 
he  declined  on  the  ground  that  he  was  not  sufficiently  developed  for  such  a 
position,  and  accepted  instead  a  vacancy  at  the  Halle  library  as  assistant 
librarian.  Two  years  later  he  refused  a  professorship  at  Gratz,  and  it  is 
believed  also  at  Strassburg,  contenting  himself  with  a  promotion  at  the 
library.  It  was  only  with  great  difficulty  that  Scherer  induced  him  to  apply 
for  the  Bopp  prize,  which  was  bestowed  on  him  in  1 877. 

On  the  death  of  his  old  teacher  of  the  Slavic  languages,  C.  W.  Smith,  in 
1 881,  Vemer,  again  only  after  persistent  urging  by  friends,  applied  for  the 
vacancy,  and  in  August,  1882,  he  was  appointed  Docent  in  the  Slavic 
Languages  and  Literatures,  six  years  later  being  promoted  to  Professor 
Extraordinarius.  A  year  before  this  promotion  he  was  granted  an  honorary 
degree  by  the  University  of  Heidelberg,  in  1888  he  was  elected  a  member  of 
the  Danish  Academy  of  Sciences,  and  in  1892  he  was  decorated  with  the  order 
of  the  Danebrog. 

Even  stranger  apparently  than  Vemer's  modest  career  was  his  comparative 
unproductiveness.  The  explanation  of  this  too  is  to  be  found  in  his  character. 
In  a  letter  to  a  friend,  Vemer  once  said, '  I  have  inclination  to  enjoy,  but  not 
to  produce.'  (Jeg  har  tilbOjelighed  til  at  nyde,  men  utilbOjelighed  til  at  yde.) 
His  enjoyment,  it  should  be  added,  was  of  a  strictly  intellectual  character. 
Like  many  men  of  genius,  too,  his  interest  was  confined  to  the  intellectual 
process ;  with  the  practical  result  he  was  not  in  the  least  concerned.  After 
having  made  a  discovery  he  seemed  to  lack  any  desire  to  put  it  in  proper 
shape  for  presentation.  His  innate  modesty,  too,  had  much  to  do  with  his 
unwillingness  to  publish  the  results  of  his  investigations.  Still  another  reason 
is  found  in  his  extreme  conscientiousness.  Although  after  his  appointment  in 
Copenhagen  his  only  publications  were  two  short  articles  in  German  journals 
and  a  number  of  short  articles  on  Slavic  subjects  in  Salmonsen's  Encyclopedia, 
Vemer  was  one  of  the  busiest  members  of  the  faculty.  His  conscientiousness 
in  preparing  and  attending  lectures  was  almost  painful.  He  practically 
rewrote  the  Slavic  grammars  used  by  his  classes,  prepared  exhaustive  original 
treatises  on  Russian  accent  and  other  subjects,  and  all  for  the  exclusive  benefit 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES.  93 

of  only  one  or  two  students  at  a  time.  Hr.  Dahlenip  states  that  Verner  left 
an  enormous  amount  of  unpublished  material  in  this  and  in  other  directions. 
It  would  be  a  matter  of  interest  to  learn  if  any  of  this  valuable  matter  will  be 
published. 

Hr.  Dahlerup  closes  his  sympathetic  and  illuminating  study  as  follows: 
"  Not  only  will  Karl  Verner  be  honored  as  one  of  the  foremost  linguists  that 
the  North  has  produced,  but  his  great  personal  amiability  will  be  remembered 
by  the  many  friends  who  with  sorrow  have  learned  of  his  early  death." 

Daniel  Kilham  Dodge. 


II  Processo  di  Verre ;  un  capitolo  di  storia  romana.  Di  Ettore  Ciccotti. 
Milano,  edito  a  cura  dell*  an  tore,  1895.    240  pp.    L.  3.50. 

M.  Tullii  Ciceronis  Oratio  in  Verrem  De  Signis,  Publiee  avec  une  Introduc- 
tion et  un  Commentaire  explicatif  par  Henri  Bornecque.  Paris,  Colin 
et  Cie.,  1896.    176  pp.    Fr.  1.50. 

The  purpose  of  Prof.  Ciccotti's  book  is  mainly  historical.  By  means  of  a 
wide  and  careful  study  of  the  original  documents  he  has  tried  to  give  a  faithful 
picture  of  Verres  and  his  times. 

The  subject  of  the  first  chapter  (pp.  3-21)  is  well  suggested  by  Juvenal's 
vivid  line,  VI  293,  which  serves  as  its  motto,  Luxuria  incuhdt  victumque  ulci- 
seitur  orbem.  It  describes  the  great  changes  in  the  public  and  private  life  of 
the  Romans  which  followed  the  extension  of  their  sway  beyond  the  natural 
boundaries  of  Italy  and  the  establishment  of  their  provincial  system.  An 
influx  of  wealth,  an  increase  of  luxury,  electoral  corruption  and  plundering  of 
the  provinces  were  among  the  first-fruits  of  that  system.  The  provincial 
governors  had  almost  unlimited  powers  and  the  home  government  provided 
no  adequate  check  on  their  excesses;  few  of  them  showed  the  scrupulous 
honesty  of  a  Piso  Frugi  or  were  content  with  the  Jeffersonian  simplicity  of  a 
Cato  Censor.  The  second  chapter  (pp.  21-37)  gives  a  brief  description  of  the 
way  in  which  the  Roman  provinces  were  governed  {gttan  quaedam  praedia 
poptiU  Romani)  and  of  the  arrangements  made  for  collecting  revenue  from 
them.  Chapter  III  (pp.  37-57)  gives  an  account  of  the  various  leges  de  repe- 
tundis^  from  the  lex  Calpumia  of  B.  C.  149  to  the  lex  Cornelia  under  which 
Verres  was  indicted.  Chapter  IV  (pp.  57-79)  treats  of  the  conquest  of  Sicily 
(insula  Cereris)  and  of  the  various  relations  in  which  the  Sicilian  communities 
stood  to  Rome ;  also,  of  the  economic  conditions  of  Sicily  and  the  powers  of 
its  governors.  Chapters  V  and  VI  (pp.  79-106  and  107-44)  give  an  account, 
closely  following  that  of  Cicero,  of  the  earlier  career  of  Verres  (homo  amens  ae 
perditus)  and  of  his  doings  after  he  was  sent  into  Sicily  {quasi  in  praedam). 
The  last  chapter  (pp.  144-235)  describes  the  trial  of  Verres. 

It  is  evident  from  In  Verrem,  I  37,  that  Verres  was  quaestor  in  Cisalpine 
Gaul  in  B.  C.  82,  the  year  of  Carbo's  third  consulship.  Hence,  ever  since  the 
days  of  Drumann,  Cicero's  statement,  I  34,  that  Verres  was  quaestor  under 
Carbo  in  B.  C.  84  has  been  generally  regarded  as  a  mistake.  Prof.  Ciccotti 
suggests  that  Verres  was  quaestor  both  in  B.  C.  84  and  in  6.  C.  82,  that  he 
either  remained  in  office  for  three  consecutive  years  or  was  reappointed  in 


Digitized  by  VjOOQ IC 


94  AMERICAN  JO URNAL    OF  PHIL  OLOGY. 

B.  C.  82, — a  very  plausible  suggestion  which  he  has  since  supported  at  greater 
length  in  the  Rivista  di  Filologia,  N.  S.  Vol.  I  (1895),  pp.  332-40. 

With  regard  to  Cicero's  voyage  to  Sicily  in  search  of  evidence,  it  is  dis- 
appointing that  an  author  who  has  the  literature  of  his  subject  so  well  in  hand 
should  make  no  reference  to  Zielinski's  ingenious  article  "Verrina:  Die 
Chronologie  des  Processes,"  etc.,  in  the  Philologus,  Vol.  LII  (1893),  pp.  248  ff. 
It  is  interesting,  however,  to  notice  that  while  accepting  Cicero's  own  state- 
ment, Pro  Scauro,  25,  that  he  was  in  Sicily  in  the  depth  of  winter,  Prof.  Ciccotti 
thinks  that  he  returned  to  Rome  towards  the  end  of  April.  He  may  have 
remained  in  Sicily  for  a  short  time  after  his  task  was  accomplished  because 
of  his  eagerness  to  take  his  witnesses  back  with  him.  In  many  cases  their 
departure  was  hindered  or  prevented  by  the  new  governor,  Metellus. 

The  most  interesting  part  of  the  book  is  contained  in  the  last  forty  pages, 
where  we  have  an  examination  of  Cicero's  assertion  that  defence  was  impossible 
for  Verres.  The  writer  is  not  holding  a  brief  for  Verres;  "la  storia  non 
accusa,  nd  difende,"  and  his  purpose  is  historical.  The  frequency  of  this 
assertion  in  the  imaginary  Second  Actio  suggests  that  the  great  orator  was, 
like  some  of  Agricola's  soldiers,  promptus  post  eventum  ac  magniloquus.  In 
spite  of  his  denial,  the  eonsuetudo  accusatoria  is  manifest  everywhere,  and  it 
would  be  as  unsafe  to  form  an  estimate  of  Verres  solely  from  the  elaborate 
fiction  of  the  Second  Actio  as  to  form  an  estimate  of  Warren  Hastings  from 
the  rhetoric  of  Burke,  or  the  brilliant  but  untrustworthy  essay  of  Macaulay* 
Cicero  himself  tells  us  how  much  allowance  had  to  be  made  for  the  coloring  of 
the  advocate  in  trials  in  which  great  political  interests  were  involved  (Pro 
Fonteio,  38-9).  He  is  careful  to  assure  his  readers  that  the  Sicilians  are  not 
ordinary  Greeks  (II  7),  but  the  majority  of  his  witnesses  were  Greeks  none  the 
less,  and  he  himself  knew  well  the  value  of  a  Greek's  testimony  in  a  court  of 
law  (Pro  Flacco,  11 -12).  The  first  book  of  the  Second  Actio  and  the  greater 
part  of  the  fifth  have  nothing  to  do  with  the  matters  for  which  Verres  was 
prosecuted ;  the  charges  of  the  second  book  are  only  connected  with  those 
matters  by  an  obviously  indirect,  conjectural  mode  of  argument,— jto/«//nif. 
cipaiiier  conuctura,  as '  the  ancient  scholiast  puts  it.  Many  of  the  misdeeds 
charged  against  Verres  are  moral  rather  than  constitutional  in  their  nature. 
The  most  important  factor  in  his  condemnation  was  the  political  situation  of 
the  time.  Had  he  been  brought  to  trial  a  few  years  earlier  he  might  have 
received  a  much  more  spirited  and  loyal  support  from  his  senatorial  peers. 
But  the  SuUan  constitution  which  had  conferred  a  monopoly  of  power  upon 
the  senatorial  nobility  was  now  nearing  its  fall.  It  was  a  time  for  compromise 
and  the  intransigent  Verres  was  sacrificed  to  the  political  exigencies  of  the  day. 

Prof.  Ciccotti  is  full  of  modern  instances.  The  inevitable  comparison  of  the 
case  of  Verres  with  that  of  Warren  Hastings  recurs  again  and  again  ;  the  body 
from  which  the  equestrian  order  of  Rome  was  drawn  reminds  him  of  the  class 
of  francs-tenafuiers,  or  of  the  country  burghers  from  whom  Cromwell  recruited 
his  squadrons  of  *  ironcoaiis^  \  the  popular  excitement  and  political  interest 
aroused  by  the  trial  of  Verres  find  a  modern  parallel  in  the  investigation  of 
the  Panama  Scandal  or  in  the  inquiry  into  the  irregularities  of  the  Roman 
Bank ;  the  retinue  which  accompanied  each  new  governor  into  his  province  is 
likened,  to  compare  small  things  with  great,  to  the  host  of  political  spoilsmen 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES.  95 

who  rise  up  after  every  presidential  election  in  America  to  clamor  for  their 
reward. 

M.  Bomecque's  little  book  deserves  a  short  notice  in  this  Journal  because  of 
its  useful  introduction.  The  study  (pp.  20-40)  of  the  works  of  art  acquired  by 
Verres  during  his  stay  in  Sicily  is  particularly  good.  The  various  statements 
as  to  the  chronology  of  the  trial  are  absurdly  inaccurate  (p.  9, 11.  3-21 ;  p.  19, 
1.  23;  p.  31,  1.  20;  p.  74,  n.  9;  p.  77,  n.  3;  p.  152,  n.  2).  The  commentary 
borrows  freely  from  the  editions  of  Thomas  and  Halm,  always  with  the  most 
ample  acknowledgment,  but  seems  to  be  intended  for  a  much  younger  class  of 

readers.    The  proof-reading  has  been  extremely  poor. 

W.  P.  Mustard. 


Digitized  by  VjOOQ IC 


REPORTS. 

Englische  Studibn.    Heransgegeben  von  Dr.  Eugen  KOlbing,  Leipzig. 

XX  Band,  1895. 

I. — L.  Kellner,  Interchange  and  Tautology :  Two  Peculiarities  of  Old  and 
Middle  English  Style.  At  the  present  time,  unity  and  consistency  are  con- 
sidered the  first  requisites  of  a  good  style,  and  we  should  condemn  such  a 
sentence  as :  *'  The  mob  is  cruel  and  they  are  ignorant."  This  principle  is, 
however,  purely  modem.  The  best  writers  of  Old  and  Middle  English  not 
only  did  not  consider  these  two  qualities  as  admirablCi  they  deliberately 
avoided  them.  Holding  that  variety  was  one  of  the  charms  of  style,  they 
sought  this  variety  in  ways  that  would  not  now  be  admissible. 

I.  Changes  of  grammatical  construction,  which  can  hardly  be  attributed  to 
carelessness  or  otherwise  accounted  for ;  e.  g.  the  interchange  in  the  same 
sentence  of  thou  and  ye,  when  this  cannot  be  explained  as  a  change  from 
formal  address  to  familiar,  or  vice  versa,  as,  "For  I  sawe  you  never  or  nowe, 
but  hau  semist  a  gentilman,"  Gesta  Romanorum,  208 ;  interchange  of  tenses, 
present  and  preterite,  or  preterite  and  perfect,  as, 

He  rydys  home  to  bat  lady  hende. 
And  told  hur  his  tale  to  ende. 

-^Ipomadon,  A  4535/6 ; 

use  of  the  infinitive  with  and  without  the  preposition,  etc. 

II.  Similarly,  tautology  was  employed,  both  in  grammatical  constructions 
and  in  phraseol(^[y.  Of  the  first,  the  double  comparative,  double  negative 
and  double  genitive  (e.  g.  "  in  despite  and  repreef  of  Sir  Tristrams,"  Morte 
Darthur,  ed.  Sommer,  p.  324,  line  34)  are  among  the  examples  given.  Phrase- 
ological tautology  is  seen  in  Old  English  in  such  phrases  as  habban  and  dgan^ 
hweorfan  and  gdn^  geseon  and  ongietan,  etc.  In  Middle  English  its  forms  are 
more  varied,  and  may  for  convenience  be  thus  grouped:  I.  Double  forms. 
II.  Synonyms  of  Germanic  origin.  III.  Synonyms  of  Germanic  and  Romance 
origin.  IV.  Synonyms  of  Romance  origin.  Examples  of  I  are  weitawei  and 
walawa,  swete  and  swoU,  kysse  or  cusse ;  of  II,  all  whole ^  e.  g.  '*  We  putten  oure 
deede  . . ,  al  holly  in  youre  goode  wille"  (Chaucer,  Tale  of  Melibeus,  190), 
both  two,  cUpen  and  namen,  etc.;  of  III,  fulfil  and  per/orm,  hap  and  fortune  \  of 
IV,  accept  and  receive^  pite  and  mercy,  etc.  The  number  of  citations  from 
Chaucer*s  Melibeus  is  very  striking. 

O.  Hoffmann,  Studies  on  Alexander  Montgomerie.  The  point  of  departure 
for  the  article  is  an  edition  of  Montgomerie,  'The  Poems  of  Alexander 
Montgomerie,  edited  by  James  Cranstoun,  LL.  D.,  Edinburgh  and  London, 
1887,'  which  for  the  first  time  made  this  little-known  author  generally  acces- 


Digitized  by  VjOOQ IC 


REPORTS.  97 

sible.     The  article   is  in  two  parts:   I.   Montgomerie's  life  and  writings; 
II.  The  strophic  structure  of  his  poems. 

I.  Montgomerie's  dates  are  not  exactly  known;  he  was  born  about  1545 
(Cranstoun,  in  the  Diet.  Nat.  Biog.,  suggests  1556,  but  doubtfully),  and 
died  between  1605  and  161 5,  so  that  the  period  of  his  poetic  activity  would 
about  coincide  with  that  of  Spenser,  Sidney,  and  Shakespeare.  He  was 
for  a  while  a  favorite  at  the  court  of  James  VI,  whose  treatise  on  Scottish 
poetry  seems  to  have  been,  if  not  inspired,  at  least  greatly  influenced  by 
Montgomerie's  poetry.  James  makes  free  use  of  his  poems,  as  illustrations  of 
his  "  reulis  and  cautelis,**  sometimes  quoting  entire  poems,  sometimes  a  few 
lines  from  them.  Later  the  poet  appears  to  have  lost  favor,  and  the  rest  of  his 
life  was  spent  in  retirement.  This  may  in  part  account  for  the  slight  impres- 
sion he  seems  to  have  made  on  his  times ;  but  another  reason  is  doubtless  the 
fact  that  much  of  his  poetry  remained  in  manuscript  until  long  after  his  death, 
so  that  the  radius  of  its  influence  was  necessarily  small. 

In  treating  the  poems,  the  two  long  ones,  The  Cherrie  and  the  Slae,  and 
The  Flyting  betwixt  Montgomery  and  Polwart,  are  dealt  with  separately; 
the  rest  are  taken  up  in  groups,  following  Cranstoun's  classification  into 
sonnets,  miscellaneous  poems,  devotional  poems,  and  a  collection  of  para- 
phrases from  the  Psalms,  called  The  Mindes  Melodic. 

Of  the  sonnets  Cranstoun  says  that  they  show  **a  cultured  taste  formed  on  a 
careful  study  of.  Italian  models."  From  this  opinion  Hoffmann  dissents, 
holding  that,  in  the  main,  Montgomerie  went  his  own  way  untouched  by 
Italian  influences  except  such  as  reached  him  through  the  works  of  one  writer, 
Pierre  de  Ronsard,  seven  of  whose  sonnets  he  translated.  He  used  some  of 
Ronsard's  metres,  and  his  ideas  and  expressions  show  Ronsard's  influence. 
This,  indeed,  in  view  of  the  great  mass  of  poetic  thought  which  was,  as  it 
were,  common  stock  among  the  writers  of  that  day,  might  not  be  so  significant, 
were  it  not  that  it  is  with  reference  to  Ronsard  alone  that  we  can  trace 
anything  of  the  sort  in  Montgomerie. 

In  spite  of  the  fact  that  he  was  contemporary  with  the  great  Elizabethans, 
Montgomerie  cannot  be  classed  with  them,  but  must  be  considered  rather  as 
the  link  between  the  representatives  of  late  Middle  English  literature  on  the 
one  hand,  and  those  of  so-called  Modern  English  on  the  other.  In  some  ways 
he  reminds  us  of  Dunbar,  while  in  others  he  is  already  modem :  he  imitates 
the  older  verse-structure  in  his  deliberate  and  extensive  use  of  alliteration, 
yet  much  of  his  verse  can  well  be  placed  alongside  of  the  best  of  the  new  era. 
His  spirit,  too,  seems  often  to  belong  to  the  past,  as  in  The  Flyting  betwixt 
Montgomery  and  Polwart,  while  in  the  love  poems,  for  instance,  or  in  his 
descriptions  of  nature,  we  come  upon  wholly  modem  thought  in  wholly 
modem  expression. 

It  is  to  be  regretted  that  thus  far  Montgomerie's  verse-structure  has  not 
been  treated  in  the  works  on  the  subject.  Schipper  omits  his  name  altogether, 
and  Lentzner,  in  his  work  on  the  sonnet,  barely  mentions  him,  although  his 
sonnets  are  quite  as  deserving  of  attention  as  Dunbar's. 

Part  II.  The  notes  on  strophic  stractnre  are  of  great  interest,  but  it  is 
scarcely  possible  to  do  more  than  indicate  their  scope  and  character,  without 
giving  them  entire.    Particularly  interesting  is  the  work  on  sonnet-structure. 


Digitized  by  VjOOQ IC 


98  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Hoffmann  assigns  Montgomerie  ft  very  high  place  among  sonnet-writers,  both 
becaase  of  the  command  of  language  and  because  of  the  originality  which 
they  display.  One  form — abba  abba  ccd  eed — ^he  was,  if  not  the  first,  at  least 
among  the  first  in  Great  Britain  to  employ.  Another — abab  bcbc  cdcd  ee — 
which  Schipper  says  is  first  found  in  Spenser,  is  used  by  Montgomerie  in  a 
large  number  of  poems,  some  of  which  certainly  antedate  Spenser's.  It  is 
probable,  moreover,  that  Spenser  knew  of  Montgomerie's  work.  For  it  is 
scarcely  supposable  that  the  author  of  *  The  English  Poet,'  a  lost  treatise  on 
poetry,  should  have  been  ignorant  of  King  James*  similar  work,  which  was, 
we  may  say,  based  on  Montgomerie. 

Aside  from  the  sonnets,  his  poems  show  great  variety  in  strophic  form.  In 
some  he  follows  familiar  Scottish  models,  in  some  he  adopts  older  English 
forms  as  used  by  Chaucer  and  others,  while  in  yet  others  he  shows  Renaissance 
influence. 

W.  France,  Syntax  of  Early  Modem  English.  This  is  a  continuation  of 
work  begun  in  vol.  XVII.  The  present  number  treats  of  prepositions.  These 
are  taken  up  in  order,  and  their  use  is  illustrated  by  citations  from  seventeenth 
and  eighteenth  century  authors. 

The  Book  Notices  contain  reviews  of  two  more  numbers  of  the  Erlanger 
Beitriige,  containing  a  Tractatus  de  Diversis  Historiis  Romanorum,  and  an 
edition  of  Chettle,  Dekker,  and  Haughton's  Pleasant  Comedie  of  Patient 
Grissil ;  then  follow : 

L.  Frflnkers  Shakespeare  and  the  Aubade,  O.  Rohde's  The  Tale  of  the 
Hermit  and  the  Angel  in  its  Historical  Development,  C.  Ferrel's  Teutonic 
Antiquities  in  the  Anglo-Saxon  Genesis,  F.  Liebermann's  The  Leges  Anglorum 
Saeculo  13.  Ineunte  Londiniis  Collectae,  O.  Sommer's  The  Kalender  of  Shep- 
herdes,  L.  Lewes'  Shakespeare*s  Women,  P.  Roden's  Shakespeare's  Tempest, 
A.  Tolman's  Shakespeare's  Part  in  the  Taming  of  the  Shrew,  E.  Groth's 
Charles  Kingsley  as  Poet  and  Social  Reformer,  and  J.  Wells'  Oxford  and 
Oxford  Life. 

The  two  new  volumes  of  the  Erianger  Beitrfige  do  not  fall  below  the 
standard  thus  far  maintained  by  the  whole  series.  The  Tractatus,  though 
not  so  important  as  the  Gesta  Romanorum,  is  of  great  interest  as  being 
another  of  those  collections  of  stories  in  which  the  middle  ages  took  such 
delight.  In  the  opinion  of  Hippe,  the  reviewer,  Herzstein  lays  too  much 
stress  on  a  relation  between  these  two  works,  for,  of  the  sixty-nine  tales  in 
the  Tractatus,  only  fourteen  are  treated  in  the  Gesta.  It  is,  however, 
probable  that  all  these  collections,  in  Italian  as  well  as  in  Latin,  drew  their 
material,  at  least  in  part,  from  some  common  source.  The  text  of  the  Trac- 
tatus has  come  down  to  us  in  bad  condition,  and  needed  much  emendation 
from  the  editor.  Of  Patient  Grissil  an  edition  has  been  greatly  needed,  as 
Collier's,  1841,  no  longer  meets  the  demands  of  modem  scholarship.  Httbsch's 
text  is  a  faithful  reproduction  of  the  edition  of  1603,  with  little  attempt  at 
emendation  except  in  the  case  of  evident  misprints.  The  notes  are  careful 
and  scholarly,  but  somewhat  too  scanty.  The  introduction  traces  the  story 
from  the  time  of  Chancer  on,  but,  unfortunately,  fails  to  give  as  exact  a 
description  as  we  could  wish  of  the  rare  works  referred  to.  He  combats  the 
theory  that  the  drama  in  question  is  connected  with  Boccaccio's  novella. 


Digitized  by  VjOOQ IC 


REPORTS,  99 

Frankel  approaches  his  subject  "in  the  full  armor  of  the  comparative 
method."  This  subject  is  the  parting  between  Romeo  and  Juliet,  act  III, 
scene  5.  Nothing  like  this  passage  is  to  be  found  in  English  literature  before 
Shakespeare,  but  in  Germany  countless  parallels  existed,  as  well  as  many  in 
Holland  and  Flanders  based  on  German  models.  It  is  with  these  that  the 
Shakespearean  lines  must  be  in  some  way  related,  though  how  the  poet  came  by 
his  knowledge  of  German  lyrics  is  not  known.  It  may  have  been  through  his 
intercourse  with  Germans  in  London,  or,  more  indirectly  still,  through  the 
close  commercial  relations  between  England  and  Holland,  which  must  have 
involved  some  literary  interchange  as  well.  Frankel  then  makes  a  line-by-line 
comparison  between  the  passage  in  question  and  the  German  lyrics  of  the 
same  class,  and  follows  this  with  a  more  general  treatment,  from  the  stand- 
point of  comparative  literature.  The  book  is,  in  Hippe*s  judgment,  not  only 
a  worthy  contribution  to  Shakespearean  literature,  but  a  welcome  and  sugges- 
tive study  in  the  history  of  the  love-lyric. 

Frankel  gives  a  cutting  review  of  Rohde's  work,  which  he  scores  as  almost 
worthless  because  of  the  author's  incomplete  knowledge  of  his  material,  and 
his  failure  to  profit  even  by  previous  critical  work  on  his  subject.  The  book 
is  an  illustration,  first,  of  the  danger  of  using  second-rate  reference  books 
and  poor  or  superseded  editions,  and,  second,  of  the  fact  that  diligence  by 
itself,  without  breadth  of  view  and  the  power  to  interpret  masses  of  facts,  will 
not  produce  a  scientific  work. 

Ferrel's  study  of  Old  English  life  offers  little  that  is  new,  but  is  valuable  as 
bringing  together  in  a  systematic  exposition  what  has  till  now  been  scattered 
through  the  literature  of  the  subject  in  incidental  remarks.  The  scope  of  the 
volume  is  sufficiently  indicated  by  its  main  divisions,  which  are  as  follows : 
I.  Mythology.  II.  Christianity.  A.  God.  B.  Angels.  C.  Heanren.  D.  Satan 
and  his  Companions.  E.  Hell.  F.  Paradise.  III.  Nature.  IV.  The  King 
and  his  Subjects.  V.  Kinship  and  Home-Life,  Manners  and  Customs.  VI. 
War-Life. 

The  Leges  Anglornm  is  a  collection  of  notes  on  English  legislation  from  the 
time  of  Ine  to  that  of  John — a  curious  work,  which  must  have  been  put 
together  between  1206  and  1239,  though  some  manuscripts  contain  additions 
from  a  later  time,  even  as  late  as  the  fourteenth  century.  The  thirteenth, 
century  collector  seems  to  have  tried  to  piece  together  from  various  sources  a 
history  of  English  legislation,  with  special  reference  to  the  city  of  London. 
His  principal  sources  were :  the  larger  law-records,  i.  e.  the  Quadripartitus, 
the  Leges  Edward!  Confessoris,  the  Leges  Henrici  I,  and  Glanville's 
Tractatus  temp.  Henrici  11.  These  are  given  only  in  part,  however,  and 
amongst  them  are  inserted  many  other  fragments,  some  also  from  older 
records,  some  apparently  fabricated  by  the  collector  himself  to  bear  out  his 
private  views.  He  must  have  obtained  his  material  in  the  City  archives  and 
the  City  library.  His  use  of  it  is  superficial  and  uncritical,  and  it  is  hard  to 
see  on  what  principle  he  worked  in  making  his  selections.  That  his  purpose 
was  not  scientific  but  utilitarian  is  evident  The  aim  of  the  work  is  the 
furthering  of  certain  great  governmental  reforms,  and,  in  particular,  the 
glorification  of  the  city  of  London.  Liebermann  has  in  some  cases  printed 
the  text  of  the  various  fragments,  in  others  he  has  indicated  where  they  are 


Digitized  by  VjOOQ IC 


100  AMERICAN  JOURNAL    OF  PHILOLOGY. 

to  be  foand.  He  has,  moreover,  pointed  oat  their  sources  where  this  is 
possible/ and  compared  these  versions  with  their  originals.  Though  in  many 
respects  faulty  and  inferior,  the  Leges  Anglorum  is,  on  the  whole,  to  be 
classed  with  Roger  of  Hoveden  as  the  richest  collection  of  early  English  legal 
documents  which  we  possess,  and  Liebermann  has  rendered  a  service  to  his 
fellow-specialists  in  making  it  the  subject  of  a  detailed  investigation.  The 
reviewer,  Maurer,  gives  a  rather  full  summary  of  the  contents  of  the  volume, 
to  which  those  who  are  interested  are  referred. 

The  Kalender  of  Shepherdes  is  of  great  interest  both  in  itself  and  by  reason 
of  its  connection  with  later  literature.  It  was  first  written  in  French,  and 
printed  in  Paris,  1493,  with  the  title  *  Le  compost  et  Kalendrier  des  bergiers.' 
In  1503  it  was  indifferently  well  translated  by  a  Scotchman,  whose  work  was 
revised  and  republished  in  1506  by  Pynson.  Both  these  editions  are  given  by 
Sommer,  the  first  in  photographic  facsimile,  the  second  in  reprint.  The 
introduction  gives  a  brief  history  of  the  subject,  a  bibliography,  and  a  some- 
what scanty  glossary.  KOlbing,  the  reviewer,  adds  a  few  notes  of  his  own  on 
one  of  the  poems  in  the  collection,  entitled  '  Of  an  assaute  agaynst  a  snayle.' 
This  poem,  which  KOlbing  reprints  to  substantiate  the  comparison,  is  evidently 
an  earlier  version  of  the  snail  episode  in  the  interlude  of  Thersites,  1537  (?). 
In  conclusion,  KOlbing  expresses  his  regret  that  Sommer's  publications  are  so 
generally  ignored  in  Germany,  and  his  hope  that  in  future  they  will  be  better 
known. 

A  comprehensive  work  like  Lewes'  has  not  of  late  years  been  undertaken 
in  Germany.  It  does  not  claim  to  give  anything  new,  but  rather  a  new 
arrangement  of  what  is  already  known.  The  book,  however,  according  to 
Koch,  does  more  than  its  title  promises,  for  it  gives  not  disconnected  portraits 
of  the  characters,  but  estimates  of  them  with  reference  to  and  by  means  of  a 
general  survey  of  the  plays  in  which  they  appear.  Koch  closes  by  quoting 
Lewes'  remonstrance  against  the  kind  of  Shakespeare  interpretation  which 
sees  in  every  play  a  central  idea,  philosophical  or  moral.  Such  an  interpre- 
tation can  never  justify  itself  except  by  doing  violence  to  Shakespeare. 

Koch  makes  this  statement  the  transition  to  a  brief  notice  of  Roden*s  work 
— a  study  which  might  well  have  served  as  the  text  for  Lewes's  protest. 
Roden  attempts  to  show  that  in  the  Tempest  there  is  portrayed  the  spiritual 
storm  which  disturbed  Europe  from  the  end  of  the  fifteenth  to  the  end  of  the 
sixteeenth  century.  Antonio  is  scholasticism,  which  is  arrayed  with  the 
church  (Naples)  against  the  true  scientific  spirit  (Prospero).  Ariel  is  the 
powers  of  nature,  controlled  by  Prospero  as  science,  and  so  forth. 

Tolman's  study  of  The  Taming  of  a  Shrew  is  interesting,  not  only  in  itself, 
but  as  coming  from  a  pupil  of  ten  Brink.  The  case  stands  thus :  The  Taming 
of  the  Shrew  (TTS),  whose  date  Tolman  thinks  lies  between  1604  and  1609, 
was  preceded  by  a  strikingly  similar  play.  The  Taming  of  a  Shrew  (TAS), 
printed  in  1594,  written  probably  not  later  than  1588.  Both  plays,  but 
especially  TTS,  contain  material  borrowed  from  Gascoigne*s  translation  (1566) 
of  Ariosto's  Suppositi.  The  problem,  then,  is  to  establish  the  relation 
between  these  three.  The  general  opinion  of  Shakespeare  critics  has  thus 
far  been  that  TAS  was  derived  from  the  Suppositi,  and  that  TTS  was  derived 
from  the  Suppositi  and  from  TAS.    Tolman  gives  a  number  of  arguments 


Digitized  by  VjOOQ IC 


REPORTS.  lOI 

making  for  this  view,  which,  he  judges,  has  fewer  difficulties  than  any  other 
theory,  with  one  exception— namely,  the  view  of  ten  Brink.  Ten  Brink's 
suggestion,  which  he  left  undeveloped,  was  that  neither  play  is  derived  from 
the  other,  but  both  from  a  third,  a  youthful  production  of  Shakespeare,  ante- 
dating A  Midsummer  Night's  Dream.  Wild  as  this  theory  seems,  and  wholly 
without  external  proof,  it  would,  if  accepted,  explain  all  the  facts  better  than 
any  other.  Tolman  does  not  give  a  final  judgment,  but  merely  presents  the 
evidence.  Koch,  who  reviews  his  work,  though  he  too  admits  the  difficulties 
of  the  question,  seems  disinclined  to  take  ten  Brink's  theory  seriously,  and 
indeed  remarks  that  ten  Brink  himself,  in  his  last  lecture  at  Frankfort, 
appeared  to  have  given  it  up. 

A  study  of  Kingsley  is  especially  welcome  because  so  little  has  thus  far 
been  done  on  the  subject,  either  in  Germany  or  England.  Groth's  work, 
however,  though  interesting  and  sympathetic,  leaves  much  to  be  desired. 
The  treatment  of  Kingsley  as  a  social  reformer,  for  example,  is  unsatisfactory, 
because  the  background — the  social  commotions  of  his  day — is  treated  too 
sketchily.  The  study  of  the  novelist  and  poet  is  much  more  successful.  The 
biography,  again,  is  weak,  containing  many  errors  and  failing  often  in  clear- 
ness or  adequacy.  The  treatment  of  Kingsley's  relations  with  Carlyle,  Char- 
lotte Bronte,  and  Wordsworth  is  not  full  enough,  while  the  influence  of  Scott 
and  Bulwer,  of  Dickens  and  Thackeray,  is  not  even  touched  upon.  Never- 
theless, the  work,  with  its  attractive  style  and  its  sympathetic  rendering  of 
the  poet's  personality,  will  doubtless  appeal  to  a  large  audience,  and  may  do 
much  to  awaken  interest  in  Kingsley. 

The  Miscellanea  contains  a  number  of  notes  of  interest  to  Old  English 
students:  LindstrOm  discusses  briefly  the  etymology  oi preosi\  Pogatscher  has 
a  note  on  the  words  bred  weall  (0\6.  English  Chronicle,  entry  for  the  year  189), 
which  he  explains  by  comparison  with  the  Latin  of  Beda ;  Swaen  discusses 
the  length  of  the  vowel  in  seppan^  or  sepan,  Bttlbring  contributes  a  study  of 
the  quantity  of  the  vowel  before  nd  in  Middle  English  verse,  e.  g.  sendan, 
rvetidafif  etc.  By  the  rime-test,  applied  to  the  three  parts  of  Robert  of 
Gloucester,  with  uniform  results,  he  seems  to  prove  that  in  these  verbs  the 
vowel  is  long  in  the  present  stem  and  short  in  preterite  and  perfect  participle. 
The  rimes  in  the  Ormulum,  also,  he  finds  to  conform  with  almost  perfect  regu- 
larity to  this  rule.  E.  Koeppel  has  some  interesting  notes  on  Chaucer  and 
Shakespeare :  (i)  Gower's  French  ballads  and  Chaucer ;  (2)  Chaucer's  Anelida, 
whom  Koeppel  would  make  queen,  not  of  Ermony  (Armenia),  but  of  Emony 
(Lat.  Haemonia  =  Thessaly) ;  (3)  a  striking  parallel  between  The  Misfortunes 
of  Arthur  and  Macbeth  ;  (4)  parallels,  almost  as  remarkable,  between  the  first 
part  of  Jeronimo  and  Hamlet. 

II. — E.  KOlbing,  Contributions  to  the  Elucidation  and  Textual  Criticism  of  the 
York  Plays.  Since  the  edition  of  the  York  Plays,  by  Toulmin  Smith,  appeared 
in  1885,  the  attention  of  scholars  has  been  attracted  to  this  important  but 
corrupt  text.  Zupitza,  Hall,  Herttrich,  and  Holthausen  have  all  done  some- 
thing towards  emendation,  but  much  work  still  remains.  The  reconstruction 
of  a  text  of  this  kind  should  be,  so  far  as  possible,  the  result  of  the  combined 
labors  of  Middle  English  scholars.    The  emendations  here  submitted  by 


Digitized  by  VjOOQ IC 


I02  AMERICAN  JOURNAL    OF  PHILOLOGY. 

KOlbing  do  not  profess  to  be  exhaustive  or  final,  but  are  put  forth  rather  as 
suggestions,  looking  towards  a  future  critical  edition  of  the  text — an  edition 
which  he  hopes  Miss  Toulmin  Smith  herself  may  superintend. 

E.  W.  Sievers,  Shakespeare  and  the  Pilgrimage  to  Canossa.  Not  Germany 
alone  has  had  to  humble  herself  in  the  dust  before  Rome.  England,  too,  has 
had  her  Canossa ;  the  Pope  was  Innocent  III,  the  King  was  John.  In  the 
reign  of  Elizabeth  this  event  must  have  been  constantly  in  men^s  thoughts, 
for  a  like  humiliation  seemed,  until  the  death  of  Mary  Stuart  and  the  defeat 
of  the  Armada,  not  only  possible,  but  imminent.  That  the  struggle  between 
Protestantism  and  Rome  would  have  made  a  deep  impression  on  Shakespeare 
we  can  easily  imagine ;  that  it  actually  did  so  is  proved  by  his  play,  King 
John.  The  play  must  be  regarded  as  the  poet's  cry  of  appeal  to  the  nation  to 
stand  by  the  principle  of  the  new  religion — that  is,  the  exaltation  of  the 
individual  as  opposed  to  the  belief  in  authority.  The  central  theme  of  the 
play  may  be  stated  as  the  struggle,  not  only  between  a  foreign  pope  and  a 
national  king,  but  between  the  principle  of  authority  and  the  principle  of 
individuality.  Pandulph  stands  for  the  one,  the  other  is  represented  fitfully 
by  John,  consistently  by  Philip  the  Bastard.  Pandulph — ^and  the  Roman 
Church — bases  his  power  on  one  of  the  ultimate  forces  of  the  human  soul, 
the  craving  for  salvation.  It  is  to  this  that  he  appeals  in  dealing  with  the 
French  king ;  in  this  alone  lies  the  efficacy  of  his  threat  of  excommunication. 
To  subdue  John,  however,  he  must  seek  other  means,  for  the  English  king  is 
untouched  by  threats  of  a  power  he  despises.  In  his  first  attitude  towards 
Rome,  then,  John  represents  the  Protestant  principle  of  resistance  to  authority. 
But  to  work  John's  downfall  Pandulph  relies  on  another  human  trait,  the 
instinct  of  self-preservation.  It  is  this,  he  foresees,  which  will  tempt  John  to 
kill  his  nephew ;  it  is  this  which  will  lead  the  nation  to  fall  away  from  their 
king.  All  turns  out  as  Pandulph  plans,  though— and  here  we  are  at  the  heart 
of  the  matter — it  need  not  have  done  so  if  John  had  been  thoroughly  imbued 
with  the  Protestant  spirit.  For,  though  he  is  a  usurper,  he  is  a  man  far  more 
fit  for  the  throne  than  the  true  claimant,  Arthur.  Realizing  this,  John  ought 
to  have  been  strong  enough  in  his  self-knowledge  to  rely  wholly  on  his  proved 
fitness  for  the  place,  and  to  see  in  the  invasion  of  Philip  and  Pandulph  his 
opportunity  brilliantly  to  justify  his  usurpation  by  arraying  himself  on  the  side 
of  the  national  feeling.  But  he  does  not  trust  this,  and  thinks  it  necessary  to 
make  his  right  legitimate  by  removing  Arthur.  Thus  it  appears  that,  though 
he  is  free  from  the  tyranny  of  authority  in  one  form,  he  is  still  under  its 
dominion  in  another — under  the  dominion,  namely,  of  the  principle  of  con- 
ventional legitimacy.  These  are,  then,  the  causes  of  John's,  and  through 
him  of  England's,  humiliation.  One  character  in  the  play,  however,  remains 
uniformly  true  to  himself,  to  his  country,  and  to  his  king :  this  is  the  Bastard, 
who  from  his  first  appearance  stands  as  the  representative  of  individual  worth, 
independent  of  the  conventional  sanctions  of  birth  and  position.     His  phrase : 

And  I  am  I,  howe*er  I  was  begot, 

is  indicative  of  his  character.  In  him,  if  anywhere,  Shakespeare  himself 
speaks.  His  soliloquy,  for  instance,  at  the  end  of  act  II  may  be  compared 
with  Aristophanes'  use  of  the  chorus.    He  it  is  who  keeps  a  steady  head 


Digitized  by  VjOOQ IC 


REPORTS.  103 

throughoat  the  vicissitudes  of  the  kingdom,  who  brings  the  nobles  again  to 
their  allegiance,  and  takes  from  England's  defeat  the  sting  of  finality.  In  his 
last  words  we  may  hear  Shakespeare's  warning  to  England  : 

This  England  never  did,  nor  never  shall, 
Lie  at  the  proud  foot  of  a  conqueror, 
But  when  it  first  did  help  to  wound  itself. 

. . .  Nought  shall  make  us  rue, 
If  England  to  itself  do  rest  but  true. 

In  support  of  this  interpretation  it  is  interesting  to  note  that  the  features  of 
the  play  from  which  it  has  been  deduced  are  peculiar  to  the  Shakespeare 
version,  although  in  many  respects  the  poet  followed  the  older  play  of  the 
same  name. 

A.  E.  H.  Swaen,  To  Dare.  That  the  verb  in  question  is  very  variable  is 
seen  in  the  fact  that  of  the  seven  authorities  quoted  by  Swaen,  no  two  are 
quite  in  agreement  as  to  its  forms.  Its  peculiarities  are  (i)  the  two  forms, 
dare  and  dares^  in  the  present  tense,  3d  singular ;  (2)  in  the  two  forms,  dared 
and  durst^  of  the  preterite,  and  (3)  in  the  fact  that  infinitives  governed  by  dare 
sometimes  take  to  and  sometimes  not.  The  bulk  of  the  paper  is  occupied 
with  citations  from  English  literature,  from  the  sixteenth  century  to  the 
present  time,  illustrating  the  use  of  the  word,  with  special  reference  to  these 
three  points.  The  citations  have  exact  references,  and  the  whole  is  so  well 
arranged  that  the  reader  hardly  needs  the  table  of  results  that  is  appended. 

The  Book  Notices  include  reviews  of  Kaluza's  Studies  in  Old  Germanic 
Alliterative  Verse,  A.  J.  Wyatfs  edition  of  Beowulf,  S.  Hewett's  The  Peasant 
Speech  of  Devon  and  Other  Matters  Connected  Therewith,  A.  Drake's  The 
Authorship  of  the  West  Saxon  Gospels,  J.  Wright's  A  Grammar  of  the  Dialect 
of  Windhill  in  the  West  Riding  of  Yorkshire,  J.  Groag's  The  Character  of 
Julius  Caesar  in  Shakespeare's  Play,  P.  Kreutzbei^'s  Brutus  in  Shakespeare's 
Julius  Caesar,  E.  Frey's  Robert  Browning's  Dramas,  and  P.  Branscheid's  Life 
of  Charles  Dickens. 

Wyatt's  edition  of  Beowulf,  with  its  scholarly  notes,  well-arranged  glossary, 
and  index  of  names,  is  highly  praised  by  Brenner,  who  foresees  that  it  will,  in 
England  at  least,  speedily  supplant  the  German  editions. 

Drake's  investigation  of  the  authorship  of  the  four  Gospels  yields  interesting 
results.  His  method  has  been  to  compare  the  use,  in  each  of  the  Gospels,  of 
such  words  as  allow  variation,  and  hence  are  well  adapted  for  this  kind  of 
test,  e.  g.  heofon  and  heofone,  or  ^tra  and  Y^ra,  By  this  test,  Mark  and  Luke 
would  seem  to  be  clearly  by  the  same  hand,  while  Matthew  and  John  are 
written,  if  not  by  the  same  author,  at  least  in  the  same  locality,  and  certainly 
not  by  the  translator  of  the  other  two. 

Brenner  praises  Wright's  Grammar  as  a  model,  despite  its  failure,  in  some 
respects,  to  conform  to  the  German  ideal.  Phonetics  is  given  its  due  space, 
and  no  more,  the  system  of  phonetic  sounds  is  remarkably  simple  and  easily 
grasped,  and  the  comparisons  with  older  English  or  Old  French  words  are 
interesting. 

The  two  papers  on  Julius  Caesar  are  reviewed  by  Koch.  Kreutzberg  has, 
in  his  opinion,  added  nothing  to  what  has  been  already  done,  but  Groag's 


Digitized  by  VjOOQ IC 


I04  AMERICAN  JOURNAL   OF  PHILOLOGY. 

work  is  of  interest.  There  has  been  a  tendency  among  Shakespeare  scholars 
to  the  view  that  Brutus,  not  Caesar,  is  the  true  hero  of  the  play,  which  there- 
fore has  not  been  properly  named,  and  which  fails  in  unity,  because  of  this 
confusion  of  interests.  In  different  phases  this  was  the  view  of  Gervinns  and 
others,  and  more  recently  of  Brandl.  It  is  combated  by  Dowden,  and  now 
by  Groag,  with  whom  Koch  seems  to  be  in  essential  agreement.  Shakespeare, 
who  everywhere  in  his  writing  exalts  the  name  of  Caesar,  had  surely  no 
intention  of  making  him  appear  weak,  either  to  lessen  the  guilt  of  the  con- 
spirators er  (Brandrs  suggestion)  as  a  foil  to  Brutus.  In  judging  Caesar's 
character  we  must  take  into  account  not  only  what  he  does  and  says  on  the 
stage,  but  what  every  one  else  says  about  him— how  the  people  feel  towards 
him.  Even  Cassius,  fbr  instance,  can  accuse  him  only  of  physical  weakness, 
while  in  the  eyes  of  all  he  is  unquestionably  the  hero,  the  man  of  destiny.  By 
this  test  he  is  restored  to  his  proper  position  as  the  greatest  as  well  as  the 
truly  central  figure  in  the  play. 

In  his  essay  on  Browning,  Frey  gives  a  careful  analysis  of  all  the  dramas,  as 
well  as  some  poems  not  in  dramatic  form — Pauline,  for  example — which  are, 
however,  as  the  reviewer.  Hoops,  remarks,  quite  as  much — and  as  little — 
dramas  as  are  some  of  those  so  entitled.  He  touches  upon  the  influence  of 
Shelley  on  Browning,  but  refrains  from  any  detailed  handling  of  sources,  or 
any  attempt  at  a  final  literary-historic  estimate  of  the  poet. 

Aronstein,  reviewing  Werner's  study  of  Thomas  May,  characterizes  it  as  in 
part  superficial,  in  part  worthless,  and  in  part  (the  section  on  May's  dramatic 
style)  tolerably  successful.  Branscheid's  life  of  Dickens,  though  by  no  means 
complete,  is  on  the  whole  good,  and,  in  Aronstein's  opinion,  well  worth  a 
careful  working  over. 

The  Miscellanea  contains  a  few  English  etymologies  by  Kluge,  and  some 
notes  by  Wttlfing  tending  to  support  the  theory  of  Hubbard  and  Wttlker  that 
The  Soliloquies  are  the  work  of  King  Alfred.  Kaluza  contributes  a  highly 
interesting  note  on  the  controversy  concerning  the  authorship  of  the  Romaunt 
of  the  Rose.  He  defends  Skeat  against  the  charge  of  having  ignored  some 
of  Lounsbury's  arguments,  and  reiterates  his  own  belief  in  the  'absurd' 
theory  of  a  dual  authorship.  E.  W.  Bowen  in  a  brief  paper  called  Confusion 
between  d  and  d  in  Chaucer's  Rimes,  gives  a  list  of  the  cases  where  such 
confusion  occurs,  and  where  the  only  explanation  seems  to  be  that  Chaucer 
gave  way  to  the  exigencies  of  the  verse.  Perhaps  the  most  interesting  contri- 
bution is  Koch's  brief  note  on  Shakespeare  and  Lope  de  Vega.  The  occa- 
sion of  it  is  the  appearance  (Berlin  and  Weimar,  1894)  of  Arturo  Farinelli's 
book,  Grillparzer  and  Lope  de  Vega.  Grillparzer  always  held  that  Shakes- 
peare had  come  under  Spanish  influence,  although  he  did  not  know  the 
language,  and  that  some  passages  in  his  writings  almost  certainly  imply  a 
knowledge  of  Lope  de  Vega  himself.  In  this  Farinelli  (and  Koch)  agrees. 
It  is  interesting  to  note  that  in  four  plays — Romeo  and  Juliet,  Winter's  Tale, 
As  You  Like  It,  and  All's  Well  That  Ends  Well— Shakespeare  handles 
themes  which  Lope  de  Vega  also  used.  Webster's  Duchess  of  Malfy  is 
paralleled  by  the  Spanish  El  Mayordomo  de  la  Duquesa  de  Amalfi,  but  the 
two  are,  according  to  Farinelli,  only  connected  by  the  fact  of  their  common 
source  in  Bandello's  novella. 


Digitized  by  VjOOQ IC 


REPORTS.  105 

III. — G.  Richtcr,  Contributions  to  the  Elucidation  and  Textual  Criticism  of 
the  Middle  English  Prose  Romance  of  Merlin.  The  Merlin  legend  has  been 
handled  three  times  in  Early  English  literature:  (i)  in  a  romance  in  four- 
stressed  couplets.  The  main  part  of  this  is  contained  only  in  the  Auchinleck 
manuscript,  in  Edinburgh.  This  earliest  English  form  of  the  story  must  be 
placed,  probably,  about  the  middle  of  the  thirteenth  century.  (2)  in  the 
fifteenth-century  poem  of  Merlin,  by  Henry  Lonelich,  preserved  in  a  single 
Cambridge  manuscript.  (3)  in  a  prose  romance  of  the  second  half  of  the 
fifteenth  century,  preserved  also  in  but  one  (Cambridge)  manuscript.  The 
first  has  been  lately  edited  by  KOlbing,  and  an  edition  of  the  second  is  being 
prepared  by  KOlbing  and  Miss  Mary  Bateson.  The  third  was  edited  for  the 
Early  English  Text  Society,  by  Henry  B.  Wheatley  (London,  1865-69 ;  revised, 
1877).  This  has,  however,  neither  introduction,  notes,  nor  glossary,  and 
Wheatley  scarcely  even  attempted  to  emend  the  text,  full  of  errors  though  it 
was.  Richter,  therefore,  has  undertaken  the  work  of  emendation,  admitting, 
however,  that  the  first  attempt  in  so  large  a  subject  must  always  contain  much 
that  is  merely  provisional.  As  to  sources,  all  three  English  versions  go  back, 
according  to  Kdlbing,  to  the  same  French  text,  first  printed  in  Paris,  1528, 
with  the  title  *  Le  premier  et  le  second  volume  de  Merlin.'  Of  this  edition 
the  English  prose  version  is  an  almost  exact  translation.  Richter  then  gives 
his  notes  on  the  first  half  of  the  romance.  Those  on  the  second  are  to  appear 
later. 

Ph.  Aronstein,  John  Marston  as  a  Dramatist.  The  article  is  the  first  part  of 
a  rather  comprehensive  treatment  of  Marston,  whose  dramas  have  thus  far 
[1894]  received  rather  less  than  their  due  proportion  of  attention.  The 
present  paper  forms  the  literary-historical  part,  and,  after  a  brief  biograph- 
ical introduction,  is  divided  into  three  chapters:  I.  Marston's  works  in  chro- 
nological order.  The  dates  for  the  works  are  given  as  nearly  as  possible, 
with  the  grounds  for  assigning  them.  There  are  also  brief  notes  as  to  the 
character  of  each  work  and  its  reception  by  the  public.  II.  Marston's  attitude 
towards  his  contemporaries.  The  bulk  of  this  section  is  formed  by  an  account 
of  the  well-known  quarrel  with  Jonson.  III.  Marston's  conception  of  the 
poetic  art,  and  his  attitude  towards  the  public.  Here  he  stands  in  marked 
contrast  to  Jonson.  As  to  his  idea  of  the  function  of  poetry,  nothing  can  be 
more  explicit  than  his  own  words: 

We  strive  not  to  instruct  but  to  delight. 

—Dutch  Courtezan,  Prol.,  1.  8. 

His  attitude  towards  the  public  was  almost  uniformly  apologetic,  but  we  may 
doubt  whether  the  modesty  was  not  at  least  in  part  assumed.  His  success 
with  his  own  times  seems  to  have  been  rather  remarkable,  and  one  of  his 
pUys,  The  Dutch  Courtezan,  was  revived  after  the  Restoration. 

Thus  far  the  treatment  of  the  subject  offers  little  that  is  new,  though  the 
compact  arrangement  of  old  material  is  rather  convenient  The  author 
proposes  in  the  next  part  to  take  up  the  individual  works,  beginning  with  the 
tragedies. 

J.  Ellinger,  Contributions  to  English  Grammar.  The  article  gives  numerous 
citations  from  nineteenth-century  prose  to  illustrate  grammatical  usages  which 


Digitized  by  VjOOQ IC 


I06  AMERICAN  JOURNAL    OF  PHILOLOGY. 

have  either  not  yet  been  noted  or  have  been  insufficiently  illustrated  by 
grammarians.  The  points  considered  are:  i.  The  omission  of  the  definite 
article  before  an  attributive  adjective  modifying  the  name  of  a  person.  2.  The 
comparison  of  adjectives  of  two  syllables.  3.  The  objective  case  of  personal 
pronouns  used  for  the  nominative.  4.  What  a,  introducing  indirect  questions. 
5.  Any  one^  no  ane^  sonu  one^  used  adjectively.  6.  Lest  followed  by  might, 
7.  Except  for,  only  for,  save  for  =  but  for.  For  a  comment  on  some  points  in 
the  article,  see  Englische  Studien,  XXII,  p.  153. 

O.  Schulze,  Contributions  to  English  Grammar.  Schulze  takes  up  Schmidt's 
treatment  of  limiting  relative  clauses,  and  shows  that  his  definition  and  expo- 
sition is  not  broad  enough  to  cover  all  cases  in  Modem  English. 

The  Book  Notices  comprise  reviews  of  J.  Ries's  What  is  Syntax?,  T.  R. 
Lounsbury*s  History  of  the  English  Language,  E.  Wttlfing's  The  Syntax  of 
the  Works  of  Alfred  the  Great,  Harrison  and  Sharp's  Beowulf  and  The  Fight  at 
Finnsburgh,  second  edition,  E.  Bormann*s  The  Shakespeare-Mystery,  W. 
Creizenach*s  History  of  the  Modem  Drama,  Schipper's  edition  of  the  Poems 
of  William  Dunbar,  and  O.  Bremer*s  German  Phonetics. 

Ries  complains  of  the  prevalent  vagueness  of  ideas  with  regard  to  exactly 
what  syntax  is.  He  defines  it  as  the  doctrine  of  the  sentence  and  of  other 
combinations  of  words.  He  objects,  moreover,  to  the  old  antithesis  between 
syntax  and  morphology,  as  well  as  to  that  between  syntax  and  semasiology, 
and  would  prefer  to  divide  the  study  of  words  into:  (i)  the  study  of  the  forms 
of  words,  (2)  the  study  of  the  meaning  of  words.  Similarly,  syntax  is  the  study, 
(i)  of  the  form  of  word-combinations,  (2)  of  their  meaning.  Ellinger,  the 
reviewer,  concludes  that,  although  Ries's  system  is  not  free  from  flaws,  inas- 
much as  it  is  (i)  too  a  priori,  (2)  more  applicable  to  descriptive  than  to  historical 
grammar,  yet  the  author  has  done  good  service  in  making  this  decided 
advance  towards  a  scientific  demarcation  of  the  field  of  syntax. 

Luick  commends  -Lounsbury's  book,  but  considers  its  usefulness  greatly 
impaired  by  the  fact  that  it  is  in  the  main  a  history  only  of  written  English. 
Phonetics  would,  of  course,  be  out  of  place  in  a  somewhat  popular  treatise,  but 
if  the  author  had  dealt  with  his  subject  from  the  phonological  standpoint  he 
would  have  avoided  many  statements  which,  as  they  stand,  give  a  false 
impression.  For  example,  he  says  that  the  language  of  Chaucer  is  very  near 
to  Modern  English.  Yet  Chaucer's  speech,  or  even  Shakespeare's,  would  be 
unintelligible  to  a  modern  Englishman.  He  was  thinking,  of  course,  only  of 
the  written  language,  and  took  no  account  of  the  sweeping  changes  that  have 
taken  place  in  the  spoken  language  since  Chaucer,  especially  since  the 
sixteenth  century.  Yet  these  changes  left,  as  Luick  says,  scarcely  one  stone 
standing  upon  another,  and  gave  us  a  language  differing,  in  sound,  from 
Chaucer's  little  less  than  did  Chaucer's  from  that  of  Alfred.  It  is  to  be  hoped 
that  in  another  edition  this,  the  only  great  defect  in  the  work,  will  be  corrected. 

Kellner  expresses  a  slight  feeling  of  disappointment  that  WQlfing's  work  on 
Alfred  does  not  ofier  more  that  is  new,  either  on  syntax  in  general,  or  on 
Alfred's  syntax  in  particular.  The  work  seems  in  the  main  to  furnish  only 
new  instances  illustrating  principles  of  syntax  already  established. 

The  fourth  edition  of  Harrison  and  Sharp's  Beowulf  shows  few  changes  in  the 
text,  rather  more  in  the  glossary  and  list  of  names.     The  contents  of  the 


"  Digitized  b^vjOOQlC 


REPORTS.  107 

appendixes  of  earlier  editions  is  now  incorporated  in  the  notes.  The  system 
of  accentuation  is  not  altered,  but  some  errors  are  corrected.  One  good 
feature  of  the  notes  is  their  constant  reference  to  other  Old  and  Middle 
English  texts. 

Frilnkel,  in  a  long  note  on  the  Shakespeare-Bacon  question,  takes  as  his 
text  Borman's  work,  which  is,  he  says,  the  ripest  and  most  scholarly  defence 
of  the  Bacon  authorship  that  has  yet  appeared.  After  reviewing  the  history 
of  the  controversy,  Friinkel  takes  up  this  book  somewhat  in  detail,  and  finally 
remarks  that  as  this  work,  the  ablest  on  the  subject,  does  not  seem  to  have 
carried  conviction  with  it,  the  adherents  of  the  theory  may  as  well  give  up 
their  attempt  to  prove  an  absurdity.  He  quotes  with  approval  Professor 
Corson's  remark :  "  If  Shakespeare  did  not  write  the  plays  attributed  to  him, 
certainly  Lord  Bacon  did  not  write  them."  In  conclusion  he  alludes  to  a 
book  about  to  appear,  which  promises  to  be  a  remarkable  work.  It  is — 
according  to  the  announcement  of  the  author,  L.  Strohl — to  prove  that  the 
works  of  Shakespeare  were  written  neither  by  Shakespeare  nor  by  Bacon,  but 
by  a  London  soap-boiler,  Shakspere.  who  also  wrote  three  plays  under  the 
pseudonym  of  Marlow. 

The  first  volume  of  Creizenach*s  work  contains,  of  course,  less  to  interest 
English  students  than  will  the  second.  He  has  treated  his  subject  under 
eight  heads.  In  English  literature  he  treats  of  the  Towneley  Mysteries, 
especially  Cain,  the  Chester  and  Coventry  Plays,  and  the  Digby  Plays. 
Creizenach  thinks  that  at  this  period  French  influence  was  less  strong  in  the 
English  drama  than  in  the  other  branches  of  English  literature. 

Schipper's  edition  of  Dunbar  goes  further  than  those  of  Laing  and  Small, 
in  that  it  is  critical  and  is,  where  possible,  chronologically  arranged.  Each 
poem  has  a  separate  introduction.  The  notes  are  limited  to  the  elucidation 
of  words  and  sentences,  without  giving  parallels  from  other  Scotch  poetry. 
KOlbing  thinks  that  in  punctuating  the  text  Schipper  has  followed  Small  too 
closely,  and  that  he  has  been  too  conservative  in  making  emendations.  The 
edition  is,  however,  far  superior  to  its  predecessor,  both  because  of  its  chrono- 
logical arrangement,  its  accuracy,  and  the  beauty  of  its  outer  form — the  print, 
etc.  KOlbing  gives  detailed  comments  on  one  of  the  poems,  *  The  tua  mariit 
wemen  and  the  wedo.' 

Bremer's  Phonetics  was  at  first  intended  as  an  introduction  to  the  study, 
but,  owing  to  the  necessity  of  supporting  its  views,  it  contains  much  that  will 
appeal  only  to  specialists.  After  its  general  introduction  the  book  is  divided 
into  two  parts:  I.  Our  vocal  organs  and  their  functions.  II.  The  acoustic 
results  of  their  activity.  By  far  the  greater  part  of  his  observations  would,  of 
course,  hold  good  of  Germanic  as  well  as  of  German  phonetics.  Nader  says  of 
the  book :  '*  It  is  a  work  of  merit,  both  because  it  furnishes  a  general  founda- 
tion for  German  dialect  study,  and  because  it  gives  a  careful  and  acute  expo- 
sition of  phonetic  questions  of  a  general  nature." 

In  the  Miscellanea  Swaen  has  a  note  on  Kellner's  study  of  tautology  in  Old 
and  Middle  English  (Englische  Studien,  XX).  KOlbing  replies  for  the  second 
time  to  the  criticisms  of  Vietor  and  Witzoldt  on  the  teaching  of  English  in 
German  universities.  Sommer  pays  a  warm  tribute  to  Richard  Morris,  the 
English  philologist,  who  died  May  12,  1894.    Sommer  describes  him  as  *'a 


Digitized  by  VjOOQ IC 


I08  AMERICAN  JOURNAL    OF  PHILOLOGY. 

man  of  modest  and  retiring  disposition, . . .  disinterested,  self-sacrificing  and 
forbearing/'  and  withal  possessed  of  "  a  great  deal  of  natural  tact."  *'  He  was, 
indeed,  the  first  Englishman  who  has  attempted  to  utilize  the  results  obtained 
in  the  past  hundred  years  in  the  field  of  historical  and  scientific  philology, 
and  to  apply  them  to  the  study  of  his  native  tongue.  What  he  has  really 
achieved  is  witnessed  by  his  contemporaries,  but  will  only  be  thoroughly 
appreciated  by  the  growing  generation  who  have  been  taught  by,  and  have 
learnt  from,  his  books." 

Albert  S.  Cook. 


Rhkinisches  Museum  fCr  Philologie,  Vol.  LI. 

Pp.  1-20  and  31 8-20.  Zwei  neu  aufgefundene  Schriften  der  graeco-syrischen 
Literatur.  V.  Ryssel.  A  translation  into  German  of  two  treatises  found  in 
a  Syrlac  MS  on  Mount  Sinai.  One  is  a  '  Treatise  (composed  by  the  Philoso- 
phers) on  the  Soul.'  the  other  a  free  version  of  Plutarch's  essay  'On  the 
Advantage  to  be  derived  from  one's  Enemies.'  An  inaccurate  translation  of 
the  latter  into  English  has  been  published  by  Eberhard  Nestle,  London,  1894. 
These  two  treatises  have  been  published  in  the  Studia  Sinaitica.  They  are 
found  in  the  same  MS  as  the  Apology  of  Aristides  for  the  Christians. 

Pp.  21-6.  Qui  orationum  Isocratearum  in  archetypo  codicum  ordo  fuerit. 
E.  Drerup.  From  a  comparison  of  the  order  given  by  the  leading  MSS  with 
that  given  by  Photios,  the  writer  concludes  that  the  speeches  of  Isocrates 
were  arranged  in  the  archetype  in  three  parts  of  seven  each,  the  letters  in 
three  parts  of  three  each.  Here  is  his  conjectural  order :  a)  Contra  Sophistas, 
Busiris,  Helena,  Euagoras,  Ad  Demonicum,  Ad  Nicoclem,  Nicocles ;  b)  Archi- 
damus,  Areopagiticus,  Plataicus  (Plataicus,  Areopagiticus  ?),  De  Pace,  Philippus, 
Panathenaicus,  Panegyricus ;  c)  De  Permutatione,  De  Bigis,  In  Callimachum, 
Aegineticus,  In  Euthynum,  Trapeziticus,  In  Lochiten;  d)  Dionysio,  Archi- 
damo,  lasonis  filiis;  Philippo,  Philippo.  Alezandro;  Antipatro,  Timotheo, 
Mytilenaeorum  magistratibus. 

Pp.  27-44.  Textkritisches  zu  Statins.  F.  VoUmer  examines  a  number  of 
passages  in  Statins.  For  the  Thebais,  the  only  reliable  authority  is  the  Codex 
Puteaneus,  and  Kohlmann's  edition  would  have  been  improved  by  a  closer 
adherence  to  the  manuscript  readings. 

Pp*  45-51  and  p.  164.  Zwei  Hermogeneskommentatoren.  K.  Fuhr.  These 
are  (i)  Eustathios,  whose  commentary  is  mentioned  by  Johannes  Doxopatres. 
He  seems  to  have  incorporated  parts  of  an  older  commentary  verbally  into  his 
work.  (2)  Phoibammon,  of  whose  person  and  time  we  know  nothing,  though 
his  name  suggests  that  he  was  of  Egyptian  stock.  His  commentary  was  freely 
used  by  Johannes  Doxopatres. 

Pp.  52-69.  De  Hippiatricorum  codice  Cantabrigiensi.  E.  Oder  describes 
a  MS  in  the  library  of  Emmanuel  College,  and  prints  the  text  of  the  fragment. 

Pp.  70-108.  Beitrfige  zur  lateinischen  Grammatik.  Th.  Birt.  I.  Ueber 
Vocalisirung  desj.    The  trisyllabic  eliam  is  usually  said  to  be  a  compound  of 


Digitized  by  VjOOQ IC 


REPORTS.  109 

et  and  jam^  but  no  parallel  has  been  cited  from  the  historical  period  of  the 
Latin  language  for  the  phonetic  change  thus  assumed.  After  a  prefix  which 
ends  in  a  consonant  j  regularly  retains  its  consonantal  force ;  it  becomes 
vocalic  only  when  it  immediately  precedes  or  follows  a  vowel  i.  In  this  case 
j  first  becomes  vocalic  and  then  disappears  altogether.  Thus  rejicic  (for 
redjUio)  becomes  reiicio,  then  rHcio\  quadrijuga  becomes  quadrijiga^quadrijga^ 
quadriga,  Etiam  must  be  formed  from  ett-jam.  This  eti  (cf.  dti^  in)  would 
correspond  to  et  as  uti  to  ut.  This  hypothesis  is  supported  by  such  combina- 
tions as  eiiam  htm,  etiam  nunc,  and  by  many  passages  in  which  etiam  obviously 
means  *  still,  yet/  which  was  its  earliest  meaning.  In  like  manner  quispiam  is 
formed  from  quispe-Jam^  through  quispi-jam ;  the  trisyllabic  nunciam  of  Roman 
comedy  from  nunce-jam^nunci-jam;  quoniam  from  quone-jam^quoni-Jam,  Inci- 
dentally Professor  Birt  suggests  (p.  83)  that  cumti  is  derived  (not  from  co-juncti^ 
but)  from  eumque^  as  quinius  from  quinque.  This  is  the  cumque  which  is  found 
in  quicumque,  Cumque  =  omnino,  cuncti  =  qui  omnino  sunt.  Its  original 
independence  in  position  is  shown  by  the  so-called  tmesis  in  qui  testamentum 
tradet  tidi  cumque  (Horace),  quod  quaiquc  quomque  incident  (Terence),  etc.  For 
the  addition  of  the  termination  -tus  to  an  adverb  like  cumque^  compare  tam^ 
tantus;  quam^  quantus. 

Pp.  109-26.  Arrians  Periplus  Ponti  Euxini.  C.  G.  Brandis.  The  document 
entitled  ^Appiavoit  ktcurroTJl  rrp6(  Tpalavdv,  kv  y  koI  irepinhjvq  Ev^eivov  UdvTov  is 
found  in  only  one  MS,  Palatinus  398.  The  first  part  of  this  document  seems 
to  be  a  genuine  letter  from  Arrian  to  the  Emperor  (Hadrian) ;  the  remainder 
is  probably  a  forgery  composed  in  the  late  Byzantine  period. 

Pp.  127-37.  Das  alte  Athen  vor  Theseus.  W.  DOrpfeld  replies  to  the 
criticism  of  J.  M.  Stahl  on  his  new  interpretation  of  Thuk.  II  15  (see  vol.  L, 
pp.  566-75 ;  A.  J.  P.  XVI  516).  DOrpfeld  maintains  that  the  lower  part  of 
the  city  of  Kekrops  was  confined  to  the  slope  of  the  hill ;  Stahl,  he  says,  must 
assume  that  it  extended  as  far  as  the  Ilissos — which  is  impossible.  Relying 
upon  the  authority  of  C.  Wachsmuth  (Ber.  der  Sfichs.  Ges.  d.  Wiss.,  phil. 
Klasse,  1877,  S.  387),  he  holds  that  imb  ri^  aKp&iroXiv  must  here  mean  half  way 
up  the  slope,  or  a  fourth  of  the  way  up.  He  thinks  it  certain  that  both  parts 
of  the  ancient  city  were  surrounded  by  the  Pelargic  wall,  and  that  Athens  in 
the  days  of  Kekrops  was  a  small  fortified  town  like  Eleusis,  or  Aphidna,  or 
Thorikos.  He  still  maintains  that  tovto  rb  pipoq  means  the  site  of  the  whole 
of  the  ancient  city,  not  merely  of  the  lower  part,  and  is  still  of  the  opinion 
that  n-p^c  here  means  '  on '  or  '  on  to.* 

Pp.  138-52.  Der  pseudoeuripideische  Anfang  der  Danae.  R.  Wttnsch. 
This  fragment,  which  has  been  regarded  as  spurious  since  the  days  of  Elmsley 
and  Jacobs,  seems  to  have  been  written  by  Markos  Musuros.  The  terminus 
post  qnem  for  its  composition  is  the  time  of  Theodorus  Prodromus,  i.  e.  the 
first  half  of  the  twelfth  century;  the  terminus  ante  qaem  is  the  year  of 
Musuros'  death  (15 17),  or  rather  the  year  of  his  final  departure  from  Venice 

(1516). 

Miscellen.— P.  153.  F.  Bacheler.  Versus  tragicus  graecus.  The  verse  is 
(A  d*  afi^v^ovrat  "SiSav  etrkXrfKdTa,  ilH  circumstipahmt  sibi  Nedam  rigidis  torridis. 


Digitized  by  VjOOQ IC 


no  AMERICAN  JOURNAL    OF  PHILOLOGY, 

It  is  gained  by  a  very  slight  change  in  the  verse  quoted  by  Hesychius  in  his 
gloss  on  tCKhiK6Ta, — Pp.  153-7.  Th.  Birt.  Zu  Antisthenes  und  Xenophon 
The  fourth  book  of  the  Memorabilia  was  an  independent  essay  7rep£  ircuSeiac^ 
and  was  intended  to  be  a  polemic  against  Antisthenes.  The  four  divisions  of 
education  are  announced  in  Memorab.  IV  3,  i,  apparently  in  the  order  of 
their  importance  in  the  eyes  of  Antisthenes;  but  in  the  discussion  which 
follows  the  practical  Xenophon  takes  them  up  in  a  different  order. — Pp.  157- 
60.  J.  M.  Stahl.  Zu  Philons  Schrift  vom  beschaulichen  Leben.  A  discussion 
of  the  passage  479  M  27-49,  ^^^^  ^  suggestion  toward  the  improvement  of  the 
text. — Pp.  160-62.  M.  Manitius.  Handschriftliches  zur  Anthologia  latina. 
Variae  lectiones  of  several  poems  of  the  Anthology  (395,  394,  639,  640,  736) 
found  in  the  Codex  Berolin.  Philipp.  1869,  s.  IX. — Pp.  162-3.  J.  Ziehen.  Zu 
Cicero  ad  Quintum  fratrem  III  i.  For  Velvinum  (v.  1.  velvinus)  in  §4  the 
writer  proposes  V  (=  Varro)  eluviem  \  for  silva  viridicata  in  §3  he  would  read 
silva  viridi  ditatam, — P.  164,  R.  Fuchs.  Nachtrag  zu  Band  L,  S.  580.  The 
word  Pov^iuv  is  explained  by  Gustav  Meyer,  Neugriechische  Studien,  II,  as 
meaning  '  elder'  (tree). — K.  Fuhr.    Nachtrag  zu  oben,  S.  48  f. 

Pp.  165-96  and  p.  466.  Ueber  die  Schriftstellerei  des  Klaudios  Galenos. 
III.  J.  Ilberg.  (Continued  from  vol.  XLVII  514 ;  see  A.  J.  P.  XV  387.)  An 
account  of  Galen's  treatises  on  pathology,  therapeutics  and  hygiene,  with  a 
table  showing  the  probable  order  of  their  composition.  Almost  all  of  them 
were  written  during  his  second  residence  at  Rome. 

Pp.  197-210.  Die  Textgeschichte  des  Rutilius.  C.  Hosius  publishes  a 
collation  of  a  MS  in  the  library  of  the  Duke  of  Sermoneta  in  Rome. 

Pp.  211-25.  Die  panathenaischen  und  eleusinischen  Isponoioi.  L.  Ziehen. 
I.  Aristotle,  *A07fi/,  IIoX.  54,  7,  says  that  the  *  annual '  UpoKotoi  superintended 
certain  sacrifices  and  all  the  quinquennial  festivals  except  the  Panathenaea. 
Ziehen  thinks  that  this  exception  was*  intended  to  include  both  the  greater 
Panathenaea  and  the  less.  The  latter  was  entrusted  to  a  special  commission, 
which  is  described  in  C.  I.  A.  II  163  lepoTrotol  oi  dtouoovvreq  to.  TLava&ivata  ra 
kut'  kviavrdv.  II.  From  the  same  section  of  the  'Aftyv.  IIoA.  it  is  evident  that 
the  quinquennial  Eleusinia  was  under  the  general  charge  of  the  Uporrotol  kqt* 
kviavT&if,  The  inscriptions  clearly  point  to  the  existence  of  a  different  set  of 
lepoKoioi  whose  duties  were  confined  to  the  temple  at  Eleusis.  These  were  at 
first  called  1.  *E?£votv66eVf  but  between  419  and  329  their  name  and,  to  some 
extent,  their  duties  were  changed,  and  they  were  known  in  later  times  as  I.  ey 


Pp.  226-39.  ^^  Verhfiltniss  der  aristotelischen  zu  der  thukydideischen 
Darstellung  des  Tyrannenmordes.  P.  Co'rssen.  In  the  i8th  chapter  of  the 
'Adfivaluv  UoXiTela  the  vengeance  of  Harmodios  and  Aristogeiton  is  referred  to 
the  behavior  of  Thessalos,  not  of  Hipparchos.  In  order  to  put  Aristotle's 
account  of  the  assassination  in  accord  with  that  of  Thukydides,  J.  M.  Stahl 
proposes  to  strike  out  the  words  kqi  tov(  irepl  "AvaKpiovra  . .  .  eiTTa}/)^  6k 
vedrepoc  noXh  (vol.  L,  pp.  382  ff.;  A.  J.  P.  XVI  514-15).  Corssen  maintains 
that  this  passage  is  genuine.  Aristotle  deliberately  differs  from  Thukydides 
in  his  account  of  this  event,  and  this  is  not  the  only  point  of  difference. 


Digitized  by  VjOOQ IC 


REPORTS,  III 

Pp.  240-72.  Beitrage  zur  lateinischen  Grammatik  (continued  from  p.  108). 
II.  Ueber  KUrzungen  trochaischer  WOrter.  Th.  Birt  rejects  the  view  of  F. 
Skutsch  (Forschungen  zur  lateinischen  Grammatik  und  Metrik,  Leipzig,  1892) 
that  Plautus  sometimes  dropped  the  final  e  of  such  words  as  nempe^  untU^  inde^ 
ille^  and  that  the  resulting  nemp^  und^  ind^  Uivitx^  often  further  simplified  to 
nem^  un,  in^  il.  Our  authorities  for  the  popular  pronunciation  of  Latin  show 
no  trace  of  such  shortened  forms.  According  to  Skutsch's  own  statistics 
nempe,  inde^  etc..  are  made  pyrrhic  in  Plautus  only,  or  most  frequently,  when 
they  are  followed  by  a  consonant,  a  fact  which  shows  the  careful  pronunciation 
of  the  final  e.  Nor  is  it  probable  that  ilU  and  ilia  were  both  reduced  to  the 
form  tV.  Birt  maintains  that  the  shortening  of  these  words  is  often  due  to  the 
fact  that  they  lack  emphasis  or  accent  in  the  sentence.  All  the  pronominal 
forms  under  consideration  are  naturally  unemphatic  and  are  apt  to  be  rapidly 
pronounced.  The  German  *n&mlich'  is  not  an  adequate  translation  of  the 
conversational  nempe\  better  is  'ja'  or  'doch/  e.  g.  Cure.  42  nempe  obloqui 
me  iusseras  *  Dn  befahlst  mir  doch  zu  widersprechen.'  The  proclitic  nature 
of  nempe  is  shown  by  its  vocalic  weakening  (cf.  igitur  for  agitur)^  whereas  the 
emphatic  namque  has  retained  its  a.  Incidentally  the  writer  discusses  the 
^exilitas'  of  intervocalic  //  in  Latin,  especially  with  an  i  preceding,  and 
explains  certain  cases  of  so-called '  elision.*  Vas  argenteis  for  iftLsis  argenteis 
and  palm  et  crinibus  for  palmis  et  crinibus  are  just  such  expressions  as  the 
German  *  ein  'Und  demselben '  for  *  einem  und  demselben,'  or  '  in  gut  und 
bOsen  Tagen.'  The  form  omnimodis  is  due  to  a  similar  conscious  omission  or 
ellipsis  of  a  syllable,  and  multimodis  is  the  natural  result  of  analogy.  Here 
belong  several  of  the  supposed  cases  of  elision  of  the  s  before  vowels  given 
by  F.  Leo  (Plautinische  Forschungen,  S.  231  ff.).  Inde  has  always  kept  its  d 
intact,  while  deinde^  prnnde  and  exinde  sometimes  lost  their  final  de.  All  these 
words  begin  with  a  preposition  which  governs  the  ablative.  In  de-in-de  the 
final  de  seemed  superfluous,  and  was  omitted  in  popular  speech ;  prvin  and 
exin  were  formed  by  analogy.  Perinde  escaped  a  like  mutilation  because /^r 
did  not  govern  the  ablative ;  subinde  did  not  appear  in  literature  until  after 
the  shortened  forms  had  fallen  into  disfavor. 

Pp.  273-80.  Die  Theosophie  des  Aristokritos.  A.  Brinkmann.  Cotelerius 
and  ToUius  have  published  the  formula  of  abjuration  dictated  by  the  Greek 
church  to  converted  Manichaeans.  In  its  present  form  this  formula  seems  to 
date  from  the  second  half  of  the  ninth  century.  The  first  half  is  directed 
against  the  original  heresy  of  the  Manichaeans;  the  second  half  seems  to 
have  been  added  about  a  century  later  for  the  benefit  of  their  direct  successors, 
the  Paulicians.  At  the  end  of  the  first  part  is  an  index  librorum  prohibitorum 
which  includes  the  *  Theosophy '  of  a  certain  Aristokritos :  nyv  'ApuTTOKpirov 
pip^mff  fjv  kniypailfe  Beoao^iav,  ev  y  iretparcu  6eiKVvvcu  rdv  *lov6diafi^  Kai  rbv 
'EX^^iOfA^  Kol  rdv  XpujTtaviafidv  koX  rdv  Mavtxa'iofidv  h>  etvoi  xai  to  airrb  Sdyfia. 
Brinkmann  thinks  that  this  work  is  identical  with  a  work  which  had  the  same 
purpose  and  the  same  title,  and  from  which  we  have  some  extracts  in  the 
Xpfpjfiol  Totf  'EAAjTVixuv  published  by  Buresch  (Klaros.  Unters.  zum  Orakelwesen, 
Leipzig,  1889,  S.  95  ff.). 

Pp.  281-302.  Die  Amtstracht  der  Vestalinnen.  H.  Dragendorff.  The 
dress  of  the  vestal  virgin  is  essentially  that  of  the  bride.    The  cingulum  and 


Digitized  by  VjOOQ IC 


f 


112  AMERICAN  JOURNAL    OF  PHILOLOGY. 

the  tnrrita  corona  were  worn  by  both;  the  flammenm  of  the  bride  and  the 
snffibulum  of  the  vestal  were  originally  identical.  The  seni  crines  of  the 
vestal  was  a  kind  of  peruke  which  almost  entirely  concealed  her  own  hair. 
By  the  formal  captio  the  vestal  is  freed  from  the  patria  potestas,  and  her 
position  is  henceforth  that  of  the  mater  familias.  The  captio  is,  as  it  were,  a 
marriage  to  the  deity,  who  is  represented  at  first  by  the  king,  later,  by  the 
pontifex ;  the  words  used  by  the  pontifex  in  this  ceremony  are  significant :  ita 
te,  amatay  capio.  So  the  Christian  maiden  renounces  earthly  marriage  to 
become  the  bride  of  Christ.  It  is  significant,  also,  that  it  was  the  pontifex 
who  punished  unchastity  in  the  vestal,  and  that  this  punishment  was  the  same 
as  was  inflicted  in  early  times  by  the  injured  husband  upon  the  adulterous 
wife. 

Miscellen. — Pp.  303-4.  F.  Solmsen.  Ein  nominaler  Ablativus  Singularis 
im  Griechischen.  This  is  the  word  foiKu  in  an  inscription  recently  discovered 
at  Delphi  and  published  by  Th.  HomoUe  (Bull.  corr.  hell.  XIX  5  ff.).  HomoUe 
explained  the  word  as  a  genitive  singular. — Pp.  304-5.  J.  Wackernagel.  Das 
Zeugniss  der  delphischen  Hymnen  ttber  den  griechischen  Accent.  A  word 
which  has  the  grave  accent  may  be  compared  with  the  syllables  of  a  single 
word  which  precede  the  accent. — Pp.  306-11.  J.  M.  Stahl.  Noch  einmal  das 
vortheseische  Athen.  A  reply  to  DOrpfeld's  paper,  pp.  127-37.  DOrpfeld's 
topographical  view  as  to  the  position  of  the  Dionysion  kv  Ai/ivaic  and  the 
Enneakrounos  is  only  an  hypothesis,  which  cannot  be  admitted  unless  it 
agrees  with  the  statement  of  Thukydides.  His  excavations  have  not  yet 
furnished  a  single  certain  proof  of  his  claim.  Stahl  is  still  skeptical  of  the 
discovery  of  the  site  of  a  Dionysion  kv  Aifivaic  which  was  about  15  m.  higher 
than  the  ancient  market-place,  and  still  rejects  D.'s  explanation  of  inr*  avr^, 
of  TovTo  rd  fUpoc  and  of  irpbc  tovto  t6  fUpoq, — Pp.  311-14.  E.  Oder.  Ad 
Simonis  Atheniensis  fragmentum  (pp.  67-9)  addendum.  Some  notes,  supplied 
by  F.  Kenyon,  on  a  MS  in  the  British  Museum.— Pp.  314-15.  L.  Rader- 
macher.  De  Phoinicis  loco.  A  proposed  correction  of  a  verse  cited  by  Athe- 
naeus,  530*.  R.  would  read  rdXavf,  'Ano^Xov,  for  xai  raXla  iroXXov, — Pp.  315-18. 
M.  Ihm.  Zu  Philodem  irepl  KoXaKeiag. — Pp.  31 8-20.  V.  Ryssel.  Nachtrag  zu 
'  Zwei  neu  aufgefundene  Schriften  der  graeco-syrischen  Literatur'  (Bd.  LI,  S. 
I).  The  writer  has  recently  discovered  the  Greek  text  of  the  *  Treatise  of 
(one  of)  the  Philosophers  on  the  Soul.'  The  Syriac  version  is  a  translation  of 
the  A&yoc  Ke^aXat66ifc  ^epi  ^IrvxvC  ^P^  TaTiav6v  written  by  Bishop  Gregorius 
Thaumaturgus  of  Neocaesarea,  who  died  about  270  A.  D. — Pp.  320-25.  E. 
Hoffmann.  Die  Fescenninen.  Horace,  Ep.  II  1. 145,  says  that  the  Fescen- 
nina  licentia  was  introduced  at  the  ancient  harvest -home ;  Ve^il,  G.  II  385, 
refers  not  to  the  festival  of  the  vintage,  but  to  the  Liberalia  which  fell  on 
March  14.  Liber,  a  native  Latin  deity,  was  the  author  of  fertility,  and  his 
symbol  was  the  fascinum.  The  name  versus  fescenmni  is  due  to  the  promi- 
nence of  the  fascinum  in  the  celebration  of  his  festival. — Pp.  325-6«  F- 
Buecheler.  Zum  Gedicht  des  Pseudosolinus.  The  title  ponHcon  may  be  a 
corruption  of  poeiUon,  To  judge  from  the  Latinity,  the  poem  was  probably 
composed  in  the  age  of  the  Antonines.«-Pp.  327-8.  C.  Weyman.  Varia. 
Notes  on  Acts,  XXVIII  16;  Juvencus,  II  754;  Damasus,  XXXII  z  ff.  (read 


Digitized  by  VjOOQiC" 


I 

i 


REPORTS,  113 

eamifices,  not  carmjUis) ;  Carm.  Ut.  epigr.  727  B ;  and  on  the  expression  difnu 
dignisj^V.  328.  S.  Frftnkel.  Zu  Band  L,  S.  587.  For  the  obscure  word 
robprrer,  cf.  Ducange,  Z591,  and  L5w,  Aram.  Pflanzenn.,  410,  Nr.  80. 

Pp.  329-80.  Die  drei  Brftnde  des  Tempels  zu  Delphi.  H.  Pomtow.  It  is 
commonly  believed  that  the  temple  at  Delphi  built  by  the  Amphictyons  after 
the  conflagration  of  548-47  remained  standing  for  more  than  700  years,  and 
that  it  is  this  temple  whose  ruins  have  recently  been  discovered  by  the  French 
excavations  at  the  village  of  Kastri.  This  belief  rests  upon  the  express  state- 
ments of  Pausanias  (X  5, 13),  Strabo  (IX  421)  and  others.  A  new  temple  of 
stone  was  begun  about  540  and  completed  about  520-15.  The  architect  was 
Spintharos  of  Corinth.  This  was  the  ddfioc  Beajr^  of  Pindar's  seventh  Pythian. 
The  temple  of  Spintharos  was  destroyed  by  fire  about  372.  The  rebuilding 
soon  began,  but  was  not  completed  before  the  end  of  the  third  century. 
Under  the  year  84  B.  C.  =  01.  174,  I,  Eusebius  says  (II,  p.  133,  Schoene): 
**  templum  tertio  apud  Delfos  a  Thracibus  incensum  et  Romae  Capitolium." 
Tko  M^f6av  in  Plutarch,  Numa,  9,  should  be  corrected  to  virb  yialSuv^  and  rd 
AeX^ucdv  in  Appian,  lUyr.  5,  is  a  mistake  for  rd  Auduvalov,  This  third  fire  may 
be  assigned  to  the  last  quarter  of  Ol.  174,  I  ==84-83,  i.  e.  to  April-June  of 
B.  C.  83.  After  this  calamity  Delphi  sank  into  a  condition  of  utter  insignifi- 
cance and  helplessness  which  lasted  for  more  than  a  century.  The  work  of 
restoration  was  perhaps  begun  in  the  time  of  Augustus,  but  little  progress  was 
made  until  the  visit  of  Nero  to  Delphi  in  the  autumn  of  67 ;  Xkyovai  &  bri 
ifieivtv  km  TToAw  XP^^  areXr^  (sc.  6  vc^),  e«f  ov  verepov  "Nipuv^  6  paatXevc 
'Fofioluv,  knTJipuatv  avr^,  knelae  irapayevdftevoc  (Schol.  to  Aeschin.  Ctesiph.  115). 

Pp.  381-400.  Zu  Ciceros  Rede  pro  Flacco.  F.  SchOU.  Textual  notes, 
especially  on  the  fragmentary  introduction. 

Pp.  40i-4a  Die  jetzige  Gestalt  der  Grammatik  des  Charisius.  L.  Jeep. 
Diomedes  must  have  known  and  used  the  work  of  Charisius.  Charisius  did 
not  borrow  directly  from  Romanus,  but  both  drew  upon  common  sources. 

Pp.  441-55.  Beitrftge  zur  Kritik  und  Erkl&rung  des  Dialogs  Axiochos. 
A.  Brinkmann. 

Pp.  456-62.  Das  Wahlgesetz  des  Aristeides.  E.  Fabricius.  Plutarch, 
Arist.  22,  mentions  a  decree  proposed  by  Aristides  in  487-86,  rove  apxotrraiQ  k^ 
'Afhp^aUjv  airdvTuv  aipeiadai.  This  statement  is  apparently  contradicted  by  the 
*Adriv<uuv  Uohreia^  but  it  need  not  be  absolutely  rejected  as  a  "groundless 
invention"  (Wilamowitz,  Aristoteles  und  Athen,  I,  S.  124.  4).  Aristides  did 
propose  such  a  decree,  not  as  a  constitutional  change,  but  as  an  exceptional 
procedure  for  that  year  to  suit  the  exceptional  circumstances  in  which  the 
Athenians  found  themselves.  Plutarch's  authority  was  probably  the  decree 
itself,  which  may  have  been  published  by  Craterus. 

Miscellen. — Pp.  463-6.  L.  Radermacher.  Varia.  Textual  notes  on  several 
passages  in  Aelian. — P.  466.  J.  Ilberg.  Ueber  Galenos,  Nachtrag. — Pp.  466-8. 
K.  Kalbfleisch.  Ueber  Galens  Schrift  Uepl  ?.tirrwoi}anc  Siair^, — Pp.  468-70. 
Th.  Birt.  Zu  Catnll  und  Petron.  GemeUi  in  Catullus,  57,  6,  means  *  testiculi; 
and  tUrique  is  dative.    There  is  a  similar  use  of  ol  Sidvfioi  in  Greek ;  cf.  Philo> 


Digitized  by  VjOOQ IC 


114  AMERICAN  JOURNAL   OF  PHILOLOGY. 

dem.,  Anthol.  Pal.  V  125  tw^  keIvov  neXinei  del  6i6ifiovc  d^AfZv.  Ammiantts 
Marcellinus,  Claudianus  and  Solinus  use  gemini  with  the  same  meaning,  and 
this  meaning  must  be  lurking  in  two  passages  of  Petronius,  35  and  39 :  **  super 
geminos  testiculos  ac  rienes/*  and  "  in  geminis  nascuntur  . .  .  colei."— Pp.  470- 
71.  O.  Hirschfeld.  Petronius  und  Lucianus.  The  oiKkry  veonhjiyrt^t  of  the 
twentieth  chapter  of  Lucian*s  essay  ttuq  del  iaropiav  avyyp&^tv  is  probably  a 
reminiscence  of  Trimalchio ;  cf.  Cena  Trim.  76,  32  and,  especially,  36. — P.  47 '• 
J.  Gilbert.  Ad  Petroni  saturas  (53).  Proposes  to  read  reliqua  enim  iaUa 
acroamata  etc.  for  reliqua  [animalia]  acroamata  etc. — Pp.  471-3.  F.  BUcheler. 
Altes  Latein  (vgl.  Band  XLVI,  S.  233).  XX.  The  epitaph  of  Encolpus, 
C.  I.  L.  VI  14672,  may  be  assigned  to  the  third  generation  after  Petronius 
and  Nero.  The  word  opter  is  a  genuine  archaism ;  for  its  formation  cf.  inter, 
pnuter,  propter,  etc. — Pp.  473-4.  M.  Ihm.  Tessera  hospitalis.  F.  Bamabei, 
Notizie  degli  scavi  for  March,  1895,  describes  a  token  recently  discovered  at 
Trasacco.  It  is  the  half  of  a  small  ram's  head  of  bronze  divided  lengthwise, 
the  cut  surface  being  inscribed  with  the  names  of  two  men  and  the  word 
*  hospes.'  It  probably  belongs  to  the  second  century  B.  C.  This  discovery 
explains  the  purpose  of  a  similar  token  now  at  Vienna.  It  also  supports  the 
old  view  that  these  tokens  were  made  by  dividing  a  single  object,  each  party 
keeping  one  part.  Cf.  Plato,  Sympos.  191  D  and  193  A;  Plant.  Poen.  1047  f. 
— Pp.  474-5.  O.  Hirschfeld.  Das  Consulatsjahr  des  Tacitus.  This  was  the 
year  97.  The  person  referred  to  in  Plin.  Panegyr.  58,  erat  in  senatu  ter 
consul,  was  not  Verginius  Rufus,  but  Fabricius  Veiento. — Pp.  475-7.  O. 
Hirschfeld.  Die  Tyrier  in  dem  zweiten  RSmisch-Karthagischen  Vertrag. 
Polybius,  III  24,  wrote  Tvpiuv  for  Kvplcjv  by  mistake. — Pp.  478-80.  F.  Skutsch. 
Randbemerkungen  zu  S.  240  if.  A  brief  reply  to  some  of  Th.  Birt's  criticism, 
pp.  253-6.  No  Roman  poet  hesitated  to  place  t%  iUa,  illam,  etc.,  before  a 
word  beginning  with  a  vowel,  because  the  gender  of  the  pronoun  was  likely 
to  be  obscured.  Plautus  did  sometimes  shorten  the  first  syllable  and  elide 
the  last  syllable  of  the  same  word  ;  e.  g.  Aulul.  708  tibi  Ylle  ibiit,  785  ^go  Ilium 
lit,  Asin.  370,  757,  Rud.  960.  etc.,  etc.— P.  480.  C.  F.  W.  M.  Zu  Band  LI, 
S.  328.  With  the  expression  digna  dignis  compare  Amob.  I  39,  p.  26,  19  Reif. 
digna  de  dignis  sentio. 

Pp.  48I--9I.  Zur  Handschriftenkunde  und  Geschichte  der  Philologie. 
(Continued  from  vol.  XL,  pp.  453  f[.;  A.  J.  P.  X  112.)  R.  Foerster.  IV. 
Cyriacus  von  Ancona  zu  Strabon.  Cyriacus  of  Ancona  had  a  copy  of  the 
seventeen  books  of  Strabo  made  for  him  by  a  friend  in  Constantinople,  and 
on  the  margin  of  this  copy  he  added  all  sorts  of  geographical,  historical  and 
linguistic  comments  with  his  own  hand.  The  first  part  of  this  MS,  containing 
the  first  ten  books,  is  in  the  library  of  Eton  College,  the  second  part  (11-16) 
is  at  Florence,  Laurent.  XXVIII  15.     The  interesting  history  of  its  fortunes. 

Pp.  492-505.  De  Properti  poetae  testamento.  Th.  Birt.  A  commentary 
on  Propert.  II  13  ^.  After  the  introductory  couplet  the  poem  falls  into  two 
parts  of  equal  length.  The  first  part  closely  follows  the  order  of  the  Roman 
funeral  rites.  The  words  funeris  acta  mei  (v.  18)  may  be  compared  with 
mandata  dt  future  sua  (Sueton.  Aug.  loi). 


Digitized  by  VjOOQ IC 


I 


REPORTS,  11$ 

Pp.  506-28.  De  Franconim  Gallornmque  origine  Troiana.  Th.  Birt  defends 
the  epithet  Gallicus  in  Propert.  II  13.  48  Gallicus  Iliacis  miles  in  a|ggeribtt8. 
The  belief  that  the  Franks  and  Ganls  were  descended  from  the  Trojans  was 
widely  spread  throaghout  the  Middle  Ages,  and  must  have  existed  at  Rome 
when  this  poem  was  written.  Cf.  Aethicus,  Cosmogr.  (ed.  Wuttke,  p.  77) ; 
l^ucan,  Phars.  I  427 ;  Amm.  Marc.  XV  9 ;  Caes.  B.  G.  I  33.  2 ;  Qaint.  Smym. 
Posthom.  VII  611.  The  use  of  this  epithet  is  very  like  the  erudite  Propertius ; 
for  its  application  cf.  Teucrum  Qtdrinum,  IV  6.  21. 

Pp.  529-43.  Neu  aufgefundene  graeco-syrische  PhilosophensprQche  aber 
die  Seele.  V.  Ryssel.  These  *  Sayings  of  the  Philosophers '  are  found  in  the 
same  Syriac  MS  as  the  '  Treatise  of  a  Philosopher  on  the  Soul*  (pp.  i  if.).  A 
German  translation  is  given. 

I^P*  544-59*  Excurse  zu  Virgil.  O.  Crusius.  I.  Entstehung  und  Compo- 
sition der  achten  Ekloge.  The  writer  refutes  the  heresy  of  the  prosaic  E. 
Bethe  (vol.  XLVII,  590  if.;  A.  J,  P.  XV  387)  that  the  two  songs  of  the  eighth 
Eclogue  were  originally  intended  to  be  independent  mimes,  not  counterparts 
for  an  agon.  II.  Zur  vierten  Ekloge.  In  v.  60  risu  can  only  mean  the  laugh 
of  the  child,  and  the  subject  of  risere^  v.  62,  must  be  the  same  as  the  logical 
subject  of  risu.  Crusius  would  read  with  Quintilian  qui  turn  risen  parenH, 
The  ttascens  puer  of  v.  8  is  not  the  child  of  any  Roman  noble ;  he  is  a  purely 
imaginary  wonder-child.  Modo  in  the  same  line  should  be  compared  with 
modo^  Aen.  IV  49  f.  The  mystic  imagery  of  the  beginning  and  close  of  the 
poem  is  of  Sibylline  origin,  e.  g.  v.  10  and  w.  50-51.  The  infant  is  to  show 
at  once  that  he  is  more  than  human  (v.  60) ;  this  idea  is  derived  from  Greek 
mysticism. 

Pp.  560-88.  Delphische  Beilagen  (S.  oben  S.  329).  H.  Pomtow.  I.  Die 
Jahre  der  Herrschaft  des  Peisistratos.  Merd  61  rairra  in  Arist.,  *A6tfv,  IIoA.  XV 
I,  should  be  changed  to  fitrd  Si  raimfv,  and  trti  uAXiara  i^6fti,)  to  fupft  fi&hjora 
eP66fu,f.  The  career  of  Peisistratos  was  as  follows:  first  tyranny,  spring  to 
autumn,  560 ;  first  exile,  autumn  560  to  the  end  of  556-55 ;  second  tyranny, 
seven  months  of  555~54 ;  second  exile,  spring  554  to  the  end  of  545-44 ;  third 
tyranny,  middle  of  544  to  spring,  528-27.  II.  Die  Datirung  der  VII.  Pythi- 
schen  Ode  Pindars.    The  date  of  the  poem  is  B.  C.  486. 

Pp.  589-95.  Textkritisches  zu  Ciceros  Briefen.  J.  Ziehen  proposes  the 
following  readings:  i)  Qu.  F.  II  14,  2  nee  labor antiqua  mea  etc.;  2)  Qu.  F.  I 
I,  II  atque  incertos  eos  quos  etc.;  3)  Att.  II  20,  I  sed  quia  hohpragmoHci 
homines  etc.;  4)  Att.  IV  11,  2  abs  te  opipare  delector  etc.;  5)  Att.  XI  23,  3 
audimus  enim  de  statua  Clodi\  6)  Brut.  I  4,  5  prorsus  alienae  etc.;  keep  the 
reading  unchanged  and  make  prorsus  ironical ;  7)  Qu.  Ill  8,  i  LaSeoni  dedisse, 
qui  adhuc  non  venerat. 

Pp.  596-629.  Ueber  den  Cynegeticus  des  Xenophon.  I.  L.  Radermacher 
concludes  from  an  examination  of  the  language  and  style  of  the  Cynegeticus 
that  it  is  not  the  work  of  Xenophon.  Even  the  mention  of  bears  (XI  i) 
becomes  a  stumbling-block,  and  so  does  the  absence  of  all  reference  to  riding 
to  hounds. 


Digitized  by  VjOOQ IC 


Il6  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Miscellen. — Pp.  630-32.  H.  Weber.  Zu  Ariston  von  Chios. — Pp.  632-6. 
E.  Ziebarth.  Zur  Epigraphik  von  Thyateira. — Pp.  636-7.  W.  Schwarz.  Die 
Heptanomis  seit  Hadrian.  By  the  founding  of  the  city  of  Antinoe,  under  the 
Emperor  Hadrian,  the  number  of  districts  in  the  Heptanomis  was  increased  to 
eight.  The  district  of  Arsinoe  was  then  separated  from  the  Heptanomis,  and 
the  domain  of  the  old  Heptanomis  was  thereafter  described  as  *'  epistrategia 
septem  nomorum  et  Arsinoitae."  This  Arsino€  was  the  city  on  Lake  Moeris, 
not  the  Arsinoe  on  the  Red  Sea.  Cf.  Orelli,  Inscr.  516 ;  C.  I.  L.  Ill  6575.— 
Pp.  637-8.  A.  Riese.  Zu  Statins'  Silven.  Proposes  to  change  calvum,  IV  3, 
19,  to  clavum. — P.  638.  M.  Ihm.  Zu  Augustins  Confessiones.  For  inspirabat 
populo  jam^  VIII  2,  3,  read  inspirabat  popuJo  Osirim,  Cod.  Bamberg,  s.  X  has 
populosirim,  —  Pp.  638-40.  F.  Buecheler.  De  inscriptionibus  quibusdam 
christianis.  Notes  on  some  inscriptions,  both  Greek  and  Latin,  recently 
found  in  the  catacombs  at  Syracuse.  They  were  written  between  the  years 
383  and  452. 

Havbkvord  Collkgb.  Wilfred  P.  Mustard. 


BeitrAgk  zur  Assyriologib  und  skmitischkn  Sprachwissenschaft,  heraus- 
gegeben  von  Friedrich  Deutzsch  und  Paul  Haupt.  Dritter  Band, 
Heft  3  (pp.  385-492).    Leipzig,  1897.1 

The  third  Heft  of  the  third  volume  of  the  Beitr&ge  contains  three  articles. 

The  first  of  these  (pp.  385-92)  is  an  introductory  paper  by  Friedrich 
Delitzsch,  embodying  some  *  preliminary  remarks'  to  the  two  following 
treatises  by  Demuth  and  Ziemer  on  legal  and  government  records,  dating 
from  the  reigns  of  Cyrus  and  Cambyses.  Delitzsch  takes  this  opportunity  to 
explain  the  method  of  transliteration  (his  own)  followed  by  both  writers,  to 
comment  on  the  reading  of  the  proper  names,  and  to  give  a  complete  table  of 
the  numbers  of  the  texts  as  they  occur  in  Strassmaier  and  in  Demuth-Ziemer's 
work. 

He  devotes  two  pages  to  an  interesting  discussion  of  the  doubtful  words 
imiUu  and  saUukku,  The  first  of  these,  which  occurs  hundreds  of  times  in  the 
legal  tablets  in  apposition  to  suluppu  *  dates,'  he  explains  with  great  ingenuity 
as  meaning  *  assessment,  valuation,*  deriving  it  from  emidu  *  to  impose/  e.  g.  a 
tax  or  duty  {^imidtu  =  imittu),  Suluppu  imittu,  therefore,  are  dates  which  are 
to  be  paid  by  the  tenant  to  the  proprietor  as  a  rent,  according  to  a  previous 
agreement  between  the  owner  and  the  lessee  of  a  field.  Delitzsch  had  already 
conjectured  that  this  was  the  meaning  of  imittu  in  his  AW.,  p.  93,  but  arrives 
definitely  at  this  conclusion  in  this  article  in  the  Beitrlige,  being  led  thereto 
by  a  passage  in  a  legal  document  which  he  cites  in  full,  where  the  word  is 
used  without  any  doubt  in  the  sense  of  *  rent' ' 

He  states  also  that  sattukku  does  not  mean  *  established  offering,*  which  is 
the  meaning  given  in  his  AW.,  p.  513,  but  rather  *the  established,  regular 
standard  of  value'  {GehaU)}  This  word  seems  to  be  an  intensive  noun -form 
from  a  stem  *]nD,  which  probably  meant  originally  *  to  stand,  to  be  perpetual.' 

>  For  the  report  on  Bd.  Ill,  Heft  a,  see  A.  J.  P.  XVII,  pp.  xax-s. 

sCf.  also  Demnth,  p.  404.  sCf.  also  Demnth,  p.  438. 


Digitized  by  VjOOQ IC 


REPORTS.  117 

Thus  we  find  the  adjective  saiiakka,  Nerigl.  ii.  12,  used  practically  synony- 
mously with  la  baflak  *  unceasing/  Sattukku  is  also  discussed  ZA.  I,  p.  3. 
The  noun  mastaku '  place  of  abode '  seems  also  to  be  a  derivatiye  from  the 
same  stem. 

The  second  article  in  the  Heft  (pp.  393-444)  is  a  transliteration  and  trans- 
lation with  philological  commentary  by  Ludwig  Demuth,  of  fifty  legal  and 
government  records  of  the  time  of  Cyrus  (538-529  B.  C). 

Among  other  interesting  legal  peculiarities  of  the  Babylonians,  the  author 
explains  (p.  400)  the  laws  in  force  regulating  the  value  of  slaves,  if  offered  as 
security  for  a  debt.  Thus,  according  to  him,  a  female  slave  and  her  daughter 
were  accepted  as  security  only  when  the  debtor  offering  them  owed  the 
interest  on  the  capital  debt,  e.  g.  the  slaves  were  expected  to  pay  the  amount 
of  interest  due  by  their  labor  for  the  creditor  during  a  fixed  period  of  time. 
If,  however,  the  debtor  owed  his  principal,  slaves  were  not  regarded  as  a 
satisfactory  security.    In  this  case  it  was  necessary  to  offer  real  property. 

The  opinion  expressed,  p.  408,  that  the  original  meaning  of  urdhi  was  not 
'assignment/  but  rather  *debt,  obligation,'  is  highly  interesting,  as  it  suggests 
the  possible  derivation  of  the  word  from  erisu  *  to  desire,  demand,'  e.  g.  urdhi 
might  have  meant  *a  demand  on  a  person,*  hence  *an  obligation.'  Demuth 
considers  ilihi  a  syhonym  of  urdhi  (p.  409). 

The  author's  remarks  on  government  slaves  (p.  417)  are  also  very  instructive. 
He  shows,  in  commenting  on  the  expression  arad-larriitu,  that  there  may  have 
been  certain  male  slaves  who  had  been  conquered  in  battle  and  who  were 
forced,  either  to  render  military  service,  or  to  work  on  the  royal  buildings 
(palaces,  temples,  walls,  etc.).  He  adds,  however,  that  it  seems  probable  that 
these  slaves  were  purchasable  by  private  persons. 

The  third  and  last  article  in  the  Heft  is  a  similar  treatise  (pp.  445-92)  by 
Ernst  Ziemer  on  the  legal  and  government  records  of  the  reign  of  Cambyses 
(529-521  B.  C). 

It  is  interesting  to  notice  that  in  Nr.  I  of  these  selections  the  fact  is 
recorded,  but  not  especially  alluded  to  by  Ziemer,  that  Cambyses  was  coregent 
while  his  father,  Cyrus,  was  still  living.  Both  Solomon  and  Ahirbdnipal^  and 
possibly  Belshazzar,  son  of  Nabonldus,  probably  exercised  similar  functions  in 
the  lifetimes  of  their  respective  fathers,  while  in  the  inscriptions  of  Antiochus 
Soter,  V  R.  66,  25,  mention  is  made  of  Seleukus  his  son  and  the  viceMng} 

Ziemer  comments  very  strikingly  (p.  449)  on  the  exact  meaning  of  the 
preposition  pHt,  which  is  translated  by  Peiser,  who  writes  it  pHt  (sic),  as 
'receipt'  (see  also  p.  398).  The  author  shows  satisfactorily  that  the  word  in 
the  contracts  is  a  preposition  with  the  force  *  for,  instead  of,  opposite  to.*  He 
might  have  added  that  this  word  is  also  used  in  the  narrative  inscriptions  in 
the  sense  of  'opposite';  cf.  Shalm.  Mon.  26  ina  pHt  dlthi  argip,  h&pHt^piUu 
is  an  abstract  formation  Utyoi  pil  'mouth,'  its  original  meaning  is  probably 
'entrance';  cf.  SainHrammdn^  iv.  41  ina piU DurpapstUtal, 

The  allusions  in  various  contract  tablets  to  Egyptians  who  appeared  as 
witnesses  of  deeds,  etc.,  as,  for  example,  that  mentioned  p.  452,  show  conclu- 

1  Prince,  Mene,  Mene,  p.  ay,  n.  Z4« 


Digitized  by  VjOOQ IC 


Il8  AMERICAN  JOURNAL    OF  PHILOLOGY. 

siyely,  as  Ziemer  states,  that  the  relations  existing  between  Egypt  and  Baby- 
lonia daring  the  reign  of  Cambyses  must  have  been  very  close.  The  Egyp- 
tians mentioned  in  most  of  the  inscriptions  had  become  entirely  Babylonian, 
bearing  Babylonian  names  and  living,  no  doubt,  according  to  the  Babylonian 
customs. 

Such  work  as  that  of  Demuth  and  Ziemer,  which  contributes  to  our  knowl- 
edge of  the  laws  and  family  customs  of  the  Babylonians,  cannot  fail  to  be  of 
the  greatest  value  both  to  Assyriology  and  general  history. 

New  York  Univbksxtt.  J.  DyneleY  PrINCE. 


Digitized  by  VjOOQ IC 


BRIEF  MENTION. 

Archaeologists'  Greek  is  not  intended  to  be  a  complimentary  expression 
(A.  J.  P.  IX  98),  but  it  can  hardly  be  expected  that  archaeologists  should  pay 
much  attention  to  grammar.  It  is  doubtless  to  archaeologists'  Greek  that  we 
owe  the  precious  statement,  still  to  be  found  in  Baedeker  and  repeated  care- 
lessly erery where,  e.  g.  in  The  Nation  of  Jan.  14,  1897,  that  Herodes  Atticus 
**  almost  exhausted  the  quarries  of  Pentelikon  in  carrying  out  [his]  magnificent 
improYement"  [of  the  Stadion].  What  Pausanias  says  (I  19  ex.)  is:  rotn-o 
hviip  'AdtpKuog  'Upi^fK  fpKoddfi^e  kcu  ol  t6  iro^  nyf  XtBoTOfiiag  r^f  llevTe^aiv  if 
nyv  oiKodofti^  ainj^Mj.  Surely  it  was  not  necessary  to  appeal,  as  has  been 
done,  to  the  actual  state  of  Pentelikon  in  order  to  correct  the  '  exaggeration ' 
of  Pausanias.  In  the  very  next  chapter  we  read  I^kbv  \^pi{jvri\  olxeoOcu  79 
Upa^iTiXei  rd  no^  ruv  ipyuv.  The  quarry  was  the  quarry  of  Herodes  as  the 
works  were  the  works  of  Praxiteles.  I  cannot  recall  whether  it  was  an 
archaeologist  or  a  philologian  who  translated  AlACfiTHPA  'by  the  Saviour,' 
but  it  was  an  archaeologist  who  settled  the  hypaethral  question  by  translating 
Strabo,  VIII  30:  dirrdfuvov  6k  axe^bv  ry  Kopv^  rfj^  ofxxp^g  'appearing  almost 
to  touch  the  uppermost  ceiling.*  And  why  should  not  a^ffoaSai  take  the  dative 
in  Strabo  as  dtyelv  does — in  Pindar?  But  apart  from  such  monstrosities  as 
these  last,  and  their  number  might  be  multiplied,  archaeologists  are  apt  to 
satisfy  themselves  too  readily  as  to  points  of  Greek  usage,  especially  on  the 
dangerous  ground  of  the  prepositions,  and  so  Professor  Rbisch,  in  the  great 
work  noticed  elsewhere,  has  a  short  and  easy  method  with  km  ok/Mk  (p.  285). 
'*Das8  das  Vorwort  M  (mit  Genetiv,  Dativ  uud  Accusativ)  nicht  nur  zur 
Bezeichnung  von  HShenunterschieden,  sondern  auch  zur  Bezeichnung  der 
Nachbarschaft  zweier  auf  gleichem  Boden  befindlicher  Dinge  verwendet  wird, 
dUrfte  wohl  bekannt  genug  sein."  Then  follows  a  list  of  quotations,  sadly  in 
need  of  sifting.  Surely  the  average  grammarian,  on  contemplating  this  *  happy 
despatch,*  cannot  but  sadly  think  how  long  he  has  disquieted  himself  in  vain 
about  this  very  preposition  e^rt,  and  especially  about  the  uses  of  km  w.  gen. 
and  kiri  with  dat.  Nay,  there  has  of  late  appeared  a  special  dissertation  on  a 
single  branch  of  the  subject,  and  perhaps  Professor  Reisch  would  not  have 
written  in  this  slap-dash  way  if  he  had  read  Dr.  Forman's  thesis  On  the 
difference  between  the  genitive  and  dative  with  kiri  used  to  denote  superposition^  67 
pp.  (1894),  inasmuch  as  the  author  has  decidedly  advanced  the  treatment  of 
the  subject  and  brings  out  the  characteristic  difference  of  the  cases  in  a 
manner  that  does  great  credit  to  his  fine  appreciation  of  syntactical  effect. 
The  general  conditions  of  the  problem  are,  after  all,  not  so  abstruse,  but  it  is 
one  thing  to  state  the  conditions,  another  to  work  them  out  with  scholarly 
care  and  discernment.  When  there  is  a  rivalry  between  the  genitive  and 
dat.-locative,  the  dat.-locative  is  the  more  plastic.    This  is  very  plain  with 


Digitized  by  VjOOQ IC 


120  AMERICAN  JOURNAL    OF  PHILOLOGY. 

vjr6  (Introduction  to  Pindar,  c),  and  km  is  ttsed  most  frequently  with  the 
dative  ''when  the  superposition  sense  makes  itself  felt"  (1.  c,  xcix ;  A.  J.  P. 
XIV  499).  M  w.  gen.  tends  to  the  phraseological,  iiri  with  dat.-loc.  to  the 
actual,  the  corporeal,  if  plastic  (Introduction,  xcri)  is  not  plain  enough ;  kiri 
with  dat.-Ioc.  tends  to  fixity,  ini  with  dat.  to  freedom  (A.  J.  P.  XI  372).  km 
TTf  ice^o^f,  to  put  the  matter  coarsely,  is  •  on  the  head,'  kiri  ry  ne^aXy  •  on  top 
of  the  head.*  Comp.  Pind.  OL  2, 12.  This  is  a  natural  deduction  from  the 
cases  themselves.  The  dative  combines  with  a  preposition  fua  locative,  the 
genitive  either  gua  ablative  or  gua  fossilized  adjective.  Such  a  fossilized 
adjective  we  have  in  the  familiar  expressions  etc,  kv^  kK  6i6aaK&kcv,  in  which 
the  genitive  may  be  replaced  by  the  proper  case  of  SidaaKaXeiov,  The  genitive 
does  not  depend  on  the  preposition,  but  on  the  local  notion  involved  in  the 
use  of  a  preposition.  There  is  no  real  ellipsis  any  more  than  in  the  corres- 
ponding English  possessives.  Tom's  may  be  Tom's  house  or  Tom's  shop  or 
what  not.  The  practical  equivalent  is  irapd,  ch€M  i  irapd  Stdaait&Xov,  irapa  diHa" 
ffjcdA^,  irapa  6i6&aKakov,  Now,  an  extension  of  this  doctrine  would  satisfy  the 
conditions  with  biri  and  elsewhere.  In  the  vast  majority  of  instances  kiri  with 
gen.  denotes  characteristic  superposition,  and  it  may  still  denote  superposition 
in  such  standing  expressions  as  ktrl  ri/ovf,  kif  ouc^fiaroc.  See  my  Justin  Martyr, 
Apol.  1  26, 15,  and  note  especially  Athen.  5, 220  D :  raig  kiri  tuv  fUKpuv  6iiaifidT(jv, 
Perhaps  a  visit  to  Pompeii  might  help  the  grammarian's  faith  in  the  behavior 
of  prepositions  and  postpone  the  divorce  of  this  km  rkyov^  from  the  km  tqv 
rtyovc  of  Lys.  3,  ii :  f^^Muic  KarioTTfoav  kiri  rov  riyovt.  The  height  of  such  a 
hut  was  well  fitted  for  the  display  of  the  wares  exposed.  But  any  stand,  any 
form  of  superposition,  will  answer  the  conditions — a  seat  in  front,  a  step  in 
the  doorway.  Still,  Dr.  Forman  has  cited  a  number  of  examples  against  such 
rule-makers  as  Rutherford  (Babrius  2, 9),  examples  in  which  kiri  cannot  strictly 
mean  superposition,  though  he  adduces  an  interesting  example  (p.  63),  Dem. 
58,  40:  kiri  Tuv  duaumfpluv  xai  rov  p^paroi,  in  which  kiri  retains  enough  literal- 
ness  for  the  second  member.  But  whatever  the  local  exigencies  may  be,  the 
phraseological,  the  adjectival  character  of  the  combination  is  unmistakable. 
ol  kiri  OK^v^  as  a  technical  term  is  simply  oi  aiaivlTai^  the  *  hutmen.'  The  rarity 
of  kiri  w.  gen.  of  mere  proximity  in  the  best  period,  the  large  possibilities  of 
the  *  upon '  element  even  then— all  this  is  abundantly  shown  in  Dr.  Forman's 
dissertation.  That  ol  aird  axrfvyf  is  more  common  than  ol  kiri  aiofv^t  a  fact  on 
which  RusCH  lays  great  stress,  is  a  very  simple  matter,  dird  etapni^  is  kiri 
eiapf^g  from  a  different  point  of  view.  Sporadic  examples  in  which  kiri  with 
gen.  seems  to  mean  '  before'  do  not  strengthen  the  ff«7*^  argument,  which  may 
quietly  repose  on  the  phraseological  use  of  kiri,  *  On  the  playhouse  side '  is 
all  the  theory  demands,  and  the  phrase  was  fixed  long  before  the  time  of  th% 
earliest  passage  cited. 

Another  preposition  which  plays  an  important  part  in  topography  is  irp6g, 
and  the  Enneakrunos  controversy  may  be  said  to  hinge  on  it.  See  Thuk.  2, 
15.  Without  going  into  that  controversy,  it  may  suffice  to  say  that  irp6c  with 
ace.  can  only  mean  '  facing,  fronting.'  It  is  the  same  preposition  that  is  used 
of  the  same  locality  in  Ar.  Lysistrata.  The  Akropolis,  or  ir6hq,  as  it  was 
popularly  called,  has  two  faces,  and  the  part  meant  in  the  Lysistrata  is  rd 
irpwriikauL  (v.  265),  the  western  end.    The  women  had  barred  the  Propylaia, 


Digitized  by  VjOOQ IC 


BRIEF  MENTION,  121 

and  the  men  had  resolved  to  burn  the  barriers  away,  so  that  the  burning 
question  of  the  Akropolis  now  was  the  burning  question  then. 


A  satisfactory  edition  of  Horace,  like  a  satisfactory  translation  of  Horace, 
is  an  impossibility.  Erery  scholar  worthy  of  the  name  has  an  edition  of  his 
own,  just  as  he  has  heartstrings  of  his  own  about  which  the  Venusian  plays. 
For  the  English  schoolboy,  perhaps  nothing  better  could  be  hoped  than  the 
abridged  triumrirate  Horace  of  Pagb,  Palmer  and  WiLKINS  just  published  by 
The  Macmillan  Co.,  and  many  American  scholars  will  welcome  it  in  spite  of 
drawbacks  which  will  be  felt  more  in  this  country  than  in  England.  Neither 
grammar  nor  metres  are  in  accord  with  recent  studies.  German  authorities  as 
late  as  1894  are  cited,  it  is  true.  Names  like  that  of  Plfiss  recur  frequently, 
but  there  has  been  no  systematic  revision.  On  egit  visen  monies  (C.  I  2,  9)  we 
are  told  that  this  infinitive  is  of  extreme  rarity,  but  Robinson  Ellis,  who 
remarks,  A.  J.  P.  V  12  (1884),  that '  it  belongs  in  the  main  to  later  Latin,'  says 
in  his  Avianus,  XXII I  (1887),  that  'it  is  common  from  the  earliest  Latin  to 
the  latest, and  that  it  is  not  unfrequent  in  Augustan  and  post- Augustan  poetry.' 
Brenons  (Hell^nismes,  p.  275),  like  Bonnet,  believes  in  Greek  influence,  but  it 
is  Greek  influence,  not  strictly  Greek  idiom.  The  note  on  ne — quaesieris  (C.  I 
II,  I)  is  decidedly  behind  the  times,  not  to  say  antiquated  (see  Elmer,  A.  J.  P. 
XV  [1894],  133  foil.),  and  one  becomes  a  little  weary  of  the  fut.  indicative  as 
a  'polite  imperative,'  against  which  I  raised  my  voice — a  vox  clamanHs  in 
deserto — ^many  years  ago.  It  is  not  a  polite  imperative,  nor  a  mild  imperative : 
it  is  a  familiar  imperative,  such  as  one  often  uses  in  English  to  a  servant  (cf. 
Hopkins,  The  Aryan  Future,  A.  J.  P.  XIII  37),  and  Horace  has  some  good 
examples  in  Epist.  I  13.  where  he  addresses  one  Vinius,  whom  he  treats 
throughout  de  kaut  en  ^oi.— To  turn  for  a  moment  to  another  sphere,  the  long 
note  on  Epist.  I  6,  51 — a  note  which,  like  many  others,  is  too  long  for  a 
schoolbook — might  have  been  shortened,  if  the  redactor,  Mr.  Page,  had 
noticed  or  had  seen  in  time  the  inscriptional  evidence  which  Mr.  Olcott 
(A.  J.  P.  XVI  79)  has  brought  to  strengthen  the  traditional  interpretation  of 
trans  pondera.  But  such  close  study  of  periodical  literature  is  hardly  to  be 
expected.  Not  unreasonable,  however,  would  be  the  demand  of  decent  proof- 
reading. It  is  hard  to  understand  how  a  Greek  scholar  could  suffer  the  jumble 
of  accents  one  finds  in  the  familiar  quotation  from  Alkaios  (C.  I  9),  and  Bera- 
eiuies,  cited  C.  1 14,  has  been  HerakUitos  to  my  certain  knowledge  since  1851, 
the  date  of  Mehler*s  edition  of  the  Homeric  Allegories,  and  ought  to  have 
been  Herakleitos  before.  Inexplicable  to  me  is  the  omission  of  the  Greek 
original  or  the  Greek  basis  of  C.  I  27.  But  perhaps  the  editor  was  too  busy 
to  notice  Anakreon,  so  bent  was  he  on  the  mild  joke,  *  The  manner  of  Mr. 
Bardell's  decease  is  strictly  classical.'  Now,  personally  I  have  no  objection  to 
the  playful  reference.  Only  I  remember  how  I  myself  was  maltreated  by  Mr. 
Page's  countrymen  for  a  like  reference  to  Mrs.  Waterbrook,  who  is  quite  as 
presentable  as  Mr.  Bardell  (cf.  A.  J.  P.  XIV  501).  In  the  same  ode,  v.  10.  we 
have  Opuntiae  Megyllae^  and  we  are  told  that  "  Megylla  is  a  fictitious  name  " 
and  that  "Opus  was  the  capital  of  Opuntian  Locris."    That  may  be  quite 


Digitized  by  VjOOQ IC 


122  AMERICAN  JOURNAL    OP  PHILOLOGY, 

enough,  but  is  it  quite  enough?  Of  course,  Horace's  Greek  girls  are  fictitious. 
No  human  being — not  even  the  author  of  the  Amores  (III  7,  26) — would  have 
been  equal  to  so  many  sweethearts.  But,  however  unreal  they  are,  it  is  the 
scholar's  business  to  find  out  the  origin  of  their  names.  Milton's  Amaryllis 
comes  straight  from  Vergil,  Milton's  Neaera  comes  straight  from  Horace,  hair 
and  all,  and  in  like  manner  Horace's  Neaera  comes  straight  from  Parthenius. 
There  was  a  Milesian  lady  of  that  name,  a  name  which  would  have  been 
tolerable  in  the  home  of  Aspasia  (see  Wilamowitz-MoUendorif,  cited  A.  J.  P. 
XVI  125),  and  we  all  know  what  manner  of  woman  the  Neaira  was  that  figures 
in  Ps.  Dem.  LIX.  Lyce's  name  is  borne  by  an  ancient  Greek  light  o*  love, 
A{>ica.  Lyde  was  the  sweetheart  of  Antimachos.  Asterie  recalls  the  famous 
epigram  aaripag  etaadpeig^  dar^p  kfidg,  and  Asterie  was  doubtless  a  fallen  star. 
In  short,  the  proper  names  of  these  improper  persons  are  clothed  or  must  have 
been  clothed  with  literary  atmosphere,  and  I  do  not  believe  that  Horace  had 
the  courage  to  invent  so  much  as  the  name  of  a  Greek  cocotU,  So  Megylla  or 
MegiUa  is  not  a  chance  name  selected  for  its  smooth  sound,  as  Orelli  would 
have  us  believe.  Megylla  was  doubtless  a  great  scamp,  and  we  find  a  HikyOCka 
playing  the  part  of  an  kroipa  in  Lucian.  Nor  is  the  adjective  Opuntiae  taken 
at  random.  Megylla  was  a  Locrian,  and  every  one  knows  what  manner  of 
songs  the  AoKpuca  gafiora  were  (see  Hanssen,  A.  J.  P.  IX  457  foil.).  Now,  I  do 
not  say  that  Horace  belongs  to  those  who  are  capable  of  alluding  to  Doll 
Tearsheet  and  Moll  Flanders  without  having  met  a  Doll  Tearsheet  or  a  Moll 
Flanders  in  the  flesh,  but  he  is  a  bookman  through  and  through,  and  Greek  is 
necessary  to  the  appreciation  of  him  even  in  a  school  edition. 


The  fourth  edition  of  the  Pirsi  Book  of  Classen's  Thukydides  has  appeared 
(Berlin,  Weidmannsche  Buchhandlung).  The  editor  is  J.  Stkup,  who  has 
made  considerable  additions  and  modifications,  so  that  the  Thukydidean 
scholar  cannot  afford  ^o  neglect  the  book.  Especially  interesting  is  Stud- 
niczka's  learned  appendix  on  the  Old  Attic  coiffure.  In  the  preface,  after 
resenting  Croiset*s  treatment  of  his  labors,  Steup  proceeds  to  mention  the 
late  Professor  MoRRis*s  ed.  of  Classen  in  the  usual  sniffy  German  way.  Now, 
Morris  worked  very  carefully  and  conscientiously.  His  basis  was  Classen,  to 
be  sure,  but  beneath  Classen  lay  what  Morris  deemed  to  be  truth,  and  he  did 
not  hesitate  to  differ  on  occasion  from  the  master  whom  he  followed  in  the 
main.  Consequently  Steup  is  forced  to  admit  that  Morris  has  often  aided 
him  in  the  correction  of  typographical  errors  and  other  small  oversights, 
though  otherwise  he  has  very,  very  seldom  received  any  furtherance  from  the 
American  editor.  The  correctness  of  the  statement  I  am  not  disposed  to 
impugn,  inasmuch  as  the  examination  of  a  few  selected  passages  has  sufficed 
to  convince  me  that,  at  least  in  matters  of  grammar,  Steup  does  not  know 
enough  to  learn  from  others,  and  I  call  attention  to  his  grudging  acknowledg- 
ment of  American  work  chiefly  to  emphasize  a  doctrine  which  it  is  the  duty  of 
an  American  Journal  of  Philology  (X  502)  to  uphold.  No  one  has  been  more 
generous  than  I  have  been  in  my  praises  of  German  learning  and  German 
methods,  but  are  we  capable  of  nothing  else  but  adaptations?    No  matter 


Digitized  by  VjOOQ IC 


BRIEF  MENTION,  1 23 

how  carefully  the  work  may  be  done,  it  will  be  considered  by  some  slavish 
and  by  others  indolent.  The  ancient  classics  speak  to  each  nationality  with 
a  different  voice,  and  surely  American  classical  philology  is  no  longer  in  its 
nonage,  and  American  editors  ought  to  stand  on  their  own  feet.  Let  hand- 
books be  translated,  if  it  must  be  so,  but  let  us  look  into  the  face  of  the 
ancients  with  American  eyes. 


Mr.  Lionel  Horton-Smith's  elaborate  article  on  ov  =  haud  reminds  me  to 
say  what  I  ought  to  have  said  long  ago,  that  while  I  welcomed  very  heartily 
Professor  Elmer's  first  treatise  on  the  I^Hn  Prohibitive^  which  appeared  in 
A.  J.  P.  XV  132-53,  and  was  glad  to  incorporate  the  main  results  in  my  Latin 
grammar  of  1894,  the  second  part,  which  resolved  nee  with  the  perf.  subj.  into 
a  potential  construction  (XV  299-328),  has  never  satisfied  me.  The  only  true 
syntactical  parallel  to  ov  is  haud,  Neque  and  neve  have,  it  is  true,  a  certain 
practical  correspondence  to  Mk  and  }jai6k  respectively,  but  ni  qtddem  and  ni^ 
quaquam  and  niquiquam  are  sturdy  monuments  of  the  old  state  of  things, 
and  prevent  the  perfect  parallelism  of  //^  and  ne.  Ni  qids  seems  to  be  sharply 
differentiated  from  nuffus,  but  how  often  does  ne  quis  occur,  how  often  the 
accusative  feminine  ?  Or  are  we  seriously  to  suppose  that  in  the  line  NuUamy 
Vare^  sacra  vite  prius  severis  arbarem^  which  is  an  instructively  close  translation 
of  Alkaios :  nff6lv  dAXo  ^vreho^  np&repov  dhdpiov  afiiri^o,  Horace  conceived  the 
nuUam  severis  as  cv6kv  av  <lnfTebaeutc.  To  be  frank,  should  we  not  be  tempted  to 
call  Horace  nequam^  if  he  had  used  nequam  as  demanded  by  the  strict  rule? 
Nor  does  it  seem  probable  that  in  passing  from  tu  ne  quaesieris  scire  ne/as  to  nee 
Bafylanios  temptaris  numeros  he  felt  any  special  lurch.  Why,  one  may  well  ask, 
if  the  feeling  was  so  strong,  did  he  not  use  neu'i  In  a  matter  that  is  largely  a 
matter  of  feeling  it  is  impossible  to  be  dogmatic,  but  a  syntactical  equation  of 
inh  and  ni  does  not  seem  tenable,  especially  when  one  considers  the  enormous 
range  of  /^  in  Greek. 


This  Journal  has  been  singularly  unlucky  in  its  efforts  to  call  attention  to 
*  epochal'  works,  and  the  notices  of  Mr.  Gawain  Hamilton's  Moods  of  the 
English  Bible  (IX  516)  and  Mr.  Philip  Skene's  Ante-Agamemnona,  a  New 
Departure  in  Philology  (XIV  258)  have  been  interpreted  unfavorably  by  the 
authors  concerned.  The  Editor's  mind  is  wide  open  to  conviction  on  all 
etymological  matters,  and  when  his  collaborators  have  refused  to  consider 
doctrines  that  they  had  condemned  in  advance,  he  has  not  hesitated  to  call 
for  a  fair  field ;  but  in  vain.  And  it  is  to  be  feared  that  the  Abb^  Espagnolle's 
laborious  work,  Le  vrai  dictumnaire  dtymologique  de  la  langue  fran^aise  (Paris, 
Klincksieck),  runs  counter  to  so  many  prejudices  that  the  bare  statement  will 
suffice.  According  to  him,  Hellenic  is  the  direct  descendant  of  Pelasgian. 
The  neo-Latin  peoples  are  of  Greek  or  Pelasgian  origin.  Modem  French  is 
a  sister-language  of  Hellenic,  and  Modern  French  exhibits  about  4500  Greek 
roots  attested  and,  as  it  were,  authenticated  by  Hellenic.  The  common  stock 
of  the  idioms  of  the  Mediterranean  is  Greek,  not  Latin.  A  few  examples  will 
serve  to  show  the  way  in  which  the  vocabulary  is  handled.    "Aooortvfr. 


Digitized  by  VjOOQ IC 


124  AMERICAN  JOURNAL    OF  PHILOLOGY. 

cortois,  x^p^^f  poll*  qui  sait  les  usages.  Lit(tr6),  a  et  oorrigere."  "Aieol 

yfr.  aioi,  at^hog,  tr^s  Ag^,  at6^toc  est  le  inline  que  aUnmoc  parceque  le  v  et  le  A 
pennutent.  Lit(tr£),  aviolUB,  Latin  forg^."  "Arraoher,  arracAo,  ap/>6aaii 
dor.  pour  avdp)jfyjati,  le  m^me  qu'avapp^v/u< J'arrache.  Lit(tr£),  abradioare." 
"ArtiBte»  6ihperUrru^je pr^re  avtc  art,fembelHs,  ^Affrlaru  est  une  forme  dor. 
d'aprtC<>>.  Lit(tr£),  ars."  The  Doric  and  Aeolian  dialects  play  a  great  rdle 
in  the  Abb^  EspagnoUe's  etymologies,  and  a  special  chapter  prepares  us  for 
the  unfamiliar  forms  which  the  Hellenic  words  assume  in  this  DUtumnaire 
Atymoiogique,  But  the  subject,  as  I  have  said,  does  not  lie  within  my  compe- 
tence, and  I  must  content  myself  with  recording  the  appearance  of  a  book 
which  will  remind  every  scholar  of  Henri  Estienne's  famous  Omformit/. 


In  the  matter  of  English  di  me  pusilH  finxerunt  ammu  I  am  conservatire 
by  nature,  by  education,  by  profession.  I  do  not  glory  in  American  English 
unless  I  am  backed  by  ancient  usage,  and  I  am  mortally  afraid  of  my  trans- 
planted fellow-countryman,  Dr.  Fitzedward  Hall.  His  *  Modem  English'  is 
alwa3rs  at  my  elbow,  but  I  am  afraid  to  consult  it  too  often,  lest  I  should  cease 
to  write  altogether,  and  writing  is  my  trade.  The  letter  A  in  the  Oxford 
Dictionary  caused  me  sleepless  nights.  To  be  sure,  I  was  comforted  for 
having  used  aloofness  in  my  Pindar,  but  I  could  not  deny  that  I  had  once  or 
twice  in  my  life  employed  aside  in  order  to  vary  the  monotony  of  apart^  and 
when  Dr.  Hall  transfixed  an  American  scholar  for  having  used  at  that^  I  felt  as 
if  his  spear  had  gone  through  me  also.  And  the  worst  of  it  is  that  in  my 
Essay  on  Platen  the  wretched  phrase  makes  a  rhyme  with '  flat,'  and  I  have 
never  been  able  to  think  of  a  good  equivalent.  Of  late,  around  has  begun  to 
haunt  me,  and  when  '  fond  memory  brings  the  light  of  other  days  around  me,' 
I  bethink  me  of  the  many  passages  in  which  I  and  other  people — notably 
English  classics — have  used  that  unnecessary  '  a.'  If,  then,  I  am  somewhat 
given  to  Biblical  phraseology,  it  is  because  I  can  cling  to  the  altar  of  the 
Authorized  Version  and  feel  myself  fairly  safe.  But  my  timidity  does  not 
keep  me  from  indulging  in  a  certain  fearful  joy  when  any  one  is  bold  enough 
to  try  conclusions  with  Dr.  Fitzedward  Hall,  and  this  is  what  Mr.  Ralph 
Olmstsd  Williams  has  done  in  Some  Questions  of  Good  EngHsh  Examined  in 
Controversies  with  Dr.  Fitudward  Hall  (New  York,  Henry  Holt  ft  Co.).  All  I 
dare  say  is  that  it  is  very  good  reading. 


*Eine  edie  Frau  giebt  es  nicht  bel  [Pindar],'  says  Wilamowitz  in  his 
recent  edition  of  the  Choephori  {Das  Opfer  am  Grabe^  p.  32).  That  is  a  hard 
saying,  especially  to  an  editor  who  has  ventured  to  call  Pindar '  a  manner  of 
Frauenlob'  (see  my  Pindar,  p.  aoi).  Surely  Alkmene  is  not  unworthy  of  her 
son  (N.  X,  50).  It  means  something  that  Cheiron  is  known  by  his  mother's 
name  and  that  the  pure  daughters  of  the  Centaur  reared  that  model  prince 
Jason  (P  4, 103).  The  Ninth  Olympian  is  given  up  to  the  Eternal  Feminine, 
and  the  lofty  realm  of  personifications  is  full  of  goddesses.  If  the  worship  of 
the  Virgin  Mary  is  a  tribute  to  womanhood  and  motherhood,  something  is  to 
be  said  for  Pindar's  shining  forms  arrayed  in  woman's  garb. 


Digitized  by  VjOOQ IC 


Digitized  by  VjOOQ IC 


126  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Rambaud  (L-B.)    La  langae  mande.     Paris,  1897.    8vo.     5  fr. 

Revilloat  (£.)  Notice  des  papyrus  demotiques  arcbalques.  Paris,  1897. 
4to,  544  pp.     70  fr. 

Viteau  (I.)  £tade  sur  le  grec  du  Nouveau  Testament.  Pans,  1897. 
8vo.     12  fr. 

Vlachos  (A.)  Dictionnaire  grec-fran^ais.  Paris,  1896.  8vo,  1000  pp. 
25  fr. 

GERMAN. 

Aegyptiaca.  Festschrift  f.  Georg  Ebers  zum  i.  IIL  1897.  Mit  i  Taf. 
in  Licbtdr.  u.  9  Fig.  im  Text.    gr.  8.    vii,  152  S.    L.,  W,  Engelmann.    m.  20. 

Aiox^Tmv  Spdfiara  ffu^dfieva  koI  diro^cj^^(Jv  airoa^daftaTa  fierd  k^ifytfTiKov  naX 
KpiTiKuv  oTjfuUtatuv  Ty  owepyaaiq.  Evyeviov  Zufiaptdov  kic6t66fieva  vnd  N.  Wecklein. 
Td/<of  II  Kal  III  Tevxog  A',  gr.  8.  Atben.  Leipzig,  O.  Ilarrassowitz  in 
Komm, — II.  7repiix('n>  Upofitfdia,  'liciTiSac  kcu  *  KiroairaafjidTLa.  .vii,  648  S .  m.  10. 
— Ill  A',  irtpikxiw  'AirooTraafjidTia,     S.  649-798.     m.  2. 

Anthologia  latina,  edd.  Franc.  Baecbeler  et  Alex.  Riese.  Pars  II.  Car- 
olina latina  epigrapbica,  conlegit  Franc.  Buecbeler.  Fasc.  II.  8.  S.  399- 
921.     L,f  B,  G.  Teubfur,     m.  5.20. 

Apulei  (Lucii)  metamorpboseon  libri  XI.  Recensuit  J.  van  der  Vliet.  8. 
xxix,  292  S.     L.,  B,  G.  Teubner,     m.  3. 

Ausgaben  u.  Abbandlungen  aus  dem  Gebiete  der  romaniscben  Pbilologie. 
VerOffentl.  v.  E.  Stengel.  95.  Hft.  gr.  8.  Marburg,  N  G.  Elwerfs  Verl. 
— 95.  Patzold  (Alfr.)  Die  individuellen  Eigentttmlicbkeiten  einiger  her- 
vorragender  Trobadors  im  Minneliede.     145  S.     m.  3.20. 

Bibliothek  der  angels£cbsiscben  Poesie,  begr.  v.  Chrn.  W.  M.  Grein. 
Hrsg.  V.  Rich.  Paul  Wtilker.  3.  Bd.  i.  Halfte.  gr.  8.  vii,  248  S.  L.,  G. 
H»  Wigand,     m.  11. 

Bibliothek  indogermanischer  Grammatiken,  bearb.  v.  F.  BUcheler,  B. 
DelbrUck,  K.  Foy  u.  a.  III.  Bd.  L.,  Breitkopf^  ZTtfr/^/.— IIL  Griechische 
Grammatik  v.  Gust.  Meyer.     3.  Aufl.     xviii,  715  S.     m.  13;  geb.,  m.  14.50. 

Breymann  (Herm.)  Die  phonetische  Literatur  von  1876-1895.  gr.  8. 
iii,  170  S.     L.,  A,  Deichert  Nachf,    m.  3.sa 

BrUnnow  (Rud.  E.)  A  Classified  List  of  all  Simple  and  Compound 
Cuneiform  Ideographs  occurring  in  the  Texts  hitherto  published.  Indices 
dazu.  gr.  4.  viii,  344  S.  Leiden,  Buchh.  u,  Druckerei  vorm.  E,  J,  Brill. 
m.  25. 

Corpus  inscriptionum  atticarum  academiae  litteicarum  regiae  borussicae. 
Appendix.  Defixionum  tabellae  atticae,  collegit,  collectas  praemissa  prae- 
fatione  edidit  Ric.  Wuensch.     Fol.     xxxii,  52  S.     B.,  G.  Reimer.    m.  9. 

Fragmenta  scaenicae  Romanorum  poesis,  tertiis  curis  recognovit  Otto 
Ribbeck.  Vol.  I.  Tragicorum  fragmenta.  8.  viii,  335  S.  L.,  B.  G,  Teub- 
ner.    m.  4. 

Gregorii  Abulfaragii  Bar-Hebraei  scholia  in  libros  Samuelis,  ed.  E. 
Schlesinger.     gr.  8.     iv,  32  S.     B.,  S,  Calvary  6f*  Co,     m.  2. 

Handbuch  der  klassischen  Altertums-Wissenschaft.  Hrsg.  von  Iwan  v. 
Mttller.  22.  u.  23.  Halbbd.  qu.  Fol.  Mttnchen,  C.  If.  Beck.  In  Mappe, 
m;  13.50.— 22,  23.  Atlas  zu  Bd.  VI :  Arcbaologie  der  Kunst  v.  Karl  Sittl. 


Digitized  by  VjOOQ IC 


RECENT  PUBLICATIONS.  12/ 

64  Taf.  m.  1000  Abbildgn.,  nebst  Inhaltsverzeichnis  u.  alphabet.  Register. 
28  S.     m.  13.50; 

Heiiiichen(Frdr.Adph.)  Lateinisch-deutsches  Schulw5rterbach.  6.  Aufl. 
V.  C.  Wagener.  Lez.-8.  zzix,  926  S.  L.,  ^9.  G,  Teubner,  m.  6.30;  geb. 
in  Halbfr.,  m.  7.50. 

H()hler  (Wilh.)  Die  Comutus-Scholien  zam  i.  Buche  der  Satiren 
Juvenals.  [Aus  *Jahrbb.  f.  class.  Philol.,'  23.  Suppl.-Bd.]  gr.  8.  63  S. 
L.,  ^.  G,  Teubner,     m.  2.40. 

Holder  (Alfr.)  Alt-celtischer  Sprachschatz.  9.  Lfg.  gr.  8.  2.  Bd.  Sp. 
1-256.     L.,  B.  G,  Teubner.     m.  8. 

Jacob  (Geo.)  Studien  in  arabiscben  Dichtern.  IV.  Hft.  Altarabische 
Parallelen  zum  Alten  Testament,    gr.8.    25  S.    B.,  Mayer  6*  Mailer,    m.  i. 

Jahresbericht,  kritischer,  Ub.  die  Fortschritte  der  romanischen  Philologie. 
Hrsg.  V.  Karl  VoUmOller.  III.  Bd.  1891-94.  2.  Halfte.  4  Hfte.  gr.  8, 
I.  Hft.  128  S.     Erlangen,  ^. /««^^.     m.  18. 

Kluge  (Frdr.)  Von  Luther  bis  Lessing.  3.  Aufl.  Mit  e.  Kartchen.  8. 
▼ii,  151  S.     Strassburg,  X.J,  Triibner  Verl.     m.  2,50. 

Krassowsky  (Walth.)  Ovidius  quomodo  in  isdem  fabulis  enarrandis  a  se 
ipso  discrepuerit.  Diss.  gr.  8.  38  S.  K()nigsberg,  Grdfe  &•  Unzer^s  Sort. 
m.  I. 

Krohn  (Frdr.)  Quaestiones  Vitnivianae.  Particula  I.  De  M.  Ceti 
Faventini  epitoma.     Diss.    gr.  8.    43  S.     B.,  Mayer  6*  Mailer,     m.  1.20. 

Kuhn  (E.)  u.  Schnorr  v.  Carolsfeld  (H.)  Die  Transcription  fremder 
Alphabete.     gr.8.     15  S.     \*.^0.  Harrassnvitz.     m.  1.20. 

Kunze  (Alfr.)  Sallustiana.  3.  Hft.  Zur  Stilistik.  i.  Thl.  Beitrag  zu  e. 
Darstellg.  der  genet.  Entwickelg.  des  Sallustian.  Stils.  gi.  8.  xiv,  95  S. 
L.,  Simmel  dr*  Co.     m.  2.50. 

Lucretius  Carus  (T.)  de  rerum  natura.  Buch  III,  erklart  v.  Rich.  Heinze. 
gr.  8.     vi»  206  S.    L.,  B.  G.  Teubner.     m.  4 ;  geb.,  m.  5. 

Ludwich  (Arth.)  Kritische  Miscellen  (I-XI).  Progr.  gr.  4.  20  S. 
KOnigsberg,  Akadem.  Buchh,  v.  Schubert  &*  Seidel.     m.  — 30. 

Memoires  de  la  soci^t^  finno-ougrienne.  IX.  gr.8.  Helsingsfors.  L., 
O.  Narrassowitz  in  Komm. — IX.  Schlegel  (Gust.)  Die  chinesische  Inschrift 
auf  dem  uigurischen  Denkmal  in  Kara  Balgassun.  Uebers.  u.  eri&utert. 
XV,  141  S.  m.  Titelbild  u.  i  Schrifttaf.     m.  6. 

Miladinoff  (Iwan  An.)  Dentsch-bulgar.  WOrterbuch.  3.  u.  4.  Hft.  8. 
S.  281-612  u.  iii  S.     Sofia.     L.,  Kdssling.     m.  4  (L  Thl.  kplt.,  m.  8). 

Mitteilungen  aus  den  orientalischen  Saznmlungen  der  k()nigl.  Museen  zu 
Berlin.  VIII.  Hft.  Fol.  B.,  W.  Spemann.^VUl.  Steindorff  (Geo.) 
Grabfunde  des  mittleren  Reichs  in  den  k5nigl.  Museen  zu  Berlin.  I.  Das 
Grab  des  Mentuhoten.    viii,  46  S.  m.  Abbildgn.  u.  13  (11  farb.)  Taf.     m.  80. 

der  vorderasiatischen  Gesellschaft.      1896.     2-4.     gr.  8.     B.,  fV. 

Peiser  Verl.  in  Komm. — 2.  Hartmann  (Mart.)  Bohtan.  Eine  topograph- 
isch-histor.  Studie.  I.  60  S.  m.  3.50. — 3.  Peiser  (F.  E.)  Skizze  der 
babylonitchen  Gesellschaft.  32  S.  m.  1.50. — 4.  Meissner  (B.)  Pallacottas. 
— MttUer  (W.  M.)  Ein  ph()niktscher  KOnig ;  Ein  neuer  HetiterkOnig.-- 
Winckler  (Hugo).  Das  Siegel  Ahlib-sar's,  u.  s.  w.— Niebuhr  (C.)  Das 
Land  Jarimuta.     36  S.     m.  2. 


Digitized  by  VjOOQ IC 


128  AMERICAN  JOURNAL    OF  PHILOLOGY, 

Mailer  (Job.)  Kritische  Stadien  za  den  Briefen  Senecas.  [Aus  *  Sitz- 
ungsber.  d.  k.  Akad.  d.  Wiss.']  gr.  8.  32  S.  Wien,  C  Gerald' i  Sohn  in 
Kontfn,    m.  — ^o. 

Mussafia  (Adf.)  Zur  Kritik  u.  Interpretation  romanischer  Texte.  2. 
Beitrag.  [Aus  <  Sitzungsber.  d.  k.  Akad.  d.  Wiss.']  gr.  8.  72  S.  Wien, 
C.  Gerald's  Sohn  in  Komm.    m.  1.60. 

Petraris  (K.)  Taschenw5rterbuch  der  neugriechischen  a.  deutschen 
Sprache.  2  Tie.  gr.  16.  L.,  O.  HoltzeU  Nachf,  Geb.  in  Leinw.,  m.  7.60. 
— I.  Neugriechiscb-Deutsch.  iv,  430  S.  m.  3.60. — 2.  Deutscb-Neugrie- 
chiscb.    ii,  553  S.     m.  4. 

Phaedri  August!  liberti  fabulae  Aesopiae,  ed.  Leop.  Hervieax.  gr.  8. 
133  S.     Paris,  Firmin-Didot  6*  Cie,    m.  3.20. 

Pi8cbel(Kicb.)  u.  Geldner  (Karl  F.)  Vediscbe  Studien.  2.  Bd.  2.  (Scbluss-) 
Hft.     gr.  8.    X  u.  S.  193-334.     St.,  W,  Kohlhammer,    m.  4.50. 

Planta  (Rob.  v.)  Grammatik  der  oskiscb-umbrischen  Dialekte.  2.  Bd. 
Formenlebre,  Syntax,  Sammlg.  der  Inscbriften  u.  Glossen,  Anbang,  Glossar. 
gr.  8.    XV,  772  S.     Strassburg,  K.  J.  TrUbner  Verl.    m.  20. 

Platonis  opera  omnia.  Kecensuit  et  commentariis  instruxit  Godofr. 
Stallbaum.  Vol.  VIII,  sect.  II.  Sopbista.  Ed.  II.  Recensuit,  prolego- 
menis  et  commentariis  instruxit  Otto  Apelt.  gr.  8.  viii,  217  S.  L.,  B,  G. 
Teubner,     m.  5.60. 

Plini  Secundi  (C.)  naturalis  historiae  libri  XXXVII.  Post  Ludov.  lani 
obitum  recognovit  edidit  Carol.  Mayboff.  Vol.  IV.  Libri  XXIII-XXX. 
8.     X,  500  S.     L.,  B,  G.  Teubner.    m.  6. 

Plutarchi  moralia,  recognovit  Greg.  N.  Bemardakis.  Vol.  VII.  Plutarcbi 
fragmenta  vera  et  spuria  multis  accessionibus  locnpletata  continens.  8. 
Ivi,  544  S.     L.,  B.  G»  Teubner,     m.  4. 

Epilogus.     8.     47  S.     L.,  B.  G,  Teubner.    m.  2. 

Preuss(Artb.)  De  yersuum  iambicorum  in  melicis  partibus  usu  Aescbyleo. 
Diss.     gr.  8.     118  S.     l^^E.Grdfe.    m.  2. 

Prosopograpbia  imperii  romani  saec.  I.  II.  Ill  edita  consilio  et  auctoritate 
academiae  scientiarum  regiae  bornssicae.  (In  4  partibus.)  Pars  I  et  II. 
Lex.-8.  B.,  G.  Reimer,  m.  44. — I.  Edidit  Elimar.  Klebs.  ix,  489  S. 
m.  24. — II.  Edidit  Herm.  Dessau,    v,  443  S.     m.  20. 

Radloff  (W.)  Versucb  e.  WSrterbucbes  der  Tttrk-Dialecte.  8.  Lfg. 
hocb  4.  2.  Bd.  Sp.  321-640.  St  Petersburg.  L.,  Voss*  Sort,  in  ICcmm, 
m.  2.50. 

Rydberg  (Gust.)  Zur  Gescbicbte  des  franz5siscben  9,  I.  Die  Entstehg. 
des  p-Lautes.  gr.  8.  67  S.  Upsala,  1896.  L.,  O,  Harrassowitz  in  Komm, 
m.  2.50. 

Scboemann  (G.  F.)  Griecbische  Altertbttmer.  4.  Aufl.,  neu  bearb.  v. 
J.  H.  Lipsius.  I.  Bd.  Das  Staatswesen.  gr.  8.  viii,  600  S.  B.,  fVeidmann, 
m.  12. 

SchGne  (Alfr.)  tJeber  die  Ironie  in  der  griecbischen  Dicbtnng.  Rede, 
gr.  8.     23  S.     Kiel,  Universitdts-Buchh,    m.  x.40. 

Scbwteger  (Paul).  Der  Zauberer  Virgil,  gr.  8.  76  S.  B.,  E,  S.  Mittler 
&*  Sohn,     m.  1.50. 


Digitized  by  VjOOQ IC 


■noT^ 


RECENT  PUBLICATIONS,  1 29 

Scriptores  latini  medii  aevi  suecani,  ediderunt  Johs.  Paulson  et  Lars 
Wahlin.  I.  Petri  de  Dacia  vita  Christinae  Stambelensis  ed.  Johs.  Paulson. 
Fasc.  2  secundum  de  vita  Christinae  librnm  continens.  gr.  8.  v,  257  S. 
Gothenburg,  Weitergren  4*  Kerber.    m.  6. 

Serrurier  (L.)  Catalogue  raisonn^  des  livres  et  des  manuscrits  japonais 
enregistr^s  k  la  biblioth^que  de  I'universit^  de  Leyde.  Lex. -8.  xiii,  298  S. 
Leiden,  Buchh,  u»  Druckerei  vorm.  E,J,  Brill,     m.  15. 

Strassmaier  (J.  N.)  Babylonische  Texte.  12.  Hft.  gr.  8.  Leipzig,  £. 
Pfeijfer, — 12.  Inschriften  v.  Darius,  KOnig  v.  Babylon  (521-485  v.  Chr.),  v. 
den  Thontafeln  des  Brit.  Museums  copirt  u.  autogr.  3.  Hft.  Nr.  452-579 
vom  17.-23.  Regierungsjahre.     S.  321-416.     m.  7.20. 

Studlen,  semitistische,  hrsg.  v.  Carl  Bezold.  12.  Hft.  gr.  8.  Weimar, 
E.  Felber. — 12.  Mordtmann  (J.  H.)  Beitrage  zur  minaischen  Epigraphik. 
Mit  22  in  den  Text  gedr.  Fcsms.    xiv,  127  S.    m.  12. 

Tacitus  (P.  Cornelius).  Ab  excessu  divi  Augusti  Buch  I  u.  IL  Fttr  den 
Gebr.  der  SchUler  erkl.  V.  Geo.  Andresen.  Text.  gr.  8.  iii,  90U.  Anmerkgn. 
53  S.     B.,  Weidmann.     Geb.  in  Leinw.  u.  geh.,  m.  1.40. 

Thukydides.  Erkl art  v.  J.  Classen,  i.  Bd.  Einleitung.  i.  Buch.  4.  Aufl. 
V.  J.  Steup.     gr.  8.    Ixxiv,  398  S.  m.  6  Abbildgn.     B.,  Weidmanu,    m.  4.50. 

Tiktin(H.)  Rumanisch-deutsches  WOrterbuch.  2.  Lfg.  Lex...8.  S.65- 
128.     Bukarest     L.,  O.  ffarrassowitn,     m.  1.60. 

ITALIAN. 

Monaci  (Ernesto).  Crestomazia  italiana  dei  prim!  secoli,  con  prospetto 
delle  flessioni  grammaticali  e  glossario.  Fascic.  IL  Citt^  di  Castello. 
L.  10. 


Digitized  by  VjOOQ IC 


BOOKS  RECEIVED. 

Addresses  delivered  at  the  Lowell  Commemoration  held  in  the  Architek- 
tenhaus,  Berlin,  Feb.  19,  1897.     Berlin,  Mayer  6*  Mutter^  1897. 

Bernstein  (Ludwig).  The  Order  of  Words  in  Old  Norse  Prose.  (Colum- 
bia  University  Diss.)     New  York,  The  Knickerbocker  Press,  1897. 

Borsy  (Bernhardus).  De  Aristotelis  iroXtrelac  *AOrivaiuv  partis  alterius, 
fonte  et  auctoritate.     Dorpat,  C,  Maitiesen,  1897. 

Ciceros  Rede  gegen  Q.  Caecilias  u.  das  vierte  Buch  der  Anklageschrift 
gegen  C.  Verres.  Far  den  Schulgebr.  herausg.  v.  Hermann  Nohl.  2te 
Attfl.    70  pf. 

Anklageschrift  gegen  C.  Verres.  Fanftes  Bach.  Far  den  Schul- 
gebr. herausg.  v.  Hermann  Nohl.    Leipzig,  G,  Freytag,  1897.     70  pf. 

Corpus  Scriptorum  Ecclesiasticorum  Latinorum,  ed.  consilio  et  impensis 
Academiae  Litterarum  Caesareae  Vindobonensis.  Vol.  XXXIL  S.  Am- 
brosii  opera.  Pars  I  ez  recensione  Caroli  Schenkl.  Fasc.  Prior.  Vindo- 
bonac,  F,  Tempsky,    Lipsiae,  G,  Freytag,  MDCCCLXXXVL     12  m.  80  pf. 

Dante  Society.  Fifteenth  Annual  Report.  Dante  in  America,  by  Theo- 
dore W.  Koch.     Boston,  Ginu  6*  Co.,  1896. 

Demosthenes.  The  First  Philippic  and  the  Olynthiacs,  with  introduction 
and  critical  and  explanatory  notes  by  John  Edwin  Sandys.  London,  Mac 
millan  6*  Co,     New  York,  The  Macmillan  Co.,  1897. 

Dictionnaire  g^n^ral  de  la  langue  franQaise  du  commencement  du  XV He 
Slide  jusqu'ii  nos  jours,  par  MM.  Hatzfeld,  Darmesteter  et  Thomas.  2oe 
livraison.  M£RCENAIRE-N]£gRILLON.  Prix  de  souscription  i  I'ouvrage 
complet,  30  fr.     Paris,  Ch,  Delagrave,  1897. 

Educational  Review.  Ed.  by  N.  M.  Butler.  March,  April,  May,  1897. 
New  York,  Henry  Holt  &*  Co,    I3  per  annum ;  35  cts.  a  copy. 

English  Dictionary  (The  Oxford).  Ed.  by  James  A.  H.  Murray.  DIS- 
TRUSTFULLY-DOOM  (vol.  HI).  FLEXUOSITY-FOISTER  (vol.  IV), 
by  Henry  Bradley.  Oxford,  At  the  Clarendon  Frets.  New  York,  The 
Macmillan  Co.^  1897.     (^  2S.  6d. 

Euripides.  The  Troades.  With  revised  text  and  notes  by  R.  Y.  Tyrrell. 
London,  Macmillan  £r*  Co,     New  York,  The  Macmillan  Co.,  1897. 

Fairclough  (H.  Rushton).  The  Attitude  of  the  Greek  Tragedians  toward 
Nature.     (J.  H.  U.  Diss.)    Toronto,  Ont.,  Ronsell  ^  Hutchinson,  1897. 

Fletcher  (W.  L)  and  Bowker(R.  R.)  The  Annual  Literary  Index,  1896. 
New  York,  Ojgice  of  the  Publishers'  Weekly,  1897. 

Galdos  (B.  P.)  Dofla  Perfecta.  With  an  introduction  and  notes  by  A. 
R.  Marsh.    Boston,  Ginn  6*  Co,,  1897. 

Gardner  (E.  A.)  A  Handbook  of  Greek  Sculpture.  Part  II.  New  York, 
Thi  Macmillan  C«. ,  1 897 .    $1.25. 


Digitized  by  VjOOQ IC 


BOOKS  RECEIVED.  I3I 

Hanssen  (Federico).  Estadiot  sobre  la  conjugacion  leonesa.  (Reprint 
from  Anaies  de  la  Universidad,)  Santiago  de  Chile,  Imprenta  Cervantes, 
1896. 

— —  Sobre  el  hiato  en  la  antigua  versificacion  castellana.  (Pablicado  en 
I08  Anaies  de  la  Universidad  di  Diciembre.)     Santiago  de  Chile,  1896. 

Hemp]  (Geo.)  Middle  English  -w^  -ur^.  (Reprinted  from  the  Journal 
6f  Germanic  Philelogy,  vol.  I,  No.  i,  1897.) 

HQbner  (Emil).  Jacobo  Zobel  de  Zangi6niz.  {Deutsche  Rundschau, 
Marz  1897.)    Separatabdruck.    Berlin,  Gebr.  PaeteL 

Inscriptionet  Hispaniae  Latinae,  ed.  Aemilins  Huebner.  C.  I.  L.  snp- 
plementam.  £z  £ph.  Epigr.  vol.  VIII,  fasc.  Ill  seorsum  expressam. 
Berlin,  Georg  Reimer,  1897. 

Johnston  (Harold  W.)  Latin  Mannscripts.  (Inter-Collegiate  Latin 
Series.)    Chicago,  Scott  Foreseman  &*  Co,,  1897. 

Jonmal  of  Germanic  Philology.  Ed.  by  Gustav  E.  Carsten.  Vol.  I,  No.  i. 
Bloomington,  Ind.,  The  Editor.     Boston,  Ginn  6r*  Co.,  1897. 

Koerting  n.  Koschwitz.  Zeitschrift  fttr  franzOsische  Sprache  a.  Litte* 
ratar,  heraasg.  v.  D.  Behrens.  Bd.  XIX»  Heft  2  u.  4.  Referate  n.  Rezen- 
sionen.     Berlin,  IVilhelm  Gronau,  1897. 

Kretschmer  (Paul).  Einleitung  in  die  Geschichte  der  griechischen 
Sprache.    G5ttingen,  Vandenhoeck  u.  Ruprecht,  1896.     10  m. 

Le  Blant  (Edmond).  750  inscriptions  de  pierres  gravies  in^dites  ou  pen 
connues.  Extrait  des  Memoires  de  I'Academie  des  Inscriptions  et  de 
Belles-Lettres.     Paris,  Lidrairie  C.  Klincksieck,  1896.     8  fr.  75  c 

Mather,  Jr.  (F.  J.)  King  Ponthus  and  the  Fair  Sidone,  an  inedited 
version  of  the  Story  of  King  Horn  in  LXVII  1-150.  Publications  of  the 
Modem  Language  Association  of  America.  Ed.  by  Jas.  W.  Bright.  Vol. 
XII,  No.  I. '  N.  s.,  vol.  V,  No.  I.     Baltimore,  The  Association,  1897. 

Modern  Language  Association  of  America  (Publications  of  the).  Ed.  by 
James  W.  Bright.  Vol.  XII,  No.  2.  N.  s.,  vol.  V,  No.  2.  Baltimore,  The 
Association,  1897. 

Mortet  (Victor).  Un  nouveau  texte  des  traites  d'arpentage  et  de  g^om^- 
trie  d'Epaphroditus  et  de  Vitruvius  Rufns.  Avec  introd.  de  Paul  Tannery. 
Tir^  des  Notices  et  Extraits  des  mss.  de  la  Bibliothique  Nationale,  XXXV 
2^.     Paris,  Lidrairie  C.  Klincksieck,  1896.     2  fr.  60  c. 

Plauti  Rudens.     (MacGill  University  Edition.)    Montreal,  1897. 

Postgate  (J.  P.)  Silva  Maniliana.  Congessit  J.  P.  P.  Cantab.  MDCCC- 
LXXXXVIL 

UpOKTusa  T^c  kv  *A6fyifaig  *Apxato^^usvC  'Eroupeioc.     'Ad^vfjotv,  1896. 

Ramsay  (W.  M.)  The  Cities  and  Bishoprics  of  Phrygia.  Vol.  I,  Part  II. 
West  and  West-Central  Phrygia.  Oxford,^/  the  Clarendon  Press.  New 
York,  The  Macmillan  Co,     21s. 

Riess  (Ernst).  Superstitions  and  Popular  Beliefs  in  Greek  Tragedy. 
(Extract  from  Transactions  of  the  American  Philological  Association,  vol. 
XXVII,  1896.) 

Sanders  (Daniel).  Encyclopaedic  English-German  and  German-English 
Dictionary.    Part  second,  German-English.    Unabridged  edition.    Fasc.  i. 


Digitized  by  VjOOQ IC 


132  AMERICAN  JOURNAL   OF  PHILOLOGY. 

A-Ambannen.  Berlin,  LangtnschHdf sche  Buchhandlung,  New  York, 
International  News  Co»    24  fasc.  @  i  m.  50  pf. 

Satura  Viadrina.  Festschrift  zam  fanfundzwanzigsten  Bestehen  des 
Philologischen  Vereins  zn  Breslau.     Breslaa,  Schottl&nder^  1896. 

Schneidewin  (Max).  Die  antike  Humanitat  Berlin,  Weidmannsche 
Buckkandlungy  1897.     12  m. 

Schoemann  (G.  F.)  Griechische  Alterthttmer.  Vierte  Auflage.  Nea 
bearbeitet  von  J.  H.  Lipsius.  Berlin,  Weidmannsche  Buchhandlungy  1897. 
12  m. 

Sweet  (Henry).  The  Student's  Dictionary  of  Anglo-Saxon.  New  York, 
The  Macmillan  Co.^  1897. 

Thucydides.  Book  VI.  £d.  by  £.  C.  Marchant.  London,  Macmillan 
<&•  Co.    New  York,  The  Macmillan  Co.^  1897. 

Thnkydides.  Erkl.  von  J.  Classen.  Erster  Band.  Einleitung.  Erstes 
Buch.  Vierte  Anfl.,  bearb.  von  J.  Steup.  Berlin,  Weidmannsche  Buch' 
handlungy  1897.     4  m.  50  pf. 

Tsountas  (Chrestos)  and  Manatt  (J.  Irving).  The  Mycenaean  Age. 
Boston  and  New  York,  Houghton^  Mifflin  &*  Co,,  1897.     $6. 

Vos  (B.  J.)  The  Diction  and  Kime-Technic  of  Hartman  von  Aue. 
(J.  H.  U.  Diss.)    New  York,  Lemche  <Sr-  BUchner,  1896. 

Weil  (Henri).  ]£tude8  sar  le  drame  antique.  Paris,  Hachette  et  Cie., 
1897. 

«—  Un  nouveau  fragment  de  Ph^recyde  de  Syros.  Revue  des  £tudes 
grecques.    (Tirage  i  part)     1897. 

Wilhelm  (Eugen).  Perser.  (Sonderdruck  aus  Jahresberichte  der  Ge- 
schichtswissenschaft.)    Berlin,  R,  Gaertner*s  Verlagsbuchhandlung^  1897. 

Zeitschrift  f&r  deutsches  Altertum  u.  deutsche  Litteratur,  herausg.  v.  £. 
Schroeder  u.  Gustav  Roethe.  XLI.  Bd.  2tes  Heft.  Berlin,  Weidmannsche 
Buchhandlungy  1897. 

Zielinski  (Th.)  Cicero  im  Wandel  der  Jahrhunderte.  Ein  Vortrag  v. 
Th.  Zielinski.    Leipzig,  B.  G.  Teubner,  1897.     2  m.  40. 


Digitized  by  VjOOQ IC 


AD  VEX  TISEMENTS. 


LeMCKE  &  BUECHNER, 

Formerly  B.  WMtMVUUiD  &  Co., 

Publishers,  Booksellers 

AND  importers, 

812  Broadway,  New  York, 

With  Offices  at  London^  Leipzig  and  Paris. 


Just  Published— the  American  Edition  of 

PLUBOBL-SCHMIDT-TANQER'S 

School-  and  Hand-Lexicon  of  the  Qerman  and  English  Languages. 
2  vols.,  Urge  8vo,  half  leather,  II4.50. 
The  German-English  part  separately,  $2.60. 

Destined  to  replace  the  venerable  Adler  and,  more  or  less,  all  other 
German-English  Dictionaries  now  used  in  Colleges  and  High  Schools.  It 
is  just  the  book  which  Professors  of  German  have  been  asking  us  to 
SQpply  for  many  years.  When  it  was  not  promptly  forthcoming,  there 
was  a  project  considered  by  these  gentlemen  for  a  time  to  compile  a 
dictionary  here.  Happily  Flnegel  now  is  ready,  and  orders  for  Schools 
can  be  filled.     Professors  will  please  communicate  with  us  direct. 

Special  Notice. 


The  stock  of  complete  sets  in  the  hands  of  the  management  of  The 

American  Journal  of  Philology  was  exhausted  some  time  ago,  but  a 

few  sets  have  been  made  up  of  late  by  purchase  and  exchange.    These  sets 

will  be  sold  at  the  regular  price,  I51  for  the  seventeen  volumes,  cash  to 

accompany  the  order.    Of  single  volumes  ($3)  Volumes  IV  and  VI  cannot 

be  furnished.    Of  single  numbers  ($1)  13,  15,  21,  22,  23,  49,  50,  52  are 

wanting.    No.  40,  which  contains  the  Indtfx  to  Vols.  I-X,  is  not  furnished 

separately,  but  a  small  edition  of  the  Index  itself  is  for  sale  at  $1.00  a 

copy.    Address, 

B.  L.  Gildersleeve,  Baltimore,  Md. 


Digitized  by  VjOOQ IC 


AMERICAN  JOURNAL  OF  PHILOLOGY. 


The  Macmillan  Company's 

New  Books  in  Foreign  Languages,  etc. 

An  Aid  of  Great  Importance  to  Students  of  Early  English, 

Student's  Dictionary  of  Anglo-Saxon  (The).     By  Hsnry  Sweet,  M.  A., 
Ph.D.,  LL.D.,  CorrespoDding  Member  of  the  Munich  Academy  of 
Sciences.     Small  4to,  cloth,  pp.  xvi  -{-  217.     Price,  $1.75  net. 
The  Anglo-Saxon  Primer  and  Reader  by  Henry  Sweet  are  so  well  known 

that  this  needs  no  introduction;    they  rank  at   once  among  the  books 

necessary  to  every  student  of  Anglo-Saxon. 

FRENCH. 

BIART.— Quand  J'etais  Petit.  Histoire  d'un  Enfant  racontee  par  un 
Homme.  By  Lucien  Biart.  Adapted  for  Use  in  Schools,  with  Notes 
and  Vocabulary  by  James  BoIelle,  B.  A.  Part  I.  i6mo,  cloth.  Pitt 
Press  Series.     Price,  50  cents  net. 

MOLIBRB — L'Avare.  Par  J.-B.  P.  MoLiiRE.  Edited  with  Introduc- 
tion and  Notes  by  £.  G.  W.  Braunholtz,  M.  A.,  Ph.  D.,  University 
Lecturer  in  French.  i6mo,  cloth,  pp.  xlvii  +  245.  Pitt  Press  Series. 
Price,  70  cents  net. 

GERMAN. 

MACMILLAN'S  FOREIGN  SCHOOL  CLASSICS.  Edited  by  G. 
Eugene  Fasnacht.  LESSING. — ^Minna  von  Bamhelm.  Edited 
with  Introduction  and  Notes  by  Kev.  Charles  Mere,  M.  A.,  Ph.  D., 
Leipzig.     i2mo,  cloth,  pp.  Ixviii -f-  224.     Price,  75  cents  net. 

AMERICANA  GERMANICA.  A  Quarterly  devoted  to  the  Comparative 
Study  of  the  Literary,  Linguistic  and  other  Cultural  Relatious  of 
Germany  and  America.  Editor,  Professor  M.  D.  Learned,  of  the 
University  of  Pennsylvania.  Vol.  I,  No.  i.  Single  number,  75  cents. 
Annual  subscription,  $2.00. 

ENGLISH. 

BARRERE  and  LELAND. — A  Dictionary  of  Slang,  Jargon,  and  Cant. 
Embracing  English,  American  and  Anglo-Indian  Slang,  Pidgin  English, 
Gypsies'  Jargon,  and  other  irregular  phraseology.  Compiled  and 
edited  by  Albert  Barr&re,  Officier  de  I'instruction  publique;  Pro- 
fessor R.  M.  A.  Woolwich,  author  of  'Argot  and  Slang,'  etc.,  etc.; 
and  Charles  G.  Leland,  M.  A,  Hon.  F.  K.  S.  L.,  author  of  <The 
Breitmann  Ballads,"  The  English  Gypsies  and  their  Language,' etc. » 
etc.  In  2  vols.  Vol.  I,  A-K,  pp.  xix+500;  vol.  II,  L-Z,  pp.  415. 
Crown  8vo.    Price,  $4.00. 

In  Preparation. 
GERMAN. 

HEINE.— Selections  from  Heine.  Edited  with  Notes  and  Introduction 
by  C.  A.  BucHHEiM,  Professor  of  German  Literature  in  King's  College* 
Uniform  with  Deutsche  Lyrik,  Golden  Treasury  Series. 


The  Macmillan  Company,  66  Fifth  Ave.,  New  York. 


Digitized  by  VjOOQ IC 


ADVERTISEMENTS. 


THE  FRIEDENWALD 

COMPANY, 

Printers, 
Lithographers, 
Book  Binders, 

BALTIMORE,  EUTAW  AND  GERMAN  STREETS, 

BALTIMORE,  MD. 


THE  LEADING  HOUSE. 


Printers  of  the  American  Journal  of  Philology,  American 
Journal  of  Mathematics,  Memoirs  from  the  Biolog- 
ical Laboratory,  Studies  in  Historical  and 
Political  Science,  issued  by  the  Johns 
Hopkins  University. 


ESTIMATES  CHEERFULLY  SUBMITTED. 


Digitized  by  VjOOQIC 


AMERICAN  JOURNAL  OF  PHILOLOGY. 


INDISPENSABLE  TO  EVERY  LATIN  TEACHER. 
QILDBRSLBBVB'S  LATIN  GRAMMAR. 

THIRD  EDITION,  REVISED  AND  ENLARGED. 

By  B.  L.  GiLDERSLEEVE,  Ph.  D.,  Professor  of  Greek  in  the  Johns  Hopkins 
University,  Editor  of  the  American  Journal  of  Philology,  and  Gonzalez 
Lodge,  Ph.  D.,  Associate  Professor  of  Latin  in  Bryn  Mawr  College,  Pa. 
550  pp.     Cloth,  9z.ao. 

For  school  and  college  use.  the  new  edition  of  Gildersleeve's  Latin  Grammar,  in  point  of 
fulness,  accuracy  and  convenience,  rivals,  if  it  does  not  surpass,  all  Latin  grammars  with 
which  I  am  acquainted.— -J.  P.  Postgatb,  Fellow  of  Trinity  College,  Cambridge,  England. 

It  is  a  superfluous  compliment  to  praise  any  work  of  Prof.  Gildersleeve,  who  never  does 
any  which  is  not  praiseworthy.— W.  W.  Goodwim,  Ph.  D.,  Eliot  Prof,  of  Greek  Literature, 
Harvard  University,  Cambridge,  Mass. 

CABSAR'S  CIVIL  WAR. 

With  Maps  and  Vocabularies.  Edited  for  the  use  of  Schools,  with  refer- 
ences to  the  Latin  Grammars  of  Gildersleeve,  Allen  and  Greenough, 
and  Harkness.  By  B.  Perrin,  Ph.  D.,  Professor  of  Greek  in  Yale 
University.    i2mo,  348  pp.     Price,  Cloth,  9z.oo. 

Prof.  Tract  Pbck,  Yale  University,  New  Haven,  says :  "  It  is  matter  for  congratulation 
that  at  last,  and  so  auspiciously,  the  'Civil  War'  has  been  made  available  and  attractive  for 
American  schools." 

Prof.  W.  A.  Packakd,  College  of  New  Jersey,  Princeton.  N.  J.,  says:  "Perrin's  'Civil 
War'  is  thoroughly  and  skillfully  edited,  and  is  a  welcome  addition  to  our  preparatory  text- 
books in  Latip.^' 

UNIVERSITY  PUBLISHING  COMPANY, 

Boston,  352  Wathington  St.         43,  45,  47  E.  Tenth  St.,  NEW  YORK. 
New  Orleans,  714-716  Canal  St.  [4T. 

ESSAYS   AND   STUDIES 

BY 

BASIL  LANNEAU  GILDERSLEEVE 
^o  pages t  small  quarto,  bound  in  cloth.    Price,  $3.50  (net). 


Educational  Essays: — i.  Limits  of  Culture.  2.  Classics  and 
Colleges.  3.  University  Work  in  America.  4.  Grammar 
and  Aesthetics. 

Literary  and  Historical  Studies: — i.  Legend  of  Venus. 
2.  Xanthippe  and  Socrates.  3.  ApoUonius  of  Tyana. 
4.  Lucian.  5.  The  Emperor  Julian.  6.  Platen's  Poems. 
7.  Maximilian ;  his  Travels  and  his  Tragedy.  8.  Occa- 
sional Addresses. 

N.  MURRAY, 

The  Johns  Hopkins  Press, 
Oct.^  1890.  Baltimore,  Md. 


Digitized  by  VjOOQ IC 


AD  VER  TISEMENTS. 


HELPS  IN  CLASSICAL  STUDY. 


HARPER'S  CLASSICAL  DICTIONARY. 

Harper's  Dictionary  of  Classical  Literature  and  Antiquities.  Edited  by  Harrt  Thur- 
ston Pbck,  M.  a.,  Ph.  D.,  Professor  of  the  Latin  Lan^ages  and  Literature  in  Columbia 
Unirersity,  City  of  New  York.  With  the  cooperation  of  many  special  contributors. 
Pp.  zvi,  1 701.  with  about  1500  illustrations.  Royal  8vo.  cloth,  $6.00;  half  leather,  |8.oo. 
Two-volume  edition,  cloth,  I7.00. 
Unquestionably,  in  many  rexpects,  the  most  anthoritative  work  of  the  kind  that  has  yet 

been  published. ...    In  short,  this  is  what  the  editor  claims  for  it— a  classical  encyclopaedia. 

There  is  no  one  rolume  on  the  market  at  once  so  complete  and  so  succinct,  and  whose  vast 

collection  of  information  is  so  readily  and  immediately  accessible A^.  Y,  Htrald, 

The  student  of  clswsical  antiquity,  whether  in  art,  life,  religion,  or  literature,  will  find  the 

volume  a  whole  library  of  information,  admirably  arranged  for  practical  use. . .  .    The  text  is 

clear,  terse,  and  pointed.— Ct>ra^<y  Inier-Octan* 

MODERN  GREEK  MASTERY. 

A  Short  Road  to  Ancient  Greek.  By  Thomas  L.  Stbdmah,  A.  M.,  M.  D.  Pp.  viii,  380. 
Pott  8vo,  cloth,  $1.50. 
Dr.  Stedman  has  filled  a  very  actual  want. ...  A  work  by  means  of  which  the  mastery  of 
the  so-called  modem  Greek  language  will  be  very  greatly  facilitated  and  that  of  the  ancient 
will  be  made  more  easy. .  .  .  The  parts  dealing  with  the  conjugation  of  the  verb,  especially 
the  sectioiu  on  the  aorist  and  future,  are  written  with  the  greatest  clearness,  and  are  readily 
comprehensible.  .  . .  There  never  hajt  been  published  in  the  English  language  fuch  a  practical 
method  for  the  mastery  of  the  Greek  tongue;  and  our  opinion  Is  that  this  important  and,  as 
far  as  that  Is  possible,  perfect  work  of  Dr.  Stedman  is  a  production  of  great  irvighi,— Atlantis, 
New  York. 

AN  ENQLISH-QREEK  DICTIONARY. 

A  Concise  Dictionary  of  the  Enelish  and  Modern  Greek  Languages,  as  Actually  Written 
and  Spoken.    By  A.  M.  Jamnaris,  Ph.  D.    English-Greek.    Pp.  xvi,436.    Post  8vo,  doth, 
I9.50. 
The  author  has  endeavored  to  render  this  work  compact  and  concise  by  employing  ajudiclous 
system  of  abbreviation  and  by  omitting  all  references  and  quotations.    The  book  is  of  interest 
to  the  student  of  Classic  Greek  who  posaenses  any  tendency  towards  philology.    The  extinc- 
tion of  the  present  infinitive,  of  the  optative,  and  of  part  of  the  imperative  will  be  noticed  in 
examining  the  verb.    Likewise  the  replacing  of  the  dative  by  the  accusative,  the  disappear- 
ance of  rho  final,  and  the  abandonment  of  the  reduplication  in  nouns  and  participles.    Archaic 
or  learned,  coUoquial,  literary  and  ecclesiastical  expressions  are  all  given,  and  are  marked 
with  different  signs.    Careful  comparison  reveals  the  Roman,  Byxantine,  or  Turkish  influence 
in  some  of  the  older  forms,  and  the  tendency  towards  learned  Greek  In  the  present  Restoration 
Period  of  the  languages. 

PROFESSOR  QILDERSLEEVE'S  PINDAR. 

Pindar:  The  Olympian  and  Pythian  Ode«.   With  Introductory  £«say, Notes. and  Indexes. 
By  Basil  L.  Gildbkslbkvk,  Professor  of  Greek  in  the  Johns  Hopkins  University. 
Pp.  cxvi,  396.    lamo,  cloth,  I1.50. 
Prepared  with  mat  exactness.— ^r^^^/^  Union, 

Never  before  did  the  appearance  of  an  American  edition  of  a  Greek  poet  so  nearly  approach 
an  'event*  in  scholastic  circles  as  the  publication  of  Gilderslceve's  ' Pindar.'— AVw  York 
Timg*. 

Pour  ceux  qui  commencent  T^tude  de  Plndare  rien  ne  saurait  remplacer  ce  livre. — Retmo  do 
Pkitologit, 

PROFESSOR  GILDERSLEEVE'S  PERSIUS. 

The  Satires  of  A.  Perkins  Flaccus.     Edited  by  Basil  L.  Gildbsslbbvb,  Ph.  D.  (Gdt- 

ttngen),  LL.  D.,  Professor  of  Greek  in  the  Johns  Hopkins  University.    Pp.  933.    lamo, 

cloth.  90  cents.    By  mail,  |i.oo. 

I  discover  in  it  at  once  ample  evidence  of  the  accurate  and  thorough  scholarship  which  I 

expected  to  find.  ...    I  have  myself  found  instruction  and  recreation  in  It.    The  '  Introduc- 

tion'  and  'Notes'  are  as  good  a  whetstone  as  the  'Satires'  themselves.— W.  S.  Tvlxr, 

Professor  of  Greek,  Amherst  College. 

JUSTIN  MARTYR. 

The  Apologies  of  Justin  Martyr.  To  which  is  appended  the  Epistle  to  Diognetus.  With 
an  Introduction  and  Notes  by  Basil  L.  Gildkr8i.bbvb,  Ph.  D.  ^Gfitt.),  LL.  D.,  Professor 
of  Greek  in  the  Johns  Hopkins  University.  Pp.  xlii,  990.  xamo,  cloth,  ^x.30.  By  mail, 
Si.44.  

Published  by  HARPER  &  BROTHERS,  New  York. 

Tko  abovt  works  art  for  salf  by  all  booksell'rs,  or  will  he  sent  by  the  Publishers  on  receipt 
0/ price,    l/ordored sent  by  mail,  xo per  cent,  must  be  addod  to  cover  cost  0/ postage.       [it. 


Digitized  by  VjOOQ IC 


THE  JOHNS  HOPKINS  PRESS, 


I.  American  Journal  of  Mathematics. 

Edited  by  T.  Craig,  with  the  co-operation  of  S.  Nkwcomb.     Quarterly. 
4to.     Volume  XIX  in  progress.     $5  per  volume. 

II.  American  Chemical  Journal. 

I.  Remsen,  Editor.   Ten  numbers  a  year.    8vo.     Volume  XIX  in  progress. 
$4  per  volume. 

III.  American  J6umal  of  Philology. 

6.  L.  GiLDERSLESVB,  Editor.     Quarterly.    8vo.    Volume  XVII  complete. 
$2  per  volume.     Index  to  Vols.  I-X,  $1, 

IV.  Studies  from  the  Biological  Laboratory.    8vo.    Volume  V  complete. 

$5  per  volume. 

V.  Studies  in  Historical  and  Political  Science. 

H.  B.  Adams,  Editor.    Monthly.    8vo.     Volume  XV  in  progress.     $3  per 
volume. 

VI.  Johns  Hopkins  UniTersity  Circulars. 

Containing  reports  of  scientific  and  literary  work  in  progress  in  Baltimore. 
4to.     Vol.  XVI  in  progress.    $1  per  year. 

VII.  Johns  Hopkins  Hospital  Bulletin.    4to.    Monthly.    Volume  VIII  in 
progress.    $1  per  year. 

VIII.  Johns  Hopkins  Hospital  Reports.    4to.    Volume  VI  in  progress.     $5 
per  volume. 

IX.  Contributions  to  Assyriology  and  Comparative  Semitic  Philology. 
(Beitr&ge  zur  Assyriologie,  etc.)    Vol.  Ill  in  progress. 

X.  Memoirs  from  the  Biological  Laboratory. 

W.  K.  Brooks,  Editor.     Volume  III  complete.     $7.50  per  volume. 

XI.  Annual  Report. 

Presented  by  the  President  to  the  Board  of  Trustees,  reviewing  the  opera- 
tions of  the  University  during  the  past  academic  year. 

XII.  Annual  Register. 

Giving  the  list  of  officers  and  students,  and  stating  the  regulations,  etc,  ol 
the  University.     Published  at  the  close  of  the  academic  year. 


In  addition  to  the  serials  above  named,  copies  may  be  obtained  of  the  works 
mentioned  below : 
The  Critical  Edition  of  the  Hebrew  Text  of  the  Old  Testament. 

Edited  by  Professor  Paul  Haupt.     Prospectus  on  application. 
Rowland's  Photograph  of  the  Normal  Solar  Spectrum.    10  plates. 

$20.00. 
Description  of  the  Johns  Hopkins  Hospital.     J.  S.  Billings,  Editor. 

116  pp.  4to.     56  plates.    $7.50,  cloth. 
The  Teaching  of  the  Apostles.    Complete  facsimile  edition.    Edited  by 

J.  Rendel  Harris.     1 10  pp.  4to.     10  plates.     $5.00. 
Reproduction  in  Phototype  of  a  Syriac  MS  with  the  Antilbgomena 

Epistles.    Edited  by  I.  H.  Hall.    I3. 
The  Oyster.     By  W.  K.  Brooks.     240  pp.  i2mo.     14  plates.    $1.00. 
Studies  in  Logic.     By  members  of  the  Johns  Hopkins  University.     C.  S. 

Pierce,  Editor.     123  pp.  i2mo.     $2.00,  cloth. 
The  Constitution  of  Japan,  with  Speeches,  etc.,  illustrating  its  significance. 

48  pp.  i6mo.     50  cents. 
Bibliographia  Hopkinsiensis.   Part  I.  Philology.   Parts  II-III.  Chemistry, 

Mineralogy   and    Geology.     Parts   IV-V-Vl.    Physics,   Astronomy  and 

Mathematics.    8vo.     30  cents  per  part. 

A  full  list  of  publications  will  be  sent  on  application. 


Communications  in  respect  to  exchanges  and  remittances  may  be 
sent  to  The  Johns  Hopkins  Press,  Baltimfia^ Maryland. 


Digitized  by  VjOOQ IC 


AMERICAN 
JOURNAL    OF    PHILOLOGY 

Vol.  XVIII,  2.  Whole  No.  70. 


I.— THE  SUBJUNCTIVE  IN  INDEPENDENT  SEN- 
TENCES IN  PLAUTUS. 

I. — Facts. 

This  paper  consists  of  three  parts:  first,  a  presentation  of  the 
facts  of  usage ;  second,  a  discussion  of  some  of  the  forces  which 
appear  to  have  affected  the  mode ;  third,  remarks  upon  the  poten- 
tial and  the  optative  uses  and  upon  the  subjunctive  in  general. 

No  precise  line  can  be  drawn  between  the  independent  sentence 
and  the  dependent  clause,  but  for  my  purpose  I  have  included 
most  of  the  cases  of  the  subjunctive  in  parataxis.  Of  indirect 
questions,  however,  I  have  taken  only  those  into  which  uis,  uin 
and  similar  words  are  inserted ;  other  indirect  questions,  though 
many  of  them  are  actually  paratactic,  are  not  included  in  the  lists. 
The  paratactic  prohibition  with  ne  passes  at  once  into  the  ne 
clause,  and  is  therefore  not  given.  No  sentences  are  included 
which  contain  a  formal  protasis ;  this  involves  the  omission  of  a 
considerable  number  of  cases  where  the  subjunctive  is  undoubt- 
edly independent  of  the  protasis,  but  they  are  in  all  respects 
similar  to  the  cases  given  in  the  lists  and  the  material  is  sufficient 
without  them. 

In  the  presentation  of  the  facts  there  is  no  classification  by 
function,  such  as  is  usually  made.  The  division  is  by  tense, 
person  and  number,  with  a  subdivision  according  to  the  form  of 
the  sentence.    This,  in  full,  would  be  as  follows : 

I.  Affirmati/e. 

A.  Indeoendent. 

a)  Non-interrogative. 

b)  Interrogative. 


Digitized  by  VjOOQ IC 


134  AMERICAN  JOURNAL    OF  PHILOLOGY. 

B.  Paratactic. 

a)  Non-interrogative. 
F)  Interrogative. 
II.  Negative  (with  the  same  sub-classes). 
For  all  practical    purposes  this  scheme  is  followed    closely 
enough  by  making  for  each  person  and  number  four  classes : — 
Independent,  Interrogative,   Paratactic  and   Negative   with  ne^ 
Under  each  class  the  uses  to  which  the  form  may  be  put  are 
described  and  illustrated,  and  this  leads  in  the  end  to  a  partial 
classification  by  function.     Full  lists  are  given  only  where  the 
usage  is  important  or  infrequent,  but  complete  statistics  will  be 
found  in  the  tables. 

Present,  i  si  sing. 

A.  Non-interrogative,  independent. — As  the  same  form  is  used 
in  the  3d  and  partly  in  the  4th  conjugation  for  pres.  subj.  and  fut* 
indie,  the  undoubted  subjunctive  forms  are  given  first  and  used 
as  a  criterion  for  distinguishing  pres.  from  fut.  where  the  form  is 
the  same. 

Three  cases  are  marked  as  subjunctive  by  uiinam,  Aul.  433,. 
Epid.  196,  Trin.  618.  These  are  all  wishes.  Pers.  575,  modo  ut 
sciam  quanti  indicet,  is  marked  by  ut,  but  in  meaning  is  like  those 
which  follow. 

The  subj.  form  is  found  in  the  following  cases :  Ba.  1049  quod 
perdundumst,  properem  perdere ;  1058  sed  crepuit  foris :  ecfertur 
praeda  ex  Troia.  taceam  nunciam;  Trin.  11 36  quid  ego  cesso* 
hos  conloqui?  sed  maneam  etiam,  opinor;  Pers.  542  uideam 
modo  mercimonium.  Cf.  575,  above,  and  Ter.  Heaut.  273  hoc 
quod  coepi  primum  enarrem,  Clitipho:  post  istuc  ueniam.  All 
express  the  speaker's  desire  or  choice  in  regard  to  an  act  of  his 
own.  The  first  three  are  in  soliloquy ;  Pers.  542  and  575  are  in 
a  dialogue  and  are  answered,  but  they  are  also  half-soliloquizing. 

All  the  remaining  cases  (except  those  of  the  3d  conjug.)  are 
hypothetical,  dealing  with  the  speaker's  action  in  a  supposed 
case.  The  phrase  non  meream  (merear')  is  found  in  Ba.  11 84 
quem  quidem  ut  non  hodie  excruciem,  alterum  tantum  auri  non 
meream,  Men.  217,  Poen.  430,  and  in  a  like  sense,  non  emanty 
Capt  274.  Almost  identical  is  floccum  (jctccum)  non  interduim, 
Rud.  580,  Trin.  994,  and,  with  negative  implication,  Aul.  672  tam 
duim  quam  perduim.  All  these  assert  the  speaker's  conviction  in 
regard  to  his  own  action  in  a  supposed  case,  corresponding  in 
general  to  the  English,  *  I  would  not  do  it  for  the  world.' 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  I35 

In  two  cases  the  speaker  assures  the  hearer  in  regard  to  his 
action  in  a  supposed  case.  Capt.  237,  quod  tibi  suadeam,  suadeam 
meo  patri,  is  not  above  suspicion  (jsuadeo^  Cam.,  Sch.),  but  Trin. 
758  is  a  clear  statement  of  intention. 

The  assurance  is  scarcely  more  than  a  view  or  opinion  in  True. 
495  sine  uirtute  argutum  ciuem  mihi  habeam  pro  praefica,  and 
Aul.  230,  Merc.  125  nimis  nili  tibicen  siem,  are  clearly  only 
expressions  of  opinion.  In  Capt.  237,  Trin.  758,  Aul.  230,  a 
clause  with  conditional  coloring  precedes,  and  the  same  thing  is 
implied  in  True.  495. 

The  text  is  doubtful  in  several  passages.  In  Aul.  570  there  is 
a  hiatus  and  Seyffert  supplies  quod . .  .  habeo.  This  makes  good 
sense,  but  I  think  it  sure  that  non  potent  could  mean  the  same 
thing  as  nolopotare^  which  Goetz  suggests  in  the  notes.  In  Trin. 
749  the  difficulty  is  in  reconciling  the  text  to  the  following  verses, 
but  adeam^  edoceam  are  correct  in  themselves,  and  precbely 
similar  to  cases  which  will  be  mentioned  in  3d  sing.,  where  a  half- 
indirect  subjunctive  is  used  in  proposing  a  plan  of  action.  Men. 
982,  if  si  (Bx.)  be  not  supplied,  is  an  extreme  case,  scarcely 
parallel  to  any  of  the  preceding.    St.  208^  is  a  gloss. 

Of  clearly  subjunctive  forms,  then  (omitting  uelim)^  PI.  uses 
only  20,  with  perhaps  two  to  be  added  from  doubtful  passages. 
Functionally,  they  are  of  four  kinds.  They  express  a)  the 
speaker's  choice  or  desire  in  regard  to  his  own  action;  V)  his 
conviction  as  to  his  (negative)  action  in  a  supposed  case ;  c)  his 
assurance  to  another  person  in  regard  to  his  action  in  a  supposed 
case;  and  d)  his  opinion  about  a  supposed  case.  In  c  and  d 
there  is  usually  a  clause  or  phrase  which  gives  hypothetical  tone. 

As  to  the  verbs  which  use  the  same  form  for  ist  sing,  of  fut. 
indie,  and  pres.  subj.,  it  may  be  said,  by  way  of  preliminary,  that 
the  fact  that  language  has  found  no  special  form  to  express 
futurity  in  the  first  person  dicam^  as  it  has  in  dices^  dicas,  is  clear 
proof  that  there  was  no  strong  distinction  in  meaning.  It  is 
therefore  an  error  to  force  every  case  of  faciam  or  dUam  either 
into  a  future  or  a  pres.  subjunctive. 

Some  of  the  frequently  recurring  soliloquies  in  which  a  speaker 
conceals  himself  as  he  hears  the  door  opening,  like  Ba.  610  sed 
hue  concedam,  Epid.  103,  Cas.  434,  closely  resemble  Trin.  1136, 
so  that  with  a  different  verb  we  might  have  had  a  clear  subj.  form, 
e.  g.  abeam.  The  situation  in  Ba.  798,  Rud.  1356  sed  conticiscam, 
is  exactly  like  Ba.  1058  taceam  nunciam.    As  the  periphrastic 


Digitized  by  VjOOQ IC 


136  AMERICAN  JOURNAL    OF  PHILOLOGY. 

form  in  Ba.  1049  (above)  helps  to  d^fm^  properem^  so  in  Cist.  657 
faciundumst  puerile  ofBdum :  conquiniscam  ad  cistulam,  the  two 
phrases  refer  to  the  same  act,  toward  which  the  speaker's  mental 
attitude  is  the  same,  and  the  periphrastic  strongly  suggests  the 
sense  'must'  for  the  subj.  Cf.  the  parataxis  with  necessest,  below. 
In  Asin.  605  sermoni  iam  finem  face  tuo :  huius  sermonem  acci- 
piam,  Aul.  405,  Cas.  516  nunc  amicine  anne  inimici  sis  imago, 
.  . .,  sciam,  the  verb  expresses  desire,  not  futurity,  and  Merc.  881, 
recipiam  me  illuc,  expresses  choice,  as  in  parataxis  with  opiumumsL 

The  modal  shading  in  these  cases  is  faint  and  it  would  no  doubt 
be  possible  to  translate  them  all  as  futures,  but  they  are  parallel 
to  the  subj.  forms  given  above  and  to  the  paratactic  uses  to  be 
given  later.  With  them,  though  less  clearly,  may  be  classed  the 
cases  of  the  subj.-fut.  forms  with  poiiuSy  like  Epid.  149  ne  feceris : 
ego  istuc  accedam  periclum  potius,  Cas.  999  hercle  opinor  potius 
nobis  credam  quod  dicitis,  Aul.  767  i  refer:  dimidiam  tecum 
potius  partem  diuidam.  There  is  a  slight  modal  shading,  deter- 
mination taking  the  form  of  preference,  but  there  is  nothing  in 
the  context  to  give  a  hypothetical  tone.  Cf.  also  the  difference 
between  uelim  and  malim^  below,  and  see  the  careful  treatment 
of  these  expressions  in  Neumann,  de  fut  in  prise  Lat. .  . .  ui  et 
usu,  Breslau,  1888. 

non  dicam  dolo,  Men.  228,  Trin.  480,  is  an  introduction  to  a 
following  remark  and  is  allied  to  the  frequent  use  of  dicam, 
eloquar  to  introduce  a  statement.  Trin.  90  haud  dicam  dolo  is 
exactly  similar,  though  the  statement  is  postponed  to  94;  it 
should  be  followed  by  a  colon,  not  by  a  period.  The  ordinary 
punctuation  makes  it  wrongly  a  reply  to  the  preceding  question. 
Of  the  same  nature  is  the  phrase  deum  (maiorum)  uirtute  dicam, 
M.  G.  679,  Pers.  390,  Ps.  581,  Trin.  346.  It  is  parenthetical  and 
introductory  to  the  statement  which  follows.*  In  Aul.  283  dicam 
seems  to  be  similar,  though  the  text  is  hopelessly  corrupt.  It  is 
possible  enough  to  translate  '  I  may  say,'  but  instead  of  trusting 
to  the  very  uncertain  test  of  translation  I  prefer  for  the  present  to 
say  that  these  cases  lie  in  the  borderland  between  the  future  and 
the  subjunctive,  having  resemblances  to  both. 

In  a  few  cases  the  context  shows  that  the  mode  expresses  an 
opinion  in  regard  to  a  supposed  case.    In  Aul.  232  there  is  an 

^  Cramer,  de  perf.  coniunct.  usu  potential!  ap.  prise,  script.  Lat.,  Marburg, 
1886,  p.  54,  calls  this  optative  and  comp./or/  tua  dixerim,  Brix  on  Trin.  346 
calls  it  future. 


Digitized  by  VjOOQ IC 


wmmm^^smsmmmmBmm^ 


THE  SUBJUNCTIVE  IN  PLAUTUS.  1 37 

ubi  clause;  in  Ps.  358  the  hypothetical  suggestion  is  in  the 
sentence  itself;  in  M.  G.  845  it  is  suggested  by  the  other  speaker ; 
Amph.  156,  161  are  in  a  long  hypothetical  passage  containing  a 
si  clause.  Men.  985  is  confused  and  Bx.  is  probably  right  in 
making  a  protasis. 
To  summarize,  there  are  in  ist  sing.  pres.  the  following : 


With  uiinam. 

3 

modo  id, 

I 

Subj.  forms— 

of  desire  or  choice, 

4 

conviction  (neg.), 

7 

assurance, 

2 

opinion, 

3 

Doubtful  forms — 

of  desire  and  choice. 

9 

vf'iih  poiius,                (?) 

non  dicam  dolo,            3 

deum  uiriute  dicam,    4 

of  opinion, 

5 

34 

To  this  number  perhaps  two  or  three  should  be  added  from 
passages  of  doubtful  text,  and  if  the  cases  -^x^poiius  were  added, 
the  number  would  be  somewhat  increased. 

The  remaining  cases  under  this  heading  are  all  cases  oiuelim 
and  compounds,  of  which  full  lists  are  given. 

uelim  with  paraiactic  subjunctive. — In  the  2d  sing,  pres.,  Cas. 
234  enicas.  II uera  dicas  uelim;  Rud.  511  pulmoneum  edepol 
nimis  uelim  uomitum  nomas;  Rud.  1067,  Men.  909.  The  taunt- 
ing verse  in  Trin.  351  quod  habes  ne  habeas  et  illuc  quod  non 
habes  habeas  is  followed  by  uelim  malum  in  A,  by  malum  in  BD. 
To  the  reasons  given  by  Bx.  Anh.  for  believing  malum  to  be  the 
gloss,  I  would  add  the  close  resemblance  of  this  passage  to  the 
others  above  and  the  fact  that  the  2d  sing,  always  has  uiinam 
or  uelim  in  wishes  except  in  certain  formulas  and  once  in  the 
marriage  song,  Cas.  822.  With  the  3d  pers.,  Rud.  877  perii.  || 
uerum  sit  uelim ;  and  six  cases  with  ueniat,  Aul.  670  nimis  hercle 
ego  ilium  coruom  ad  me  ueniat  uelim,  Cas.  559,  Most.  1074  nunc 
ego  ille  hue  ueniat  uelim,  Poen.  1288,  Ps.  1061,  True.  481.  In 
five  of  these  nunc  is  used  and  the  person  is  sometimes  in  the 


Digitized  by  VjOOQ IC 


138  AMERICAN  JOURNAL    OF  PHILOLOGY. 

nom.  with  ueniat,  sometimes  in  ace.  (Aul.  670,  Cas.  559,  Ps.  1061). 
The  order  is  always  ueniai  tielim.  With  the  3d  pers.  of  the  perf., 
Ba.  334  nescit  quid  faciat  auro.  ||  mihi  dederit  uelim,  Poen.  1206 
.  . .  quod  haruspex  . . .  dixit.  ||  uelim  de  me  aliquid  dixerit,  Rud. 
662,  Poen.  570.  Once,  in  Most.  632  nihilo  plus  peto.  ||  uelim 
quidem  hercle  ut  uno  nummo  plus  petam,  an  id  is  inserted 
between  uelim  and  the  subjunct.,  with  the  same  challenging  and 
hostile  sense  that  appears  in  most  cases  with  paratactic  subjunctive. 

In  ten  cases  uelim  takes  an  iniin.  Without  subject  ace,  Cist. 
497  di  me  perdant — 1|  quodcumque  optes,  tibi  uelim  contingere  ; 
Asin.  274.  With  me^  Ba.  530,  Cas.  287,  Epid.  120,  Most.  218  in 
anginam  ego  nunc  me  uelim  uorti.  With  te,  Aul.  120  uelim  te 
arbitrari  med  haec  uerba .  . .  tuae  rei  causa  facere;  Fragm.  41 
uelim  ted  arbitrari  factum.  With  other  subjects,  St.  587  edepol 
ne  ego  nunc  mihi  medimnum  mille  esse  argenti  uelim ;  Trin.  433. 
With  a  perfect  participle,  Aul.  504  moribus  praefectum  mulierum 
hunc  factum  uelim ;  Ba.  603  sufflatus  ille  hue  ueniet.  ||  disruptum 
uelim ;  Cas.  326  ego  edepol  illam  mediam  disruptam  uelim ; 
Cure.  83,  St.  191,  613.  With  an  adj.,  Amph.  834  uera  istaec 
uelim ;  cf.  Rud.  877,  above.  With  a  direct  object,  Amph.  1058 
animo  malest,  aquam  uelim ;  Most.  266  nimis  uelim  lapidem,  qui 
.  . .  diminuam  caput;  Ps.  598,  Rud.  211.  Absolute,  Cas.  464  ut 
tibi,  dum  uiuam,  bene  uelim  plus  quam  mihi !  (This  is  the  only 
passage  where  «/,  which  here  shows  plainly  its  exclamatory 
character,  is  used  with  uelim)  \  Most.  742  (R.^  L.'*  uellem),  Pers. 
629,  Ps.  1070,  Trin.  58,  Fragm.  38. 

As  to  the  meaning  of  these  44  cases,  the  following  points  may 
be  noted :  i)  With  a  few  exceptions,  the  speaker  does  not  expect 
that  the  expression  of  his  desire  will  bring  about  the  realization 
of  the  desire.  This  is  the  general  characteristic  which,  with 
some  modifications,  is  common  to  all  the  various  forms  of 
optation.^  The  exceptions  are  Men.  909,  perhaps  Rud.  1067, 
Aul.  120,  Fragm.  41,  in  all  of  which  the  person  addressed  is  to  be 
the  actor.  An  expression  of  desire  thus  directly  addressed  to 
the  person  who  has  it  in  his  power  to  act  is  of  necessity  very  little 
removed  from  a  direct  expression  of  will,  and  Men.  909  adeas 
uelim  is  only  a  trifle  more  polite  than  adeas  or  adeas  uolo.  But 
in  Rud.  1067  .  . .  ne  uideas  uelim  it  is  not  in  the  power  of  the 
second  person  to  avoid  seeing,  and  in  Aul.  120,  Fragm.  41,  where 

^  I  venture  to  use  this  word  in  a  technical  sense,  in  order  to  avoid  confusion 
from  the  different  meanings  of  the  English  word  wish. 


Dlylilz^d  by 


Geogte- 


THE  SUBJUNCTIVE  IN  PLAUTUS.  1 39 

the  dependent  phrase  is  te  arbiirari^  the  thinking  or  believing  is 
not  strictly  an  act  which  depends  on  the  will  of  the  believer. 
Possibly  Most.  742  should  also  be  called  an  exception,  but  the 
passage  is  peculiar  in  other  ways  also.  2)  In  more  than  half  of 
these  cases  the  content  of  the  sentence  is  a  curse  or  a  wish  which 
involves  a  threat.  3)  The  paratactic  subjunctive  is,  with  perhaps 
two  exceptions,  the  kind  of  subjunctive  which  in  other  connections 
would  be  recognized  as  optative. 

In  general,  the  understanding  of  a  modal  usage  must  depend 
upon  a  correct  interpretation,  and  full  lists  of  uelim  have  been 
given  in  order  to  enable  the  reader  to  see  for  himself  that  the 
ordinary  explanation  of  uelim  as  a  potential  or  a  subjunctive  of 
modesty  or  of  mild  assertion  is  absolutely  inconsistent  with  the 
facts  of  usage  in  Plautus.  To  translate  it '  I  should  like '  is  simply 
to  introduce  confusion  by  the  use  of  a  peculiar  English  idiom. 
The  subjunctive  is  optative  in  character,  but  the  fuller  discussion 
of  this  must  be  postponed  until  the  other  paratactic  verbs  ai-e 
given. 

mauelim,  malim.  With  paratactic  subjunctive,  Poen.  11 50 
abeo  igitur.  ||  facias  modo  quam  memores  mauelim,  Poen.  11 84. 

With  ut  clause,  Trin.  762  malim  hercle  ut  uerum  dicas  quam 
ut  des  mutuom. 

With  infin.,  Asin.  811  emori  me  malim,  quam  haec  non  eius 
uxori  indicem,  Ba.  465,  490,  514,  519  (a  gloss),  Men.  720,  Merc. 
356, 889,  Pers.  4,  Poen.  827,  True.  260, 743  (a  somewhat  uncertain 
conjecture),  Vid.  no.  The  subject  is  mCt  expressed  or  implied, 
except  in  the  last  case. 

With  direct  object,  Poen.  151  istuc  mauelim. 

With  adj.  or  ptc,  implying  esse,  Aul.  661,  Epid.  119,  Poen. 
1 214,  True.  742. 

With  dependent  phrase  implied  in  the  context,  though  not 
expressed,  Capt.  858,  Rud.  570,  True.  422. 

It  will  be  noted  that  malim  takes  the  infin.,  with  me  implied  or 
expressed,  much  more  generally  than  uelim,  and  that  it  has  the 
paratactic  subjunctive  only  twice,  in  2d  sing.  pres.  With  these 
exceptions,  the  relationship  to  uelim  appears  clearly ;  the  desire 
(preference)  is  expressed  without  expectation  that  the  expression 
will  lead  to  its  realization,  as  in  all  optations,  and  a  large  number 
of  cases  contain  a  kind  of  self-curse  (emori  me  malim,  quam; 
perire  me  malim;  mendicum  malim  mendicando  uincere;  arare 
mauelim,  etc.).    The  composition  with  mage,  however,  by  reduc- 


Digitized  by  VjOOQ IC 


Pant.  lubj. 

ut 

Infin 

uelim^        1 6 

I 

10 

malim,        2 

I 

13 

noitm, 

2 

peruelifn^ 

2 

140  AMERICAN  JOURNAL    OF  PHILOLOGY. 

ing  the  wish  to  a  preference  somewhat  obscures  the  optative 
sense  and  gives  more  apparent  reason  for  calling  malim  potential. 
Compare  what  is  said  later  on  the  effect  oi  poUus. 

nolim  is  used  three  times :  Amph.  86  and  sppt.  943  with  infin., 
Merc.  539,  absolutely. 

pemelim  also  occurs  three  times :  with  infin.,  Cas.  862,  Epid. 
536 ;  with  ptc.  {esse  implied),  Cure.  102. 

In  tabular  form  these  uses  are : 

Infin.        Ptc.  and  adj.         Obj.  Absol. 

7  4  6  =  44 

4  I  3  =  24 

I  =    3 

I  =    3 

74 

B.  InterrogaHve  sentences^  with  fires,  subj.,  ist  sing. — Since  the 
effect  of  the  interrogative  form  of  sentence  upon  the  meaning  of 
the  mode  must  come  up  later  for  discussion,  the  statement  of 
usage  is  made  as  brief  as  possible. 

I.  Questions  with  quis^  arranged  (with  the  exception  of  the 
dicam  questions)  according  to  the  form  and  construction  of  quis. 

The  pronoun  occurs  5  times  in  the  accus.  {qium  3,  qtiam  i, 
quod  i) ;  two  of  these  are  in  soliloquy  and  the  question  is  delib- 
erative, the  others  follow  an  impv.  or  its  equivalent  and  seek  to 
learn  the  desire  of  the  person  addressed,  quam  ob  rem  (3  cases, 
if  we  include  Mil.  Glor.  360)  follows  and  repudiates  a  suggestion 
or  command,  quo  modo  (J>acio)y  3  cases,  always  implies  nullo 
tnodo,  and  in  M.  G.  1206  seems  to  imply  *can.' 

The  adverbs  of  place  are  like  the  pronominal  forms,  not 
idiomatic,  ubi,  3  times  in  soliloquy  with  quaeram,  requiram^ 
inuetnam,  all  deliberative,  once  in  a  question  as  to  the  will  of  the 
person  addressed,  quo  follows  an  impv.  4  times  and  an  implied 
suggestion  once,  in  true  questions;  in  soliloquy,  deliberative,  3 
times,  unde  (2),  once  after  an  impv.,  with  rejecting  force,  once  in 
soliloquy. 

The  adverbs  which  mean  'how*  or  'why'  are  more  idiomatic. 
qui  is  used  only  in  the  formula  qui  ego  isiuc  credam  (Jibt)  ?,  Cure. 
641,  Merc.  627,  902,  'how  do  you  expect  me  to  believe  that?,' 
with  repudiating  force,  quitiy  M.  G.  426,  is  repudiating,  uij  Ba. 
149,  is  more  nearly  hypothetical,  under  the  influence  of  lubens. 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  I4I 

quor  {cur)  is  used  13  times.  The  verb  is  always  of  mental  action, 
poshUem,  rogem,  negem,  curem  (2),  mirer^  mentiar^  tniniiery 
quaeram  (2),  suscei^eam,  adflicter  (Ps.  1295),  perpeirem  (Cas. 
701).  A  few  of  the  questions  are  addressed  directly  to  the  other 
speaker,  but  nearly  all  have  a  half-soliloquizing  tone,  and  all 
imply  a  rejection,  as  in  English  when  the  why  is  made  emphatic 
with  a  falling  inflection.    The  negative  is  nan. 

Of  the  139  quis  questions  with  the  ist  sing,  pres.,  81  are  intro- 
duced by  quid. 

quid  is  the  direct  object  in  54  cases.  Of  the  18  cases  with 
other  verbs  than  faciam  and  agam  (which  are  given  separately 
below),  about  half  repeat  an  impv.,  as  in  Aul.  651  redde  hue.  || 
quid  reddam?,  with  a  tone  which  varies  from  repudiation  to  a 
distinct  question  as  to  the  desire  of  the  speaker  (cf.  True.  789 
quid  loquar  ?  with  Epid.  584  quid  loquar  uis  ?).  About  half  a 
dozen  are  in  soliloquy  and  are  deliberative. 

quid  faciam  serves  so  well  to  illustrate  the  history  and  meaning 
of  quis  questions  that  I  give  the  lists  in  full. 

a)  quid  faciam  f  Ba.  634,  Cist.  63,  301,  Epid.  98,  Merc.  207, 
565,  M.  G.  459,  Most.  523,  Pers.  42,  Poen.  357.  Of  these,  Ba. 
634  and  Merc.  207  are  in  soliloquy  and  are  deliberative ;  the  rest 
mean  *  What  do  you  want  me  to  do?' 

b)  quid  nunc  faciam  f  Men.  834,  Ps.  1229,  both  asking  for 
advice  or  direction. 

c)  quid  ego  faciam  f  Cure.  589  (deliberative),  Pers.  26 
(addressed  to  another  person,  but  half-soliloquizing). 

</)  quid  ego  faciam  nuncf    Epid.  255,  like  Pers.  26. 

e)  quid  ego  nunc faciamf  Ba.  857  {nunc  ego),  Cas.  549,  Cure. 
555»  Men.  963,  M.  G.  305,  Most.  371.  Both  Ba.  857  and  Most. 
371  are  answered,  but  the  question  was  not  necessarily  addressed 
to  the  other  person ;  the  rest  are  in  soliloquy. 

f)  quid  faciam  with  other  added  words :  aliud  Merc.  568,  hoc 
.  .  .postea  Most  346,  tibi  Ps.  78,  huic  homini  Ps.  1316.  The  last 
is  in  pretended  deliberation. 

There  are  also  some  cases  with  appended  protasis. 

quid  agam  is  used  in  the  same  way.  a)  quid  agamf  in 
dialogue,  Aul.  636,  M.  G.  363.  b)  quid  nunc  agamf ,  Amph. 
1046,  Cas.  952,  Poen.  351  {agam  iiunc),  all  in  soliloquy,  c)  quid 
^go  agamf.  Most.  378  (in  soliloquy),  Trin.  981  (repudiating  an 
impv.).  d)  quid  ego  nunc  agamf,  Aul.  274,  447,  Cist.  528,  all  in 
soliloquy.  /)  quid  ego  nunc  cum  illoc  agamf.  Men.  568,  for 
advice. 


Digitized  by  VjOOQ IC 


142  AMERICAN  JOURNAL    OF  PHILOLOGY. 

In  the  accus.  of  compass  and  extent  quid  is  generally  indistin- 
guishable from  quid  *  why.*  Capt.  556  .  . .  etiam  huic  credis  ?  || 
quid  ego  credam  huic  ?  ||  insanum  esse  me,  is  clearly  the  pronoun, 
and  where  the  pronominal  force  is  distinct  the  questioning  force 
is  also  most  distinct.  But  in  general  these  cases,  20  in  all,  are 
like  those  with  quor,  above,  in  implying  that  there  is  no  reason 
for  acting  as  the  other  person  desires,  or  at  least  in  implying  a 
disinclination  toward  the  action.  Three  cases,  Capt.  536,  Rud. 
447,  Trin.  1024,  are  in  soliloquy,  and  Amph.  41  is  in  a  prologue. 

In  quid  ni  (7  times),  quippe  ni  (once)  the  ni  has  been  suffi- 
ciently shown  to  be  negative,  not  conditional,*  and  these  sentences 
are  merely  the  negative  form  of  those  given  above. 

Cas.  454  deosculer.  ||  quid  'deosculer*?  is  not  really  a  question 
with  subjunctive  verb,  and  in  Epid.  281,  Merc  887  the  text  is 
entirely  uncertain. 

Beside  these  there  are  13  cases  in  which  dicam  is  inserted  and 
one  similar  case  vrith  praedicem.  They  are  introduced  by  various 
forms  of  quis.  True.  689  quam  esse  dicam  hanc  beluam  ?,  Cas. 
616  qua,  abl.;  quid,  ace,  Ps.  744  sed  quid  nomen  esse  dicam  ego 
isti  seruo?,  Asin.  587,  Merc.  516,  Pers.  400;  quid  'why,'  M.  G. 
1 201  quid  te  intus  fuisse  dicam  tam  diu?,  St.  288;  quo,  Capt.  533 
quo  ilium  nunc  hominem  proripuisse  foras  se  dicam  ex  aedibus?. 
Cure.  I,  12;  unde,  Ps.  966,  Rud.  264,  Ps.  1305  f.  sed  die  tamen, 
.  .  .  unde  onustam  celocem  agere  te  praedicem  ?  In  the  last  case 
it  does  not  matter  whether  the  question  is  regarded  as  direct  or 
indirect;  see  Becker  in  Studemund's  Studien,  I,  p.  160,  and  cf. 
Ps.  709.  The  introducing  word  in  these  questions  is  for  the  most 
part  of  a  kind  that  does  not  greatly  influence  the  meaning  of  the 
mode,  and  the  questions  differ  from  those  given  above  only  in 
the  fact  that  dicam  with  the  infin.  is  used  as  a  periphrasis  for  the 
simple  verb.  In  the  simpler  form  these  questions  would  be  quae 
est  haec  beluaf  quid  nomen  est  isti  seruo  f  quid  intus  fuisti  iam 
diuf  The  inserted  dicam  expresses  the  same  thought  as  the 
English  *What  kind  of  a  creature  am  I  to  suppose  this  to  be?' 
*  What  would  you  have  me  call  that  slave  of  yours  ?  *  That  dicam 
is  subj.  and  not  future  is  plain  from  Ps.  1306  and  from  the  thought. 

II.  Sentence  questions  are  for  the  most  part  exclamatory, 
corresponding  in  form  to  exclamations  with  the  indicative  rather 
than  to  true  interrogations. 

*0.  Brugmann,  Ueber  den  Gebrauch  des  condicionalen  ni,  Leipzig,  1887. 


Dig  itEed-b>^V3-0  OQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS,  1 43 

egone  is  used  26  times,  including  Cas.  117  egon  quid  faciam 
tibi  ?  and  True.  276,  where  Schoell  reads  tene  ego.  All  are  repu- 
diating exclamations,  and  the  will  of  the  other  person,  which  is 
repudiated,  is  frequently  expressed  in  a  preceding  imperative. 
The  verb  is  usually  a  repetition  of  the  preceding  (. . .  mecum  i 
potatum.  II  egone  earn  ?)  or  an  amplification  of  it  (Cure.  10  lautus 
luces  cereum.  ||  egon  apicularum  congestum  opera  non  feram  ?). 
In  a  few  cases  a  verb  of  speaking  or  keeping  silence  is  used,  or 
the  force  of  the  repudiation  is  intensified  hy  patiar^  possim  (Asin. 
810  egon  haec  patiar  aut  taceam?  emori  me  malim). 

Other  forms  of  pronouns  with  -ne  {tene  2,  eamne^  tuane^  mene^ 
uosnCy  meosne)  have  exactly  the  same  meaning,  ^nd  patiar  is  used 
in  3  of  the  7  cases. 

Parallels  to  these,  in  which  the  indicative  occurs  in  exclamatory 
repudiation  of  a  statement  of  fact,  may  be  found  in  any  play. 

Questions  with  -ne  appended  to  a  noun  are  not  idiomatic,  either 
with  the  indie,  or  with  the  subj.  Pers.  26  deisne  follows  quid  ego 
faciam  ?  and  is  half-deliberative.  In  Poen.  730  quid  tum  ?  homi- 
nemne  interrogem  . . .  ?  would  be  a  question  for  advice,  but  -ne 
is  a  conjecture  of  Ritschl,  following  A;  the  passage  is  given 
below  under  parataxis  {censen  for  quid  tiim,  with  Pall.). 

The  two  cases  with  etiamne  are  for  advice,  Rud.  1275,  1277. 

Without  a  particle,  there  are  8  cases  of  the  verb  alone  or  with 
introductory  quidf  repeating  and  repudiating  an  expressed  impv., 
as  in  Most.  579  abi  quaeso  hinc  domum.  ||  abeam  ?,  Merc.  749 
abi.  II  quid,  abeam  ? 

Four  cases  with  non  at  or  near  the  beginning  of  the  sentence 
repeat  and  repudiate  a  negative  suggestion,  Epid.  588  quor  me 
igitur  patrem  uocabas?  ||  non  patrem  ego  te  nominem,  ubi . . .  ? 

With  the  verb  at  the  beginning  of  the  sentence,  the  exclamatory 
and  rejecting  force  is  especially  clear.  Such  sentences  begin  like 
those  which  consist  of  the  verb  alone  (e.  g.  Asin.  838  an  tu  me 
tristem  putas?  ||  putem  ego  quern  uideam  esse  maestum  ...?), 
but  run  off  into  added  details.  Where  the  verb  is  not  at  the 
beginning  the  sentence  is  short  (Jibiego  demf  loricam  adducafnf)y 
because  the  exclamatory  tone  cannot  be  long  sustained.  In  three 
or  four  cases  (Ba.  903  hodie  exigam  aurum  hoc  ?  ||  exige  ac 
suspende  te,  Men.  539  dicam  curare  ?  ||  dicito,  and  Pers.  26,  after 
quid  ego  faciam  f)  the  question  is  not  repudiating.  The  first  two 
are,  of  course,  not  to  be  distinguished  from  futures ;  the  last  is 
partially  deliberative.     In  Most.  664  GS.  rightly  use  a  period. 


Digitized  by  VjOOQ IC 


144  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Most  556  quid  nunc  faciundum  censes  ?  ||  egon  quid  censeam  ?  is 
an  indirect  quotation,  and  censeam  is  not  properly  the  verb  of  the 
question. 

Amph.  813  quor  istuc,  mi  uir, .  . .  ex  te  audio?  ||  uir  ego  tuos 
sim?  is  defended  by  Ter.  Andr.  915,  Hec.  524.  It  repudiates  the 
claim  implied  in  mi  uir. 

The  one  case  with  anne.  Cist.  518,  is  an  impatient  demand  ;  the 
disjunctive  questions  (4  cases)  are  all  deliberative. 

Of  the  73  cases  of  sentence  question,  four  are  asked  for  the 
sake  of  getting  advice  or  direction,  and  four  (the  disjunctive 
questions)  are  deliberative.  Beside  these  the  only  deliberative 
questions  are  the  two  in  Pers.  26,  which  in  form  seem  to  delib- 
erate, but  in  content  (deisne  aduorser,  cum  eis  belligerem)  are 
plainly  rejecting. 

C.  Present^  \st  singular^  in  parataxis. — The  value  of  these 
examples  for  the  interpretation  of  the  subjunctive  is  so  great  that 
full  Ibts  are  given. 

a)  Depending  upon  an  impv. — Upon  sine,  Ba.  29  (24  GS.)  sine 
te  amem;  1027  sine  perlegam,  1176,  1199,  Cas.  136,  Cist.  454, 
Ep.  204,  M.  G.  1084  (jsinite).  Most.  1180,  Pers.  750,  Poen.  142, 
261,  Poen.  375  (3  cases),  Ps.  61,  239  (2  cases).    Total,  18. 

These  are  all  short  sentences,  usually  only  sine  and  the  verb. 

With  facy  Epid.  567  fac  uideam ;  Poen.  893  fac  ergo  id  *  facile* 
noscam  ego ;  the  rest  are  all  fac  sciam,  taking  the  place  of  the 
impv.  of  a  verb  meaning  *to  cause  to  know*;  Cure.  414,  617, 
Men.  890,  M.  G.  277,  Ps.  696,  Rud.  1023,  Trin.  174.  The  verbs 
are  all  of  knowing.    Total,  9. 

With  caue,  St.  37  tace  sis :  caue  sis  audiam  ego  istuc  posthac 
ex  te.    True.  942  is  a  conjecture. 

There  is  a  small  but  remarkable  group  in  which  a  paratactic 
subj.  goes  with  the  impv.  of  a  verb  of  action.  Cure.  313  uin 
aquam  ?  ||  si  frustulentast,  da,  obsecro  hercle,  obsorbeam ;  True. 
367  deme  soleas:  cedo  bibam ;  Most.  373  uigila.  ||  uigilo :  cedo 
bibam  (MSS  cedo  ui  bibam').  Cf.  Verg.  Aen.  IV  683  f.  date 
uolnera  lymphis  abluam,  where  date  is  not  equivalent  to  siniiCy  as 
Ladewig  thinks.  Most.  849  mane  sis  uideam,  and  perhaps  such 
cases  as  Cure.  427  concede  inspiciam  quid  sit  scriptum,  though 
this  is  usually  printed  with  a  colon  after  concede.  The  impv.  in 
these  uses  expresses  the  action  which  is  necessary  as  an  antecedent 
to  the  subj.  verb.  With  the  impv.,  32. 


Digitized  by  VjQOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  I45 

b")  With  an  impersonal  phrase. — optumumsiy  Asin.  448  nunc 
adeam  optumumst;  Cas.  949,  950,  Epid.  59  sed  taceam  optu- 
mumst;  Rud.  377  capillum  promittam  optumumst  occipiamque 
hariolari. 

With  necesse  est,  Poen.  1244  pro  hoc  mihi  patronus  sim 
necessest.    True.  817  is  a  conjecture. 

With  concessum,  datum,  Amph.  12  nam  uos  quidem  id  iam 
scitis  concessum  et  datum  mi  esse  ab  dis  aliis,  nuntiis  praesim  et 
lucro. 

With  decretumsi,  Poen.  501  profestos  festos  habeam  decretumst 
mihi. 

With  ceriumsiy  Asin.  248,  Aul.  681,  Ba.  382,  Capt.  779  {certa 
res),  Cas.  448.  These  are  all  verbs  of  3d  conjug.,  but  they  are 
not  quite  futures. 

Possibly  Cist.  519  non  remittam  definitumst  is  similar. 

The  single  case  of  licet  with  ist  sing.,  Asin.  718  licet  laudem 
Fortunam,  tamen  .  .  .  shows  by  the  use  of  tamen  that  licet  is  felt 
as  a  conjunction.  With  impersonals,  14. 

c)  With  indicatives. — Rud.  681  quae  uis  uim  mi  adferam  ipsa 
adigit,  Trin.  681  meam  sororem  tibi  dem  suades  sine  dote. 
Amph.  9  is  in  a  dependent  clause  in  a  long  sentence,  uti  bonis 
uos  uostrosque  omnis  nuntiis  me  adficere  uoltis,  ea  adferam,  ea 
uti  nuntiem,  and  I  have  no  doubt  that,  with  colloquial  freedom, 
the  infin.,  an  2^  clause,  and  the  paratactic  subj.  adferam  are  used 
as  parallel  constructions. 

The  effect  of  indirect  discourse,  which  will  be  felt  in  these  cases, 
will  appear  also  where  other  persons  and  numbers  are  used 
paratactically  with  an  indie,  verb.  With  indicative,  3. 

d)  Paratactic  questions. — In  quis  questions  uis  is  inserted,  Aul. 
634  redde  hue  sis.  ||  quid  tibi  uis  reddam?  (cf.  651  redde  hue.  || 
quid  reddam  ?),  Ba.  692  nunc  hoc  tibi  curandumst, . . .  ||  quid  uis 
eurem?,  Epid.  19,  584  quid  taces?  ||  quid  loquar  uis?,  Merc.  158 
quid  uis  faciam  ?,  M.  G.  300,  Most.  578,  St.  115. 

uis  stands  between  quid  and  the  verb  except  in  Epid.  584. 
The  function  of  uis  is  evidendy  to  bring  out  more  clearly  the 
inquiry  as  to  the  will  of  the  person  addressed  which  is  contained 
by  implication  in,  e.  g.,  quid  reddamf  With  uis  inserted,  8. 

In  sentence  questions,  which  are  almost  invariably  repudiating, 
uin  is  inserted  or  prefixed  in  order  to  emphasize  the  true  inter- 
rogative character  of  the  sentence. 

Capt.  360  uin  uocem  hue  ad  te  ?  ||  uoca ;  858  uin  te  faciam 
ibrtunatum?;  Asin.  647,  Cas.  272,  544,  Men.  606,  Merc.  486  (2), 


Digitized  by  VjOOQ IC 


146  AMERICAN  JOURNAL    OF  PHILOLOGY. 

721,  M.  G.  335,  1399  (but  the  text  is  not  sure),  Pers.  575,  Poen. 
439,  990i  1226,  Ps.  324,  522,  St.  397,  486,  Trin.  1092,  True.  502 
uin  adeam  ad  hominem  ?  ||  uolo,  924  (2).  In  Trin.  59  uin  conmu- 
temus  ?  ego  tuam  ducam  et  tu  meam  ?  the  force  of  uin  passes 
over  to  ducam.  In  Merc.  728  etiam  uis  nomen  dicam?  the 
question  is  introduced  by  etiam^  and  -ne  is  not  needed.  In  Capt. 
121  the  position  oi-ne  is  changed  for  emphasis  to  mene  uis  dent. 
In  Poen.  730  GS.,  following  Pall.,  have  censen  hcmiinem  interro- 
gemt    A  has  quid  turn. 

Of  these,  Men.  606  has  something  of  repudiating  force  (men 
rogas?  II  uin  hunc  rogem  ?),  but  none  is  either  an  exclamation  or 
a  deliberative  question,  uin  is  prefixed  in  order  to  exclude  the 
ordinary  meaning  of  sentence  questions  with  the  subjunctive. 

With  uiny  27. 

The  ist  person  sing,  is  not  used  with  ne. 

Present  subjunctive,  2d  person  singular. 

A.  Non-interrogative,  independent. — Wishes  with  uiinam  occur 
Men.  1 104  utinam  efBcere  quod  poUicitu's  possies,  Cist.  555  utinam 
audire  non  queas,  both  with  verbs  meaning  '  to  be  able.'  In  the 
marriage  song,  Cas.  821  uir  te  uestiat,  tu  uirum  despolies,  the 
circumstances  give  something  of  optative  force,  which  comes  out 
more  clearly  in  the  following  verses.  Trin.  351  will  be  given 
under  parataxis. 

saluos  (salud)  sis  is  used  17  times  as  a  form  of  greeting  and 
ualeas  7  times  in  parting.    The  meaning  gives  them  optative  force. 

Postponing  for  a  moment  the  hypothetical  uses  and  the  indefi- 
nite 2d  person,  there  remain  121  cases  of  the  subjunctive  express- 
ing some  kind  of  will  or  desire.  In  a  broad  sense  of  the  word 
these  might  be  called  jussive,  but  not  more  than  a  tenth  of  the 
number  are  true  commands  and  about  as  many  more  are 
demands.  Advice,  serious  or  sarcastic  or  urgent,  is  the  most 
common  kind  of  use,  not  far  from  50  cases  coming  fairly  under 
this  head.  There  are  15  or  20  requests  and  about  as  many 
expressions  of  permission.  Invitation,  challenge,  petition,  expres- 
sion of  obligation,  curse,  are  used  each  a  few  times,  and  there  is 
one  asseveration.  Most.  182.  But  where  the  form  is  unchanged  it 
is  useless  to  make  purely  functional  distinctions.'  I  prefer  to  note 
the  usage  of  certain  of  the  more  common  verbs,  accipias  (4 
cases)  is  used  only  in  requests  and  advice ;  agcLS  (2)  and  uel  aias 
uel  neges  (2)  in  a  challenging  demand ;  dicas  (6)  is  in  all  cases 


--Digitized  b^i^Ogk- 


THE  SUBJUNCTIVE  IN  PLAUTUS.  I47 

but  one  advice,  as  part  of  a  plan;  habeas  (11)  is  generally  a 
sarcastic  permission ;  ignoscas  (3)  is  a  petition ;  iubeas  (3),  advice ; 
face  as  (6)  a  command  in  all  but  one  case. 

Hypothetical  uses  are  rare,  and  in  every  case  some  preceding 
or  accompanying  phrase  gives  the  hypothetical  tone.  In  Capt. 
599  sapias  magis  is  in  answer  to  the  question  quid  st.,.  iusserim  f 
In  Rud.  1229  si  sapias,  sapias:  habeas  quod  di  dant,  habeas  is 
only  a  continuation  and  expansion  of  the  apodosis.  In  Aul.  231 
the  ubi  clause  contains  a  protasis  and  in  Asin.  180,  Trin.  554, 
quouis  and  quamuis  prepare  for  the  hypothetical  use. 

una  opera^  in  its  peculiar  Plautine  sense,  is  used  three  times 
'^\\}[i  posiules  and  twice  with  iubeas y  all  hypothetical. 

There  are  21  cases,  also,  in  which  the  subject  seems  to  be  the 
indefinite  second  person:  Aul.  506,  517,  520,  Capt.  420,  Cas.  562, 
M.  G.  94,  689,  761,  Most.  278,  Poen.  585,  831,  836,  1416,  Ps.  137, 
1 176  (?),  Trin.  671  (2  cases),  914, 1031, 1052,  1054.  Two  or  three 
of  these,  which  are  not  in  soliloquy,  might  be  questioned.  If 
they  are  not  indefinite,  they  should  be  added  to  the  list  of  hypo- 
thetical passages.  In  many  of  the  cases,  some  phrase  or  clause 
precedes  which  sets  the  hypothetical  tone.  Thus  in  Aul.  506 
quoquo  uenias,  Cas.  562  quom  aspicias,  Ps.  1176  ubi  aspicias. 
The  verbs  are  uideas  (5),  censeas  (3),  audias  (2),  nescias  (2), 
scias^  inuenias,  conspicias^  cupiaSy  uelis^  desideres^  all  of  mental 
action,  and  noceas  and  perdas.  The  last  two  are  preceded  by 
quomferias  and  by  duarum  rerum  exoriiur:  uel  perdas  . . .  uel 
• .  .  amiseris.  2d  sing,  independent,  179. 

B.  Qtusiions. — There  are  only  5  quis  questions,  and  the  small 
number  and  sporadic  character  make  a  precise  interpretation 
difficult.  Rud.  1322  quid  dare  uelis,  qui  istaec  tibi  inuestiget 
indicetque?  and  Asin.  558  edepol  uirtutes  qui  (how)  tuas  nunc 
possis  conlaudare,  sicut  ego  possim  ?  appear  to  be  hypothetical, 
but  the  use  of  the  auxiliary  verbs  as  expansions  of  quid  des^  qui 
conlaudes,  complicates  the  phrases.  Pers.  638  quid  (why)  eum 
quaeras  qui  fuit?  is  like  Rud.  1322,  i.  e.,  might  have  been 
expanded  for  greater  clearness  into  quid  quaerere  uelis,  Epid. 
693  quid  ago  ?  ||  quid  agas  ?  mos  geratur  means  '  what  should  you 
(ought  you  to)  do  ? '  as  the  mode  of  mos  geratur  shows.  M.  G. 
554  fateor.  ||  quid  ni  fateare, . . .  ?  goes  with  other  cases  of  quid 
ni,     Rud.  767  is  not  a  quis  question,  but  a  relative  qui  with  ne. 

Sentence  questions  all  repudiate  the  expressed  or  implied 
desire  of  the  person  addressed.    There  are  8  cases  with  -«^,  all 


Digitized  by  VjOOQ IC 


148  AMERICAN  JOURNAL    OF  PHILOLOGY, 

tun  or  iuin  (gen.),  one  with  an  (or  four,  if  Asin.  813  is  read  an 
iu),  and  10  without  a  particle.     Aul.  431  is  an  indirect  question. 

C.  Parataxis. — a)  With  imperative,  fac  (13),  Amph.  976  hue 
fac  adsis;  Capt.  439  fac  fidelis  sis  fideli;  Cas.  421  et  quamquam 
hoc  tibi  aegrest,  tamen  fac  accures.  ||  licet;  Cure.  521,  Merc.  498, 
M.  G.  812  (^face  follows  subj.),  1360,  Pers.  196,  198,  Poen.  1035 
{face  follows),  Ps.  236,  481. 

With/arj/^,  Asin.  238  syngraphum  facito  adferas;  Cas.  523, 
Most.  216,  Poen.  1084,  1278,  1418,  1414  leno,  tu  autem  amicam 
mihi  des  facito  aut  mihi  reddas  minam;  Trin.  485.  The  impv. 
precedes  the  subj.  in  17  of  the  20  cases. 

With  sine^  Asin.  902  sine  reuenias  modo  domum :  faxo  scias . . . 
GS.  punctuate  sine:  reuenias ^  but  cf.  Cas.  437  sine  modo  rus 
ueniat;  Most.  11  sine  modo  adueniat  senex,  with  exactly  the 
same  threatening  tone,  sine  has  lost  something  of  its  verbal 
force. 

With  uide^  Asin.  755  adde  et  scribas  uide  plane  et  probe ;  Poen. 
578  uide  sis  calleas. 

With  cauey  10  cases.  Capt.  431  caue  tu  mi  iratus  fuas;  439 
caue  fidem  fluxam  feras;  Cas.  530,  Epid.  437,  Most.  810,  1025, 
Pers.  51,  816  caue  sis  me  attigas,  ne  tibi .  .  .  malum  magnum 
dem  ;  Rud.  704.  Aul.  660  is  also  a  case  of  caue,  though  the  rest 
of  the  sentence  is  confused.  With  imperatives,  33. 

d)  With  indicatives. — uo/o,  Capt.  383  ergo  animum  aduortas 
nolo;  388,  430,  M.  G.  546,  Poen.  279, 1197,  Rud.  1414,  Trin.  372. 
uoio  follows  in  6  of  the  8  cases. 

Ttolo,  Cas.  233  ted  amo.  ||  nolo  ames;  Most.  1176  sine  ted  exo- 
rarier.  ||  nolo  ores.  ||  quaeso  hercle.  ||  nolo,  inquam,  ores.  ||  nequi- 
quam  neuis ;  Pers.  245,  Trin.  945.  nolo  precedes  the  verb  in  all 
cases. 

malOf  Ps.  209  taceo.  ||  at  taceas  malo  multo  quam  tacere  dicas. 

faxo,  Asin.  876  iam  faxo  ipsum  hominem  manufesto  opprimas ; 
Men.  113  faxo  foris  uidua  uisas  patrem;  644  faxo  scias;  Most. 
1 133  ego  ferare  faxo;  Ps.  949,  Trin.  62  ne  tu  hercle  faxo  haud 
nescias  quam  rem  egeris ;  882  faxo  scias.  In  Cure.  587  BJ  have 
faxo  reperiaSy  E  reperies ;  the  future  is  of  course  possible,  but 
the  subjunctive  is  perfectly  good.  In  Asin.  902  the  MSS  have 
faxo  ut  scias,  and  as  this  construction  is  not  infrequent,  there  is 
no  sufficient  reason  for  omitting  ut.  In  7  of  the  8  cases  faxo 
precedes. 


/Google 


THE  SUBJUNCTIVE  IN  PLAUTUS.  I49 

c)  With  uelim  diTidfaxim. — Cas.  234  enicas.  ||  uera  dicas  uelim, 
Men.  909,  Rud.  511,  1067.  Compare  especially  Poen.  1150  abeo 
igitur.  II  facias  modo  quam  memores  mauelim,  with  Ps.  209,  quoted 
above  with  male. 

With /axtm,  Amph.  511  ilia  si  sciat .  . .,  ego  faxim  ted  Amphi- 
truonem  esse  malis,  quam  louem. 

d)  With  impersonals. — ofiiumumsi,  Aul.  568  tum  tu  idem  optu- 
mumst  loces  ecferendum.  With  iicei,  Epid.  471  estne  empta  mihi 
istis  legibus?  ||  habeas  licet;  Most.  713,  Trin.  1179.  Rud.  139  is 
especially  noteworthy  because  saluos  sis  is  usually  so  distinctly 
optative:  me  periisse  praedicas.  ||  mea  quidem  hercle  causa  saluos 
sis  licet. 

e)  In  questions. — Most.  322  uisne  ego  te  ac  tu  me  amplectare  ? 
Here  the  parataxis  is  really  due  to  the  omitted  ampiectar.  With 
poHn,  Cas.  731  potin  a  med  abeas?,  Pers.  297. 

C.  Present,  2d  person  singular,  with  ne. — The  distinction 
between  the  independent  sentence  and  the  dependent  clause  is 
nowhere  more  difficult  than  in  sentences  with  ne.  No  thought  is 
really  independent  of  the  preceding  thought,  and  the  connection 
may  increase  in  closeness  until  it  is  one  of  real  dependence  with- 
out finding  expression  in  language.  It  is  only  when  language 
begins,  so  to  speak,  to  run  in  ruts,  to  form  fixed  phrases  giving 
evidence  of  dependence,  that  we  know  that  the  line  has  been 
crossed. 

Such  a  phrase  has  been  formed  in  Plautus  in  the  clauses  in 
which  ne  is  used  with  a  verb  of  mistaking,  of  thinking  wrongly. 
nefrusira  sis  occurs  7  times,  nepostuUs  6  times,  ne  censeas  twice, 
ne  erres  once,  ne  speres  twice,  and  other  forms  (jexistumes,  arbi- 
irere,  opinere,  etc.)  with  verbs  of  thinking  and  saying  {praedices, 
dicas'),  and  even  occasionally  with  other  verbs  (meiuas,  territes  (?), 
quaeras),  are  found  once  or  twice  each.  These  may  fairly  be 
excluded  as  semi-dependent,  though  they  of  course  show  some- 
thing of  prohibitive  force. 

molesius  ne  sis  (10  times)  shows  in  a  few  cases  (Asin.  469 
abscede  hinc,  molestus  ne  sis,  Aul.  458),  especially  where  it 
follows  immediately  after  another  command,  a  tendency  to 
dependence,  and  the  same  beginnings  of  a  feeling  of  purpose  may 
be  suspected  in  other  cases,  e.  g.  M.  G.  1361  i,  sequere  illos :  ne 
morere :  Pers.  318  emitte  sodes,  ne  enices  fame :  sine  ire  pastum ; 
M.  G.  1 215  moderare  animum, — ne  sis  cupidus.    But  whether 


Digitized  by  VjOOQ IC 


ISO  AMERICAN  JOURNAL    OF  PHILOLOGY. 

these  are  included  or  not,  they  would  not  change  the  result.  In 
the  use  of  the  pres.  2d  sing,  with  ne  there  is  the  same  range  of 
function  as  in  independent  uses  without  ne.  The  cases  vary  from 
sharp  and  emotional  warning  {ne  atiigas  mCy  adportum  ne  biias, 
dico  iam  iibi,  molestus  ne  sis)  to  prayers  (Amor,  amicus  miki  ne 
fuaS)  and  mild  warning  and  advice.  So  far  as  I  can  see,  the 
proportion  of  emotional  cases  is  not  smaller  than  in  the  corres- 
ponding uses  without  ne,  where  also  the  prevailing  tone  is  one  of 
advice  or  suggestion.  The  number  given  in  the  table  (61) 
includes  the  partially  dependent  cases. 

Present,  ^d person  singular. 

A.  Non-interrogative ^  independent. — Certain  forms  of  wish  are 
so  well  marked  in  the  3d  pers.  that  they  can  be  set  apart  with 
precision,  differing  in  this  from  wishes  in  ist  or  2d  pers. 

With  utinam  there  are  4  cases,  Asin.  418,  M.  G.  1009  f.,  Most. 
233,  Rud.  158,  all  general  in  content,  not  like  the  specialized 
forms  of  wish  to  be  given  below. 

With  ut,  Cas.  238  ut  te  bonus  Mercurius  perdat,  an  unusual 
kind  of  wish  with  ut.  In  Poen.  912  ualeas  beneque  ut  tibi  sit, 
two  forms  of  wish  are  put  together.  Pers.  290,  Cure.  257  have  ui 
in  an  expression  of  desire,  not  a  wish. 

The  phrase  qua£  res  bene  {male')  uortat,  quod  bonum  aique 
fortunatum  sit,  etc.,  is  found  10  times.  It  is  introductory,  as  in 
classical  Latin,  in  only  3  cases. 

Other  impersonal  forms  of  wish  are  bene  {male)  sit,  uae  tibi  sit, 
bona  pax  sit,  male  istis  euenat,  in  all  7  cases. 

The  wishes  which  contain  the  namje  of  a  god  are  especially  well 
marked.  Mars  adiuuet,  me  facial  quod  uolt  luppiter,  luppiter  te 
seruet,  Hercules  te  infelicet  (after  a  repetition  of  licet),  and 
especially  luppiter  te  perdat  {perduit)  ;  13  cases.  There  are  also 
6  cases  of  asseveration,  ita  .  . .  amet]  Most.  182  has  a  lover's 
name  instead  of  a  god's,  and  Capt.  877  f.  combines  ita  amabit 
with  ita  condecoret. 

Beside  these,  Pers.  269  uapulet  is  a  curse,  Most.  374  pater 
aduenit .  .  .  ||  ualeat  pater  is  the  3d  pers.  of  ualeas,  and  Cas.  822 
tua  uox  superet  tuomque  imperium  :  uir  te  uestiat  is  defined  as  a 
wish  by  the  fact  that  it  is  used  in  a  marriage  song.       Wishes,  45. 

The  formal  contract  read  by  the  parasite  in  Asin.  751  ff.  con- 
tains 9  cases  of  the  3d  sing.     These,  like  the  other  subjunctives 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS,  151 

in  that  passage,  express  that  kind  of  obligation  which  is  involved 
in  a  contract  and  deserve  separate  mention. 

The  remaining  cases  (except  the  hypothetical)  require  a  some- 
what careful  analysis,  because  they  imply  in  use  much  more  than 
the  verb-form  is  capable  of  expressing,  and  the  implied  but 
unexpressed  elements  lead  to  a  considerable  extension  of  the 
meaning  of  the  mode.  Nothing  in  the  verb-form  defines  in  any 
way  the  relation  of  the  hearer  to  the  will  or  to  the  action,  though 
that  relation  may  be  both  real  and  close,  and  the  variety  of 
possibilities  in  regard  to  the  subject  of  the  verb  (a  person,  a 
thing,  impers.,  etc.)  further  complicates  the  matter. 

When  the  subject  of  the  verb  is  a  definite  person,  the  relation 
of  the  hearer  to  the  action  suggests  the  following  groups  of 
usage: — a)  The  hearer  is  to  convey  the  speaker's  will  to  the 
third  person,  the  actor.  Amph.  951  euocate  hue  Sosiam  :  guber- 
natorem  .  .  •  Blepharonem  arcessat,  i.  e.  'tell  him  that  I  want  him 
to  call  Blepharo';  Poen.  905  manu  eas  adserat,  suas  popularis, 
liberali  causa.  Cf.  M.  G.  1037  adeat,  siquid  uolt.  (i.  e.  '  tell  her 
to  come  here')  ||  siquid  uis,  adi,  mulier,  where  the  wish  (permis- 
sion) is  immediately  conveyed  to  the  actor. 

3)  The  hearer  is  to  bring  about  the  performance  of  the  action 
by  the  third  person.  Most.  920  octoginta  debentur  huic  minae  ?  || 
.  • .  Ijhodie  accipiat  'see  that  he  gets  them';  M.  G.  1304  omnia 
composita  sunt  quae  donaui :  auferat  *  have  her  carry  them  off' ; 
M.  G.  1 100,  Cas.  697.  With  these  should  go  the  large  number  of 
cases  in  which  the  speaker  is  advising  the  hearer  as  to  the  way  in 
which  a  third  person  is  to  act  in  order  to  carry  out  a  plan.  Pers. 
151  sed  longe  ab  Athenis  esse  se  gnatam  autumet;  Trin.  764  if. 
scitum  consilium  inueni :  homo  conducatur  . . . :  is  homo  graphice 
exornetur  . . . :  salutem  ei  nuntiet  uerbis  patris ;  M.  G.  792,  Ps. 
753  f..  St.  299. 

c)  The  speaker  wishes  the  hearer  to  permit  the  third  person  to 
act  The  will  may  not  extend  to  the  third  person,  who  may  be 
ready  of  himself  to  do  the  act.  Merc.  989  redde  filio:  sibi 
habeat.  ||  iam,  ut  uolt,  per  me  habeat  licet ;  Merc.  991,  Pers.  447, 
Rud.  1 1 21  aliud  quidquid  ibist,  habeat  sibi. 

d^  The  speaker  expresses  his  indifference  in  regard  to  an  act 
of  the  third  person  and  implies  that  the  hearer  also  is  to  be 
indifferent.  Poen.  264  erus  nos  . .  .  mantat.  ||  maneat  pol :  mane, 
i.e.  'never  mind  if  he  is  waiting';  Ba.  224  adueniet  miles.  || 
ueniat  quando  uolt. 


Digitized  by  VjOOQ IC 


152  AMERICAN  JOURNAL    OF  PHILOLOGY. 

e)  In  some  cases — the  large  majority,  probably,  in  formal  style 
— the  second  person  is  merely  a  hearer,  the  recipient  of  the 
speaker's  confidence.  Ba.  502  ilium  exoptauit  potius?  habeat; 
Amph.  300  clare  fabulabor :  hie  auscultet  quae  loquar,  where  LG., 
ed.  crit.,  suggest  ui  hie,  unnecessarily,  I  think.  In  such  sentences 
the  expression  of  speaker's  desire  is  the  important  thing,  and  its 
effect  upon  the  action  of  the  third  person  is  of  little  consequence, 
so  that  these  sentences  may  approach  a  wish;  St.  711  modo 
nostra  hue  amica  accedat. 

As  only  definite  persons  are  involved,  the  nature  of  the  will — 
command,  advice,  permission — is  much  the  same  as  in  2d  sing. 

With  definite  actor,  48. 

Passages  in  which  the  actor  is  not  a  definite  person  fall  also  into 
several  groups,  but  the  difference  in  meaning  is  slight.  When 
the  subject  is  described  in  a  relative  clause,  the  meaning  is  almost 
the  same  as  when  the  subject  is  definite.  M.  G.  81  qui  autem 
auscultare  nolet,  exsurgat  foras;  Rud.  486;  with  quisque^  Pers. 
373  dicat  quod  quisque  uolt :  ego  non  demouebor.  In  a  number 
of  cases  the  subject  is  an  ideal  or  typical  person,  a  true  lover  (Ps. 
307  det,  det  usque :  quando  nil  sit,  simul  amare  desinat),  a  genuine 
woman,  by  the  standards  of  comedy  (M.  G.  190  qui  arguat  se, 
eum  contra  uincat  iure  iurando  suo),  an  ideal  slave  (Amph.  960 
proinde  eri  ut  sint,  ipse  item  sit :  uoltum  e  uoltu  conparet ;  Ba. 
656,  Aul.  599  f.).  In  Pers.  125  cynicum  esse  egentem  oportet 
parasitum  probe:  .  . .  pallium,  marsuppium  habeat,  the  decline  of 
the  subjunctive  in  these  cases  to  a  mere  sense  of  artistic  propriety 
is  illustrated  by  the  parallel  of  esse  oportet.  In  nearly  all  of  these 
cases  the  second  person,  if  one  is  present,  is  disregarded  and  the 
speaker  addresses  the  audience;  i.  e.  both  the  actor  and  the 
hearer  are  indefinite.  In  all,  25  cases. 

If  the  subject  is  a  thing  (which  occurs  rarely)  or  if  the  verb  is 
passive,  the  subject  is  not  the  actor. 

When  the  subject  is  a  definite  person,  the  hearer  is  usually  to 
be  the  actor,  and  the  expression  of  will  may  be  essentially  the 
same  as  in  the  cases  above  under  b  or  even  in  2d  sing.  So 
Fragm.  50  (Carbon.  II)  patibulum  ferat  per  urbem,  deinde  adfi- 
gatur  cruci ;  Trin.  767  is  homo  exornetur ;  M.  G.  1401  iamne  ego 
in  hominem  inuolo  ?  ||  immo  etiam  prius  uerberetur  fustibus ;  Capt. 
609,  M.  G.  141 8. 

The  cases  in  which  the  subject  is  a  thing  are  nearly  all  in  the 
plural,  but  beside  a  few  verbs  of  passive  meaning,  maneat,  stety 


Digitized  by  VjOOQ IC 


^smmmmmmmmm 


THE  SUBJUNCTIVE  IN  PLAUTUS.  1 53 

supersiiy  defiai,  the  phrase  cena  deiur  occurs  twice,  tnos  geratur 
4  times  zndfiat  27  times.  Of  the  last,  three  cases  have  a  subject, 
but  the  rest  are  all  in  connection  with  an  expression  of  desire,  to 
which  y^/  gives  assent.  In  these  cases  the  previous  speaker,  the 
second  person,  has  already  expressed  the  desire,  and  the  speaker 
oifiai,  who  is  really  to  be  the  actor,  selects  a  form  which  leaves 
the  person  who  wills,  the  hearer,  the  actor  and  the  nature  of  the 
act  indefinite,  so  that,  in  truth,  nothing  is  expressed  except  assent 
to  the  desire,  as  if  he  said  'your  will  is  mine.'  This  leaves yfa/  so 
weak  that  it  is  little  more  than  a  future,  and  in  i^xXfiet  is  also  used 
(Men.  186,  Merc.  302,  M.  G.  908)  in  the  same  sense,  as  mos  UH 
geretur  is  used,  Ps.  22,  for  the  subjunctive.  With  passives,  45. 
The  3d  person  singular  is  also  used,  though  not  frequently,  of 
a  supposed  case.  In  six  or  eight  passages  the  hypothetical  tone 
is  set  by  a  protasis  and  is  continued  through  the  following 
sentences.  These  passages  are  not  counted.  There  are  also  12 
cases  where  there  is  no  distinct  protasis  in  the  immediate  context. 
Ba.  139  non  par  uidetur  neque  sit  consentaneum, .  . .,  praesens 
paedagogus  una  ut  adsiet,  is  an  excellent  illustration  of  the 
meaning  of  this  use  of  the  mode,  par  uidetur  is  equal  to  sit 
consentaneum,  the  meaning  of  uideri  exactly  expressing  the 
opinion  or  view  which  in  the  second  phrase  is  expressed  by  the 
mode.  Ba.  97  ego  opsonabo :  nam  id  flagitium  meum  sit,  mea  te 
gratia  •  . .  facere  sumptum;  True.  221  stultus  sit,  qui  id  miretur; 
M.  G.  736  qui  deorum  consilia  culpet,  stultus  inscitusque  sit ;  St. 
24  ioculo  istaec  dicit:  neque  ille  sibi  mereat  Persarum  montes, 
. . .,  ut  istuc  faciat  (cf.  non  mereani) ;  M.  G.  691  hoc  numquam 
.  .  .  audias:  uerum  priusquam  galli  cantent,  .  . .,  dicat  'da,  mi 
uir';  Capt.  208  nos  fugiamus?  quo  fugiamus  ?  ||  in  patriam.  || 
apage,  baud  nos  id  deceat  fugitiuos  imitari ;  True.  907  numquam 
hoc  unum  hodie  ecficiatur  opus,  quin  opus  semper  siet ;  Ps.  432 
fors  fuat  an  istaec  dicta  sint  mendacia;  Amph.  1060  nee  me 
miserior  feminast  neque  ulla  uideatur  magis.  In  Trin.  441  hie 
postulet  frugi  esse:  nugas  postulet,  the  first  verb  is  in  sense  a 
protasis  and  the  subjunctive  is  not  hypothetical ;  the  second  verb 
might  perhaps  be  omitted  from  this  list,  as  being  influenced  by 
the  protasis.  Asin.  465  Sauream  non  noui.  ||  at  nosce  sane.  ||  sit, 
non  sit :  non  edepol  scio,  is  very  peculiar  and  perhaps  unparal- 
leled in  Plautus,  but  the  meaning  is  clear;  cf.  Capt.  964,  St.  31  ff. 
In  Most.  984  possiet  is  a  conj.  of  Cam. 


Digitized  by 


Google 


154  AMERICAN  JOURNAL    OF  PHILOLOGY. 

B.  Questions, — The  quis  questions  are  introduced  by  quis  (3), 
quid (j7^)y  qui  'how*  (2},  unde  (i),  and  are  all  of  one  pretty  well 
defined  class,  implying  impossibility.  The  only  cases  which  call 
for  notice  are  quid  hoc  sit  hominisf,  Amph.  576,769,  and  quid 
hoc  sit  negoti . .  .f,  Asin.  407  (Cam.  reads  est^  but  sit  is  defended 
by  the  other  cases).  In  these  the  relation  to  the  subjunctive  of 
desire  is  somewhat  more  apparent  than  in  some  of  the  other 
cases,  though  none  is  strictly  potential. 

The  sentence  questions  have  ne  in  3  cases,  and  are  without  a 
particle  in  8.  All  are  repudiating  exclamations,  but  all  show  the 
same  leaning  toward  the  potential  which  appears  in  the  quis 
questions,  and  non  is  found  three  times.  Rud.  728,  where  dei^ 
not  det,  is  the  correct  reading,  and  Men.  763  are  not  included. 

Questions,  20. 

C.  Parataxis, — As  the  vagueness  of  the  3d  pers.  sing,  of  the 
subjunctive  is  chiefly  in  the  undefined  relation  of  the  hearer  to 
the  will  and  the  action,  it  will  be  found  that  the  leading  verb 
serves  mainly  to  define  what  the  mode  alone  leaves  undefined. 
The  classes  below  are  arranged  in  the  same  order  as  those  above, 
under  A. 

<£)  With  iube^  indicating  that  the  hearer  is  to  convey  the 
speaker's  will  to  the  actor.  Most.  930  die  me  aduenisse  filio.  || 
...  II  curriculo  iube  in  urbem  ueniat ;  Pers.  605  iube  dum  ea  hue 
accedat  ad  me ;  Rud.  708  iube  modo  accedat  prope ;  with  a  more 
polite  addition,  Most.  680  euoca  dum  aliquem  ocius,  roga  circum- 
ducat.     Cf.  Amph.  951  euocate  Sosiam  :  .  .  .  arcessat . . . 

V)  With  fax^facitOyfacite.  The  hearer  is  to  cause  the  subject 
of  the  verb  to  act.  Rud.  12 19  et  tua  filia  facito  oret:  facile 
exorabit;  Pers.  445  facito  mulier  ad  me  transeat;  Most.  854  age 
canem  istam  a  foribus  aliquis  abducat  face.  Some  cases  with/a^ 
have  a  verb  of  passive  sense,  often  with  a  thing  for  the  subject, 
and  correspond  to  the  uses  with  passive  verbs  in  which  the  hearer 
is  the  real  actor.  So  Ps.  157  aquam  ingere :  face  plenum  ahenum 
sit  coco;  Men.  866  facitote  sonitus  ungularum  appareat;  Pers. 
438,  Men.  867,  992,  Rud.  621,  12 15. 

c)  With  sine ;  the  second  person  is  to  permit  the  third  person 
to  act.  M.  G,  1244  sine  mulier  ueniat,  quaeritet,  desideret, 
exspectet ;  Cas.  206  sine  amet,  sine  quod  lubet  faciat ;  Cist.  734, 
Ep.  36,  Ps.  478.  With  sine  modo,  Amph.  806,  Cas.  437,  Most.  11. 
In  Ps.  159  at  haec  (securis)  retunsast.  ||  sine  siet,  and  Asin.  460  ne 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  155 

duit,  si  non  uolt.    sic  sine  adstet,  the  speaker  is  indifferent,  as  in 
the  following  class.     Cf.  Poen.  264  erus  nos  mantat.  ||  maneat  pol. 

d)  With  licet;  the  speaker  is  willing  or  indifferent.  Merc.  989 
redde  filio :  sibi  habeat.  ||  iam,  irt  uolt,  per  me  habeat  licet ;  Capt. 
303  the  speaker  is  helpless. 

e)  With  uolo,  emphasizing  the  speaker's  will  and  leaving  the 
second  person  out  of  the  action.  Ps.  11 23  leno  argentum  hoc 
nolo  a  me  accipiat  atque  amittat  mulierem,  Asin.  77,  Rud.  1332, 
True.  473.  But  in  Poen.  1151  patruo  aduenienti  cena  curetur 
nolo,  as  the  verb  is  passive  with  a  thing  for  subject,  the  hearer  is 
to  be  the  actor  and  the  will  is  a  command.  Pers.  832  at  enim 
quod  ille  meruit,  tibi  id  obsit  nolo,  continues  the  curse  expressed 
in  831,  and  therefore  approaches  the  meaning  of  the  subj.  with 
uelim. 

The  cases  with  malo  (M.  G.  1333,  with  somewhat  uncertain 
text)  and  nolo  (Merc.  107,  Ps.  436,  St.  734)  call  for  no  comment. 

uelim  with  the  subjunctive  has  been  given  above.  The  passages 
are  Aul.  670,  Cas.  559,  Most.  1074,  IJoen.  1288,  Ps.  1061,  Rud. 
877,  True.  481.    With  malim  following  an  asseveration,  Poen.  289. 

With  faxOt  emphasizing  the  speaker's  determination  to  bring 
about  the  act,  Amph.  972,  Ba.  864,  Most.  68,  True.  643.  With 
faxtm  in  an  ut  clause.  True.  348.    With  faciam,  Amph.  63,  876. 

Scattering  cases  are  St.  757  si  quidem  mihi  saltandumst,  tum 
uos  date  bibat  tibicini  (cf.  da  bibani)\  Merc.  1004  nihil  opust 
resciscat;  Cure.  461  leno,  caue  mora  in  te  sit  mihi.  Capt.  961 
quod  ego  fatear,  credin  pudeat  quom  autumes  ?  is  the  only  para- 
tactic  question.  It  gives  a  good  basis  for  interpreting  M.  G.  614 
quodne  nobis  placeat,  displiceat  mihi?  and  shows  that  these 
questions  deal  with  an  opinion.  In  parataxis,  60. 

D.  With  ne, — The  contract  in  Asin.  751  ff.  contains  17  cases 
with  ne^  and  a  proclamation  in  Poen.  prol.  17  ff.  has  3  more. 
The  rest  are  nearly  all  ne  quis^  quisquam,  and  call  for  no  remark. 

With  ne,  28. 

Present^  \st  person  plural. 

A. — The  hortatory  use  is  so  well  marked  and  so  well  known 
that  nothing  need  be  said  of  it  here.  There  are  94  cases.  The 
verb  eamuSy  with  its  compounds,  is  used  42  times  (evidently 
because  of  a  recurring  dramatic  situation),  agamus  is  used  4 
times  and  other  verbs  once  or  twice  each,    utinam  is  used  only 


Digfe^  Google 


IS6  AMERICAN  JOURNAL    OF  PHILOLOGY. 

once,  Asin.  615.  The  use  of  age^  agUe^  and  of  a  vocative  {eamus^ 
mea  germancL)^  and  especially  of  tu  (True.  840  eamus  tu  in  ius), 
show  that  the  sense  of  the  2d  person  was  felt. 

B. — Questions  occur  only  3  times:  with  quo^  Capt.  208,  with 
uieTy  deliberative,  St.  696,  and  in  a  repudiating  exclamation,  Capt. 
208. 

C.  Parataxis. — ^With  uolo,  Ba.  708,  St.  670  nolo  eluamus  hodie 
peregrina  omnia.  With  censeOy  Merc.  1015  immo  dicamus  senibus 
legem  censeo.  With  suades^  Asin.  644  proinde  istud  facias  ipse, 
quod  faciamus  nobis  suades;  cf.  Trin.  681  dem  suades.  With 
aranty  Amph.  257  uelatis  manibus  orant  ignoscamus  peccatum 
suom.    There  are  2  questions  with  «/«,  St.  736,  Trin.  59. 

The  proportion  of  paratactic  to  independent  uses,  7  out  of  105^ 
is  much  smaller  than  in  other  persons. 

D. — There  is  one  case  with  ne^  Poen.  251. 

Present^  2d  person  plural. 

A. — Poen.  623  fortunati  omnes  sitis  is  a  wish;  M.  G.  1341  is  a 
petition  or  request ;  Cure.  632  quid  istuc  ad  uos  attinet  ?  quaeratis 
chlamydem  et  machaeram  banc  unde  ad  me  peruenerit,  follows 
two  requests  that  he  should  tell  where  he  got  a  certain  ring,  and 
is  ironical,  *ask  me  where  I  got  my  cloak.'  But  the  addition  of 
una  opera  (as  with  postules)  would  bring  out  clearly  the  under- 
lying sense,  'you  might  as  well  ask  me.' 

B. — There  are  no  questions. 

C.  Parataxis. — The  cases  are  so  exactly  like  those  in  the  sing, 
that  I  give  only  the  numbers,  uolo  3,  facite  3,  oro  obtestor  2, 
modo  facialis  oro  i,  caue  i. 

D. — With  ne^  13  cases,  of  which  8  are  addressed  to  the  audi- 
ence; ne  expectetis  5,  (jad^miretnini  2,  uereamini  i. 

Present^  ^id  person  plural. 

The  differences  between  3d  sing,  and  3d  plur.  are:  i)  the 
absence  of  cases  in  which  the  subject  of  the  verb  is  a  definite 
person;  2)  the  large  number  of  wishes  and  curses;  3)  the 
peculiar  paratactic  uses. 


■  uigitizea 


bTVjOOglc 


^■a 


THE  SUBJUNCTIVE  IN  PLAUTUS.  1 57 

A.  Nofi'inierrogeUive,  independent. — Wishes.  With  utinam, 
Asin.  841,  Pers.  289,  Ps.  108,  all  general  in  character.  With  ut 
and  diperdant  {perduint'),  Aul.  785,  Merc.  710,  Pers.  298,  Fragm. 
(Boeot.)  21.  Rud.  prol.  82  ualete,  ut  hostes  uostri  difBdant  sibi 
(cf.  Poen.  912  ualeas  beneque  ut  sit  tibi) ;  Cist.  202  (proL)  ualete 
et  uincite  . . .  seruate  . . .  socios, .  .  .  parite  laudem  et  lauream  : 
ut  uobis  uicti  Poeni  poenas  sufferant;  perhaps  also  Poen.  prol. 
128  ualete  atque  adiuuate:  ut  uos  seruet  Salus,  though  it  is 
usually  punctuated  adiuuate  ut.  These  seem  to  be  all  cases  of  ut 
in  a  wish,  but  I  fear  that  my  list  is  incomplete. 

Wishes  and  curses  with  di  {deaeque)  or  the  names  of  gods. 

There  are  25  cases  of  di  te  (illum^  istam)  perdant  and  5  of  di 
me  perdant)  11  of  di  ie  (istunit  illos)  perduint  and  i  with  me. 
Other  forms  are  di  te  infelicent  (5),  malum  quod  isti  di  deaeque 
duini(^i),  di  deaeque  . . .  te  . . .  excrucient.  Of  good  wishes  there 
are  dite  ament{i^),  di  ie  seruent{i)^  sospttent{i),  dibene  uartant 
(7),  di  tibi  dent  quaequomque  optes  (quae  uelis)  (10),  di  duint 
quaequamque  optes  (2),  omnia  optata  offer  ant  (1),  di  bene  (male, 
melius)  faciani  (8).  Asseverations  with  ita  are  iia  me  di  ament 
(20),  ita  me  di  seruent(i),  and  wishes  ita  difaciant  occur  4  times. 

Two  other  cases,  Cure.  575  ita  me  machaera  et  clypeus  . .  . 
bene  iuuent  and  M.  G.  1316  saluae  sient,  show  by  their  content 
and  relation  to  other  phrases  that  they  are  wishes. 

The  whole  number  of  wishes  in  3d  plur.  is  127,  of  which  118 
contain  the  names  of  gods  or  the  word  di.  With  these  utinam  is 
used  3  times  (in  wishes  of  a  general  character),  ut  4  times,  qui  7, 
quin  2,  at  6,  o  once,  ah  once,  ita  27  times. 

Beside  the  wishes  the  subjunctive  is  used  21  times  in  other 
expressions  of  will  or  desire.  Most  of  these  are  active,  and  the 
subject  is  never  inanimate  and  never  definite  individuals,  but 
always  a  class,  reges,  haruspices,  matronae,  inimici,  or,  more 
vaguely,  alii,  omnes,  or  a  class  described  in  a  qui  clause.  The 
person  addressed  is  usually  the  audience  or  a  person  present  on 
the  stage  and  treated  as  a  representative  of  the  public ;  in  a  few 
cases  it  might  be  said  that  the  third  persons  are  really  the  persons 
addressed,  indirectly  and  impersonally,  as  in  laws  (e.  g.  Poen. 
prol.  32  ff.  matronae  tacitae  spectent  etc.).  As  both  actors  and 
persons  addressed  are  thus  vaguely  conceived,  there  is  no  possi- 
bility of  cases  of  direct  command,  such  as  appear  in  3d  sing. 
The  3d  plur.  expresses  only  the  more  general  kinds  of  desire, 
amounting  usually  to  no  more  than  a  statement  of  obligation  or 


Digitized  by  VjOOQ IC 


IS8  AMERICAN  JOURNAL    OF  PHILOLOGY. 

propriety.  In  two  cases,  Asin.  671  . . .,  ni  genua  fricantur.  ||  quid 
uis  egestas  imperat:  fricentur,  Ba.  11 33  cogantur  quidem  intro, 
where  the  verb  is  passive,  the  person  addressed  is  really  the  actor, 
and  these  express  more  direct  forms  of  will,  almost  equivalent  to 
friceSi  cogamus.  So  also  in  Capt.  115,  where  uti  adserueniur 
expresses  a  command  and  is  not  a  dependent  clause. 

The  hypothetical  uses  are  found  chiefly  in  four  long  passages, 
Amph.  155  ff.,  Aul.  228  ff*.,  Merc.  407  ff.  (10  verbs),  Rud.  978  ff., 
where  a  protasis  has  preceded  or  has  been  plainly  implied.  In 
M.  G.  1369  f.  the  protasis  is  implied  in  caue  isiuc  feceris  \  in  Asin. 
602  qui  sese  parere  adparent  huius  legibus  suggests  a  protasis  ; 
in  Trin.  703, 740,  743  the  implication  is  plain  before  the  subjunctive 
verb  is  reached. 

Counting  all  the  verbs  there  are  23  cases,  occurring  in  8  passages. 

B. — The  3d  plural  is  used  only  twice  in  questions.  Poen.  860 
is  a  repetition  of  a  preceding  di  ament;  Ps.  205  is  a  repudiating 
exclamation. 

C.  Parataxis. — The  cases  are  arranged  as  in  3d  sing. 

a)  iuhe,  Men.  956  tu  seruos  iube  hunc  ad  me  ferant ;  St.  396  i 
intro  . .  . :  iube  famulos  rem  diuinam  mi  apparent  (cf.  for  ace. 
some  cases  of  ueniat  uelim). 

b)  With  faCy  facito,  faciiCy  Aul.  402  (BDJ,  but  Non.  has  a 
different  tradition,  which  GS.  follow) ;  Aul.  407  facite  totae  plateae 
pateant ;  Cas.  521  fac  uacent  aedes ;  Cas.  527  fac  habeant  linguam 
tuae  aedes.  liquid  ita?||quom  ueniam,  uocent;  Ps.  166,  181, 
Fragm.  70  (Cornic.  VII).  With  uide^  Amph.  629  uide  ex  naui 
efferantur. 

c)  With  sincy  Ba.  1134  sic  sine  adstent. 
d^  With  licet  there  are  no  cases. 

e)  With  uolOy  Pers.  293  eueniant  uolo  tibi  quae  optas;  nolOy 
True.  585  uasa  nolo  auferant ;  with  faxoy  Amph.  589,  Men.  540  ; 
with yaa'am,  M.  G.  1399  uin  faciam  quasi  puero  in  collo  pendeant 
crepundia?  (notice  the  double  parataxis);  with/a^r/w,  Aul.  495, 
Merc.  829,  Pers.  73,  Trin.  221,  222. 

Thus  far,  with  three  or  four  exceptions,  the  verbs  are  passive  in 
meaning  or  in  form  and  the  subjects  are  things.  In  the  indepen- 
dent uses,  on  the  contrary,  the  verbs  are  nearly  all  active  and  the 
subjects  are  classes  of  persons.  The  paratactic  uses  in  3d  plur., 
therefore,  are  not  parallel  to  the  independent  uses,  extending 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS,  1 59 

them  along  the  same  lines,  but  supplementary,  expressing  ideas 
which  the  independent  uses  express  rarely  or  not  at  all.  Plautus 
did  not  ^ayfac  sese  domi  caniineant,fac  mairariae  tacitae  specteniy 
because  the  subjects  of  the  verbs  were  both  the  persons  addressed, 
though  indirectly,  and  the  actors ;  nor  did  he  say  uaceni  aedes  or 
ex  naui  efferarUur  qutie  imperaui  omnia^  without  fac  or  uide^ 
because  such- phrases  would  omit  the  actor  entirely. 

A  few  other  scattering  cases  are  found.  Ps.  938  si  exoptem, 
quantum  dignus,  tantum  dent;  Capt.  694  nil  interdo  dicant  (or 
tnierdico  aiani,  Fl.  GS.);  Ps.  207  prohibet  faciant  (a  gloss); 
Poen.  prol.  22  decet . . .  stent . . .  temperent ;  Ba.  1033  caue  . .  . 
fuant.  Amph.  632  utinam  di  faxint,  infecta  dicta  re  eueniant  tua 
is  a  paratactic  wish.  St.  31  ff.  ipsi  interea  uiuant,  ualeant,  ubi 
sint,  quid  agant,  ecquid  agant,  neque  participant  nos  neque 
redeunt  presents  two  curious  indirect  questions,  uiuant^  ualeant^ 
depending  paratactically  upon  participant  without  an  interrogative 
particle. 

D. — With  fu  there  are  only  three  cases,  Poen.  prol.  23,  29,  38, 
entirely  like  the  other  independent  uses  in  this  passage. 

Imperfect,  1st  person  singular, 

A. — In  independent  non-interrogative  uses  only  wishes  with 
utinam  and  expressions  of  desire  with  uel/em,  mauellem  occur. 

With  utinam,  Amph.  575  utinam  ita  (i.  e.  ebrius)  essem ;  Rud. 
533  utinam  fortuna  nunc  hie  anetina  uterer.  Both  express  a 
present  wish  contrary  to  the  fact,  impossible  of  fulfilment. 

uelUm  is  used  9  times: — with  infin.,  Asin.  589,  Poen.  681,  Cist. 
93 ;  with  perf.  ptc.  Cist.  506 ;  with  adj.,  Most.  980,  Ps.  309 ;  with 
obj.,  St.  713;  with  parat.  subj.,  Poen.  1066  patrem  atque  matrem 
uiuerent  uellem  tibi.  ||an  mortui  sunt  ?  ||  factum ;  St.  312  nimis 
uellem  haec  fores  erum  fugissent.  Of  these  cases,  Poen.  681 
uidere  equidem  uos  uellem,  quom  huic  aurum  darem  expresses  a 
desire  in  the  past,  but  still  felt  by  the  speaker,  in  regard  to  a 
future  act  (cf.  682  illinc  procul  nos  istuc  inspectabimus).  The 
rest  are  all  Uke  uelim  in  that  they  express  a  wish,  but  the  impli- 
cation that  it  will  not  be  fulfilled  is  clear  in  some  cases  (Ps.  309, 
Poen.  1066,  Cist.  506)  and  possible  in  all. 

mauellem  (jncUlem)  is  used  8  times: — with  infin.,  Amph.  512, 
Ba.  198, 452,  Ps.  1057 ;  with  perf.  infin.  act.,  Cure.  512 ;  with  infin. 
implied,  Ba.  1201,  Ps.  131;  with  parat.  subj.,  Ba.  1047  ^^  i^^ 


Digitized  by  VjOOQ IC 


l6o  AMERICAN  JOURNAL    OF  PHILOLOGY. 

edepol  Ephesi  multo  mauellem  foret, . . .,  quam  reuenisset  domum. 
The  sense  of  desire  or  will  is  present  in  all ;  some  express  a  curse 
(Ba.  198,  Ps.  1057,  131);  in  a  few  cases  (Cure.  512,  Ps.  131,  Ba. 
1047)  the  suggestion  of  non-fulfilment  is  quite  distinct. 

B.  Questions. — ^There  are  two  quis  questions,  both  referring  to 
a  past  obligation  or  desire  on  the  part  of  the  second  person. 
Cist.  94  I  should  take  as  a  dependent  clause. 

Sentence  questions  are  mainly  rejections  of  an  expression  or 
implication  of  obligation  in  the  past.  In  Trin.  177  pcUerer  is 
balanced  by  indicare  me  aeqtcomfuii  in  the  preceding  verse.  In 
Most.  183  ita  ego  istam  amarem?  is  an  exclamatory  repetition  of 
ita  Philolaches  tuos  te  amet  in  182.  The  negative  in  Most.  455 
is  nan, 

C. — In  parataxis  there  is  only  one  case,  St.  177  hoc  nomen 
repperi  eo  quia  paupertas  fecit  ridiculus  forem. 

Itnperfeciy  2d  singular. 

A. — ^With  uHnam  there  is  one  case,  Rud.  494  f.  utinam  ...  in 
Sicilia  perbiteres,  with  distinct  past  reference. 

In  one  case  the  subjunctive  is  plainly  hypothetical ;  Men.  160 
edepol  ne  tu,  ut  ego  opinor,  esses  agitator  probus.  The  reference 
to  the  past  is  certainly  not  clear,  but  cf.  Merc.  125  nimis  nili  tibicen 
siem,  where  the  future  reference  is  apparent. 

In  all  other  independent  2d  sing,  cases  (13)  the  subjunctive 
expresses  an  obligation  which  the  actor  should  have  felt  in  the 
past.  Merc.  633  ff.  quid  ego  facerem  ?  || . . .  men  rogas  ?  requae- 
reres,  rogitares;  Merc.  637,  Poen.  387,  391,  Rud.  842  quin  occi- 
disti  extemplo  ?  ||  gladius  non  erat.  ||  caperes  aut  fiistem  aut  lapi- 
dem;  Trin.  133  ff.  (4  verbs);  Pers.  710  animus  iam  in  nauist 
mihi.  II  eras  ires  potius,  hodie  hie  cenares  deserves  special  mention, 
because  it  expresses  a  past  obligation  in  regard  to  a  future  action ; 
*  You  should  have  made  up  your  mind  (then)  to  go  to-morrow 
(not  to-day).'  Ba.  432  . . .  ubi  reuenisses  domum, ...  in  sella 
apud  magistrum  adsideres  is  in  a  passage  describing  old  customs  ; 
the  other  verbs  are  in  the  impf.  indie,  and  the  subj.  adsideres  is 
due  to  the  ubi  clause.  Ps.  494  should  be  printed  with  a  period, 
as  in  GS.,  otherwise  there  is  no  justification  for  the  past  tense. 

B.  Questions. — With  quis^  Merc.  884  quo  nunc  ibas  ?  ||  exula- 
tum.  II  quid  ibi  faceres?    Goetz,  ed.  crit.,  supplies  «/,  but  the  text 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  l6l 

as  it  stands  (GS.)  gives  the  proper  sense,  'What  were  you 
intending  to  do  there?' 

There  are  no  sentence  questions.  Capt.  713  is  a  continuation 
of  a  conditional  sentence.     For  Ps.  494  see  above. 

C  Parataxis, — Asin.  503  si  esses  percunctatus  . . .,  scio  pol 
crederes  is  not  counted,  and  Ba.  635,  if  nan  is  dropped  from  the 
text,  is  exacdy  similar.  The  only  clear  case  is  St.  624  ueni 
(impv.).  II  hucine?  ||  immo  in  carcerem.  ||  quid  igitur?  ||  dixi  equi- 
dem  in  carcerem  ires.  This  is  really  a  quotation,  dixi  *  in  carcerem 
t,'  remarkable  for  the  use  of  dixi  instead  of  iussi, 

D. — With  ne  there  are  3  cases,  expressing  past  obligation :  Ps. 
437,  Ba.  29,  30  (16  f.  in  ed.  crit.). 

Imperfect^  3^  singular, 

A. — With  uiinam^  Merc.  823  utinam  lex  esset  eadem,  quae 
uxorist,  uiro.    There  are  2  cases  in  Rud.  379  f.  of  past  obligation. 

There  are  3  passages  where  the  mode  appears  to  be  hypo- 
thetical, Cas.  910,  Rud.  1262  (2),  Ba.  314.  In  the  last,  nimio  hie 
priuatim  seruaretur  rectius,  a  slight  sense  of  obligation  is  produced 
by  recHus. 

B. — The  one  quis  question,  Rud.  379,  is  a  question  in  regard 
to  a  past  obligation. 

The  sentence  questions  are  all  associated  wiih  a  protasis  in  the 
context,  though  Ps.  288  is  a  repudiating  exclamation.  The  others 
are  Trin.  178,  Capt.  714,  Trin.  954. 

C — In  parataxis,  mauellem  is  used  with  foret^  Ba.  1047,  in  a 
wish.  Trin.  115  si  inimicus  esset,  credo  baud  crederet  is  not 
counted.  The  rest  are  all  of  the  nature  of  indirect  quotations. 
Ba.  551  ille,  quod  in  se  fuit,  accuratum  habuit,  quod  posset  mali 
faceret  in  me,  inconciliaret  copias  omnis  meas.  The  MSS  have 
ineonciliare,  which  in  such  colloquial  style  is  not  impossible; 
faceret  expresses  a  past  intention,  quoted  by  accuratum  habuit, 
Pers.  634  tactus  lenost,  qui  rogarat  [rogabat?],  ubi  nata  esset, 
diceret;  that  is,  the  leno  asked  {dic^  ubi  tu  nata'sf  Trin.  591 
tandem  impetraui  abiret  is  as  if  for  uolui  (iussi)  abiret  et  tandem 
impetraui ;  Merc.  536  f,  inter  nos  coniurauimus, . . .,  neuter  stupri 
causa  caput  limaret;  in  Epid.  316  me  iussit  senex  conducere 
aliquam  fidicinam  . . . :  dum  rem  diuinam  faceret,  cantaret  sibi, 


Digitized  by  VjOOQ IC 


1 62  AMERICAN  JOURNAL    OF  PHILOLOGY, 

there  are  various  conjectures,  but  I  take  cantarei  to  be  a  part  of 
the  ordtT,  Jidicinam  conducas:  ea  mihi  caniety  as  often  in  pres. 
tense. 

D. — The  only  case  oine  has  a  protasis. 

There  are  no  cases  of  ist  or  2d  plur. 

Imperfecta  yi plural. 

A. — With  utinam,  Capt.  537  utinam  te  di  prius  perderent, 
quam  periisti  e  patria  tua,  with  distinct  reference  to  the  past. 
Trin.  1028  f.  may  perhaps  also  refer  to  the  past,  but  is  better 
taken  as  an  ordinary  unfulfilled  wish. 

The  other  cases  all  occur  in  passages  where  the  speaker  is 
giving  the  details  of  a  plan  made  in  the  past.  In  M.  G.  731  f. 
itidem  diuos  dispertisse  uitam  humanam  aequom  fuit :  . . .,  uitam 
ei  longinquam  darent :  .  . .,  is  adimerent  animam  cito,  and  in 
Epid.  386  aequom  fuit  clearly  defines  the  meaning  of  the  mode. 
Poen.  1 1 39  C  hodie  earum  mutarentur  nomina,  facerentque  . .  . 
quaestum  corpore  is  part  of  a  plan  made  by  the  leno^  not  by  the 
speaker,  and  would  be  introduced  paratactically  by  lenoni  decre- 
tumsi  or  some  similar  phrase. 

There  are  no  hypothetical  or  interrogative  uses. 

C. — In  parataxis,  Poen.  1066  depends  upon  uellem ;  Merc.  prol. 
52  (48  G.)  pater  clamitare  .  .  .  et  praedicere  (histor.  infin.),  omnes 
timerent  m^ituitanti  credere ;  M.  G.  54  at  peditastelli  quia  erant, 
siui  uiuerent. 

.  The  temporal  force  of  the  imperfect  is  plain  in  those  uses  which 
most  nearly  resemble  the  direct  expressions  of  will  in  the  present ; 
all  expressions  of  obligation  refer  to  the  past,  even  when  the  act 
to  be  performed  is  still  in  the  future.  But  in  wishes  the  shift  of 
temporal  force,  by  which  unfulfilled  conditions  and  wishes  in  the 
present  take  an  imperfect  subjunctive,  had  already  begun.  Cases 
have  been  noted  above  where  the  reference  to  the  past  seemed 
most  distinct.    See  also  Blase,  Geschichte  des  Irrealis,  pp.  3-5. 

Perfect,  ist  person  singular. 

A. — Asin.  491  praefiscini  hoc  nunc  dixerim.  Cramer,*  pp.  47 
ff.,  doubts  the  genuineness  of  the  passage,  but  calls  it  a  subjunc- 

^F.  Cramer,  de  perfecti  coniunctiui  usa  potential!  ap.  prise,  script.  Lat. 
Marburg,  1886. 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS,  1 63 

tive  of  will,  correctly,  translating  **  Dies  wiinsche  ich  jetzt  unbe- 
rufen  zu  sagen/'     Cf.  the  like  uses  ofdtcam,  haud  dolo  dicam^  etc. 
True.  349  follows  a  si  clause  in  344  ff.  and  is  hypothetical. 

B. — In  quis  questions,  Amph.  748  ubi  ego  audiuerim  ?  repeats 
and  rejects  audiuisiin.  There  are  3  cases  of  nauerim.  Cure.  423, 
Men.  299,  M.  G.  923,  which  are  classed  here,  though  naui  is 
present  in  sense. 

In  True.  625  there  is  a  repudiating  repetition  of  a  subjunctive. 

There  are  beside  these  few  cases  many  perfects  ist  sing,  which 
are  called  potential  (v.  Brix,  n.  on  Capt.  309),  but  they  are  all  in 
subordinate  clauses  or  in  conditional  sentences. 

Perfect^  2d  singular. 

The  functions  of  perf.  subj.  and  fut.  perf.  indie,  differ  so  slightly 
in  independent  uses,  in  any  person  except  the  ist  sing.,  that  one 
set  of  forms  suffices  for  both.  The  confusion  which  might  be 
expected  from  this  is,  however,  much  less  than  that  between  fut. 
indie,  and  pres.  subjunctive. 

A. — In  independent  expressions  of  will  memineris  is  used  twice 
(M.  G.  807,  Pers.  856)  and  naueris  once  (True.  163  dum  uiuit, 
hominem  noueris :  ubi  mortuost,  quiescat),  both  really  present  in 
sense.  Ba.  840  meretricemne  esse  censes  ?  ||  quippini  ?  ||  frustra's.  || 
quis  igitur  opsecrost  ?  ||  inueneris,  i.e.  *find  out  for  yourself.* 
Trin.  1053  duarum  rerum  exoritur  optio:  uel  illud  quod  credi- 
deris  perdas  uel  ilium  amicum  amiseris  is  exactly  like  a  number 
of  cases  mentioned  in  the  pres.  tense. 

Other  cases  have  more  of  future  meaning.  They  are  Capt. 
1028  (see  Brix,  note).  Cure.  665,  Trin.  760,  Most.  1152  (see 
Lorenz',  note),  Trin»  61.  In  Capt.  1028  and  Cure.  665  there  is  a 
hypothetical  tone,  though  it  is  not  distinct  enough  to  find  expres- 
sion in  a  clause;  Trin.  61  is  influenced  by  the  preceding  ya;i:^ 
dederis.    Most.  1152  and  Trin.  760  are  the  most  distinct  futures. 

B. — There  is  only  one  question,  a  repudiating  exclamation, 
Amph.  818  tun  mecum  fueris? 

C. — In  parataxis,ya;r<?  dederis  (Trin.  60)  and  faxo  haud  come- 
deris  (Men.  521)  lie  between  the  subjunctive  and  ihe  clearly 
future  uses.  So  also  memineris  faciio  St.  47,  though  it  is  more 
closely  related  to  the  independent  uses  oi  memineris. 

The  remaining .  cases  are  all  prohibitions,  either  with  caue  or 
with  ne. 


Digitized  by  VjOOQ IC 


l64  AMERICAN  JOURNAL    OF  PHILOLOGY, 

The  verbs  used  with  caue  ^x^  feceris  (Cas.  332,  M.  G.  1368, 
Poen.  1023,  St.  285,  Trin.  513),  dixeris  (Pers.  389,  Trin.  555), 
sit^eris,  siris  (Ba.  402,  Epid.  400,  Most.  401),  praeuorieris  (Merc. 
113),  reitulerh  (Epid.  439),  sumpseris  (Cist.  300), /w^rw  (Aul. 
618),  iniramiseris  (Aul.  90),  respanderis  (Amph.  608). 

D. — With  ne  and  its  compounds  the  verbs  ds^  fueris  (Asin. 
839,  Epid.  595),  dixeris  (Cist,  no,  Merc.  ^02),  feceris  (Epid.  148, 
Men.  415),  parseris  (Pers.  572,  Poen.  993),  aitigeris  (Pers.  793), 
induxeris  (Trin.  704),  desiiteris  (Trin.  1012),  osienderis  (Rud. 
ii55)»  inierueneris  (?  M.  G.  1333). 

Other  forms  of  prohibition  are  nilmonueris  Cure.  384,  minume 
feceris  Most.  272.  In  Rud.  1135,  Pers.  395,  the  futures  in  the 
context  seem  to  show  that  nullum  osienderis^  nullum  acceperis 
are  also  future. 

Perfect^  ^d  singular. 

A. — There  are  two  cases  with  utinam^  Cas.  398  f.,  Poen.  799, 
both  referring  to  the  past,  and  one,  Trin.  753  nam  certo  scio, 
locum  quoque  ilium  omnem,  ubi  situst,  comederit,  which  is 
hypothetical. 

B. — There  are  two  quis  questions,  M.  G.  925,  Trin.  1050. 

C. — In  paratiELxis:  with  uelim^  Ba.  334,  Poen.  1206;  with/ia^r^, 
Aul.  578,  Poen.  346  (either  might  be  called  fut.  perf.)  and  Capt. 
801  (so  far  as  the  corrupt  text  makes  a  judgment  possible);  with 
caue^  Men.  994  caue  quisquam  nostrum  .  .  .  fecerit,  really  a  second 
person.  The  text  of  Cure.  27,  M.  G.  926,  True.  429  is  too  uncer- 
tain for  use. 

There  are  no  cases  of  the  ist  person  plural. 

Perfect,  2d  plural. 

The  only  case  is  M.  G.  862  ne  •  . .  dixeritis,  addressed  to  the 
audience. 

Perfect,  2td  plural. 

Aul.  542  meminerint  is  an  expression  of  propriety,  like  those*  in 
pres.  3d  plur.;  St.  385  perierint  is  a  curse,  and  both  verbs  are  in 
reality  present  in  meaning.     Poen.  617  is  a  future  perfect. 

In  parataxis  there  are  two  cases  with  uelim,  Poen.  570,  Rud. 
662,  both  curses. 

ne  di  sirint  (siuerini^  is  used  in  Ba.  468,  Merc.  613,  Merc.  323. 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS,  165 

Of  the  forms  with  perfect  ptc,  Amph.  979  fac  conmentus  sis  is 
deponent  and  perfect.  The  rest  are  all  presents:  paraius  sis 
Trin.  1189,  occlusal  sint  Asin.  759,  curata  fac  sint  Aul.  273, 
Amph.  g^i yfaciio  opsanaium  sit  Ba.  (^6y/acite  deducius  siei  Capt. 
'J2l^j/ace  occlusal  sieni  Most.  ^00,  fac  sit  delaium  Ps.  190.  They 
do  not  differ  in  use  from  active  forms. 

Pluperfect. 

In  1st  sing.,  Rud.  497  f.  utinam  cubuissem.  In  2d  sing.,  utinam 
parsisses  True.  375;  fuisses  (conj.),  hypothetical,  M.  G.  11 12. 
In  3d  sing.,  Cas.  996,  Poen.  1252  are  useless.  M.  G.  721  si  ei 
forte  fuisset  febris,  censerem  emori:  cecidissetue  ebrius  aut  de 
equo  uspiam,  metuerem  ne  ibi  diffregisset  crura,  is  a  protasis 
without  si,  under  the  influence  of  the  preceding  si  (cf.  Pseud.  863). 
Epid.  628  is  hypothetical.  In  3d  plur.  Amph.  386  is  a  wish  with 
utinam  and  St.  312  with  uellem. 

Subjunctive  forms  in  -s-. 
1st  person  singular, 

haud  (non)  ausim,  Aul.  474,  Ba.  1056,  Poen.  1358;  all  of 
hypothetical  statement. 

haud  negassim,  Asin.  503. 

non  empsim,  Cas.  347,  M.  G.  316;  also  hypothetical. 

ausim  in  questions,  Merc.  154  egon  ausim  tibi  usquam  quic- 
quam  facinus  falsum  proloqui ;  Most.  923  f.  egone  te  ioculo  modo 
ausim  dicto  aut  facto  fallere?  ||  egone  aps  te  ausim  non  cauere, 
.  .  .?;  Poen,  149.  These  are,  as  the  form  of  question  shows, 
repudiating  exclamations,  and  ausim  is  an  insertion  for  fuller 
expression;  in  such  a  phrase  as  egon  tibi  usquam  quicquam 
facinus  falsum  proloquar  ?  the  idea  of  wishing  or  desiring  (the 
proper  sense  of  audeo)  is  latent.  Cf.  the  similar  insertion  of 
patiar  and  of  uis,  uin.  But  Merc.  301  sed  ausimne  ego  tibi 
eloqui  fideliter  ?  is  a  true  question,  answered  by  audacter. 

faxim  is  used  in  male  faxim  lubens  Poen.  1091,  1093,  2i"d  is 
hypothetical,  though  no  protasis  is  found  in  the  context.  In  all 
other  cases  oi  faxim  a  protasis  is  expressed  in  the  context  (Trin. 
221,  cf.  217-20)  or  in  the  sentence  itself  (Amph.  511,  Aul.  494, 
Merc.  814  G.  (826  GS.),  Pers.  73) ;  these  cases  are  included  in  the 
list  for  completeness,  though  strictly  they  would  be  excluded  by 
the  protasis.  In  True.  63'  the  text  is  somewhat  uncertain,  though 
faxim  is  sure.    True.  892  hostissim  is  a  conjecture  and  of  no  value. 


Digitized  by  VjOOQ IC 


l66  AMERICAN  JOURNAL    OF  PHILOLOGY. 

All  cases  of  forms  in  -siin  are,  it  will  be  noticed,  hypothetical. 

2d  person  singular. 

All  forms  are  in  prohibitions,  either  with  ne^  neqtu,  numquamy 
nilj  or  in  parataxis  with  caue.  They  are  as  follows :  Asin.  839  ne 
dixiSy  Aul.  744  ne  dixis,  Capt.  149  numquam  dixis  neque  animum 
tnduxtSy  Men.  611  ne  comessis  (Bx.;  comesses  MSS  GS.),  M.  G. 
283  ne  dixiSy  1007  nil  amassis,  Most  526  nil  curassis,  1097  ne 
occupassiSj  11 15  ne  faxis^  Poen.  553  ne  curassis,  Ps.  79  ne  parsis^ 
232  nil  curassis,  Rud.  1028  neque  indicassis,  St.  149  neque 
celassiSy  Trin.  627  neque  occuliassis,  True.  606  ne  resp<msis» 

With  caue^  Asin.  256  /axis,  467  supplicassis,  625  faxis,  Aul. 
608  indicassisy  Ba.  910  par  sis  ^  1188  amissis,  Cas.  404  obiexis, 
Merc.  484  rfiorw,  M.  G.  11 25, 1245,  iyj2faxis,  Most.  523  respexis, 
SoS  /axiSy  True.  943 /axis,  Vidul.  83  ^i^rw,  91  demuiassis.  In 
Rud.  982  a2M/j  (Seyff.  Sch.)  would  have  no  precise  parallel ; 
Sonnenschein  reads  ausu's.  Most.  518  is  a  conjecture.  M.  G. 
669  optassis  is  conjectural,  as  apodosis  to  a  si  clause,  and  is  not 
counted.^ 

2^dpers<m. 

haud  ausit,  M.  G.  11,  is  hypothetical  and  exactly  like  haud 
ausim.    In  Ba.  697  non  ausit  has  an  expressed  protasis. 

All  other  cases,  sing,  and  plur.,  are  in  wishes  or  asseverations. 
Capt  622  Ha  rex  deorum  /axit,  Cist.  742  at  uos  Saius  seruassii, 
Most  398  Ha /axil  luppiter,  Ps.  14  luppiier  prokibessii,  923  iia 
/axil  luppiter,  Pers.  330/tfr^w/ajj//. 

In  the  plur.  all  are  with  names  of  gods  or  di,  diui.  With 
uiinam,  Aul.  50  adaxini,  Amph.  632  /axint  Also  di  ie  {ilium) 
/axini.  Most.  463,  Vidul.  86 ;  di  ie  seruassint,  Asin.  654,  Cas.  324, 
Ps.  37,  Trin.  384;  iia  di  me  seruassini,  St  505;  iia  di /axini, 
Aul.  149,  257,  788,  Capt  172,  Pers.  652,  Poen.  909,  911;  di 
/axini,  Cist.  151 ;  di  meliora  (melius) /axini,  Poen.  1400,  Ps.  315, 
Merc.  285;  me  amassini,  Cure.  578.  There  is  a  protasis  with 
Cist  523. 

^Of  the  34  cases  of  prohibition  with  perfect  forms  (19  with  caue,  15  with  ne 
or  compounds),  all  but  one,  Cure.  384  nil  monueris,  are  of  the  3d  conjugation. 
Of  the  33  prohibitions  with  sigmatic  aorist  forms,  10  are  of  the  ist  conjugation. 
In  other  words,  the  perfect  in  -ui  is  used  in  prohibitions  only  4  times  (nwnueris, 
siueris,  siris  2)  out  of  67,  though  verbs  which  make  their  perfects  in  -m  are 
used  14  times.  This  can  scarcely  be  accidental;  it  must  indicate  some 
relation  between  the  aoristic  forms  and  the  prohibitory  use. 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS. 


167 


It  should  be  noted  that  of  the  forms  in  -j-  all  in  the  ist  pers. 
are  hypothetical,  all  in  the  2d  pers.  are  prohibitions,  and  all  but 
one  in  3d  pers.  are  optative. 

Present, 


Singuidr 

Plural, 

xst. 

ad. 

3d. 

xst. 

ad. 

3d. 

A.  Of  will. 

17 

148 

172 

94 

3 

148 

582 

uelim^ 

74 

74 

Hypothetical, 

17 

31 

12 

0 

0 

23 

83 

B.  Questions, 

212 

24 

20 

3 

0 

2 

261 

C.  ParaUctic, 

84 

69 

60 

7 

10 

30 

260 

D.  With  ne. 

0 

61 

28 

I 

13 

3 

106 

404 

333 

292 

ImperfecL 

105 

"26 

206 

1366 

A. 

2 

14 

3 

7 

26 

uelUm, 

17 

17 

Hypothetical. 

0 

I 

4 

0 

5 

B. 

8 

I 

6 

0 

15 

C. 

I 

I 

6 

3 

II 

D. 

0 

3 

0 

0 

3 

28 

20 

19 
Perfect. 

0 

0 

10 

77 

A. 

I 

5 

2 

2 

10 

Hypothetical. 

I 

0 

I 

0 

2 

B. 

5 

I 

2 

0 

8 

C. 

0 

19 

6 

2 

27 

D. 

0 

15 

0 

0 

I 

3 

19 

7 

.    40 

II 
Pluperfect. 

0 

I 

7 

66 

A. 

I 

I 

I 

2 

5 

Hypothetical. 

0 

I 

I 

0 

2 

B. 

0 

C. 

0 

D. 

0 

I 

2 

2 

0 

0 

2 

7 

With  perf.  ptc. 

9 

16 

Forms  in  -s- 

■. 

A. 

0 

6 

21 

27 

i 

Hypothetical. 

14 

I 

0 

15 

> 

B. 

4 

'     i^ 

/^ 

C. 

0 

16 

Jr 

D. 

0 

16 

/l 

18  • 

32 

7 

0 

0 

J 

/" 

w 

78 
1603 

i 

E.  P.: 

^fORRIS. 

Digitized  by  VjOOQ IC 


II.— TEXTUAL  NOTES  AND  QUERIES  ON  PLAUTUS.^ 

I.   THE  MOSTELLARIA. 

Argiim.,  vs.  5 :  et  inde  tprimum  6migratum  etc. 

So  the  editio  minor.  I  would  read  JUium  for  primum ;  -ilium 
and  -MUM  in  a  capital  manuscript  are  rather  close  in  ductus.  The 
liability  of  P  and  F  to  confusion  is  shown  by  vs.  151,  where  all  the 
manuscripts  read  filia  for  pila.  For  this  confusion  we  must 
suppose  the  P  to  be  only  a  little  defective  in  its  curve.  The  error 
may,  however,  be  of  semi-uncial  origin,  arising  from  a  confusion 
of  F  with  one  of  the  ligatures /^r  (Most.,  vs.  320  W),pro  (ib.  95), 
pr(a)e  (Rud.  947  B).  A  ligature  for  pri  I  have  not  observed 
in  any  Plautus  manuscript,  but  we  have/rwttw  iot primum  at  vs. 
397  (D). 
vs.  5 :  exi  inquam  fnidore  cupinam  quid  lates  ? 

So  the  editio  minor.  The  best  correction  of  this  verse  is 
Pylades*s  nidor  e  culina^  quid  lates ^  but,  after  all,  nidor-e-culina 
is  a  questionable  epithet  I  suggest  that  the  line  stood  as  follows 
in*P: 

EXIINQUAMEXI  JaUDIO  JhEMCUPINNAMQUIDLATES. 

Now,  supposing  this  archetypal  manuscript  to  have  had  the 
K-form  of  H  (c£  also  Lindsay,  The  *  Palatine*  Text  of  Plautus, 
p.  18),  the  mistake  of  AUDIOKE  for  NIDORE  is  accounted  for 
by  their  similar  ductus.  The  greatest  difficulty  for  my  reconsti- 
tuted line  comes  from  the  omission  of  the  blank  spaces,  left,  we 

*  In  the  following  study  reference  will  be  made  to  each  separate  editor  of 
the  Triumvirate  edition.  The  practical  consensus  of  all  the  Palatine  manu- 
scripts will  be  spoken  of  as  the  reading  of  *P.  References  to  Leo's  text  of 
Plautus  will  simply  run:  Leo  reads,  etc.;  Klotz's  Grundzttge  altrdmischer 
Metrik  will  be  abbreviated  to  Klotz.  All  other  citations  will  be  full  enough 
to  indicate  their  source  at  once.  The  manuscript  variants  and  emendations 
of  previous  scholars  that  come  up  for  discussion  are  derived  from  the  critical 
apparatus  of  the  Triumvirate  or  from  Leo's  edition.  The  Teubner  text 
edition  of  Goetz  and  Schoell  I  will  call  the  editio  minor.  Its  readings  for  any 
passage  will  be  generally  understood  as  the  practical  consensus  of  the  manu- 
scripts extant  for  that  passage. 


Digitized  by  VjOOQ IC 


TEXTUAL  NOTES  AND  QUERIES  ON  PLAUTUS,       169 

may  presume,  in  *P  as  in  A,  for  the  subsequent  insertion  of  the 
notae  personanim  by  the  rubricator.  Still,  the  reverse  process  is 
the  more  difficult ;  that  is  to  say,  to  insert  a  nota  personae  in  one 
of  the  minuscule  manuscripts  is  easier  than  to  drop  one. 

For  the  repetition  oiexi  I  compare  Aul.  40  exi  inquam  age  exi, 
and  Cure.  275  Heus  Phaedrome  exi  exi  exi  inquam  ocius.  A 
parallel  to  audio  is  Miles  217  vigila  inquam,  expergiscere  inquam ; 
lucet  hoc  inquam  |  Xatidio.  Jviden  hostis  tibi  adesse  ?  For  hem 
introducing  the  question  cupin  I  cite  Asin.  445  non  etiam  ?  Jhem 
non?  There  is  a  possible  play  between  audio  'Well,  Tm  not 
deaf  and  haud  eo  *I  won't  come.'  Thus  hem  cupin  means  *You 
won't,  eh?'  I  note  that  cupio'l  will'  answers  the  question  Vin 
*  Will  you?'  three  times  in  Poen.  159  sq. 

We  might  read  our  line 

exi  inquam  <exi>  nid6re — <em>  cupin?  etc., 

in  which  case  cupin  implies  a  refusal  from  within,  audible  to  the 
questioner  but  not  to  the  audience.  I  prefer,  however,  either  exi 
inquam  <exi>  XduA\6  t^e<m>  ciipin  etc.,  or  exi  <6xi>  inquam 
taiidi6  etc.  In  either  case  there  is  hiatus  with  change  of  persons 
at  audio, 

vs.  6:  quid  tibi  malum  hie  ante  aedis  clam<it>atiost. 

Here  <ii>  was  inserted  by  Acidalius,  following  Camerarius. 
I  propose  to  mend  the  line  by  reading  malum  <me>  hie  etc. 
A  precise  parallel  is  vs.  34  quid  tibi  malum  me  .  .  .  curatiost  ?  I 
compare  Cas.  91  Quid  tu  malum  me  sequere? 

The  statistics  of  the  expletive  malum  render  this  suggestion 
certain,  I  would  say,  as  elision  of  either  syllable  of  malum,  cannot 
be  proved :  i)  Before  consonants  malum,  occurs  13  times  with  the 
metrical  value  ma  \  lilfky  where  ^lum  is  7  times  in  the  second  thesis 
of  a  troch.  sept  and  i  time  (Most.  368)  in  the  fourth  thesis  (the 
second  after  the  beginning  of  a  new  speech).  In  the  senarius 
^liiWi  forms  a  second  thesis  2  times  (Cas.  91,  Most.  34)  and  a 
fourth  2  times  (Cas.  472,  Rud.  492);  in  the  iamb.  sept.  (Rud. 
945  ?)  ^fourth  thesis  i  time ;  quid  hoc  mdlum  makes  a  proceleus- 
matic  I  time  in  an  anap.  dim.  catal.  (Ps.  242  b).  Malum  is  a 
pyrrhic  arsis  in  the  second  anap.  of  the  clausula  Reiziana  i  time 
(Aul.  429) ;  it  forms  the  first  arsis  at  Bacch.  696,  and  the  fifth 
(with  change  of  speaker  at  the  fifth  thesis)  at  Stich.  597,  both 
verses  being  troch.  sept. 

^  By  the  makron  over  a  consonant  I  indicate  length  by  position. 


Digitized  by  VjOOQ IC 


I/O  AMERICAN  JOURNAL    OF  PHILOLOGY. 

2)  Before  vowels  we  might  read  mdl&m  without  elision  in  \he/irsi 
arsis  of  troch.  sept  at  Am  ph.  626,  Cas.  262,  Merc.  184,  True.  801, 
where  the  pyrrhic  mdlam  is  on  the  same  footing  as  at  Bacch.  696, 
when  a  consonant  follows;  in  the  fifth  arsis  at  Epid.  710,  Pseud. 
1 165  (with  change  of  person  as  at  Stich.  597),  and  in  Men.  793 
(without  change  of  person) ;  in  the  third  arsis  (with  change  of 
person  on  the  third  thesis)  at  Mil.  446.  At  Poen.  261  quid  hie 
malum  may  be  read  as  a  proceleusmatic  with  malum  in  the  first 
arsis  (cf.  Pseud.  242  b).  The  only  occurrence  of  malum  in  arsis 
before  a  vowel  outside  of  troch.  sept  is  Ps.  1295,  an  anap.  sept. 
Here  the  3d  and  4th  feet  are  composed  of  the  words  quid  tu 
malum  in  os,  and  even  here  there  is  only  quasi-elision,  for  -um 
is  merged  into  the  nasal  syllable  in.  Such  quasi-elision  we  have 
even  with  the  monosyllabic  vae  in  the  phrase  vae  aetati  tuae 
(Capt.  885,  Stich.  594).  Thus  malum  is  followed  4  times  by  im 
(in)  and  i  time  by  ham-.  The  vowel  e  follows  malum  2  times 
(Merc.  184,  True.  801),  a  (in  cLstas)  2  times  (Mil.  446,  Poen.  261) 
and  i  i  time  (Pseud.  1165). 

Still  another  element  comes  into  consideration :  quid  tu  malum 
me  occurs  at  Cas.  91  (iamb,  sen.),  Most.  368  (troch.  sept.),  and 
quid  tibi  malum  me  at  Most  34  (iamb,  sen.) ;  at  Rud.  945  we 
have  quid  tu  malum  nam,  me.  The  phrase  quid  tu  malum  with- 
out a  following  me  is  found  only  at  Aul.  429  (claus.  Reiz.)  and 
Ps.  1295  (anap.  sept),  and  both  these  are  rare  measures.  This 
constitutes  a  further  ground  for  reading  quid  tibi  malum  <me'> 
in  our  verse  (iamb.  sen.).  My  restitution  is  much  better  founded 
palaeographically  than  the  restoration  of  <//>,  now  accepted, 
and  constitutes  besides  a  lectio  difficilior;  clamaiio  and  clam- 
<it>atio  are  alike  nonce-words. 

vs.  13:  frutex.  This  word  seems  to  occur  in  the  literal  sense 
of 'stump'  in  Suet  Vesp.  5  quercus  singulos  repente  ramos  a 
frutice  dedit.  Typical,  however,  for  the  class  frutex  were  vioUu^ 
rosae^  arundines  (Col.  Arb.  I  2),  and  'stump*  seems  to  me  an 
impossible  definition.  It  is  a  small  change  to  alter  afrutice  to 
ABRADICE,  and  even  easier  to  suppose  that  ramos  was  glossed  by 
fruiices  'shoots'  or  vice  versa.  For  our  present  passage,  seeing 
^^Xfruiex  is  parried  in  vs.  15  by  urbanus  scurra,  I  propose  to 
interpret  it  as  'green-horn':  cf.  rudis  'rod  switch'  and  rudis 
'green'  for  the  same  metaphor.  Columella  is  cited  iox  frutex 
olerum  (cf.  Lewis  and  Short,  Lat.  Diet.,  s.  v.).     I  compare  the 


Digitized  by  VjOOQ IC 


TEXTUAL  NOTES  AND  QUERIES  ON  PLAUTUS.       17I 

French  chou  (Molifere,  L'fetourdi,  I  xi  fin.)  and  our  'cabbage- 
head/  used  figuratively  for  *  dunce,'  etc. 

vs.  21 :  corrumpe  erilem  adulescentem  optumum. 

Leo  retains  this,  the  manuscript  reading,  admitting  two  hiatuses. 
We  must,  it  seems  to  me,  admit  the  hiatus  after  adulescentem^  for 
it  recurs  in  the  same  metrical  position  at  vs.  84  {ddulescinie 
dptumo)  and  at  Capt.  169  (jidulesdniem  Aleum).  Bentley's 
observation  that  Roman  comedy  showed  a  tendency  to  make  the 
metrical  ictus  correspond  with  the  prose  accent  may  be  aptly 
illustrated  for  Plautus  fi-om  dd&lis\cenSj  57  times  in  the  nomina- 
tive, but  oblique  cases,  say  dd&lis  \  ciniem,  43  times.  There  are 
15  cases  of  the  verse-ending  ddules\cens  and  4'  cases  of  the  same 
accentuation  in  initial  position.  Thus  the  deviations  from  the 
prose-accent  in  the  nominative  seem  to  be  restricted  to  definite 
metrical  positions.  The  accentuation  adulis\ceniem  is  found  but 
6  times ;  at  Asin.  833  and  Capt.  169  the  ictus  may  be  made  to 
coincide  with  the  word-accent  by  admitting  hiatus  before  proper 
names  (cf.  Klotz,  p.  109).  At  Trin.  771  the  ictus  will  coincide 
with  the  word-accent  by  admitting  hiatus  in  the  semiquinaria  of 
the  senarius.  This  leaves  only  three  offending  passages:  Rud. 
664,  where  adulis\cenie  is  initial  in  the  senarius;  no  mitigating 
circumstance  occurs  to  me  for  Rud.  11 97  and  True.  99.  With 
this  state  of  things,  it  seems  to  me  venturesome  to  disturb  the 
coincidence  of  ictus  and  accent  in  this  word  in  order  to  banish  a 
hiatus. 

I  do  not  believe,  however,  that  Plautus  could  have  written 
erilis  adidescens  for  'master's  young  son.'  I  propose,  therefore, 
to  read  corrumpe  eri  <fi>li<u>m^  comparing  vs.  27: 

ut  eri  sui  corrumpat  et  rem  et  filium. 

The  corruption  of  ri<,fi>  It  was  approximately  haplographic. 

vss.  38-9 :  quam  confidenter  loquitur  fue  Jat  te  luppiter 

dique  omnes  perdant  oboluisti  alium. 

So  the  manuscripts,  but  the  modern  editors  generally  transpose 

fue  to  a  position  before  oboluisti,  where  Goetz  and  Schoell  read 

<fufae>  and  Leo  /«•    The  metre  of  vs.  39  is  good  as  the 

manuscripts  read  it,  with  semi-hiatus  in  the  5th  arsis.    In  vs.  38  I 

propose  to  read  loquitur  hii,  etc.     For  the  change  from /ut  (i.  e. 

*  Men.  loas,  1066,  Pers.  660,  Trin.  968. 


Digitized  by  VjOOQ IC 


172  AMERICAN  JOURNAL    OF  PHILOLOGY. 

phuil)  to  huiYiQ  have  to  reckon  with  a  confusion  of  native  Latin 
hui  and  Greekish  phy.  I  note  Terence,  Ad.  411-12,  where  hui 
and  phy^TQ  alternating  exclamations  of  astonishment.  The  Latin 
grammarians  also  state  the  fact  of  this  relationship  (cf.  Diomedes 
in  Keirs  Grammatici  Latini.  I,  p.  419,  10).  If  hui  be  read,  its 
tone  is  the  same  as  in  Pers.  801  Hui  babae  basilice  te  intulisti  etc. 
The  elision  of  hui  ought  not  to  be  more  surprising  than  of  O  /  (jd, 
Richter  in  Studemund*s  Studien,  I,  p.  598),  or  we  may  operate 
with  semi-hiatus  as  with  ei/  (ib.,  p.  469). 

We  ought  perhaps  to  read /f^/  here  and  also  at  Pseud.  1295. 
For  the  latter  line  A  seems  to  read  di  te  ament  Pseudole 
Jhae  Ji  in  malam  crucem.  In  B  there  are  no  breaks,  but  p/ui 
stands  for  the  hae  of  »A,  where  the  /  is,  I  surmise,  derived  from 
a  P  of  the  rubric  of  the  manuscript  prior  to  B,  which  may  have 
indicated  the  character  of  Pseudolus  by  a  P  as  B  does.  At  Cas. 
727  *P  almost  certainly  had  FYFY,  and  A  seems  to  read  edepoL 
Thus,  in  two  cases  the  scribe  of  A  seems  to  have  been  baffled  by 
FY,  FVI,  possibly  because  <^v  was  written  in  Greek  characters  in 
his  exemplar,  though  Greek  script  occurs  in  A ;  or  perhaps  the 
scribes  were  at  a  loss  how  to  register  an  onomatopoetic  word 
amounting  to  a  stage-direction. 

It  is  perfectly  clear  at  Pseud.  1295  that  whatever  Pseudolus 
may  have  said,  what  he  did  was  to  retch,  and  at  Cas.  727  the 
interjection  simulated  vomiting.  The  outburst  in  vs.  39  of  our 
text  seems  hardly  justified  by  quam  confidenier  loquitur,  but  if  to 
this  be  added  fy,  which  we  may  render  by  the  stage-direction 
'pretending  to  vomit,'  the  outburst  is  accounted  for.  Here  by 
fy  we  must  understand  *  your  talk  is  sickening,'  putting  it  on  a 
footing  W\ih  fy/yfoeiet  iuos  mihi  sermo  of  Cas.  727.  In  all  three 
of  the  passages  under  discussion  fy  (Jyfy')  has  no  necessary 
metrical  value,  and  may  be  merely  equivalent  to  a  stage- direction, 
*  retching.'  * 

'  This  suggestion,  though  original  with  myself,  proves  to  be  not  altogether 
new.  See  Spengel,  Reformvorschlage,  etc.,  p.  80,  note.  Along  with  Spengel 
and  Ussing,  I  accept  it,  spite  of  the  words  *'  insulsam  Gruteri  explicationem 
recoquentes"  flung  by  Schoell  at  these  two  scholars  (Appendix  to  Most.,  p. 
153).  The  English  interjection  of  disgust  '\%  foh!,  and  in  Congreve's  Old 
Bachelor,  IV iv,  the  stage>direction  'spits'  accompanies  fohl  I  do  not  know 
that  there  are  any  stage-directions  in  Latin  comedy,  but,  considering  the 
divergence  of  A's  Hae  and  *P*s  pfui  at  Pseud.  1295,  one  is  tempted  to  see  in 
pfyi  pfui  (so,  in  fact,  C  and  D),  a  possible  mistake  for  spui<,t>  *he  spits,* 
taken  up  from  a  marginal  stage-direction  or  from  a  gloss. 


Digitized  by  VjOOQ IC 


TEXTUAL  NOTES  AND  QUERIES  ON  PLAUTUS,       173 

VS.  40:  g^ermana  inluvies  rusticus  hircus  hara  sui<s>. 

So  the  manuscripts,  but  riis  \  iicus  offends  the  metrical  law  that 
two  shorts  ending  a  polysyllabic  word  may  form  neither  arsis  nor 
thesis  in  iambic-trochaic  metres.  Bergk's  correction  of  rusticus 
to  siircus  is  not  a  bad  emendation,  but  rucius  seems  to  me  better, 
as  a  lectio  difficilior,  and  fadges  precisely  with  the  interpretation 
offered  just  above  for^  (vs.  38).  I  note  the  use  of  rucius  in  the 
similar  situation  of  Pseud.  1294-1301. 

If  Leo's  contention  that  final  s  is  treated  like  final  m  in  the  pre- 
classic  period  holds,  then  we  might  scan  without  change  rustic^ 
hircus, 

vs.  56 :  stimulis,  si  hue  reveniat  senex. 

So  the  manuscripts.  Leo  reads  stimulis  Kcarnufices^^  and 
Schoell,  much  more  plausibly,  si  hue  <re  bene  gestd>  etc.  V 
propose  stimulis  si  h<,oc  eveniit>  ur  reveniat  senex ^  where  the 
assumed  haplography  seems  to  me  better  warranted  than  in 
Schoell's  emendation.  For  the  construction  eveniet  ut  I  compare 
Pers.  535. 

vs.  63 :  data  es  inonestis  etc. 

So  B.  Here  Leo,  and  before  him  Ritschl,  read  most  plausibly 
date  si  nan  estis  (i.  e.  editis).  The  other  minuscule  manuscripts 
read  inhonestis.  I  take  my  cue  for  the  emendation  of  the  passage 
from  vs.  604-5  • 

daturin  estis  faenus  actutum  mihi  ? 
da/2^r  faenus  mihi  ? 

So  A,  but  in  *P  da/^  mihi  faenus.  where  the  variant  reading 
doubtless  comes  from  a  ligature  for  -/«r.  At  True.  247  da/ar  is 
unquestionably  the  reading,  but  *P  has  da/«r.  Combining  these 
fects,  I  propose  to  read  here  da/^r  es  in<h>o-nestis  *you  are 
generous  enough  to  your  wicked  associates.'  For  this  sense  of 
dator  es  compare  True.  244  sq.  Thus  there  is  no  difficulty  in  my 
reading  except  the  change  of  number,  and  this  is  not  more  violent 
than  the  change  in  vs.  603  below'. 

vs.  73 :  venire  quod  moleste  quam  illud  quod  cupide  petas. 

So  the  manuscripts.  Much  nearer  to  the  MSS  than  anything 
yet  proposed  is  veni\re  quod  \  moleste  <est>  \  quam  ilhid  qtiod  \ 
cupide  I  petas.  For  mdlisfisi  I  refer  to  Klotz  (p.  82) ;  ^w"*"  lllud 
qu{fd~ seems  warranted  by  the  trochee  non  illuc  (Epid.  715).  See 
also  below  on  vs.  204. 


Digitized  by  VjOOQ IC 


174  AMERICAN  JOURNAL    OF  PHILOLOGY. 

VS.  84 :  vide6|  corruptum  ex  idulescente  6ptumo. 

Thus  I  would  read  the  verse  with  the  manuscripts,  allowing^ 
hiatus  in  the  5th  foot  (see  above  on  vs.  21). 

vs.  104:  sibi  quisque  simile  suo  is  sua  sumptu  operam  parcunt 
suam. 

This  is  B's  reading.  It  makes  sense  and  good  metre  to  read 
the  verse  as  follows:  sibi  quisque  simile <s>  suo  ^^5^  sumptu; 
<n6> OPERAM  parciint  suam;  or,  as  CD  read  sumptu^  we  may 
represent  the  archetypal  reading  as  sumptunonoperam,  and  the 
error  consists  only  of  haplography  of  nono.  The  reading  suo 
issua  arose  from  suo  esse  suo,  in  which  the  repetition  of  suo 
represents  a  shift  in  position  such  as  we  have  in  vs.  235  quidem 
absumpta  quidem  (BCD),  vs.  311  cum  amica  cum  (BCD),  vs.  529 
ui\\A  hodie  «/ (BCD). 

vs.  119:  <id>  dfcere  ut  hominis  a6dium  esse  similis  arbitr6mini. 

The  insertion  I  propose  (taking  id^  of  course,  in  the  sense  oiid 
quod  dixi  and  defining  dicere  by  'explain')  makes  the  verse 
metrical.  Leo  scans  it  dicire  ui  hdmines^  etc.,  with  hiatus  between 
-re  and  ut\  As  I  read  the  verse  the  2d  foot  is  a  proceleusmatic 
and  is  not  objectionable  on  the  score  of  metre  (cf.  Klotz,  p.  353). 

vs.  124 :  et  tit  in  |  ustim  |  boni  et  in  spe|ciem. 

I  propose  to  read  this  verse  as  an  iambic  dimeter,  like  vs. 
902  b.  The  question  is  whether  bon*  et  in  spe-  can  be  read  as  a 
proceleusmatic  The  shortening  of  in  here  is  on  precisely  the 
same  basis  as  the  shortening  of  a  in  Asin.  59,  where  there  is  a 
tribrach  it  d  m*  in-  (cf.  Klotz,  69),  but  in  our  passage,  to  be  sure, 
in  would  get  its  shortening  from  the  word-accent  on  et^  not  from 
an  ictus  there,  as  in  quid dni\'  (Capt.  206). 

vss.  129-30:  ad  legionem  tcomita  adminiculum  eis  danunt, 
ftum  iam  iliquem  cognattim  suom. 

If  we  read  comita<,nt>,  vs.  129  becomes  an  iambic  senarius, 
and  the  sense  is  '  the  fathers  accompany  them ;  they  give  them 
help  (money  ?)  and  also  (?)  one  of  their  relatives  (as  a  com- 
panion).' It  is  doubtful,  however,  whether  tum  iam  may  be 
taken  in  the  sense  of  etiam^  though  turn  does  approximate  to  et 
(cf.  Lewis  and  Short,  Lat.  Diet.,  s.  v.  tum,  C  i). 


Digitized  by  VjOOQ IC 


TEXTUAL  NOTES  AND  QUERIES  ON  PLAUTUS.       17$ 

VS.  139  sq.:  ha6c  verectindiam  mi  6t  virtutis  modum 
d^turbavit  ftexit  det6xitque  a  me  flico 
p6stilla  optfgere  ea[m]  n6glegens  fui. 
This  is  the  reading  of  the  editio  minor.  Now  it  happens  at  vs. 
583  of  our  play  that  A  reads  domum  and  *P  modo^  and  at  vs.  432 
one  of  two  successive  lines  ending  in  modo  has  been  corrected  to 
domum ;  and  I  propose  in  vs.  139  domum  for  modum^  and  would 
retain  earn  in  vs.  141.  There  has  been  an  elaborate  comparison 
up  to  this  point  between  a  man's  character  and  a  house,  and  our 
verbs  are  mosUy  literally  used  in  connection  with  a  house.  We 
must,  however,  take  vs.  162  sq.  into  the  count :  modestiam  omnem 
deiexit^  tectus  qua  fui;  and  the  same  figure  seems  to  recur  at 
Trin.  317 :  sarta  tecta  tua  praecepta  usque  habui  mea  modestia. 
On  the  other  hand,  modesHam  might  be  the  occasion  of  domum 
having  been  changed  to  modum  in  our  verse.  This  proposal 
demands  verecundtae,  perhaps,  in  vs.  139,  and  domus  verecun- 
diae  is  hardly  a  bolder  figure  than  ciedes  aurium  (Pseud.  469). 
In  vs.  140  I  suggest  ex<,ci'>U  for  iexity  with  the  orthography 
exHii\  cf.  Cure.  295,  where  B's  extiam  is  probably  for  exciam 
(so  Leo).  As  to  definition,  exciii  is  an  intensive  ^i^r^  =  'has 
shaken  up  thoroughly.*  Metrically,  vs.  140  will  be,  with  my 
reading,  a  trochaic  septenarius,  like  vs.  145 ;  and  vs.  141  a  cretic 
dimeter + a  trochaic  tripody  catal. 

vs.  146:  atque  6depol  ita  ha6c  tigna  timid6  <ex>p(itent:  n6n 
uide5<r>  mihf. 

I  suggest  <ex>puient  on  the  basis  of  exputescunt  at  Cure. 
242,  and  scan  the  verse  as  an  iambic  octonarius  like  the  next 
verse.  The  hiatus  at  the  end  of  the  4th  foot  is  normal.  The 
close  syntactical  connection  of  umide  and  exputeni  does  not 
hinder  hiatus  (cf.  Klotz,  p.  147).  We  should  possibly  write 
<e>puient  like  epoto^  though  in  Plautus  manuscripts  only 
expoto  is  preserved. 

vs.  159:  eventus  rebus  omnibus  vdut  homo  messis  magna 
fuit. 

I  am  inclined  to  ask,  recalling  the  steriles  omi  of  Vergil 
(Georg.  II  III),  whether  omo  did  not  stand  here  originally;  to 
be  sure,  homo  would  be  the  lectio  difficilior,  but  might  have  crept 
in  from  some  grammarian's  handling  of  the  text.  Nonius,  s.  v. 
(121.  7),  cites  Lucilius,  but  does  not  cite  Plautus.    I  cannot  find. 


Digitized  by  VjOOQ IC 


176  AMERICAN  JOURNAL    OF  PHILOLOGY, 

however,  that  the  orntts  was  proverbially  disappointing,  the  sense 
in  which  it  would  stand  here,  for  yielding  flowers  and  the  promise 
of  fruit,  but  no  fruit.  If  we  read  omo  it  suggests  ornate^  and  the 
sentiment  is  '  all  your  beautifying  will  come  to  nothing  in  the  end.' 

vs.  200:  nihilo  ego  quam  nunc  tu  amata  sum  atque  uni  modo 
gessi  morem. 

Exception  has  been  taken  to  the  construction  here ;  the  minus 
that  seems  to  be  lacking  cannot  be  supplied  without  hurt  to  the 
metre.  The  editio  minor,  however,  admits  the  construction. 
Perhaps  we  should  read  NIMIO  or  NIMIU  for  NIKILO,  sup- 
posing H  to  have  had  the  K-form  as  in  A,  and  then  change 
quam  nunc  tu  to  quanrum  iu<m>,  but  the  admission  of  e^o 
between  nimium  quantum  I  cannot  support  by  a  parallel.  If  we 
read  nimium  ego,  quam.  nunc  tu,  amata  sum,  the  tam  correlative 
to  quam  is  to  seek. 

vss.  204-5:   [solam]  illi  me[o]  soli  censeo  esse  oportere  obse- 
quentem, 
solam  ille  me  soli  sibi  suo  <sumptu>  liberavit. 

So  the  editio  minor,  which,  however,  inverts  the  order  of  the 
lines  without  a  cogent  reason,  so  far  as  I  can  see.  The  motive 
for  clipping  solam  from  the  text  is  precisely  counter  to  that  for 
clipping  '0  from  meo,  for  me  stands  precisely  below  it.  Schoell 
questions  solam  as  follows:  "Philolachis  erat,  non  Philematii 
censere  *  solam*  illam  esse  oportere  obsequentem,"  but  solam  in 
the  next  verse  is  exposed  nominibus  mutatis  to  the  same  objec- 
tion. I  do  not  see  why  our  verse  does  not  mean  *  I  ought  to  be 
solely  devoted  to  P.  only.' 

The  assumption  that  me  has  been  corrupted  to  meo  seems  to 
me  to  move  on  the  lines  of  greatest  resistance,  for  the  omission  of 
the  personal  pronoun  with  oportere  is  common  in  Terence 
(Heaut.  200,  247,  Ad.  214). 

According  to  Klotz  (p.  62  sq.),  the  cretic  may  stand  wherever 
the  dactyl  may  stand  in  iambic- trochaic  measures.  Thus  sdt"^ 
llli  is  an  allowable  ist  foot. 

The  further  question  arises  whether  till  may  not  have  the 
iambic  shortening  in  this  foot,  and  so  be  equivalent  to  a  pyrrhic  ? 
Above,  in  vs.  73,  lllad  seemed  to  be  a  pyrrhic.  Further  cases 
are  illdm  (Merc.  380)  and  llliifh  (Trin.  792).  As  for  Jf/A,  the 
motive  for  shortening  may  be  derived  from  lllius  (Bacch.  494  et 


Digitized  by  VjOOQ IC 


'     TEXTUAL  NOTES  AND  QUERIES  ON  PLAUTUS,       1 77 

a].).  All  of  this  is  called  in  question  by  Mueller  (Plautinische 
Prosodie,  p.  337  sq.). 

For  vs.  205  I  propose,  not  <sumpiu>y  but  PhilolacheSj  scan- 
ning 

solam  ills  me  soli  sibi  su6  <Phil6laches>  liberdvit. 

For  suum  in  the  sense  of  property — especially  with  sibi — I  note 
Trin.  156  reddam  suom  sibi.  In  the  previous  verse  meo  is 
euphemistic,  like  quod  suom  esse  nolit  in  vs.  247 ;  we  may  also 
compaire  pecuiium  in  its  bad  sense  (vs.  253,  cf.  Pers.  192). 

vs.  213.  For  the  unmetrical  vlillena  I  propose  viti<^i'>-lena 
from  an  earlier  viii<iu>-lenay  just  such  a  compound  as  sociu- 
fraude  (Pseud.  362). 

vs.  241 :  edepol  si  summo  loui  fbo  arg6nto  sacruficassem. 

This  is  the  reading  of  the  editio  minor,  following  B.  Here  D 
reads  uiuo  and  C  reads  ioui  for  the  bo  of  B.  The  confusion  of  B 
and  V  is  common  even  in  A,  and  B's  ioui  bo  may  well  be  haplo- 
graphic  for  ioui  uiuo,  though  the  variant  in  C  renders  this  less 
probable.  I  believe  that  in  *P  we  must  assume  a  text  Ioui  uiuo 
argenio  uiuo,  with  the  adjective  repeated,  a  not  infrequent 
phenomenon,  and  I  would  emend  the  line  to  read : 

edepol  si  stimmo  Ioui  arg^nto  uiuo  sdcruficdssem. 

Here  there  would  be  semi-hiatus  with  liul  in  the  thesis.  The  verse 
is  broken  by  the  seminovenaria  caesura,  and  so  a  spondaic  4th 
foot  is  permissible.  For  the  sense  of  argento  uiuo  I  cite  Cicero's 
commercial  phrase,  de  uiuo  detrahere  (resecare)  aliquid  *  to  take 
something  from  the  capital.'  Plautus  is  here  probably  playing 
on  caput  in  the  next  verse,  by  way  of  double  entendre  between 
the  senses  'person  capital.' 

vss.  274-5 :  nam  istae  ueteres,  quae  se  unguentis  unctitant,  inter- 
poles 
vetulae  edentulae. 
B  first  read  isles  ueleres,  and  further  on  spells  uelule  edenluU, 
with  e  for  ae,  as  commonly.     I  propose  to  read  islae  suelae 
(assuelae  ?  Asin.  217, 887)  [resl,  defining  suelae  by  *  experts.'    The 
reading  ueteres  was  due  to  a  false  division  (as  in  B)  or  to  a  gloss. 

vss.  284-5  •  •  •  •  is  nequid  emat,  nisi  quod  tibi  placere  censeat 

Schoell  has  corrected  tibi  to  sibi  and  is  followed  by  Leo  and 
the  editio  minor.     I  cannot  follow  them ;  vss.  287-9  ^^^^  us  that 


Digitized  by  VjOOQ IC 


178  AMERICAN  JOURNAL    OF  PHILOLOGY. 

*  beauty  unadorned  's  adorned  the  most'  in  the  lover's  eyes,  and 
surely  the  lover  will  not  here  be  commanded  to  buy  adornments 
to  please  himself. 

vs.  301.  The  oprobarier  of  the  MSS  has  been  corrected  to 
opprobrarier,  but  we  know  the  Romans  found  difficulty  with  a 
succession  of  r's,  and  reduced  percrebruit^  say,  to  percrebuU. 
Here  we  should  read  opprobarier,  if  we  would  reproduce  Plautus's 
probable  pronunciation.  So  also  at  Pers.  193  *P  spells  opprobari 
(but  cf.  True.  280  opprobras  A,  where,  however,  there  are  only 
two  r's). 

vss.  313-14:  aduorsum  uenire  mihi  ad  Philolachem 
uolo  temper!. 

The  editors  change  uenire  to  ueniri,  which  is  not  necessary  for 
the  metre  nor  for  the  syntax  either,  if  we  may  suppose  that  uolo 
'order'  lapsed  over  to  the  construction  of  iubeo^  where  the 
standing  ellipsis  of  the  natural  subject  of  the  dependent  infinitive 
amounts  to  using  the  active  form  as  a  passive  (c£  the  author  in 
Am.  Jour.  Phil.  XV  221).  I  note  the  close  association  of  uolo 
and  iubeo  in  the  legal  formula  uelUis  iubeaHs. 

vs.  319 :  Ecquid  tibi  uideor  mam-ma-madere. 

Here  *P  seems  to  have  read  Hecqutd,  and  at  vs.  339  the 
drunken  Callidamates  says  Eecquis^  (B)  or  Hecquis  (CD).  The 
orthography  seems  to  me  possibly  to  reproduce  the  hiccough  and 
stammer  of  a  drunken  man,  while  the  ho-ho-hocellus  of  vs.  325 
is,  I  believe,  certainly  explicable  in  this  way.  Forms  of  ecquis 
with  initial  h-  are  rather  common  in  MSS  of  this  play,  though  in 
some  of  the  cases  they  have  been  corrected :  vs.  900  hecquis 
ecquis  hue,  vs.  445  heus  hecquis  hie  est,  vs.  907  haecquid  placent  ,* 
but  ecquis  is  found  in  vss.  354,  899,  988. 

Taking  the  statistics  of  ecquis  and  ecquid  for  all  the  rest  of 
Plautus.  there  is  no  h-  in  67  cases.  At  Bacch.  580  D  reads 
hecquis  4  times,  but  also  hostium  (for  ostium)^  and  the  h-  possibly 
has  some  mimetic  intention  here;  at  Men.  163  hecquid  (B) 
possibly  represents  the  ze/^iffing  of  odors ;  and  at  Mil.  993  h-  may 
be  due  to  ze^Aispering ;  {heus^  hecquis  hie  is  found  at  Men.  674  (B), 

^  I  note  an  interesting  lusus  renim  from  True.  505  .  .  .  ehecquit  mei  similest  ? 
rogas?  I  quin  ubi  natust  <ma-ma>  machaeram-fuscebat.  If  <m<i-m<i>  were 
not  an  ingenious  insertion  of  Schoell,  the  orthography  of  ehec  would  conclu- 
sively indicate  stammering. 


Digitized  by  VjOOQ IC 


TEXTUAL  NOTES  AND  QUERIES  ON  PLAUTUS.       Ijg 

MU.  902  (B),  1297  (D),  Trin.  870  (B),  and  at  Cas.  166  hecquis 
haec ;  in  all  these  cases  h-  may  be  due  to  an  alliterative  impulse. 
The  following  passages  suggest  to  me  no  reason  for  their  h-i 
Bacch.  980  (DC),  Pers.  108  (B),  Pseud.  370  (A),  746  (D),  748  (D), 
True.  584  (DC).  Thus  there  are  some  13  cases,  taking  all  the 
manuscripts  into  account,  of  the  spelling  with  A-,  so  that,  after  all» 
we  may  not  make  any  cogent  inference  regarding  the  h-  of  vss. 
319,  339- 

vss.  320-21 :  semper  istoc  modo  moratus  . . . 
uitae  debebas. 
So  the  editio  minor.  I  propose  utie<r>  for  uitae ^  and 
construe  isioc  modo  as  the  ax^/xa  atth  koivov  with  moratus  and  again 
with  uH€<r>,  The  metre  is  trochaic  dimeter  acatal.  with  troch. 
tripody  acatal.  The  infinitives  in  -ter  in  Plautus  are,  however^ 
restricted  to  final  position  (see  Lorenz  on  Most.  963),  and  the 
only  deviation  (Men.  1006)  hardly  fiirnishes  a  warrant  for  the 
present  passage,  and  nothing  more  can  be  claimed  for  the 
proposed  viie<r>  than  for  any  emendation  ad  sensum.  Perhaps 
we  should  read  uH  e<o>  debebas,  making  eo  refer  to  isioc  modo. 
The  lost  Ko>  would  be  due  to  haplography  with  D  in  a  capital 
manuscript. 

vs.  328 :  sine  sine  cadere  me  ^sino  <%>  sed  hoc  quod  mihi  in 

manu[se]st. 

So  the  editio  minor.    No  instance  of  a  repeated  sine  is  known 

to  me  (at  Poen.  375  each  of  the  three  sinews  has  its  own  dependent 

verb),  and  I  suggest  si-si-sine,  letting  the  drunken  man  keep  up 

Del. 

his  stammering.  Further  in  the  verse  B  reads  me  sinof  &  hoc^ 
and,  recalling  the  form  semol  of  the  inscriptions,  I  propose  . . .  me* 
<Del.  sino.  Cal.>  semol  et  hoc  etc,  which  seems  to  me  a  litde 
nearer  the  manuscripts  than  Schoell's  simiiu  or  Hermann's  situs 
et. 

vs.  334 :  quod  ego  eam. 

The  reading  of  *P  was  indubitably  qtu>d  and  not  quo,  and  a 
precise  parallel  is  Asin.  864  hoc  ecastor  est  quod  ille  it  ad  cenam 
cottidie.  The  same  locution  is  used  by  Vergil  (Aen.  II  664),  hoc 
erat  alma  parens,  quod  me  per  tela  . . .  eripis.  In  all  these  cases 
I  take  quod  to  be  terminal.  I  do  not  agree  with  Lindsay  (Lat 
Lang.,  p.  568),  who  takes  the  terminal  adverbs  to  be  originally 
ablatives.     I  believe  them  to  be  either  datives  with  the  same 


Digitized  by  VjOOQ IC 


l80  AMERICAN  JOURNAL    OF  PHILOLOGY. 

paragogic  d  shown  by  the  accusatives  me-d  ie-dy  or,  more 
plausibly  perhaps,  datives  in  -o  plus  the  enclitic  preposition 
-d{e),  which  is  retained  also  in  Homer  in  a  few  archaic  formulae. 
Possibly  Most.  786  belongs  here :  quod  me  miseras,  adfero  omne 
impetratum,  though  we  may,  to  be  sure,  interpret  quod  me  (sc.  ui 
facerem)  miseras  etc. 

vss.  334-5:  quo[d]  ego  earn  an  scis?     CAL.  scio:  in  mentem 

venit  [DEL.]  modo: 
nempe  domum  eo  co[m]missatum  tCAL.  immo  istuc 

quidem  iam  memini. 
So  B,  substantially,  in  the  editio  minor,  but  with  differences  in 
the  division  of  the  lines.  Instead  of  \_DEL.'\  I  propose  to  read 
the  vocative  <Deiphium>,  which  makes  the  line  a  very  good 
troch.  sept.,  if  we  read  sctd  in  thesis  with  semi-hiatus.  In  the 
next  verse  CAL  is  put  in  above  the  line  by  B,  and  I  would 
eliminate  it  altogether,  making  istttc  refer  to  some  gesture  on 
Delphium's  part.  This  verse  is  also  a  septenarius  with  diaeresis 
and  hiatus  after  the  4th  foot.  In  the  7th  foot  we  have  a  dactyl, 
which,  though  rare,  is  allowable. 

vs.  358 :  ubi  aliqui  quique  denis  hastis  corpus  transfigi  solet. 

So  the  manuscripts,  but  aliqui  quique  seem  certainly  corrupt. 
We  have  in  Cicero  (Div.  II  50,  104)  aliquidquam  and  in  Livy 
(41.  6)  alicuiquam^  but  both  have  been  emended  by  the  editors. 
I  venture  to  propose  here  aliquiquam^  which  is  orthographic  for 
aliquoiquam  (cf.  Pers.  489  A). 

vs.  365:  quid  ita?  Jpater  adatesi  Jquid  ego  e<x>  te  audio? 
;|;absumpti  sumus. 

The  adaiest  of  the  manuscripts  I  correct  to  ad-ad-est  and  B 
divides  ad  at  est  Tranio  is  stammering  with  fright,  and  punning 
besides  on  attat  I  agree  with  Leo  in  reading  <adest>  adest  at 
vs.  363,  where  Philolaches  announces  the  coming  of  dainties  with 
the  reiteration  of  joy. 

Now,  at  vs.  366  the  metre  is  again  defective,  and  the  defect 
seems  to  consist  of  an  omission  before  the  same  word  adesU  The 
verse  runs : 

pater  inquam  tuos  venit  Jubi  is  est  obsecro?  TR. . . .  adest. 

So  the  editio  minor.  I  propose  to  read  <  Tranio^  before 
obsecro  (cf.  Poen.  1322,  True.  503),  and  possibly  <ad'>adesty 
with  'Crd  in  semi-hiatus. 


Digitized  by  VjOOQ IC 


TEXTUAL  NOTES  AND  QUERIES  ON  PLAUTUS,       l8l 

VS.  376 :  quaeso  edepol  exsurge. 

I  suggest  <ie>  exparge^  and  compare  Ep.  732  . . .  lumbos 
porgite  (*P  surgiie)y  and  Pseud,  i  exporgi  meliust  lumbos.  Still, 
ie  exporgt  means  'stretch  yourself/ and  not  'get  up.'  Perhaps 
we  should  read  ie  surge  or,  with  Ritschl,  ie  exsurge^  though  I 
can  furnish  no  citations  of  a  transitive  surgere  earlier  than  Vergil. 

vss.  412  sq.:  uerum  id  uidendumst,  id  uiri  doctist  opus, 
quae  dissignata  sint  et  facta  nequiter, 
niquid  potiatur,  quam  ob  rem  pigeat  uiuere, 
tranquille  cuncta  et  ut  proveniant  sine[mo]  malo. 
The  order  of  the  two  last  lines  has  been  needlessly  inverted,  it 
seems  to  me,  by  the  editors.    If  we  invert  with  the  editors,  or 
follow  the  manuscripts,  an  <ea>  after  ifanquille  or  after  cuncia 
is  easy  to  insert  palaeographically  and  lightens  the  syntax.    I 
would  take  ut  as  subordinate  to  niquid  potiatur^  not  co-ordinate. 

vss.  451-2 :  . . .  natus  nemo  in  aedibus 

seruat,  neque  qui  recludat  neque  qui[s]  respondeat. 
I  propose,  after  Bothe,  to  drop  quis  entirely  from  the  text. 
For  the  absolute  use  of  respandeai  I  cite  Cic  de  Or.  3.  49.  191 
respondebunt  non  vocati.     Seyffert  notes  Rudens  226,  where 
responsorem  means '  ostiarius.' 

vs.  469 :  vos  quoque  terram  ^obsecro  hercle  quin  eloquere. 

So  the  editio  minor.  If  we  may  read  hircli  (cf.  Klotz,  p.  47), 
I  propose  to  read  this  verse  as  follows : 

uos  qu6que  terram  [J]  obsecro  h6rcle  <t>  quin  eloquere 
<6bsecro>, 
and  compare  Cure.  308 : 

eloquere,  obsecro  hercle  Jeloquere  te  obsecro  etc. 

The  metre  can  also  be  mended  by  inserting  <iu>  before  eloquere^ 
if  we  read  with  semi-hiatus  and  a  shortened  /-  (cf.  Klotz,  p.  73). 
For  quin  <iu>  cf.  Asin.  659,  868,  and  iu  following  quin  would 
be  liable  to  haplographic  loss. 

vs.  663 :  nisi  ut  in  uicinum  hunc  proxumum  mendacium. 

I  would  correct  the  mendacium^  of  *P  to  mendax  siem.    After 

t  ^ 

proximutn  A  has  in  Studemund's  Apographum  D — RDIE.    For 

000 

^  SchoeU  would  have  it  that  mitidacium  is  picked  up  from  vs.  665. 


Digitized  by  VjOOQ IC 


1 82  AMERICAN  JOURNAL   OF  PHILOLOGY. 

the  first  D  alternative  letters  are  P,  and,  with  less  probability, 
E  I  T,  which  shows  that  the  perpendicular  only  of  the  D  is 
clear,  and  this  might  be  as  well  the  first  stroke  of  an  M.    Of  RD 

o  o 

only  the  lower  half  remains,  so  X  is  a  plausible  substitute  for  R 

(x)  [d]  t 

(cf.  Epid.  19  R),  and  S  for  D  (cf  Most  722  0,  Mil.  34  D),  though 

o  00 

Studemund  omits  S  here  in  giving  CGQTE  as  possible  readings  of 
D.  The  words  mendax  stem  do  not  violate  the  only  letters  reported 
certain  by  the  Apograph,  and  offer  a  plausible  substitute  for  *P's 
mendacium,  Nettleship  is  entitled  to  priority  in  point  of  this 
correction,  but  he  has  construed  in  here  with  the  ablative.  I 
do  not  see,  however,  that  in  with  the  ace.  is  necessarily  of  hostile 
intention,  and  so  need  not  mean  more  than  'put  a  lie  on.*  Plautus 
elsewhere  uses  aduorsum  (Aul.  690,  Poen.  400). 

vs.  701 :  nam  et  cenandum  ei  et  cubandum  est  male. 

So  A ;  B  omits  ei  and  reads  cubandumst  ni  irahis  male ;  the 
editio  minor  reads  nam  et  cenandum  ei  cubandumst  ei  male^ 
which  does  not  account  for  *P*s  ni  trahis.  This  we  can  do  by 
treating  A's  ei  as  dittographic  for  et  and  reading  intra  his  (sc 
aedibuSy  iaedis).  The  omission  of  the  noun  with  his  can  be 
supplied  by  a  gesture.  The  difficulties  are  that  intra  with  the 
ablative  seems  not  to  have  occurred,  and  while  his  could  be 
justified  for  a  nominative,  no  accusative  his  is  known.  Perhaps 
we  should  read,  then,  ifttra  has  or  intra  hie,  accounting  for  the 
variant  by  the  division  ni  trahis.  We  must  then  scan  the  end  of 
our  line  -diimst  intra  hds  male,  a  troch.  trip,  catal.,  such  as  we 
have  in  the  previous  verse. 

vss.  709-10 :  — haec  sat  scio  quam  me  habeat  male 

peiius  posthac  fore  <et>  quam  fuit  mihi. 
Leo  would  have  a  gap  between  these  lines  and  the  editio  minor 
suspects  quam  habeat  male  (omitting  me  with  B).  The  difficulty 
Leo  makes  is  removed  by  the  insertion  of  et  after  fore  [or,  with 
Bothe,  after  male?:  cf.  Rud.  1169,  Mil.  1132  (A)].  The  post- 
ponement of  et  after  /ore  is,  however,  harsh,  unless  we  can 
consider  it  relieved  by  the  position  of  ^/  at  the  end  of  the  second 
cretic. 

vs.  731 :  uitam  <iam>  cdlitis  Jimmd  <ita>  uit<a>  antehac  erat. 

I  insert  iam  and  ita.    I  note  Epid.  12  minus  iam  furtificus 

quam  antehac.    By  supplying  ita  the  retort  means  *  Nay,  that's 


Digitized  by  VjOOQ IC 


TEXTUAL  NOTES  AND  QUERIES  ON  PLAUTUS,       183 

how  we  used  to  live.'  That  this  must  be  the  sense  the  next 
verse,  cited  below,  goes  to  prove.  The  subsequent  retort  (vs. 
733)»  ^^^  oppide  occidimus  omneSy  plays  upon  iia  uiia — erat, 
taking  erai  in  the  sense  oifuiU  In  vs.  731  something  like  iia  is 
certainly  needed  for  the  predicate  of  uiia  erai. 

vs.  732 :  nunc  nobis  tcommunia  haec  exciderunt. 

So  the  editio  minor.  Here  the  metre  seems  undoubtedly 
corrupt.  The  verse  will  scan  as  a  trochaic  hexapody  catalectic, 
like  verses  704-5  above  in  the  same  passage,  if  emended  as  follows : 

nunc  nobis  communia  haec  <hic  c>e[x]cidSrunt — 

if  we  may  employ  Vergil's  3d  plur.  pf.  ending.  For  the  sense, 
communia  haec  means  'the  common  end,'  as  communis  locus 
(Cas.  19)  means  'the  common  place,'  i.  e.  Hades. 

vs.  756 :  quid  consomniavit. 

Here  the  metre  is  defective.  Ritschl  proposed  <hem>  quid^ 
but  as  D  reads  qui%  and  C  con  sommavii^  I  propose  quid  <se> 
cum  somniavii,  and  compare  Ter.  And.  442  secum  reputavit,  and 
Cic.  Off.  3.  I.  I  secum  loqui,  as  general  analogues  for  the  con- 
struction. This  correction  would  entirely  banish  the  nonce-word 
consomniare. 

vs.  873 :  bonis  sum  improbis  sunt  malus  fuit. 

So  B,  but  the  line  is  neither  metrical  nor  sensible.  Now,  in  a 
minuscule  manuscript /f^iV  may  well  stand  for/tW,  and  sum  in  a 
capital  manuscript  is  close  to  sini  in  ductus.  I  therefore  propose 
to  read  the  verse  as  follows : 

bonf  sint  <  bonis  si>  improbi  sint,  mail  sint 

'the  masters  would  be  good  to  good  slaves,  but  if  they  were  bad, 
bad.'    I  compare  Amph.  659-61 : 

atque  ita  seruom  par  videtur  sese  insdtuere: 

proinde  eri  ut  sint,  ipse  item  sit . . . 

tristis  sit  si  eri  sint  tristes ;  hilarus  sit,  si  gaudeant, 

which  approximates  our  construction.  A  further  parallel  at 
Bacch.  660 : 

. .  .  bonus  sit  bonb,  malus  sit  mails. 

It  would  conform  to  these  parallels  better  if  we  should  read  donus 
sii . .  .  mains  sii,  but  there  is  no  cogent  reason  for  the  singular 
in  this  truism. 


Digitized  by  VjOOQ IC 


1 84  AMERICAN  JOURNAL    OF  PHILOLOGY. 

VS.  890 :  ferocem  facis  quiate  eratus  amat  uha 
So  B.     I  propose  to  emend  as  follows : 

fer6cem  facis  <te>  quia  effdrtus  ama<s>  tu  <  J>  ha. 

As  far  as  the  ductus  goes  quiaeffartus  is  most  similar  to 
QUIATEERATUS,  though  there  is  an  inversion  (metathesis  perhaps  ?, 
cf.  Lindsay,  Latin  Language,  p.  97)  of  ar.  The  inserted  <te> 
would  have  been  lost  owing  to  the  conversion  of  quia  ef-  to  quia  te. 
The  bacchius  qut  ifftrtus  is  far  from  impossible  (cf.  Klotz,  pp. 
343»  352;  33)'  I  interpret  the  verse:  *You  are  playing  the  r61e 
of  one  swollen  with  anger  because  you  love  jw^/Zm^^-with-food.' 
The  noun  effarius  (jefferius)  would  be  formed  like  partus, 
Plautus  uses  effercio  above  (vs.  65)  and  uses  efferius  as  a  parti- 
ciple 3  times  (twice  in  the  superlative).  At  vs.  169  above,  he 
uses  the  simplex  fartum  (MSS/arlim,  but  see  the  editio  minor, 
Praefatio,  sud  versu),  where  vestis  fartim  seems  to  mean  *the 
stuffing  of  the  clothes.'  For  the  sentiment  I  compare  Bacch. 
580,  where  a  parasitus  says  to  his  boy -attendant :  comesse  panem 
tris  pedes  latum  potes. 

There  can  be  no  objection  to  reading  ha  (i.  e.  ah,  cf  Epid.  540 
in  A)  for  uah  at  the  beginning  of  the  next  line  on  the  score  of 
metre  or  meaning  (cf  True.  525,  where  ah  is  used  "ubi  dolorem 
corporis  mulier  simulat":  Richter  in  Studemund's  Studien,  I,  p. 
401). 

vs.  904:  quid  tibi  visumst  mercimoni  t<totus>  totus  gaudeo. 

Gruter's  insertion  of  <totus>  is,  I  believe,  correct;  cf.  Cas.  621 
tota  tota  occidi. 

vss.  905-6 :  . . .  nunquam  edepol  ego  me  scio 
vidisse  umquam  abiectas  aedis  etc 
I  would  let  nunquam — unquam  remain ;  it  is  certainly  no  worse 
tautology  than  nemo  homo  (Pers.  211)  and  homo  nemo  (Most.  901 
in  A,  "ut  videtur"). 

vs.  926 :   EGOENIMCAVIREC  .  .  .  AMBISGRATIAMATQ-ANIMOMEO. 

So  A,  while  B  reads  Earn  dehis  graiiam,  dehis  being  probably 
a  spelling  of  dis  as  above  at  vs.  563.  That  A  confuses  B  and  D 
is  clear  from  aebis  for  aedis  at  True.  252.  I  propose  to  read  our 
verse  as  follows : 

ego  enim  caui  recte ;  /am  dis  gratia[m]  atque  animo  meo. 


Digitized  by  VjOOQ IC 


TEXTUAL  NOTES  AND  QUERIES  ON  PLAUTUS.       18$ 

B's  Earn  for  tarn  is  responsible  for  gratiava.  The  phrase  tarn  dis 
gratia  means  *  great  thanks  to  the  gods.'  I  compare  Ep.  lo  huic 
gratia  ^thanks  to  this  hand  of  mine';  at  Men.  387,  Stich.  472  tarn 
graHast  means  'no,  many  thanks/  'no'  being  inferred  as  in 
German  ich  danker  but  ick  danke  is  also  positive.^  I  note  in 
Terence  the  parenthetic  {es()  dis  gratia  (Ad.  121,  138). 

vs.  984:  Tranio:  is  uel  Herculi  conterre  quaestum  potest. 

So  the  manuscripts,  but  B'  has  corrected  the  second  r  of 
conterre  from  an  original  e.  I  believe  that  *P  had  the  trisyllabic 
infinitive  canierery  and  I  would  read  the  verse : 

Tranio:  is  uel  Herculi  <illi>.conterer  quaestum  potest. 

To  supply  <illi>  makes  a  perfect  diaeresis  here,  and  it  is  thor- 
oughly Plautine  (cf.  above  ille — luppiter^  vs.  398).  The  apoco- 
pated infinitive  dicer  is  demanded  by  the  metre  at  Merc  282  (cf. 
Sonnenschein  in  Transactions  of  the  American  Philological  Asso- 
ciation, XXIV,  p.  14),  and  biber  for  bibere  is  certainly  genuine 
(cf.  Charisius  124.  i,  in  KeiPs  Grammatici  Latini);  while  Stolz's 
explanation  of  the  infinitive  in  -ier  as  apocopated  from  -iere  is 
after  all  the  best.' 

vs.  1012:  quid  a  Tranione  seruo?    SI.  tn^ulto  id  minus. 

So  the  editio  minor.  I  propose  to  read  57.  <immo>  multo 
etc.,  assuming  a  haplography  of  Simo  immo,  I  note  vs.  807, 
where  the  manuscripts  give  SL  immo.  The  metrical  value  immd 
occurs  again  at  vs.  1091. 

vs.  1081 :  quid  iam?  tscio  iocaris  tu  nunc  [tu]  nam  ille  <illud> 
quidem  baud  negat. 
It  is  obvious  that  both  iu's  cannot  stand,  and  they  are  most 
easily  accounted  for  as  arising  from  a  shift  in  order  (cf.  supra  on 
vs.  104).  Leo  inserts  <iedepol>  after  quidem^  but  restitutions  of 
edepoldSid  its  like  seem  to  me  idle  where  haplography  cannot  be 
made  plausible.     I  propose  <illud>,  meaning  '  his  own  act.' 

^  There  is  such  an  inferred  *no'  at  Epid.  638  non  me  nosti?  {quod  quidem 
nunc  veniat  in  mentem  mihi. 

'I  note  my  supplement  to  this  explanation  in  the  Classical  Review,  X  i83» 
where  I  incorrectly  declared  vid^  to  have  manuscript  warrant  at  Epid.  62, 
citing  a  previous  error  of  mine  in  Am.  Jour.  Phil.  XV,  p.  372.  A  better 
example  is  our  present  passage,  where  conterer  approximates  to  manuscript 
warrant. 


Digitized  by  VjOOQ IC 


1 86  AMERICAN  JOURNAL    OF  PHILOLOGY. 

VS.  1089:  dat  profecto  <X>  quin  tet  ilium  in  ius  si  veniam  mane. 
So  the  editio  minor.  I  believe  the  verse  to  make  good  sense 
and  metre  as  it  stands.  After  ilium  occurs  the  normal  hiatus  in 
the  diaeresis.  I  interpret  the  verse:  *he  doesn't  intend  to  go  to 
court  any  more  than  I  do  (who  as  a  slave  am  debarred  from 
court)';  or,  literally,  'expect  him  too  when  (if)  I  come  into 
court.' 

vs.  1090:  <TH.>  experiar  ut  opinor  TR.  tcertumst,  mihi  homi- 
nem  cedo. 
So  A.    Schoell,  following  a  cue  of  Gruter's,  reads  ut  opino 
TR,  <^opino''i>  etc.    Leo  assigns  certumsi  to  TH.  and  reads 
TR,  <immo>  mihi  etc.     I  propose  the  following : 

<TH.>  experior  ut  opinor.    TR.  certum<n>est?    <TH.  cer- 
tumst  TR.>  mihi  hominem  cedo, 

and  cite  in  illustration  Stich.  482  certumnesif  Xcerium  ,  and 
Merc.  461  certumnesif  ;|;censen  cerium  esse?  .  1  note  also 
the  entirely  satisfactory  restoration  of  Camerarius  at  Most.  639 
eerie f      <cerle^'>  inquam 

vs.  1091 :  uel  [hominem]  iube  aedis  mancipio  poscere  {immo  hoc 
primum  uolo. 

The  sense  here  is  excellent  if  mancipio  may  be  construed  with 
poscere^  on  the  analogy  oi  mancipio  CLccipere  *take  possession,' 
mancipio  dare  'give  possession.'  Leo  suggests  that  hominem 
has  crept  in  from  vss.  1090,  1093.  If  we  eliminate  hominem^  as 
Camerarius  did,  the  verse  becomes  metrical  with  only  one  ques- 
tionable foot,  the  sth — cir*lmmd.  The  pyrrhic  value  of  immd 
seems  certain  (cf.  Lindsay,  1.  c,  p.  603),  and  the  only  question  is 
whether  immd  with  hiatus  or  immd  with  its  final  shortened  by 
semi-hiatus  may  stand  in  the  arsis  of  a  foot. 

vs.  1 1 13:  ndnquam  edepol  hodi6  <di  med>  invituif^  destindnt 
tibi. 
So  I  propose  to  read  this  verse ;  besides  the  insertion  it  only 
changes  MS  invilus  to  invilum.  Schoell  reads  . . .  hodie  <hinc 
abibo>  invitus:  desine  aul  abi,  and  Leo  .  . .  hodie  <hincsivivo> 
invilus  desislam  abu  For  the  usage  of  deslinanl  in  our  passage 
I  cite  Aen.  II  129  me  destinant  arae,  and  Cic.  Off.  Ill  45  cum 
alteri  diem  necis  destinavisset. 


Digitized  by  VjOOQ IC 


TEXTUAL  NOTES  AND  QUERIES  ON  PLAUTUS,       1 87 

VS.  II 14 :  iam  iubeo  ignem  et  sarmen  <arae>  carnifex  circumdari. 
So  the  editio  minor,  following  Seyifert's  ingenious  emendation, 
B  reads  hibeo^  C  lubo  and  D  iube\  these  best  resolve  into  a 
minuscule  luber^  and  I  would  amend  the  line  thus : 

iam  lubet  <tibi>  ignem  et  sarmen  carnifex  circumdari  ? 

In  this  line  the  haplography  of  libet  and  Hbi  probably  took  place 
in  a  capital  manuscript.  I  construe  Hbi  with  circumdari  according 
to  the  <r)fifia  dir6  koivov, 

VS.  II 34:  age  mitte  ista  t^cto  ad  me  ad  cenam  tdic  venturum, 
quid  taces  ? 
So  the  editio  minor.    I  propose  to  emend  as  follows : 
age  mitte  ista,  a^<i>to  ad  me  cenam  etc., 

where  the  insertion  of  <i>  is  justified  by  the  following  /,  and  the 
interchange  of  /^  and  c  is  of  common  occurrence  in  capital  manu- 
scripts. Ussing's  ingenious  correction  ac  ie  involves  shifting  die 
veniurum  to  the  first  speaker.  While  no  great  attention  can  be 
paid  to  the  notae  personarum,  still,  dropping  one  out  demands 
some  justification. 

It  is  not,  I  take  it,  a  jejune  reiteration  of  hie  apud  nos  hodie 
ccnes  (vs.  1129),  when  the  would-be  host  renews  his  invitation  by 
agiio  cenam  Tm  having  a  big  dinner'  (cf.  Pers.  28  agitare 
eleutheria,  ib.  769  —  meum  natalem,  Asin.  834  —  convivium). 
The  [afl?]  I  have  dropped  before  cenam  came  in  with  the  error 
in  cLcto, 

vs.  1 141 :  numquid  aliud  feci<t>  nisi  quod  [faciunt]  summi<s> 
gnati  generibus. 
'So  the  editio  minor.  I  propose  to  read  faciuni  and  dso^feci. 
There  is  no  more  difficulty  in  inferring  the  verb  of  the  leading 
clause  from  the  dependent  than  vice  versa  (cf.  Livy  34,  46  and 
2,  32),  and  a  diaeresis  sSttx faciunt  would  be  easier,  I  take  it,  than 
after  quod. 

vs.  1 166 :  dispudet  < t>  tistam  ueniam :  quid  me  fiet  \X\  nunciam  ? 

So  the  editio  minor.    Assuming  that  the  nota  personae  had 

been  lost  in  *P,  I  suppose  that  manuscript  to  have  read  DISPU- 

^The  following  are  salient  cases  of  the  confusion  inToWed  here:  Pseud. 
1054  tube  ftttfir  (A),  lubet  nunc  (*P);  ib.  1125  lubet  (CD),  iubet  (B);  True.  585 
iubet,  corrected  by  Camerarius  to  lubet^  by  Buecheler  to  iube. 


Digitized  by  VjOOQ IC 


I  88  AMERICAN  JOURNAL    OF  PHILOLOGY. 

DETEISTAM  etc.,  and  would  read  here  dispudei  <Xem>  is'tam 
etc.,  referring  for  this  metrical  value  of  em  isiam  to  Richter  in 
Studemund's  Studien  (I,  p.  498),  though  at  both  the  places  cited 
(Cure.  212,  Merc.  206)  we  might  read  em  1st — ,  a  resolved  thesis, 
because  of  elisions  complicated  with  change  of  speaker. 

vs.  1 172:  mftte  quaes<o>  istum  <J>  fe  viden  ut  astat  fdrcifer? 
So  the  editio  minor,  but  restai  in  the  manuscripts.    I  propose 
to  read  as  follows : 

mitte,  qua6so,  istum  <t>  6<m>  viden  fit  res<is>tat  f(ircif6r? 

There  is  no  palaeographic  difficulty  in  reading  e<m>.  For 
e<m>  viden  I  cite  Terence,  Hec.  316  em  sensistin?  and  Pseud. 
872  em  subolem  sis  vide.  Perhaps,  though,  we  should  read  hem 
viden,  like  hem  cupin  above  (vs.  5),  and  the  passage  cited  from 
Terence  has  variants  in  hem.  It  is  quite  certain  that  resistere  is 
used  by  Plautus  in  the  same  sense  as  asiare  (cf.  Cas.  727). 

The  questionable  point  about  the  reading  I  propose  is  whether 
quaeso  may  be  read  with  hiatus  or  semi-hiatus.  It  seems  to  be 
so  read  at  Cist.  554  and  Cure.  629. 

lbximgtok,  va.  Edwin  W.  Fay. 


Digitized  by  VjOOQ IC 


III.— SUPERSTITIONS  AND  POPULAR  BELIEFS  IN 
GREEK  COMEDY. 

As  to  the  scope  of  this  paper  and  its  governing  principles,  I 
refer  the  reader  to  the  introduction  of  my  paper  on  *  Superstition 
in  Greek  Tragedy.'^  One  word,  however,  should  be  said  about 
the  material  on  which  it  is  based.  The  extended  use  which 
Plautus  and  Terence  both  have  made  of  Greek  originals  might 
lead  to  the  expectation  that  their  plays  have  likewise  been  made 
use  of  in  the  following  discussion.  This  has  not  been  done, 
partly  because  I  do  not  wish  to  extend  this  paper  to  an  undue 
length,  partly  because  I  think  it  better,  for  reasons  of  method,  to 
discuss  this  borrowing  and  adapting  of  beliefs  in  its  proper  place, 
viz.  the  discussion  of  Roman  beliefe. 

I. — In  trying  to  define  the  extent  of  superstition^  in  antiquity  I 
called  attention  to  the  &ct  that  the  word  dtiaidaifMovia  shows  the 
origin  to  have  been  an  exaggerated  fear  of  divine  spirits.  From 
the  comparatively  late  appearance  of  the  word  in  Greek  literature, 
and  from  the  conditions  of  the  age  when  it  first  appears,  I  ven- 
tured to  argue  that  it  had  its  origin  about  400  B.  C,  in  round 
numbers,  in  that  seething  cauldron  of  religious  sectarianism  which 
is  marked  by  the  preponderance  of  Orphicism  and  by  the  growing 
acceptance  of  the  cult  of  Cybele  by  the  lower  classes.  I  tried 
also  to  show  that  such  transformation  of  religious  belief  into 
superstition  was  not  the  product  of  isolated  circumstances,  but 
that  it  is  subject  to  a  law  which,  therefore,  must  be  expected  to 
work  wherever  similar  conditions  exist.  To  this  I  believe  I  am 
able  to  adduce  additional  testimony.  Even  in  Menander's  time 
the  word  dtiaidaifmy  was  not  yet  firmly  established.  In  fact,  the 
word  ISko\o£  repeatedly  occurs  in  its  stead,  not  only  here,  but  also 
in  the  Old  Comedy.'  Now,  this  word  can  only  refer  to  the 
enthusiastic  shouts  of  some  religious  service — in  one  word,  to  the 
cult  of  the  Great  Mother.     It  is  well  known  that  Cybele  priests 

1  Transactions  Am.  Phil.  Assoc.  XXVII  5  ff.  *  Ibidem,  XXVI  40  ff. 

'Menand.  Aetaidai/tuv,  112  K.,  inc.  1046  K.;  Theopomp.  Tiaafiev6c,  61  K. 


Digitized  by  VjOOQ IC 


190  AMERICAN  JOURNAL    OF  PHILOLOGY, 

were  no  mere  functionaries  in  the  ritual,  but  that,  beyond  this, 
they  were  busily  engaged  in  miraculous  cures/  It  can  also  be 
proved  from  Menander's  comedies.'  That  the  okokvy^s  played  a 
prominent  part  in  the  rites  of  the  Great  Mother  would  be  inferred 
from  their  enthusiastic  character.  But  for  this,  too,  we  have  an 
express  testimony  in  the  verses  of  Menander's  MKroyvvor': 

c^vo/icy  dc  frevrdi»r  r^ff  fffupas 
€KVfiPd\iCop  d   €frra  Bfpdiraivai  icvkXai, 
al  d   aX6\v(ov, 

The  history  of  the  word  firjrpayvfmjty  likewise,  tends  to  show  that 
at  a  comparatively  early  stage  the  more  exaggerated  features  of 
the  cult  had  fallen  into  deep  contempt.  The  rivalry  between 
deio-tduifuov  and  oXoXoff  enables  us  to  witness,  as  it  were,  the  slow 
decay  of  a  true  religion,  or  rather  of  some  of  its  features,  into 
superstition,  an  opportunity  not  often  offered  in  antiquity. 

It  is,  however,  unsafe  to  speak  too  confidently.  As  I  have 
urged  in  the  paper  referred  to,*  the  individuality  of  the  writer  is 
of  the  greatest  moment  in  these  questions,  and  we  might  sadly 
misjudge  popular  feeling  by  making  the  comedian  our  standard. 
An  example  of  this  is  furnished  by  Aristophanes.*  In  his  Amphi- 
araos  he  had  censured  the  credulity  of  the  Athenians  as  regards 
the  miraculous  cures  wrought  in  the  Amphiaraion  at  Oropos. 
This  appeared  to  him  to  be  superstition.  But  did  the  Athenians 
in  general  regard  it  as  such?  The  sacred  precinct  of  Amphiaraos 
certainly  continued  to  give  birth  to  innumerable  miracles  and  to 
form  the  centre  of  a  fervent  worship  down  to  much  later  times,  as 
the  well-known  records  unearthed  at  Oropos  prove  beyond  doubt. 
As  to  Menander  himself,  not  only  the  fact  that  he  wrote  a  comedy 
called  Aeio-idatfuioir,  but  also  his  verses  about  the  Syrian  fish-taboo,' 
show  that  he  was  considerably  in  advance  of  his  own  age  as  far 
as  popular  beliefs  are  concerned.    Now,  this  should  certainly 

1 E.  Rohdc,  Psyche,  336  ff. 

*  Menand.  'lipeia,  245  K.  About  the  time  of  the  Rhea-Cybele  cult  cp.  also 
Preller-Robert,  I  651. 

'  326  K.  =  inc.  601  K.  It  is  significant  that  the  passage  occurs  in  the 
Mtadywoc.  Women  at  all  times  hare  been  the  chosen  agents  of  superstition 
and  its  most  influential  proselytisers.  Compare  the  influence  of  the  freed- 
women  in  imperial  Rome,  as  shown  in  the  poems  of  Horace,  Ovid,  Tibullus. 
See  also  Index  s.  prostitute. 

♦Transact.  Am.  Ass.  XXVI  43,  50.  *See  Bergk,  Ar.  fgm.,  p.  951. 

*  Menand.  inc.  544  K. 


Digitized  by  VjOOQ IC 


SUPERSTITIONS  IN  GREEK  COMEDY,  I9I 

warn  us  to  exercise  the  greatest  discretion  in  relegating  a  belief 
to  the  sphere  of  superstition.  It  cannot  be  repeated  too  often 
that  the  line  of  demarcation  between  superstition  and  religion 
fluctuates. 

2. — Aristophanes, 'Hpwcff,  306  K.: 

lu\Tt.  noddviiTTpov  Bvpa{^  cjt;(Ctrc  firfrt  Xovrpiop. 

At  first  blush  this  looks  like  a  measure  of  sanitation  which  must 
have  been  very  welcome  indeed  where  the  streets  were  still 
unpaved.  But  the  fragment  had  its  place  in  a  comedy  which,  by 
its  very  name,  dealt  with  'souls.'  It  is  a  hackneyed  fact  that 
such  apparently  sanitary  measures  almost  always  had  their  origin 
in  religion.  So  it  is  here  also.  The  sacredness  of  the  door^ 
hardly  needs  proving.    It  will  be  sufficient  to  quote  Menander': 

fiapTvpofxai,  vol  fxh  r6y  *Air6k\n  rovrovi 
Koi  rat  Bvpai* 

This  sacredness  would  be  a  sufficient  reason  for  the  prohibition. 
However,  its  root  seems  to  lie  deeper.  At  the  door  the  souls 
have  one  of  their  habitual  haunts,  though  I  hardly  recollect 
any  reference  to  it  from  Greek  soil.'  But  among  the  Hebrews 
the  seat  of  the  Elohtm  of  the  family  was  at  the  door.  When  the 
Hebrew  slave  declared  his  willingness  to  remain  a  slave  rather 
than  to  make  use  of  the  liberty  offered  him  with  the  return  of  the 
jubilee  year,  he  was  led  to  the  door  and  his  ear  was  fastened  to 
the  door-post  with  an  awl.*  Moreover,  firom  German  superstition 
we  know  of  a  custom  that  will  admirably  help  to  explain  the 
Greek  prohibition.  Immediately  after  the  coffin  has  been  carried 
out  of  the  house,  the  pail  of  water  with  which  the  corpse  was 
washed  is  poured  out  after  it.  This  is  to  prevent  the  return  of 
the  soul.*  Thus,  to  pour  out  the  water  of  the  bath  would  drive 
the  spirits  from  their  abode.    Hence  its  prohibition  by  the  ^pttcr. 

*  Cp.  H.  C.  Trumbull,  The  Threshold-Covenant. 
'  Menand.  inc.  740  K. 

'  Except,  perhaps,  the  threshold  sacrifice  in  Mag.  Pap.  V.  Ill  27. 

*  Deuteron.  15, 17.  About  the  meaning  of  this  ceremony  cp.  Pauly-Wissowa, 
I  30,  60. 

^  Wuttke,  Volksabeigl.*,  §732.  Cp.  Grimm,  Mythol.,  App.,  846.  A  similar 
custom  prevails  in  modern  Greece :  C.  Wachsmuth,  Das  alte  Griechenland  im 
Neuen«  119, 129.  The  slang  expression  *to  kick  the  bucket,'  so  widely  used 
among  English-speaking  peoples,  may  possibly  be  connected  with  this  custom. 


Digitized  by  VjOOQ IC 


192  AMERICAN  JOURNAL    OF  PHILOLOGY. 

3. — Aristophanes,  TcXfucr^y,  532  K.: 

frivaiciVicoy  Sirvpop  Ix0vrip6v. 

Api(rro<t>dinjf  dc  coixc  diaiptiv  rivas  dnvpovt  Koi  ifinvpovs  irivaKtaKovs,^ 
Bergk  compares  Phrynichus':    anvpov  mvaKiaKOp'    kqivov  fifirrw  jrvp) 

7rpotr€»riv€yfA«pop,  We  may  accept  Phrynichus*  explanation  as 
correct ;  but  Aristophanes  certainly  did  not  make  this  distinction 
from  sheer  arbitrariness.  The  Telmessians  was  a  comedy  dealing 
with  the  superstitious  practices  of  the  Telmessians,  a  people  noted 
throughout  antiquity  for  its  witchcraft.  The  newness  of  earthen- 
ware is  an  important  factor  also  in  the  magical  papyri  where  the 
xvrpa  Kaivfi  is  repeatedly  mentioned.'  A  possible  explanation  of 
the  word  airvpos  would  also  be  that  the  vessel  was  to  consist  of 
unbaked  clay.  This,  too,  finds  its  warrant  in  the  magical  papyri.* 
In  either  case  the  reason  for  the  prohibition  lies  in  the  enmity 
between  fire,  the  pure,  ethereal  element,  and  magic* 

4. — Aristophanes,  TcX/ucr^y,  530  K.: 

rpdireCav  fipiv  ti(r<l>€p€ 
Tp€i£  ir6da£  Uxova-aVf  rtrrapat  di  firj   x*'*'^^ 
— Koi  fr6B€v  €ya  rplirovv  rpdn^Cav  Xrj^opai/ 

In  a  paper  on  *  Folk-lore  in  Artemidoros,*  •  I  have  tried  to  show 
that  tables  were  fitted  with  three  legs,  not  for  any  practical- reason, 
but  on  account  of  some  mysterious  religious  idea.  The  Aris- 
tophanic  passage  just  quoted,  while  it  shows  that  four-legged 
tables  must  at  least  have  been  considerably  more  frequent  than 
the  others,  still  confirms  my  view.  For  the  express  command, 
coupled  as  it  is  with  an  express  prohibition  and  occurring,  in  all 
probability,  in  an  act  of  witchcraft,  can  have  no  other  reason  back 
of  it  but  a  superstitious  one. 

5. — Alkaios,  rawfi^d^r,  4  K.: 

Difficult  as  it  is  to  interpret  half  a  line  from  an  ancient  comedy 
which  may  have  treated  of  everything  under  the  sun,  I  neverthe- 

1  Pollux,  X  82.  » In  Bekker's  An.  Gr.  I  14. 

'See  the  index  of  Wessely's  Zauberpapyrus,  in  Wiener  Denkschriften, 
XXXVI  and  XLII,  s.  v.  Katv6c, 
*  Paris.  Pap.,  vs.  900  3,  ed.  Wessely. 
»Cp.  also  Soph.  Trach.  607.  •  Rhein.  Mus.  XLIX,  p.  184. 


Digitized  by  VjOOQ IC 


SUPERSTITIONS  IN  GREEK  COMEDY,  I93 

less  think  that  this  particular  half- verse  allows  of  an  interpretation. 
Indeed,  it  enables  us  to  state  the  existence  of  a  belief  in  the  fifth 
century  B.  C.  which  hitherto  has  been  known  only  from  the 
monuments  and  the  literature  of  later  times.  I  refer  to  the  belief 
in  the  Nereids  as  evil  demons.  This  assumption  of  their  char- 
acter has,  in  fact,  mostly  been  based  upon  their  modem  signifi- 
cance.^ It  is  only  a  few  years  since  O.  Crusius,'  by  interpreting 
a  'Hellenistic'  relief,  furnished  proofs  of  the  existence  of  similar 
beliefs,  at  least  in  Alexandrian  times.  Our  fragment,  however, 
takes  us  still  farther  back.  The  drastic  treatment  of  the  daughter 
of  Nereus  is  eminently  one  of  the  means  which  were  used  against 
the  evil  eye,  and  as  such  reveals  itself  as  good  against  evil  spirits.' 
Why  the  beautiful  sea-nymphs  became  dangerous  I  do  not  know. 
Some  inference,  however,  may  be  drawn  from  the  behavior  of 
water-nymphs  towards  Hylas^  and  from  the  more  modern 
mermaids. 

6. — K rates, 'HpttCff,  10  K.: 

What  right  have  I  to  refer  this  passage  to  necromancy'?  It  is 
only  a  very  slight  justification  that  it  occurs  in  a  comedy  of  the 
title  *Hpa>cr.  But  it  seems  to  me  that  the  verse  might  almost 
stand  as  subscription  to  the  well-known  Teiresias  vase.'  On  this 
monument  of  the  fourth  century  B.  C.  we  see  Odysseus  sitting 
on  a  heap  of  stones ;  to  his  right  and  left  his  companions  are 
standing,  and  between  his  feet  the  head  of  the  ram  killed  as  a 
sacrifice  is  lying.  At  his  feet,  furthermore,  the  shadow  of 
Teiresias  is  visible.  That  is  to  say,  we  see  his  head  and 
shoulders  rising  from  the  depth.    He  is  ceriainly  represented  t6v 

avx*y  €K  yrjs  awicar  th  airrov  ffktirwvt  SO  mUch  SO  that  from  crown  tO 
chin  there  is  an  almost  horizontal  line.  It  has  long  since  been 
surmised  that  this  masterly  painting  goes  back  to  some  cele- 

'  See  espec.  B.  Schmidt,  Volksleben  d.  Neugriechen,  pp.  105, 118  ff.«  123. 

«  Philologus,  L  97  ff. 

'O.  Jahn,  Leipz.  Sitzgsber.  1855,  86  ff.  Cp.  the  way  in  which  Luther  used 
to  treat  the  attacks  of  the  arch-fiend.  About  the  demoniacal  nature  of  the 
evil  eye  see  the  remarks  by  Bloomfield,  this  Journal,  XVII  400. 

*  This  point  has  been  made  by  Schmidt,  too,  1.  c,  p.  99,  n.  7.  Cp.  also 
V.  Laistner,  Rtltsel  der  Sphinx,  and  Class.  Rev.  X  413. 

^  See  Index  s.  necromancy. 

•  Mon.  Inst.  IV  19  =  Baumeister,  Denkm.  II,  fig.  1254. 


Digitized  by  VjOOQ IC 


] 


194  AMERICAN  JOURNAL    OF  PHILOLOGY. 

brated  original,  albeit  unknown  to  us.^  The  close  resemblance 
between  the  verse  of  an  Athenian  poet  of  the  fifth  century  and 
the  Athenian  vase  seems  to  me  one  more  proof  of  the  correctness 
of  such  an  hypothesis. 

7. — Strattis,  *om<r<ra4,  46  K.  =  Aristoph.  N^aoi,  389  K.: 

"  Es  liegt  ein  tiefer  Sinn  im  kind'schen  Spiel.*'  These  words 
of  the  German  poet,  so  often  borne  out  by  facts,  are  proved  anew 
by  our  two  passages.  Nobody  doubts,  nowadays,  that  children's 
games  and  nursery  rhymes  are  often  the  last  refuge  for  older 
religious  rites  and  songs.'  The  same  process  has  taken  place 
here.  With  all  other  nations  of  the  earth,  the  Greeks  too 
believed  that  eclipses  of  sun  or  moon  were  caused  by  some  bad 
spirit  or  human  being  trying  to  attack  or  swallow  the  great 
luminaries.'  It  behooved  man,  therefore,  to  come  to  their  rescue. 
This  was  done  by  making  a  hideous  noise  or  by  shouting  incan- 
tations and  songs.  Now,  in  Pollux,^  where  the  better  of  the  two 
accounts  of  this  game  has  been  preserved,  we  read :  ^  dc  t^x^  ^ 

P€<f>os  ^v^dpafxjj  t6p  ^edv.  There  can  be  no  doubt,  to  my  mind,  that 
in  the  cloud  we  must  see  the  last  incarnation  of  the  evil  spirit, 
which  shall  be  disturbed  and  beaten  off  by  the  noise  and  shouts 
which  are  raised  against  it. 

8. — Aristoph.  Ran.  298  fF.: 

airdkovfieB  ,  &va(  'H/MicXciff. — ov  fifj  xoXccff  ft 

— At<$vuf7€  Toipvv, — rovT*  KB*  ^irrop  Baripov* 

There  seems  to  underlie  these  verses  the  well-known  idea  of 
the  power  of  the  name,  as  old  as  the  oldest  incantations  which 
we  possess.^  When  you  know  the  name  of  a  demon,  you  have 
gained  complete  control  over  him,  and,  contrariwise,  you  must 
take  good  care  lest  the  sprites  learn  yours,  or  they  will  control 

^  Baum.  II,  p.  1040. 

'  Grimm,  Mythol.,  passim.    Cp.  Alice  Gomme,  Traditional  Games. 

'  Pauly-Wissowa,  I  41, 4  ff.;  cp.  £.  Rohde,  Psyche,  379 ;  W.  Roscher,  Selene. 

*  Pollux,  IX  7. 

*  E.  g.  cp.  A.  Erman,  Aegypten,  p.  359  ff. 


Digitized  by  VjOOQ IC 


^aL..i,,,iJUj^b^l^-^W:'*^m 


^H 


SUPERSTITIONS  IN  GREEK  COMEDY,  I95 

you.'  Even  the  o^ir  pseudonym  of  Odysseus  goes  back  to  this 
belief,  however  remotely.  Dionysos  is  just  now  on  his  way  to 
the  lower  world  and  particularly  anxious  to  avoid  the  dangers 
which  beset  his  path.  It  is  only  natural  that  he  should  observe 
the  rule;  for,  as  Hall  Caine  puts  it,  custom  must  be  indulged 
with  custom  or  custom  will  weep. 

9« — Com.  anon.  85  K.:  BXcircdai/iODi',  h  ducTpafi/Atvos  ras  ^^c(f  Kal 
olop  vn6  dai/Aovos  n€ir\rjy&s.*     ffk,,  6  vn6  v6€rov  icarca«cXi;«cci>ff  Koi  KaK6xpov£ 

im6  daifi6p»v,^  "  Larvae  similis/'  says  Kock.  But  all  these  expla- 
nations are  at  variance  with  grammar.  For  compounds  which 
are  formed  by  a  verbal  stem  with  noun  following  are  objective 
compounds,  viz.  the  noun  is  governed  by  the  verb.*  The  meaning 
of  iBXeirffda(V«>y,  therefore,  would  be,  not  6  p\€w6^€yos  imo  dalfiovosy  but 
6  pXtnov  daifjLopa.  The  first  part  of  Eustathius'  gloss  gives  us  the 
clue  to  its  significance.  Squinting  has  always  been  regarded  as 
one  of  the  surest  marks  whereby  to  recognize  the  * jettatore.' * 
This  much  once  granted,  and  fiXtntdaifnop  is  easily  explained  as 
*  he  who  looks  the  demon.'  *  Many  circumstances  prove  the  *evil 
eye '  to  have  been  regarded  as  the  action  of  a  demon.  We  need 
not  go  back  to  Demokritos'  c(d»Xa  as  the  cause  of  vision,  or  to 
the  clearly  protective  character  of  the  amulets  against  the  evil 
eye.  Foremost  among  the  marks  of  the  fiacKwos  is  the  miros  in 
his  eye';  that  is,  the  figure  of  a  horse,  believed  to  be  discernible 
in  his  pupil.  O.  Jahn,  it  is  true,  thinks  this  due  to  the  confusion 
with  the  name  of  a  peculiar  eye-disease,®  a  theory  foreshadowing, 
on  a  kindred  field,  M.  Miiller's  mythological  'disease  of  the 
language.'  However,  matters  must  be  reversed.  The  very  name 
of  the  sickness  proves  that  its  origin,  too,  was  ascribed  to  the 
presence  of  a  horse-shaped  demon.*  Neither  is  it  a  far  cry  from 
this  explanation  of  pkfjrtdaifuop  to  the  second  part  of  the  passages 

'  Cp.  Laistner,  Rfttsel  der  Sphinx 

'  Eustath.,  p.  206,  27.  '  Hcsych.,  s.  v. 

*  Ktthner,  Gr.  Gramm.  I  33S,  5.  Cp.  also  H.  Osthoff,  Das  Verbum  in  der 
Nominalcomposition. 

'See  Tuchmann,  La  Fascination  in  Melusine,  IV  ff. 

•Cp.  *torva  yidens*  in  Latin. 

^  O.  Jahn,  Leipz.  Sitzgsber.  1855,  35.  For  modem  instances  see  Tuchmann, 
I.e. 

»L.  c,  p.35,  26. 

*0n  sickness,  incarnated  in  the  bodies  of  beasts,  cp.  Bienkowski,  Malocchio, 
in  Eranos  Vindob. 


Digitized  by  VjOOQ IC 


196  AMERICAN  JOURNAL    OF  PHILOLOGY. 

from  Eustathios  and  Hesychios ;  for  the  possessor  of  the  evil  eye 
is,  of  course,  himself  possessed  by  the  spirit/  and,  as  frequently 
happens  in  such  cases,  himself  harmed  to  some  degree  by  him. 

Index. 


Abstinence,  Menand.  inc.  544  K. 

acorn,  Nikochares  inc.  15  K.;  s.  cabbage. 

QiV^oXoff,  bird  of  bad  luck,  Alkaios  Ganym.  3  K. 

Akko,  Hermippos  Ath.  gon.  7  K. 

Amaltheia,  horn  of,  Philemon  Pter.  65  K. 

amulet,  Aristoph.  Ach.  964-965  (Gorgo) ;  Eupolis  Bapt.  87  K. 

Diphilos  Pall.  59  K.  (cunnus) ;  Menand.  Parakat.  387  K. 
(snake) ;  s.  bracelet. 

avakxnaiy  Magnes  Lyd.  4  K.;  s.  dreams ;  6y€ip6/iavn£» 

antipathy,  schol.  Aristoph.  Eq.  539 ;  s.  cabbage. 

ants,  gold-digging,  Eubulos  Glauk.  20  K. 

aphrodisiacs,  Xenarchos  Bupal.  i  K.;  Alexis  inc.  279  K. 

avoTp67raioy,  schol.  Ahstoph.  Ach.  284 ;  s.  amulet,  Gorgo,  evil  eye, 
squilla,  Ephesia  grammata. 

Artemis  =  Hekate,  Diphilos  inc.  124  K. 

asphalt,  in  purification,  Diphilos  inc.  126  K. 

astrology,  popular,  Damoxenos  Syntr.  2  K.;  Nikomachos  Eil. 
I  K.;  Sosipater  Kataps.  i  K.;  Menand.  inc.  531  K.  (doubt- 
ful); Philemon  Babyl.  16  K.  (doubtful). 

B. 

Babylonian,  as  astrologer,  Philemon  Babyl.  16  K.  (doubtful); 

s.  astrology. 
^(TKdpiov,  s.  evil  eye. 

ffatTKavos,  Aristoph.  Plut.  571 ;  Pherekrates  inc.  174  K.;  s.  evil  eye. 
bathing-water,  must  not  be  poured  out  at  the  door,  Aristoph. 

Heroes  306  K.;  s.  door;  s.  p.  191. 
birds,  and  barbarians,  Aristoph.  Av.  199-200;  1681 ;  Ran.  679- 
682. 
and  soul,  Aristoph.  Vesp.  49-51 ;  s.  soul, 
bracelet,  as  amulet,  Diphilos  Pall.  59  K.;  Menand.  Parakat.  387  K.; 
s.  amulet. 

'  O.  Jahn,  l.  c,  34. 


Digitized  by  VjOOQ IC 


SUPERSTITIONS  IN  GREEK  COMEDY.  igj 


Cabbage  and  wine,  schol.  Aristoph.  Eq.  539 ;  Eupolis  Bapt.  74  K.; 

Nikochares  inc.  15  K.;  Telekleides  Pryt.  27  K.;  Alexis  inc. 

286  K.;  Amphis  inc.  37  K.;  Anaxandrides  inc.  58  K.;  Eubulos 

inc.  127  K.;  s.  acorn;  antipathy, 
cathartes,  Aristoph.  Vesp.  1043 !  s*  incubus ;  sickness, 
cathartics,  Aristoph.  Pax  1250;  Vesp.  119. 

charm,  Aristoph.  Thesm.  430;  Antiphanes  Dipl.  86  K.;  Menand. 
Her.  213  K.;  inc.  559  K.,  702  K.;  monost.  313. 
and  counter-charm,  s.  moly. 
or  medicine,  Aristoph.  inc.  872  K.;  s.  ^kvt6kiov, 
produces  mania,  Aristoph.  Thesm.  561. 
charm-song,  Aristoph.  Ran.  1033  (s.  Musaios) ;  Amphiar.  29  K.; 

Daidal.  188  K.;  Anaxandrides  Od.  33,  12  K.;  Antiphanes 

Philotheb.  217,  15  K.;  com.  dub.  1206  K. 
Charon,  identified  with  death,  Antiphanes  Dipl.  86  K. 
xprjtrros  =  pure,  Antiphanes  inc.  272  K. 
Circe,  as  witch,  Aristoph.  Plut.  302. 

cock,  untimely  crowing  of,  ominous,  Com.  anon.  341  K.;  s.  omen, 
copper,  material  of  a  ring,  Aristoph.  Danaid.  250  K.;  s.  magical 

ring, 
cross-roads,  Eupolis  Dem.  120  K.;  Charikleides  Halys.  i  K.;  s. 

Hekate. 
crumbs  from  the  table,  Aristoph.  Heroes  305  K.;  s.  Heroes, 
cunnus,  Diphilos  Pall.  59  K.;  s.  amulet. 

D. 

Day,  lucky  or  unlucky,  Eupolis  Kol.  174  K.;  Menand.  Leuc.  315  K. 
fourth  of  the  month,  Ameipsias  inc.  28  K.;  Arist6nymos 
Hel.  rig.  4  K.;  Plato  Peis.  100  K.;  Sannyrion  Gel.  5  K. 
dead,  must  not  be  slandered,  Aristoph.  Pax  648-656 ;  Dionysios 
Soz.  6  K. 
intercourse  with  living,  Aristoph.  Dram.  278  K.;  s.  Mai- 
makterion. 
death  =  Charon,  Antiphanes  Dipl.  86  K. 

early  d.  is  lucky,  Menand.  Dis.  Exap.  125  K.  =  monost.  425. 
deiaecation,  to  drive  away  demons,  Alkaios  Ganym.  4  K.;  s. 

Nereid ;  s.  p.  192. 
demon,  sends  sickness,  Eupolis  Mar.  191  K. 

and  fate,  Alexis  Asotod.  25  K.;  Menand.  inc.  550  K.; 
Philemon  inc.  191  K. 


Digitized  by  VjOOQ IC 


198  AMERICAN  JOURNAL    OF  PHILOLOGY. 

demon  and  life,  Anaxandrides  Anchis.  4  K. 

?  AtoXaor,  a  spectre,  Kratinos  inc.  402  K. 

dio<njfitiat  Aristoph.  Ach.  170-17 1 ;  Nub.  579;  s.  omen. 

dog,  incarnation  of  Hekate,  Aristoph.  inc.  82  M. 

donkey,  Aristoph.  Av.  721 ;  s.  vvfjfioXoi. 

door,  sacred,  Menand.  inc.  740  K.  (oath). 

seat  of  souls,  Aristoph.  Her.  306  K.;  s.  bathing-water ;  s. 

p.  191. 

dreams,  significance  of,  Aristoph.  Vesp.  24-25 ;  Kratinos  inc.  363 

K.;  Alexis  Cithar.  103  K.,  inc.  272  K.;  Menand.  inc. 

534>  734  K.;  Com.  anon.  185  K. 

interpreted,  Aristoph.  Eq.  809 ;  Vesp.  52-53 ;  s.  dpaXvrai, 


Earthquake,  ominous,  Aristoph.  Eccles.  791-793;  s.  fire,  yaXij. 
east,  in  purification,  Kratinos  Cheir.  232  K. 
eclipse,  ominous,  Aristoph.  Nub.  584-586. 

ccpcorifloin},  kept  in  the  house,  Aristoph.  Vesp.  398-399 ;  cp.  schol. 
Eq.  720. 

Aaiofuxyreca,  SChol.  AristOph.  Ach.  1 1 28. 

Empusa,  Aristoph.  Eccles.  1056-1057 ;  Ran.  288-296. 

=  Hekate,  Aristoph.  Tagen.  500-501  K. 
Ephesia  grammata,  Anaxilas  Lyrop.  18  K.;  Menand.  Paid*  371  K.; 

s.  wedding. 
Eudemos,  vendor  of  magical  implements,  Aristoph.  Plut.  883-884 ; 

Ameipsias  inc.  27  K. 
eunuch,  foreboding  ill.  Com.  an.  350  K.;  s.  crvfipoXoi. 
Eurykles,  ventriloquist,  Aristoph.  Vesp.  1019. 
evening,  time  of  spectres,  Aristoph.  Av.  1484-1489;  s.  spectres, 
evil  eye,  Aristoph.  Eq.  103-104,  inc.  592  K.;   Pherekrates  inc. 

174  K.;  Alexis  Troph.  238  K.;  Antiphanes  Did.  80,  8  K.; 

Misop.  159  K.;  Dionysios  inc.  7-1 1  K.;  Timokles  Synerg. 

31   K.;   Menand.  inc.  540  K.;    Nikomachos  Naum.  2  K.; 

Philemon  inc.  131  K.;  Com.  anon.  85  K.,  160  K.,  359  K.; 

s.  ^cTKayos,  squinting ;  s.  p.  195. 

F. 

Fig-tree,  wood  of  bad  luck,  Com.  anon.  7  K.;  s.  portent,  sixteen, 
fire,  ominous,  Aristoph.  Eccles.  792 ;  s.  earthquake. 

ordeal  by,  Aristoph.  Lys.  133-134. 

destroys  magical  power,  Aristoph.  Telmes.  532  K.;  s.  p.  192. 


Digitized  by  VjOOQ IC 


SUPERSTITIONS  IN  GREEK  COMEDY,  1 99 

flatfish,  fairy-tale  about  its  origin,  Aristoph.  Lys.  115-116. 
footwashing,  Aristoph.  dub.  914  K.;  s.  right  and  left. 
?  foundation-sacrifice,  Aristoph.  Danaid.  245  K. 
four,  in  burial  ritual,  Aristoph.  Eccles.  1031 ;  s.  vine. 

fourth  day,  s.  day. 
furnace,  protection  of,  Aristoph.  inc.  592  K.;  s.  evil  eye. 

G. 

Vakr\,  ominous,  Aristoph.  Eccles.  792 ;  s.  omen,  cvfiPoXoi. 

and  loss  of  speech  ?,  Aristoph.  inc.  664  K. 
yaarpofULVTtiai  S.  Kurykles. 
yXovieor  (fish),  ominous,  Nausikrates  Naucl.  i.  2  K.* 

and  magic,  Nausikrates  Naucl.  i.  2  K.;  s.  omen, 
Persephone,  Sicily,  rptyXij. 
gods,  their  appearance  accompanied  by  light,  Aristoph.  Av.  1709- 

1713. 
by  scent,  Aristoph.  Av.  1715- 
1716. 
Foi^rfff,  title  of  a  comedy  by  Aristomenes  5  K. 
Gorgo,  on  a  helmet,  Aristoph.  Ach.  964-965 ;  s.  amulet,  a'noTp6' 

natov, 

Gorgoneion,  Aristoph.  Pax  560-561 ;  Lys.  560;  s.  shield. 

H. 

Hades'  cap,  Aristoph.  Ach.  386-390;  s.  wizard. 

haruspicy,  Aristoph.  Telm.  540  K.;  Euphron.  Ad.  i  K.;  Theor. 

7K. 

Hekate,  her  chapel  in  every  house,  Aristoph.  Vesp.  805. 
dog-shaped,  Aristoph.  inc.  594  K. 
invocation  of,  Charikleides  Hal.  i  K.  (s.  cross-roads). 
=  Artemis,  Diphilos  inc.  124  K.  (s.  Artemis). 
=  Empusa,  Aristoph.  Tag.  500-501  K.  (s.  Empusa). 
title  of  comedies  by  Nikostratos  11.  12  K.;  by  Diphilos 
28  fr.  K. 
hellebore,  Aristoph.  Vesp.  1489;    Kallias  inc.  28  K.;    Diphilos 
Helleb.  31  K. 
from  Antikyra,  Diphilos  inc.  126  K. 
against  mania,  Menand.  Aul.  69  K. 
Hermes,  protects  from  evil,  Aristoph.  Plut.  1153-1154;  Philemon 

dub.  226  K. 
Heroes,  seat  of,  at  doors,  Aristoph.  Her.  306  K.;  s.  door. 


Digitized  by  VjOOQ IC 


200  AMERICAN  JOURNAL    OF  PHILOLOGY, 

Heroes  do  evil,  Menand.  Syneph.  459  K.;  Com.  anon.  257  K. 

punish  neglect,  Aristoph.  inc.  692  K. 

live  on  crumbs,  Aristoph.  Her.  305  K. 

=  souls,  Aristoph.  Her.  305.  306  K.;  Myrtilos  Titanop. 
2  K.;  s.  souls,  spectres. 

title  of  comedies  by  Aristoph.;  by  Chionides,  Kock,  v.  I, 
p.  4 ;  by  Philemon  30  K. 
Heros,  title  of  comedies  by  Diphilos  47  K.,  by  Menand.  209  ff.  K. 
Hieronymos,  Atheniaij  wizard  ?,  Aristoph.  Ach.  386. 
Hippalektryon,  sign  of  a  ship,  Aristoph.  Ran.  932-933. 

I.J. 

'lierti/off,  protection  against,  Aristoph.  Av.  500-503. 

incubus,  Aristoph.  Vesp.  1 037-1 042;  s.  cathartes;  sickness. 

Isis,  oath  by  I.  and  sickness,  Ophelion  inc.  6  K. 

ivy,  Aristoph.  Vesp.  480 ;  s.  oplyapop. 

iynx,  Aristoph.  Lys.  iiio-iiii. 

K. 

KtpKOfjMVTfla,  Aristoph.  Pax  1054-1055. 
KiBapos  (fish),  Pherekrates  Dulod.  39  K. 
Kkfjdmv,  Diphilos  inc.  100  K.;  s.  omen. 

KOfrKwoiMVTtiay  Aristoph.  Nub.  373  (doubtful) ;  Philippides  inc.  37  K. 
Kotytto.  Eupolis  Bapt.  83  K. 

Kybele,  her  priests  as  miraculous  healers,  Menand.  Hier.  245  K.; 
s.  p.  190. 

L. 

Lamia,  Aristoph.  Nub.  555-556 ;  Menand.  Plok.  403  K. 
her  5px<«i  Aristoph.  Pax  758  =  Vesp.  1035." 
breaks  wind  when  caught,  Aristoph.  Vesp.  1177;  Krates 

Lam.  18  K. 
title  of  comedy  by  Krates. 

lamp,  its  sputtering  prophesies  rain,  Aristoph.  Vesp.  260-263. 

lentils,  healing,  Menand.  inc.  530  K. 

lettuce,  makes  impotent,  Eubulos  Astyt.  14  K. 

light,  accompanies  the  appearance  of  gods,  Aristoph.  Av.  1709- 

1713- 
lightning,  ominous.  Com.  anon.  49  K. 

-stroke  makes  sacred,  Aristophon  latr.  3  K.;  Anaxippos 
Ker.  3  K. 


Digitized  by  VjOOQ IC 


SUPERSTITIONS  IN  GREEK  COMEDY,  201 

lightning-Stroke  and  perjury,  Antiphanes  inc.  233  K. 

-tower  (legend),  Aristoph.  Av.  1537-1538;  cp.  Aeschyl. 
Eum.  812-814. 
love-charm,  Aristoph.  Ach.  1065-1066;  Alexis  Mandr.  141   K. 

(s.  mandragora);  Menand.  inc.  259  bM.,  646  K. 
Xv;^Waff,  Plato  Soph.  146  K. 

M. 

Magic,  evil  ?,  Aristoph.  inc.  793  K.  =  schol.  Vesp.  288. 

and    Samothracian    mysteries,  Aristoph.   Pax    277-279; 
s.  prayer, 
magical  book,  Aristomenes  Goet.  9  K. 

herbs  and  metamorphosis,  Aristoph.  A  v.  654-655. 
ring,  Aristoph.  Plut.  883-884 ;  Danaid.  250  K.;  Eupolis 
Bapt.  87  K.;  Antiphanes  Omph.  177  K.;  s.  copper, 
Eudemos,  Phertatos. 
magician,  Ameipsias  inc.  27  K.  (Eudemos) ;  Theopompos  Alth. 
2  K.;  Anaxandrides  Pharmakom.  81  K.  (doubtful) ; 
Antiphanes  Omph.  177  K.  (Phertatos);  Mnesima- 
chos  Pharmakop.  6  K.  (doubtful);  Diphilos  inc. 
126  K. 
social  estimation  of,  Alexis  Tarant.  222,  7  K. 
Maimakterion,  mouth  of  the  dead,  Aristoph.  Danaid.  278  K.; 

s.  dead, 
mandragora,  Alexis  Mandr.  141  ff.  K. 
mania,  produced  by  spectres  ?,  Menand.  Phas.  502  K. 
Man-etff,  title  of  comedy  by  Alexis  146  ff.  K. 
medicine,  popular,  Aristoph.  Acli.  862-863. 

and  charms,  Anaxandrides  Pharmakom.' 81  K. 
and  magic,  Menand.  inc.  530  K. 
Megara,  seat  of  magicians,  Theopomp.  Alth.  2  K. 
miraculous  cures,  Menand.  Hier.  245  K.;  s.  music, 
moly,  as  counter-charm?.  Com.  anon.  641  K. 
monkey,  forebodes  ill.  Com.  anon.  350  K.;  s.  crvfi^oXoi, 
moon,  influences  action,  Aristoph.  Ach.  83-84. 

drawn  down  by  magic,  Aristoph.  Nub.  749-752 ;  Menand. 
Thess.;  s.  Thessaly. 
Mormo,  Aristoph.  Av.  1245;  Eq.  693;  Pax  473-474;  Ran.  925; 

Krates  Her.  8  K. 
Mormolykeion,  Aristoph.  Thesm.  417 ;    Amphiar.  31  K ;    Ger. 
131  K. 


Digitized  by  VjOOQ IC 


202  AMERICAN  JOURNAL   OF  PHILOLOGY. 

mouse,  its  nibbling  ominous,  Com.  anon.  341  K.;  s.  omen. 

Musaios,  and  charm-songs,  Aristoph.  Ran.  1033. 

music,  subdues  the  gods,  Menand.  Hier.  245  K.;  s.  miraculous 

cures, 
ftvpoii^a,  its  relation  to  snakes.  Com.  anon.  219  M.  (doubtful). 

N. 

Name,  power  of,  Aristoph.  Ran.  298-300;  s.  p.  194. 
necromancy,  Aristoph.  Av.  1553-1564;  Krates  Her.  10  K.;  s.  p. 

193;  Alexis  Thespr.  89  K. 
Nereid,  as  evil  demon,  Alkaios  Gan.  4  K.;  s.  defaecation ;  s.  p. 

192. 
nursery-rhymes,  Aristoph.  Nes.  389  K.;  Strattis  Phoen.  46  K.;  s. 

p.  194. 

O. 

*OicvT«J#cia,  Aristoph.  Thesm.  504;  inc.  872  K.;  s.  charm. 

omen,  Aristoph.  Ach.  170-17 1 ;  Av.  720;  Eccles.  791-793;  Eq. 

24-29.  638-640;    Nub.   579  f.  584-586.   1 1 28-1 1 29; 

Vesp.  1086;  Hermippos  inc.  81  K.;  Theopompos  inc. 

74  K.;  Anaxandrides  Agr.  i  K.;  Nausikrates  Naufel. 

I.  2  K.;  Diphilos  inc.  100  K.;  Menand.  inc.  534  K.; 

Philemon  inc.  100  K.;  Com.  anon.  49  K.,  341  K.;  s. 

cock,  diooi^/ACUi,  yaX^,  yXavieof ,  Kkifiiav^  mOUSe,  Owl. 

'accipio,'  Eupolis  Dem.  119  K. 

absit,  Aristoph.  Av.  61 ;  Lys.  146-147 ;  Plut.  114-116. 359. 

855;  Vesp.  161.  535-536;  Anaxandrides  Agr.  i,  4  ft. 

K.;  Antiphanes  Omph.  177  K.;  Menand.  Deisid.  109 

K.;   Methe  321  K.     Cp.  Transact.  Am.  Phil.  Ass. 

XXVII  34. 
omission,  Myrtilos  Titanop.  2  K.;  s.  silence,  tombs. 

6vtipOfJMtfT€i£,  S.  dyaXvrai, 

onions,  Aristoph.  Eccles.  1091-1092;  Com.  anon.  484  K. 
6piyavoy,  Aristoph.  Eccles.  1030;   Ran.  602;  Vesp.  480;  s.  ivy; 

Anaxandrides  Pharmakom.  81  K. 
6ppieofjMVT€iaf  Aristoph.  Av.  593-601.  716-722 ;  Plut.  63. 
Orpheus,  as  healer,  Antiphanes  Orph.  180  K.  (very  doubtful); 

s.  magic  and  mysteries. 
owl,    Aristoph.   Vesp.  1086;    Menand.    inc.   534  K.;    s.    omen, 

ovfifioXoi. 


Digitized  by  VjOOQ IC 


SUPERSTITIONS  IN  GREEK  COMEDY.  203 


Petosiris,  Aristoph.  Dan.  257  K. 

*apfuiK6fiavTi9,  title  of  a  comedy  by  Anaxandrides  49  ff.  K. 

Phertatos,  manufacturer  or  vendor  of  magical  rings,  Antiphanes 

Omph.  177  K.;  s.  magician, 
phthisis,  sent  by  a  god,  Aristoph.  Vesp.  158-160. 
nofnrvcrfji6s,  Aristoph.  Plut.  732. 
iropdrf,  Aristoph.  Plut.  618. 

portent,  Plato  inc.  257  K.;  Com.  anon.  7  K.;  s.  fig-tree ;  sixteen, 
possessed,  Aristoph.  Vesp.  119;  Alexis  87  K.;  Menand.  Heaut. 
140  K. 
women  prophesy,  Menand.  Theoph.  223  ff.  K. 
prayer  of  fivarai  can  bind  the  steps  of  the  runner,  s.  magic  and 

mysteries, 
prostitutes,  as  witches?.  Com.  anon.  220  K. 
purification,  Kratinos  Cheir.  232  K.;  Araros  12  K.;  Diphilos  inc. 

126  K.;  Menand.  inc.  530  K. 
purity,  of  water,  Aristoph.  Amphiar.  32  K. 

nvpofuurrtia,  AristOph.  Pac.  IO26. 

R. 

Red,  in  miraculous  cures,  Aristoph.  Plut.  730-732. 
rejuvenation,  in  comedies  by  Philemon  8  K.;  by  Philippides  5  ff.  K. 
right  and  left,  Aristoph.  dub.  914  K. 
rumor,  Aristoph.  Av.  720. 


Salt,  in  medicine,  Menand.  inc.  530  K. 

sardonyx?,  Philemon  inc.  73  K.,  216  K. 

scent,  accompanies  appearance  of  gods,  Aristoph.  Av.  1715-1716 ; 

s.  gods. 
Scythia,  seat  of  witches?,  Xenarchos  Scyth.  12  K. 
sea-water,  in  purification,  Diphilos  inc.  126  K. 
servant,  meeting  a  s.  forebodes  ill,  Aristoph.  Av.  721 ;  s.  o-vfi/SoXoi. 
seven,  Aristoph.  Lys.  698. 

shield  :  device,  Aristoph.  Ach.  574  (Gorgoneion) ;  Lys.  560  (do.), 
shoe,  and  right  foot,  Aristoph.  dub.  914  K.;  s.  right  and  left. 
Sicily,  seat  of  wizards,  Nausikrates  Naukl.  i.  2  K.;  s.  yXawot. 
sickness,  as  demon,  Aristoph.  Vesp.  1037-1042 ;  s.  incubus, 
healed  by  charm-song,  Aristoph.  Ran.  1033. 


Digitized  by  VjOOQ IC 


204  AMERICAN  JOURNAL    OF  PHILOLOGY. 

sickness,  sent  by  demons,  Eupolis  Mar.  191  K. 

caused  by  witchcraft,  Xenarchos  Scyth.  12  K. 
silence,  Myrtilos  Titanop.  2  K.;  s.  omission,  tombs, 
sixteen,  Com.  anon.  7  K.;  s.  fig-tree. 

snake,  healing,  Aristoph.  Plut.  733-736 ;  Kratinos  Troph.  225  K. 
(doubtful), 
as  amulet  on  bracelet,  Menand.  Par.  387  K. 
and  stick.  Com.  anon.  486  K.;  cp.  Transact.  Am.  Phil.  Ass. 
XXVI  53. 
sneezing,  ominous,  Aristoph.  Av.  720;    Menand.  inc.  534  K.; 

Philemon  inc.  100  K. 
soul,  and  star,  Aristoph.  Pax  832-833. 

as  bird,  Aristoph.  Vesp.  49-51 ;  s.  Heroes, 
spectres,  size  of,  Krates  Her.  1 1  K. 

strike  men,  Aristoph.  Av.  1490-1493. 
evening  their  time,  Aristoph.  Av.  1484-1489. 
hover  in  or  over  the  coffin.  Com.  anon.  1151  K. 
title  of  comedies  by  Menand.  501  if.  K.;  by  Philemon 
84  K.;  by  Theognetos,  Kock,  v.  IH,  364. 
spitting,  Aristoph.  Pax  528. 
squill,  buried  ^t  the  door-sill,  Aristoph.  Dan.  255  K. 

purifying,  Kratinos  Cheir.  232  K.;  Diphilos  126  K. 
squinting,  sign  of  evil  eye.  Com.  anon.  85  K.;  s.  p.  195. 
star,  and  soul,  Aristoph.  Pax  832-833. 
storm,  sacrifice  to,  Aristoph.  Ran.  847-848. 
sulphur,  used  in  purification,  Araros  Kamp.  12  K.;  Diphilos  inc. 
126  K. 
used  in  healing,  Menand.  inc.  530  K. 
superstition,  castigated  in  Aristoph.  Amphiaraos  and  in  Menand. 

Deisidaimon;  s.  p.  190. 
superstitious  (^oXoOf  Theopomp.  Tisam.  61  K.;  Menand.  Debid. 

112  K.;  Misog.  326  K.,  inc.  601  K.,  1046  K.;  s.  p.  189  if. 
swallows,  language  of,  Nikostratos  27  K. 

arv/ui/3oXoi,  Aristoph.  Av.  721 ;   Eccles.  792 ;  Alkaios  Gan.  3  K.; 
Menand.  inc.  534  K.;  Philemon  inc.  100  K.;  Com.  anon.  350  K. 


Table,  three-legged,  in  witchcraft,  Aristoph.  Telm.  530  K.;  s.  p. 

192. 
Telmessians,  title  of  comedy  by  Aristoph.  528  if.  K. 
Thessalian,  title  of  comedy  by  Menander  229  if.  K. 


Digitized  by  VjOOQ IC 


SUPERSTITIONS  IN  GREEK  COMEDY,  20$ 

Thessaly,  seat  of  witches,  Aristoph.  Nub.  749  ff.;  s.  moon. 

theurgy,  Menand.  Hier.  245  K. 

three,  feet  of  a  table,  Aristoph.  Telm.  530  K. 

the  divine  number,  Antiphanes  Myst.  165  K. 

calls  of  the  dead,  Aristoph.  Ran.  1175-1176. 

springs,  water  from,  Menand.  inc.  530  K. 

third  wave  the  highest,  Menand.  inc.  536  K. 
threshold,  protection  of,  Aristoph.  Dan.  255  K.;  s.  squill, 
thunder,  makes  idiotic,  Aristoph.  Eccles.  793;  Menand.  Georg. 

100  K.;  Philemon  Moich.  44  K.;  Com.  anon.  965.  995  K. 
tombs,  passed  in  silence,  Myrtilos  Titanop.  2  K.;  s.  omission, 
torch,  in  purification,  Diphilos  inc.  126  K. 

rptyXi;  (fish),  sacred  to  Hekate,  Plato  Phaon  19.  20  K.;  Charikleides 
Halys.  I  K. 
to  Persephone,  Nausikrates  Naukl.  i.  2  K.;  s.  yXovKor. 

V. 

Vine,  tendrils  used  in  prothesis,  Aristoph.  Eccles.  1031 ;  s.  four. 

W. 

Water,  against  demoniacal  diseases,  Aristoph.  Vesp.  119. 

and  dreams,  Aristoph.  Ran.  i33S-i34a 

from  three  springs,  Menand.  inc.  530  K. 
wedding,  and  Ephesia  grammata,  Menand.  Paid.  371  K. 
white,  lucky  color,  Menand.  Leuc.  315  K.;  s.  day. 
wind-eggs,  Aristoph.   Daid.   185.  186  K.;    Plato  Daid.  19  K.; 

Araros  Caen.  6  K.;  Menand.  Dact.  104  K. 
witch,  s.  prostitute,  Thessaly,  women. 

and  sickness,  Xenarchos  Scyth.  12  K. 
witchcraft,  Aristoph.  Thesm.  534 ;  Menand.  inc.  535  K. 
wizard,  Aristoph.  Ach.  386 ;  s.  Hades,  Hieronymos. 
women,  as  witches,  s.  Thessaly,  prostitute. 

as  miraculous  healers,  Menand.  inc.  530  K. 
wood  of  unlucky  trees,  Eupolis  Dem.  120  K.;   s.  cross-roads; 

Com.  anon.  7  K.;  s.  fig-tree, 
word,  power  of,  Aristoph.  A  v.  646-647 ;  cp.  KKifiw\  s.  prayer. 

Ernst  Riess. 


Digitized  by  VjOOQ IC 


IV.— ON  THE  DEFINITION  OF  SOME  RHETORICAL 

TERMS. 

During  a  course  of  reading  in  the  Latin  writers  on  rhetoric  I 
collected  the  following  words  and  definitions,  either  omitted  or 
incorrecdy  given  in  our  dictionaries.  The  dictionaries  consulted 
were  those  of  Lewis  and  Short ;  White  and  Riddle,  3d  edition ; 
Forcellini,  English  edition  by  J.  Bailey;  and  Georges  Hand- 
worterbuch,  7th  edition. 

Cornificius  and  Cicero  are  cited  according  to  the  recension  of 
Friedrich,  the  minor  rhetoricians  according  to  Halm's  Rhetores 
Latini  Minores. 

Adfictio,  paronomasia.  lul.  Ru£  de  Schem.  Lex.  Halm,  p. 
51 :  Uapovofui<ria  est  secundum  praedictum  verbum  positio  alterius, 
ipso  poscente  sensu,  ut  apud  Terentium : 

Nam  inceptio  est  amentium,  hand  amantium. 

Latine  dicitur  adnominatio  vel  adfictio.    G.  alone  records  this 
word  and  refers  to  adfiominoHo. 

Antici  PATIO,  the  anticipation  of  an  opponents  argument  or 
objection;  a  translation  of  irpo/earoXij^cf.  De  Schem.  Dianoeas, 
Halm,  p.  60:  npoicaroXi^^iff  est  schema  dianoeas,  cum  id  quod 
adversarius  arrepturus  est  atque  objecturus,  praesumimus  ac 
praecipimus,  ut  illud : 

neque  me  Argolica  de  gente  negabo : 
Hoc  primum.     Et : 

Scio  me  Danais  e  classibas  unum 
Et  bello  Argolicos  fateor  petiisse  penates. 

Latine  haec  figura  dicitur  praeceptio  vel  anticipatio. 

Articulus,  asyndeton.  Ad  Her.  IV  19,  26:  Articulus  dicitur 
cum  singula  verba  intervallis  distinguuntur,  caesa  oratione,  hoc 
modo :  **  Acrimonia,  voce,  voltu,  [adversarios]  perterruisti."  Item : 
"Inimicos  invidia,  iniuriis,  potentia,  perfidia  sustulisti."  Corni- 
ficius limits  articulus  to  the  omission  of  conjunctions  between 
single  words,  and  uses  dissolutum  to  denote  the  absence  of 
conjunctions  between  clauses.    See  Ad  Her.  IV  30,  41.    L.  and 


Digitized  by  VjOOQ IC 


DEFINITION  OF  SOME  RHETORICAL  TERMS,  20/ 


Digitized  by  VjOOQ IC 


208  AMERICAN  JOURNAL    OF  PHILOLOGY. 

ac  contrarium  vertitur. .  .  .  Latine  conversio  dicitur.  Martianus 
Capella,  Halm,  p.  478 :  ' Ano<rTpo<l>^  est  in  aliquem  disiricta  conversio, 
frequens  apud  Ciceronem  ac  nobilis  figura. 

CoNVERSUM,  a  figure  of  speech  in  which  the  last  word  of  a 
clause  is  repeated  at  the  end  of  succeeding  clauses.  Aquila 
Romanus,  Halm,  p.  33:  dvTKrrptxfiri,  conversum.  Species  huius 
figura  cum  eandem  fere  vim  habeat,  contraria  est  superior!,  eo 
quod  ibi  ab  eadem  parte  orationis  saepius  incipitur,  hie  in  eandem 
partem  desinitur.  An  example  from  pro  Fonteio  4,  8  follows. 
Cornificius,  Ad  Her.  IV  13,  19,  calls  this  figure  conversion  a  name 
which  is  also  used  by  Cicero  and  Quintilian. 

Defectus,  the  omission  of  a  word;  ordinarily  denoted  by 
detractatio,  Schem.  Dian.  Halm,  p.  75 :  "^ExXcc^if ,  defectus.  Ver- 
gilius :  **  Haec  secum  ";  deest  loquitur. 

Denominatio,  paronomasia.  Schem.  Dian.  Halm,  p.  75: 
UapovofAatria  est  denominatio,  quae  similitudinem  verbi  conflectit 
ad  auditoris  affectum.  Cicero  in  invectivis :  Qui  de  huius  urbis 
atque  adeo  de  orbis  terrarum  exitio  cogitant  (Catil.  I  9),  et  Ter- 
entius:  Nam  increpatio  est  amentium,  haut  amantium  (And.  218). 
Terence  wrote  inceptiost. 

DepreCatio,  the  reply  of  the  defendant.  Ad  Her.  I  11,  18: 
Constitutio  est  prima  deprecatio  defensoris  cum  accusatoris  insi- 
mulatione  coniuncta.  Cic.  de  In.  I  10,  13:  Atque  hoc  eodem 
urgebitur,  sive  constitutionem  primam  causae  accusatoris  confir- 
mationem  dixerit  sive  defensoris  primam  deprecationem ;  nam 
eum  eadem  omnia  incommoda  sequentur.  Q.  Ill  6,  13:  Alii 
statum  crediderunt  primam  eius,  cum  quo  ageretur,  depreca- 
tionem. Forcellini  says  simply :  Interdum  simpliciter  pro  depul- 
sione  sine  precibus,  and  quotes  Cic.  pro  Rab.  9,  26 :  Huic  quidem 
afferet  aliquam  deprecationem  periculi  aetas  ilia  qua  turn  fuit,  and 
Quintilian.  While  in  both  passages  the  meaning  of  the  word 
corresponds  to  his  definition,  it  has  not  the  same  meaning  in  the 
latter  passage  as  in  the  former,  inasmuch  as  Q.  used  it  in  a 
technical  sense.  The  other  lexicographers  overlook  this  meaning 
entirely. 

Detractatio,  a  mocking,  a  satirizing,  Schem.  Dian.  Halm, 
p.  75:  Aia<rvpfi6t  est  delusio  vel  detractatio,  cum  inludentes  ea 
quae  ab  adversariis  sunt  prolata  disolvimus,  ut  est  pro  Murena  in 
Sulpicium  de  iure  civile:  Quoniam  mihi  videris  istam  iuris 
scientiam  tanquam  filiolam  complecti  tuam. 

DisiUNCTUM,  a  constrtution  in  which  several  successive  clauses 


Digitized  by  VjOOQ IC 


DEFINITION  OF  SOME  RHETORICAL   TERMS.  2O9 

are  each  concluded  with  the  appropriate  verb.  Aq.  Rom.  Halm, 
p.  36 :  Ai€(€vy fitvov,  disiunctum.  Haec  figura  ita  oraat  et  amplificat 
orationem,  ut  diversis  redditionibus  verborum  membra,  quae 
vocamus  K&\a,  disiungat  ac  separet,  sive  duo  sive  plura,  hoc 
modo:  Capuam  colonis  deductis  occupabunt,  Atellam  praesidio 
communient,  Nuceriam,  Cumas  multitudine  suorum  obtinebunt, 
cetera  oppida  praesidiis  devincient.  Ad  Her.  IV  37 :  Disiunctum 
est,  cum  eorum,  de  quibus  dicimus,  aut  utrumque  aut  unum  quid- 
que  certo  concluditur  verbo,  sic :  Populus  Romanus  Numantiam 
delevit,  Karthaginem  sustulit,  Corinthum  disjecit,  Fregellas  evertit. 
The  usual  reading  in  the  last  example  is  disiunctio,  and  the 
passage  is  cited  under  that  word  in  the  dictionaries.  Friedrich 
reads  disiunctum,  and  this  seems  to  have  the  better  MSS  support. 
Disiunctum  would  therefore  appear  to  be  the  better  reading, 
supported  as  it  is  by  Aq.  Rom.  supra,  and  Mart.  Cap.  Halm,  p. 
482:  Ai€CtvyfjLivov  disiunctum  appellamus,  cum  diversis  redditio- 
nibus verborum  cola  disiungimus,  sive  duo  sive  plura,  hoc  modo : 
[same  ex.  as  Aq.  Rom.]. 

DiSPARSUM,  as  a  translation  of  tijipi\fUvov.  Carmen  de  Figuris, 
Halm,  p.  65 :  AijjpfjfAipov 

Disparsum  reddo,  quod  sparsum  uno  ordine  reddo. 
"  Ambo  lovis  merito  proles,  verum  ille  equitando 
Insignis,  Castor,  catus  hie  pugilamine,  Pollux." 

DissoLUTio,  asyndeton.  Quint.  IX  3,  50:  Et  hoc  autem 
exemplum  et  superius  aliam  quoque  efficiunt  figuram,  quae,  quia 
coniunctionibus  caret,  dissolutio  vocatur.  .  .  .  Hoc  genus  et 
ppaxvXoylav  vocant,  quae  potest  esse  copulata  dissolutio..  Contra- 
rium  id  est  schema  quod  coniunctionibus  abundat :  illud  acrvpdtrov, 
hoc  noXvavpdtTov  dicitur.  W.  and  R.  and  L.  and  S.  define  wa7it 
of  connection,  and  cite  this  passage.     Georges  defines  correctly. 

DiSTRiBUTio,  discourse  delivered  with  frequent  pauses.  Ad 
Her.  Ill  13,  23 :  Contentio  dividitur  in  continuationem  et  [in]  distri- 
butionem.  .  .  .  Distributio  est  [in  contentione]  oratio  . . .  frequens 
[cum  raris  et  brevibus]  intervallis  [acri  vociferatione].  Ad  Her. 
Ill  i4>  25 :  Cum  autem  contendere  oportebit,  quoniam  id  aut  per 
continuationem  aut  per  distributionem  faciendum  est,  ...  in  dis- 
tributione  [vocem]  ab  imis  faucibus  exclamationem  quam  claris- 
simam  adhibere  [oportet]  et  quantum  spatii  per  singulas  excla- 
mationes  sumpserimus,  tantum  in  singula  intervalla  spatii  consu- 
merc  [iubemur].  Ad  Her.  Ill  15,  27:  Si  contendimus  per  con- 
tinuationem ...  sin  contentio  fiet  per  distributionem,  porrectione 


Digitized  by  VjOOQ IC 


2IO  AMERICAN  JOURNAL    OF  PHILOLOGY. 

celeri  bracchii,  inambulatione,  pedis  dexteri  rara  supplausione, 
acri  et  defixo  aspectu  uti  oportebit, 

Divisio,  the  dilemma.  Ad'  Her.  IV  40,  52 :  Divisio  est  quae 
rem  semovens  ab  re  utramque  absolvit  ratione  subiecta,  hoc 
modo:  Cur  ego  nunc  tibi  quicquam  obiciam  ?  Si  probus  est,  non 
meruisti;  si  improbus,  non  commovebere.  Quint.  V  10,  64: 
Divisio  et  ad  probandum  simili  via  valet  et  ad  refellendum. 
Probationi  interim  satis  est  unum  habere,  hoc  modo :  ut  sit  civis, 
aut  natus  sit  oportet  aut  factus.  This  meaning  is  not  in  any  of 
the  dictionaries  unless  it  is  thought  to  be  sufficiently  covered  by 
the  definition  logical  or  rhetorical  division, 

EvACUATio,  a  refutation  of  arguments;  a  translation  of  ayao-Rcv^. 
De  Schem.  Dian.  Halm,  p.  61 :  'Ayaaxeui;  est  superiorum  proxima 
figura  qua  ab  adversariis  maxima  proposita  destruimus  ac  redar- 
guimus  velut  falsa  . . .  Latine  dicitur  destructio  vel  evacuatio. 
L.  and  S.,  citing  evacuatio  fideiy  Tert.  adv.  Marc.  4,  24,  star  the 
word. 

ExPEDio  in  the  sense  explain^  narrate  is  cited  in  prose  before 
Tacitus  only  Sail.  J.  5,  2 ;  Asin.  Pollio  ad  Fam.  X  33,  5.  Add 
Cornificius  ad  Her,  H  26,  42;  HI  20,  33;  IV  18,  26;  IV  54,  68. 
L.  and  S.  say  it  does  not  occur  in  this  sense  in  Cicero.  Krebs- 
Schmalz,  Antibarb.,  refers  to  Cicero  ad  Brut.  I  15,  i,  where  the 
reading  is  doubtful,  and  adds :  '* jedenfalls  ist  es  nicht  sicher  fiir 
Cicero  erwiesen."  It  is  found,  however,  Cicero  de  Or.  HI  66: 
Sed  ea  si  sequamur,  nullam  umquam  rem  dicendo  expedire 
possimus. 

ExQUisiTio,  a  figure  of  speech  consisting  of  question  ana 
answer.  Schem.  Dian.  Halm,  p.  74 :  *££cra(rfu>r  est  exquisitio,  cum 
res  complures  divisas  cum  interrogatione  exquirentes  singulis 
quae  conveniunt  applicamus,  ut  Cicero :  Quid  tandem  te  impedit  ? 
mosne  maiorum?  at  persaepe  etiam  privati  in  hac  re  publica 
perniciosum  hostem  morte  multarunt.    Aut  leges  etc. 

HoMOEON,  a  simile  in  which  the  resemblance  is  confined  to 
certain  parts  of  the  objects  compared.  lul,  Ruf.  Halm,  p.  44 : 
Homoeon.  Haec  figura  fit,  cum  ex  partibus  aliqua  similitudo 
colligitur,  ut  Vergilius : 

Sic  oculos,  sic  ille  manus,  sic  ora  tenebat. 

Etiam  in  actu  fit  homoeon,  ut  idem  Vergilius : 

Non  aliter,  quam  si  immissis  mat  hostibus  omnis 
Carthago  aut  antiqua  Tyros. 


Digitized  by  VjOOQ IC 


DEFINITION  OF  SOME  RHETORICAL  TERMS.  211 

Icon,  simile.  lul.  Ruf.  Halm,  p.  44 :  Icon  fit,  cum  perfectae 
formae  similes  conferuntur.  Vergilius:  Talis  Amyclaei  domitus 
PoUucis  habenis.  Cf.  Beda  de  Tropis,  Halm,  p.  618:  Icon  est 
personarum  inter  se  vel  eorum  quae  personis  accidunt  comparatio, 
ut :  Vidimus  gloriam  eius,  gloriam  quasi  unigeniti  a  patre.  W. 
and  R.  and  L.  and  S.  do  not  recognize  this  word  in  a  rhetorical 
sense.  Georges  gives  it,  citing  Apuleius  min.  de  Nota  Aspira- 
tionis.    Forcellini  cites  for  it  Diomedes. 

Iniunctum,  a  zeugma  in  which  several  phrases  depend  upon  a 
common  verb  at  the  end.  Aquila  Romanus,  Halm,  p.  36 :  'YTreCevy- 
fifVoy,  iniunctum. .  . .  Quale  est  hoc :  Quorum  ordo  ab  humili, 
fortuna  a  sordida,  natura  a  turpi  oratione  abhorret.  Hoc  enim 
postremum  abhorret  ad  tria  refertur.  Cf.  Mart  Cap.  Halm,  p. 
482 :  ^KjfTtCtvytiivov  iniunctum.  Haec  figura  a  superiore  hoc  differt 
etc.  [same  ex,  as  above].    The  reading  ^ hyrtCtvyiUvov  is  doubtful. 

Interrogatum,  interrogation  (rhet.  fig.).  Aquila  Romanus, 
Halm,  p.  25 :  'Ep^n^/ia,  interrogatum.  Eo  utimur  ubi  exacerbando 
aliquid  interrogamus  et  augemus  eius  invidiam,  hoc  modo :  Fuis- 
tine  ne  illo  in  loco  ?  dixistine  haec  ita  gesta  esse  ?  renuntiastine  ea 
quibus  decepti  sumus  ? 

Interruptio,  parenthesis.  Jul.  Ruf.  de  Schem.  Lex.  Halm, 
p.  51 :  UaptpOtaig  est,  cum  ordinata  ac  legitima  sententia  interrum- 
pitur  per  alienum  extrinsecus  diversamque  sententiam,  ut :  ... 
Latine  haec  figura  dicitur  interruptio  vel  interiectio. 

Oppositum,  antithesis.    Carm.  de  Fig.  Halm,  p.  64:  'Am^croy 

Oppositum  dico,  contra  cum  opponimu'  quaedam. 
*'  Doctor  tute,  ego  discipulus ;  tu  scriba,  ego  censor ; 
Histrio  tu,  spectator  ego ;  adque  ego  sibilo,  tu  exis." 

Permutatio,  transposition  (rhet.  fig.).  Carmen  de  Figuris, 
Halm,  p.  64  •  'AvTifuraPoK^ 

Permutatio  fit  vice  cum  convertimu'  verba. 
"  Sumere  iam  cretos,  non  sumptos  cemere  amicos. — 
Quod  queo,  tempus  abest;  cui  tempus  adest,  nequeo,  inquit." 

This  is  noted  only  by  Georges.  It  is  used  in  another  sense  by 
Cornificius  ad  Her.  IV  34,  46 :  Permutatio  est  oratio  aliud  verbis 
aliud  sententia  demonstrans.  Ea  dividitur  in  tres  partes :  similitu- 
dinem,  argumentum,  contrarium.  Per  similitudinem  ...  sic: 
"Nam  cum  canes  fungentur  officiis  luporum  quoinam  praesidio 
pecua  credemus  ? "  Per  argumentum  . . .  ut  siquis  Drusum 
"Gracchilm  nitorem  obsoletum"  dicat.    Ex  contrario  ducitur  sic. 


Digitized  by  VjOOQ IC 


212  AMERICAN  JOURNAL    OF  PHILOLOGY. 

utsiquis  hominem  prodigumet  luxuriosum  [imprudens]  ''parcum 
et  diligentem"  appellet.  It  is  defined  by  the  lexicographers  as 
follows :  Forcellini  quotes  the  first  two  sentences  of  Cornificius' 
definition.  White  and  Riddle:  "An  exchanging  of  one  expres- 
sion for  another ;  permutation."  Lewis  and  Short:  "A  substitu- 
tion of  one  expression  for  another,  permutation."  Georges :  "die 
Vertauschung  der  Ausdriicke."  None  of  these  definitions  con- 
veys any  clear  idea  to  the  mind.  From  an  examination  of  the 
examples  we  first  arrive  at  a  definite  idea  of  what  Cornificius 
meant.  I  would  define  as  follows :  Allegory  in  its  broad  sense, 
including  enigma  and  irony, 

Praeceptio,  the  anticipation  of  an  opponents  argument  or 
objection;  a  translation  of  frpoieordXiT^ir.  De  Schem.  Dian.  Halm, 
p.  60 :  npoieardXi;^cr  est  schema  dianoeas,  cum  id  quod  adversarius 
arrepturus  est  atque  obiecturus,  praesumimus  ac  praecipimus,  ut 

iUud: 

neque  me  Argolica  de  gente  negabo : 
Hoc  primum. 

Latine  haec  figura  dicitur  praeceptio  vel  anticipatio. 

Principium,  a  kind  of  exordium,  the  direct  beginning,  opp.  to 
insinuation  Ad  Her.  I  4,  6 :  Exordiorum  duo  sunt  genera :  prin- 
cipium, quod  Graece  npoolfuov  appellatur,  et  insinuatio,  quae  tijxfdos 
nominatur.  Principium  est  cum  statim  animum  auditoris  nobis 
idoneum  reddimus  ad  audiendum.  Cic.  de  In.  I  15,  20:  Igitur 
exordium  in  duas  partes  dividitur,  principium  et  insinuationem. 
Principium  est  oratio  perspicue  et  protinus  perficiens  auditorem 
benevolum  aut  docilem  aut  attentum.  Quint.  IV  i,  42:  Eo  qui- 
dam  exordium  in  duas  dividunt  partes,  principium  et  insinua- 
tionem. 

Pronuntiatio,  as  a  figure  of  speech,  a  translation  of  {moKpiois. 
De  Schem.  Dian.  Halm,  p.  61 :  'Yfrdie/Ko-ir  est  figura  sententiae, 
cum  adversarium  gestu  et  pronuntiatione  extollimus  vel  abicimus 
et  spernimus,  ut  in  illo : 

Non  ego  Daphnim 
ludice  te  metuam. 

Scilicet  mibi  iniquus  es  et  non  recte  iudicaturus.  Latine  dicitur 
pronuntiatio. 

Regressio,  the  repetition  of  the  last  word  of  a  clause  or  verse 
as  the  first  word  of  the  next.    lul.  Ruf.  Halm,  p.  50:  naXcXoyca  est 


Digitized  by  VjOOQ IC 


DEFINITION  OF  SOME  RHETORICAL  TERMS, 


213 


cum  verbum,  quod  in  prima  sententia  est  ultimum,  in  sequente 
est  primum,  ut :  • 

Pierides :  vos  haec  facietis  maxima  Gallo, 
Gallo,  coins  amor  tantum  mihi  crescit  in  horas 

Latine  dicitur  regressio.  The  dictionaries  give  regressio  only  = 
fv6»t)bw  from  Quint.  IX  3,  35  and  lul.  Ruf.  de  Schem.  Lex.  §19 
(in  Halm  §21). 

SoLUTUM,  asyndeton,  Aquila  Rom.  Halm,  p.  35:  Solutum ; 
sic  enim  voco  quod  davvdcroy  Graeci  vocant.  Mart.  Cap.  Halm, 
p.  482:  'Ao-vvdcroy  est  solutum,  cum  demptis  coniunctionibus 
quibus  verba  aut  nomina  conectuntur  etc. 

Traductio,  the  use  of  words  of  like  form  btU  different  in 
meaning.  Ad  Her.  IV  14,  21 :  Ex  eodem  genere  est  exomationis 
cum  idem  verbum  ponitur  modo  in  hac,  modo  in  altera  re,  hoc 
modo :  '  Cur  eam  rem  tam  studiose  curas,  quae  tibi  multas  dabit 
curas?'  Item :  Nam  amari  iucundum  sit,  si  curetur,  ne  quid  insit 
amari.  Item :  *  Veniam  ad  vos,  si  mihi  senatus  det  veniam.'  The 
name  traductio  is  also  given  to  the  repetition  of  the  same  word  in 
the  sentence,  and  this  is  the  only  definition  given  by  the  lexi- 
cographers. 

TrANSITUS,    as    a    translation    Ol   /icraoraac?    or    fitrafiaai^.     De 

Schem.  Lex.  Halm,  p.  54:  Mrrdaraait  est  vel  fAtrdfiaais,  cum  a 
loquentis  persona  ad  personam  aliam  transitum  facimus,  ratione 
aliqua  vel  affectu,  ut : 

Non  haec  Evandro  de  te  promissa  parenti 
Discedens  dederam. 


Haec  figura  dicitur  variatio  aut  transitus. 


WisTntM  Rbsbhvb  Unxvbksitt, 

CLBVSL4MD. 


V.  J.  Emery. 


Digitized  by  VjOOQ IC 


REVIEWS  AND   BOOK  NOTICES. 

Beitrage  zur  historischen  Syntax  der  griechischen  Sprache,  herausgegeben  von 
M.  SCHANZ.  Band  III,  Heft  3  u.  4.  Geschichte  des  Pronomen  Reflexivum, 
von  Dr.  Adolf  Dyroff.  Erste  Abteilung,  Von  Homer  bis  zur  attischen 
Prosa.  Zweite  Abteilung,  Die  attische  Prosa  und  Schlussergebnisse. 
WUrzburg,  1892  u.  1893. 

The  stem  sve-  (parallel  form  seve-)  originally  had  a  signification  which  was 
nearly  identical  with  that  of  the  English  *  self.*  It  seems  that  this  stem  was 
not  specially  invented  for  reflexive  purposes,  but  that  it  had  a  wider  scope, 
which  was  gradually  narrowed  down  to  that  of  a  pure  reflexive.  The  conclu- 
sion of  this  narrowing  process,  or,  in  other  words,  the  emergence  of  a  distinct 
reflexive,  antedates  the  period  of  the  breaking  up  of  the  original  mother- 
tongue.  But  whilst  the  reflexive  signification  of  the  stem  sve-  can  readily  be 
proved  for  the  sister- tongues  of  the  Greek,  it  is  a  remarkable  fact  that  in  the 
oldest  documents  of  the  Greek  language — the  Homeric  poems — the  prenoun 
of  the  3d  person  is  essentially  anaphoric.  Still,  there  are  traces  of  the 
reflexive  use  of  the  substantive  pronoun  even  in  Homer,  and  these,  together 
with  the  exclusively  reflexive  use  of  the  adjective  forms  (df,  ^(5f,  etc.),  point  to 
the  reflexive  nature  of  the  pronoun  in  the  pre-Homeric  language.  The  other 
dialects  agree  with  the  Homeric  in  the  reflexive  use  of  the  pronominal  adjec- 
tive, the  rare  non-reflexive  use  of  individual  forms  belonging  to  a  much  later 
date.  In  Attic,  the  substantive  pronoun  is  undoubtedly  reflexive,  and  there 
are  indications  that  this  was  true  also  of  the  other  dialects.  Even  in  Homer 
the  parallel  form  hi  is  a  reflexive.  Furthermore,  the  plural  forms  ff^oc  and 
ai^epog,  which  are  derived  from  the  non-reflexive  forms  <t^  and  at^iv,  and  the 
plural  forms  of  the  substantive  pronoun  were  originally  reflexive,  and  this 
cannot  be  otherwise  explained  than  on  the  supposition  that  the  singular  forms 
(that  is,  what  became  the  singular  forms  after  the  differentiation  of  the 
numbers),  to  which  the  above-mentioned  plural  formations  attached  them- 
selves, must  have  been  likewise  originally  reflexive.  Finally,  the  substitutes 
for  the  simple  reflexive  pronoun  and  the  forms  by  which  it  was  supplanted  in 
various  dialects,  show  the  correctness  of  the  theory  of  the  reflexive  nature  of 
the  simple  pronoun  of  the  3d  person  in  the  pre-Homeric  language.  For,  not 
to  mention  other  attempts  that  were  unsuccessful,  it  was  this  very  pronoun 
that  was  used  in  juxtaposition  with  a  following  avr6^^  or  else  merged  with 
avrd^^  to  form  either  complex  or  compound  reflexive  forms.  Neither  does  the 
existence  of  similar  complex  and  compound  reflexive  forms  of  the  1st  and  of 
the  2d  person  weaken  the  force  of  this  argument,  for  these  forms,  when 
reflexive,  are  merely  analogical  formations,  built  after  the  pattern  of  the  forms 
of  the  3d  person. 

After  this  preliminary  statement  regarding  the  origin  and  nature  of  the 
Greek  reflexive  pronoun,  there  follows  a  historical  survey  of  the  use  of  the 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES.  21$ 

reflexive  in  the  various  departments  of  Greek  Literature  from  Homer  down 
to  Attic  Prose,  information  being  given  as  to  the  forms  of  the  simple  pronoun 
of  the  3d  person,  the  use  of  the  complex  {p^v  avrov^  etc.)  and  of  the  compound 
{iavTov,  etc.)  reflexive  of  the  3d  person,  the  limitation  of  the  pronominal 
adjective,  the  scope  of  reflexives  of  the  ist  and  2d  persons,  including  the 
possessive  adjective,  the  nature  and  degrees  of  reflexion  and  the  free  use  of 
the  reflexive. 

Homer. 

The  language  of  Homer  is  characterized  by  a  wealth  of  pronominal 
forms  of  the  3d  person.  In  the  first  place,  there  are  two  stems,  the  one  a 
demonstrative  fuv  (495  ^)  and  the  other  a  reflexive.  The  reflexive  is  split  up 
into  two  branches,  the  one  a  dissyllabic  stem  represented  by  the  forms  iol  (2), 
ii  (2)  and  i6g  (90),  and  the  other  a  monosyllabic  stem  with  differentiated  forms 
for  the  plural  and  represented  by  the  forms  io  (11),  61  (753),  e  (73),  a<j>iuv  (4), 
a<^iat{v)  (48),  a<^ac  (aa),  8f  (ao6)  and  ff^<Jf  (la).  Furthermore,  some  of  the  cases 
have  duplicate  forms,  belonging  to  diff'erent  periods  of  the  language.  So  to 
has  by  its  side  an  older  flo  (2),  a  younger  ev  (5)  and  an  ablatival  €6ev{i'j); 
a^uv  has  by  its  side  a<^iuv  (i)  and  of^v  (2) ;  <T^tm(v)  has  the  parallel  form  a<^i{y) 
(141) ;  with  a^aq  is  coupled  a^c  (i),  and  with  the  possessive  cr^r,  the  form 
a^repog  (9).  a^epog  was  perhaps  a  dual,  which  number  is  certainly  repre- 
sented by  the  substantive  forms  a<puiv  (8),  and  a^k  (5)  with  parallel  form  a^  (4). 

The  reflexive  has  lost  its  strong  force  of  'self.'  Hence  the  pronoun  is 
sometimes  strengthened  by  a  postpositive  avrdq,  rarely  by  a  prepositive  avrdq. 
This  combination  is  used  preferably  when  the  pronoun  is  used  reflexively,  but 
in  all  the  20  instances  of  the  reflexive  combination  there  is  no  governing 
preposition,  and  avT6q  directly  follows  the  personal  pronoun  and  forms  a 
complex  with  it.  The  adjective  pronoun  is  also  occasionally  accompanied  by 
a  genitive  of  amdq^  there  being  three  instances  of  k(f»  avrov,  one  of  ^  avrov  and 
one  of  airrap  at^ipyaiv. 

The  Homeric  epos  lacks  a  general  substantive  reflexive,  but  the  adjective 
pronoun  of  the  3d  person  refers  in  four  instances  to  the  1st  person  and  in  five 
to  the  2d.  As  this  use  belongs  to  the  period  preceding  the  time  of  the 
differentiation  of  the  numbers,  it  is  only  the  forms  e6q  and  5c  that  are  thus 
found.  In  but  one  instance — to  wit,  A  142 — does  8f  stand  for  the  possessive 
plural — that  is,  it  is  equivalent  to  iftirepoc,  or  rather  af^irepog.  It  is  principally 
in  stereotyped  expressions  that  the  general  use  of  the  reflexive  has  been 
preserved.  The  reason  why  the  usage  occurs  only  in  the  case  of  the  adjective 
pronoun,  and  not  in  the  case  of  the  substantive,  is  due  to  the  fact  that  in 
Homer  the  substantive  pronoun  is  almost  exclusively  anaphoric,  while  the 
adjective  remains  strictly  reflexive. 

At  this  stage  of  the  epic  literary  language,  as  has  been  stated,  the  simple 
pronoun  of  the  3d  person  is  essentially  anaphoric,  for,  with  but  two  exceptions, 
the  direct  reflexive  use  is  found  only  in  prepositional  phrases,  and  these 
phrases,  as  is  shown  by  the  frequent  preservation  of  the  effect  of  the  original 
initial  of  in  lengthening  a  preceding  syllable,  belong  to  a  period  that  antedates 
the  main  bulk  of  the  Homeric  poems. 

1  This  and  the  following  numbers  in  parentheses  indicate  the  number  of  occurrences  of  the 
form  in  Homer. 


Digitized  by  VjOOQ IC 


2 16  AMERICAN  JOURNAL    OF  PHIL  OLOGY. 

The  personal  pronoun  is  the  only  means  at  Homer's  disposal  to  express 
simple  anaphora,  for  mrrdq^  except  in  an  insignificant  number  of  passages  in 
later  portions  of  the  poems,  regularly  preserves  its  intensive  force. 

airroiv  is  also  used  as  a  reflexive  in  Homer. 

The  possessive  pronoun,  as  noted  above,  is  strictly  reflexive.  For  the 
anaphoric  expression  of  the  relation  of  the  possessor,  the  Dative  of  Interest 
is  used. 

The  Genitive  of  Possession  is  as  yet  very  rare. 

Hesiod  and  ths  Homeric  Hymns. 

Hesiod  and  the  Homeric  Hymns  show  a  reduction  in  the  wealth  of  forms. 
hi  and  ik  do  not  occur,  but  the  corresponding  genitive  idv  is  found  in  Hesiod, 
Theog.  401.  The  genitive  singular  of  the  monosyllabic  stem  is  practically 
dead,  inasmuch  as  kdev  is  missing  and  elo  and  eo  occur  only  in  borrowed 
expressions.  The  genitive  plural  is  rare  in  Hesiod  and  found  only  in  the 
form  a^uv,  while  it  is  entirely  wanting  in  the  Hymns.  Dual  forms  are  not 
used.  Hesiod  employs  the  forms  ff^c  and  a^  side  by  side  with  a^ac  and 
atftirepoi  respectively,  and  the  doublets  c  and  fiiv^  a^lv  and  a^'taiv^  and  c6f  and 
hq  are  found  both  in  Hesiod  and  the  Hymns. 

The  simple  pronoun  lacks  the  intensive  signification.  Hence  avrd^  is 
added  to  the  personal  pronoun,  but  only  when  the  latter  is  used  as  a  reflexive. 
The  position  of  avT6g  is  after  the  pronoun.  In  two  passages  compound  forms 
of  the  reflexive  are  found:  iavr§,  Theog.  126,  and  kavr&v^  Hymn.  Ill  239. 
The  simple  personal  pronoun  of  the  ist  and  2d  person,  whether  reflexive  or 
not,  may  be  combined  with  a  following  avrd^^  and  kfjik  is  used  as  a  direct 
reflexive.  In  Hesiod,  the  possessives  e6q  and  aiftirepoc  are  usually  emphatic 
and  therefore  are  not  combined  with  avrov  and  avruv,  while  dc  is  used  as  an 
unemphatic  possessive.  In  the  Homeric  Hymns,  the  possessive  has  lost  its 
intensive  force  ;  bg  is  once  combined  with  a  following  avroi),  and  kfi6c  and  adq, 
a  number  of  times. 

The  substantive  pronoun  is  not  used  as  a  general  reflexive.  Interchange 
of  persons  in  the  case  of  the  possessive  occurs  twice  in  Hesiod  and  once  in 
the  Batrachomyomachy ;  interchange  of  numbers  is  found  four  times  in  Hesiod 
and  once  in  the  Batrachomyomachy.  That  the  feeling  for  the  original  use  of 
the  general  reflexive  had  been  lost  is  shown  by  Hesiod's  use  of  a^epoc  for 
vfUrepoq  and  of  the  plural  possessives  a^g  and  ai^krepoQ  for  the  singular  6f.  In 
the  Hymns  anaphoric  h  is  used  for  a^aq  4,  267,  and  anaphoric  ff^cv  takes  the 
place  of  o\  in  19,  19. 

The  substantive  pronoun  is  anaphoric.  Remnants  of  the  direct  reflexive 
use  are  found  only  in  the  formulae  ec  o^oq^  knl  o^ag^  anb  to^  dfi^i  e  and  irapd 
c^iai,  and  the  indirect  reflexive  use  is  preserved  in  the  phrases  fierd  elo  and 
fierd  a^aq, 

avrdg  almost  everywhere  is  intensive. 

Both  Hesiod  and  the  Hymns  use  avroi)  as  a  reflexive.  In  addition  Hesiod 
uses  both  the  complex  and  the  compound  reflexive,  but  the  Hymns  lack  the 
complex  forms. 

The  possessive  pronoun  is  regularly  a  direct  reflexive,  but  it  is  twice 
found  in  Hesiod  as  an  indirect  reflexive. 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES,  21/ 

Lyric  Poetry. 

Melic  Lyric  poetry  introduces  new  forms  from  the  dialects,  especially 
from  the  Doric.  These  forms  are  kov^^  'iv,  viv  and  a^6^ ;  edev  also  has  been 
preserved  through  the  influence  of  the  dialects.  Elegy  and  Iambic  poetry  fall 
in  line  with  Hesiod's  usage,  but  omit  antiquated  forms,  c  is  very  rare,  a^ia^ 
is  found  only  in  Archilochus  and  Simonides  Cens,  and  a^  is  confined  to 
Theognis,  Simonides  and  Pindar,  viv  and  \v  are  the  only  new  forms  that 
Pindar  uses  in  common  with  the  other  Lyric  poets. 

The  simple  pronoun  is  generally  unemphatic,  but  in  four  instances  Pindar 
uses  non-reflexive  o\  and  e  with  special  emphasis,  avrov  is  used  as  an  emphatic 
reflexive  of  the  3d  person,  kfiavrov  and  aavrov  being  used  for  the  1st  .and  2d 
singular,  iavrov  is  found  in  Simonides.  The  difference  between  the  strong 
and  the  weak  stem  is  kept  alive  by  Pindar  in  his  use  of  the  forms  id^  and  be 
The  possessive  is  nowhere  accompanied  by  airrov  or  avrov. 

The  instances  of  the  erroneous  interchange  of  numbers  are  further 
augmented  by  the  plural  use  of  the  demonstrative  viv  and  by  the  singular  use 
of  01^  and  a^df,  a^epog  is  freely  used  for  idc,  and  Alcman  even  employs 
a^6g  for  non-reflexive  a^repog. 

The  substantive  pronoun  is  predominantly  anaphoric  in  Pindar.  It  is 
nowhere  used  by  him  as  a  direct  reflexive  and  very  rarely  as  an  indirect 
reflexive.  The  other  Lyric  poets  preserve  the  reflexive  use  in  the  combina- 
tions ait'  iovCf  &Tep  eOev,  less  frequently  in  edev  and  a^tv  unaccompanied  by  a 
preposition. 

of^iatv  avToic  and  avrov  are  found  as  reflexives  in  Pindar,  elsewhere  only 
avTov  is  used. 

The  possessive  is  as  yet  reflexive,  but  o^  and  a^6c  are  each  once  used 
anaphorically. 

The  Possessive  Genitive  of  reflexive  avrov  does  not  occur  in  Pindar, 
though  that  use  is  common  in  the  other  Lyric  poets. 

Tragic  Poets. 

In  the  use  of  the  forms  of  the  simple  pronoun,  the  Tragedians  show 
Lyric  influence.    The  following  is  a  tabular  exhibit  of  this  use : 

Aesch.     —  I        — 

Soph.         2        —  ] 

Eur.         —        —        — 

The  absence  of  e  and  a(^v  and  the  disappearance  of  i6g  are  to  be  noted. 

Only  the  rare  forms  of  the  simple  pronoun  when  used  reflexively  have 
special  emphasis.  These  forms  are  edev  in  Aeschylus, !  and  ov  in  Sophocles, 
and  ol  in  Euripides.  The  possessive  pronoun  is  emphatic  in  Aeschylus,  but 
lacks  special  emphasis  in  Sophocles  and  in  Euripides.  Sophocles  once 
combines  the  genitive  avrov  with  dc,  and  in  like  manner  kfidg  and  a6c  are  once 
each  combined  with  airrov  by  the  same  author. 

The  free  use  of  the  pronoun  is  quite  extended,  atpi  is  frequently  used 
for  viv,  less  frequently  viv  for  o^i,  the  metre  being  the  determining  factor. 


01 

<r^t<r»v   <r^tV 

<r^£c 

ir^ 

ytp 

5« 

9^CT«P<K 

I 

I          5 

I 

16 

48 

— 

2 

4 

I          6 

5 

M 

87 

5 

— 

I 

—          2 

5 

57 

248 

2 

— 

Digitized  by  VjOOQ IC 


2l8  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Aeschylus  (once)  and  Sophocles  (twice)  use  a^iv  for  ol,  a  use  that  was  noted 
also  for  the  Homeric  Hymns.  In  his  use  of  the  word  a^tpoq  Aeschylus  also 
preserved  its  use  as  a  substitute  for  5^,  thereby  showing  that  he  was  subject  to 
Lyric  influence.    Euripides  in  a  choral  passage  uses  5f  in  a  plural  sense. 

(70dc  and  c^iv  are  exclusively  anaphoric  and  c^k  and  viv  are  predominantly 
so.  ol  is  anaphoric  in  the  one  instance  in  which  it  is  used  by  Aeschylus  and 
in  two  choral  passages  and  one  trimeter  passage  of  Sophocles,  but  in  another 
trimeter  passage  of  Sophocles  and  in  the  only  instance  furnished  by  Euripides 
it  is  an  indirect  reflexive  as  in  Attic.  Sophocles  follows  Attic  rule  in  using 
a^iaiv  as  an  indirect  reflexive,  but  Aeschylus  uses  it  anaphorically  with  refer- 
ence to  the  subject  of  the  leading  verb.  In  dependent  sentences  i^tv  is  used 
by  Aeschylus  and  Z  by  Sophocles  as  an  indirect  reflexive.  The  only  instance 
of  a  direct  reflexive  is  that  of  ov  in  Sophocles. 

There  Is  no  complex  reflexive  of  the  3d  person.  Aeschylus  once  uses  the 
combination  avraX  vfidg  avrdg,  and,  for  metrical  reasons,  Sophocles  once 
employs  the  singular  oi  r*  avr^v,  vtv  avrdg  is  once  used  by  Euripides,  but  not 
as  a  reflexive. 

The  compound  forms  iavr&v  and  airrov,  oeavrov  and  oavrov,  and  kfMvrov  gain 
in  frequency.  Dissyllabic  avrov  and  aavrov  are  preferred  to  trisyllabic  iavrov 
and  (Tcavroi),  which  are  only  occasionally  used  when  required  by  the  metre. 
The  plural  of  iavrov  does  not  occur ;  that  of  airrov  is  used  by  Aeschylus  in 
the  Genitive  only  (4  times),  by  Sophocles  in  the  Dative  only  (2  times),  and  by 
Euripides  twice  in  the  Genitive  and  four  times  in  the  Accusative.  Sophocles 
once  uses  the  dual  of  avrov. 

avTov,  both  in  the  singular  and  in  the  plural,  is  used  as  a  free  reflexive,  and 
is  used  to  represent  both  the  ist  and  the  2d  person.  Euripides  is  sparing  in 
this  use,  showing  only  two  instances  of  it. 

iavToi)  and  airrov  are  almost  exclusively  direct  reflexives.  As  an  indirect 
reflexive,  iavrov  is  once  found  in  Aeschylus ;  eairrov  and  avrov  are  thus  used  in 
Sophocles  in  phrases  that  are  the  equivalents  of  sentences;  in  Euripides, 
eavroit  is  an  indirect  reflexive  once,  and  airrov  five  times,  in  dependent  clauses. 

Whilst  the  reflexive  forms  are  the  rule  in  the  case  of  the  reflexive  use  of 
the  pronoun  of  the  3d  person,  the  reflexive  forms  are  not  always  used  when 
the  pronoun  of  the  ist  or  2d  person  is  reflexive.  In  direct  reflexion  the 
compound  forms  are  the  rule  and  the  signification  *self'  is  prominent. 
Enclitic  forms  are  rarely  admitted  instead.  So  Aeschylus  once  uses  fxe  in  a 
choral  passage,  and  Euripides  uses  fie  and  001  even  in  the  trimeter.  The 
convenient  conversational  phrase  doK6>  fioi  is  first  met  with  in  Euripides.  Both 
Aeschylus  and  Euripides  use  unemphatic  aidev  as  a  direct  reflexive,  Aeschylus 
doing  so  twice  in  choral  passages,  and  Euripides  25  times  as  a  metrical 
necessity  at  the  end  of  verses,  especially  to  aflbrd  a  light  close  for  the  trimeter. 
More  frequently  used  are  the  orthotone  forms  of  the  simple  pronoun  in  sharp 
contrasts.  For  the  indirect  reflexive,  the  simple  pronoun  is  the  rule ;  only 
Sophocles  thrice  uses  the  compound  forms  in  dependent  sentences  and 
Euripides  8  times  in  phrases  that  are  the  equivalents  of  a  sentence. 

The  possessive  of  the  3d  person  is  generally  a  direct  reflexive.  Aeschylus 
once  uses  it  as  an  indirect  reflexive  in  a  dependent  proposition  and  Sophocles 
so  uses  it  once  in  a  declarative  sentence. 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES,  219 

The  possessive  adjective  of  the  3d  person  is  rare  as  compared  with  the 
possessive  genitive  of  the  compound  pronoun,  the  ratio  being  that  of  9  to  48. 

In  the  1st  and  2d  person,  the  simple  possessives  e/<^  and  a6q  are  far  more 
frequent  as  direct  reflexives  than  the  corresponding  possessive  genitives  of  the 
compound  pronoun. 

Aristophanrs. 

The  simple  pronoun  is  not  used  in  Aristophanes,  except  for  purposes  of 
parody.  The  complex  reflexive  of  the  3d  person  likewise  does  not  occur. 
The  compound  forms  are  used  as  follows :  efxavrov  48  times,  aavrcv  and  asavroiv 
71  and  27  times  respectively,  avrov  and  kavrov  44  and  25  times  respectively. 
The  plural  of  the  form  kavrov  is  used  in  Aristophanes  for  the  first  time,  and  is 
relatively  more  frequent  than  that  of  avrov.  Except  in  the  formula  fJtoi  doKu^ 
the  reflexive  forms  are  the  rule  also  for  the  1st  and  2d  persons  singular  in 
direct  reflexion.  The  possessive  adjective  3d  person  does  not  occur  in  Aris- 
tophanes except  in  parody,  the  possessive  genitive  of  the  reflexive  being  used 
instead,  twenty  times  in  the  singular  and  twice  in  the  plural.  kfi6c  as  a  direct 
reflexive  occurs  29  times,  possessive  kfiavrov  11  times;  aS^  (re<$c«  <^^)  ^  a  direct 
reflexive  occurs  13  times,  possessive  aawov  18  times  and  aeavrov  3  times. 

Herodotus. 

The  simple  pronoun  is  19  times  used  as  a  direct  reflexive :  a^(,)v  twice,  a^iai 
16  times,  and  a^iag  once.  Of  the  indirect  reflexive  use  there  are  about  400 
instances.  The  purely  anaphoric  use  of  the  simple  pronoun  (not  counting 
fiiv)  is  represented  by  more  than  one  thousand  examples.  The  form  a^/at  is 
never  anaphoric  in  Herodotus,  but  is  always  either  a  direct  or  an  indirect 
reflexive. 

The  complex  form  of  the  reflexive  pronoun  is  not  found  in  the  singular. 
The  plural  a^uv  airruv  occurs  21  times,  a^i  avrolai  11  times,  and  at^a^  avrov^ 
33  times.  Of  these  65  instances,  only  8  belong  to  the  indirect  reflexive  use, 
the  simple  pronoun  and  the  compound  forms  being  better  adapted  for  that 
purpose.  Complex  forms  of  the  ist  and  2d  plural  occur  14  times  and  are 
always  used  as  direct  reflexives.  In  addition  to  these  plural  forms,  the 
singular  ako  avrov  occurs  once — 1, 124. 

In  the  singular  number  the  compound  pronoun  of  the  3d  person  is  the  only 
form  used  for  direct  reflexion,  and  it  is  the  predominating  form  for  indirect 
reflexion.  In  the  plural  number,  however,  the  compound  forms  occupy  a 
subordinate  position,  especially  in  the  direct  reflexive  use.  iuvrdv  is  regularly 
used  as  a  possessive  genitive  (46  times)  rather  than  a^civ  avrav  (4  times). 
Compound  reflexive  forms  of  the  1st  and  2d  person  are  also  in  use,  but,  as  in 
the  case  of  the  3d  person,  the  simple  forms  are  preferred  in  indirect  reflexion. 

Herodotus  once  uses  ^cnrrov  for  the  1st  person  and  a^iai  avrolai  for  the  2d. 

Of  the  possessive  adjective,  only  the  form  ai^rtpoq  seems  to  have  been  used, 
\i  rjv  1^  20s  is  to  be  emended,  a^repoq  occurs  64  times  (42  times  as  a  direct 
reflexive,  including  a^tpo^  avrov  twice),  while  possessive  ionrrav  is  used  46 
times,  a^epo^  is  always  plural.  The  possessive  singular  is  represented 
exclusively  by  the  possessive  genitive  iowrov,  ifiSc  is  a  direct  reflexive  19 
times,  but  ifuuvrov  as  a  possessive  genitive  is  used  only  6  times ;  a6^  is  used 
as  a  direct  reflexive  5  times,  while  possessive  atuvrov  is  used  17  times. 


Digitized  by  VjOOQ IC 


220  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Attic  Inscriptions. 

In  the  pre-Euclidean  inscriptions,  the  form  a^v  is  twice  used  as  a  direct 
reflexive.  Complex  forms  are  fonnd  9  times,  only  in  the  plural  and  only  as 
direct  reflexives.  There  are  no  certain  instances  of  compound  forms.  &c 
occurs  once  and  a^repoc  twice,  in  poetry ;  in  prose  inscriptions  atf^irepoc  airav 
is  found  5  times. 

In  post- Euclidean  inscriptions,  the  simple  pronoun  is  wanting.  Of  the 
complex  reflexive,  ai^iaiv  avroig  is  the  only  form  used  and  that  only  in  the 
earlier  inscriptions.  The  compound  forms  are  found  both  in  the  singular  and 
in  the  plural.  The  ratio  of  iavrov  to  avrov  is  that  of  31  :  23.  a^epog  no 
longer  occurs,  iavrov  is  a  possessive  genitive  8  times  and  possessive  iavrov 
(avTov)  is  found  10  times.  In  the  absence  of  reflexion,  avrov  and  avrov  are 
used  as  possessives. 

[Xenophon]  de  rspublica  Athenibnsium. 

a<pav  and  a^i  occur  as  indirect  reflexives.  The  complex  and  the  compound 
forms  are  used  both  as  direct  and  as  indirect  reflexives.  The  adjective 
pronoun  is  represented  by  two  instances  of  off^irepot  avrov,  both  direct 
reflexives,  and  by  one  instance  of  a^repog  used  as  an  indirect  reflexive. 
There  are  two  examples  of  the  possessive  genitive,  iavrov  and  a^. 

Thucydides. 

As  the  simple  pronoun  of  the  3d  person,  the  complex  forms  and  the  com- 
pound forms  are  all  used  as  direct  and  as  indirect  reflexives,  the  following 
exhibit  of  the  relative  frequency  of  their  use  will  be  of  interest. 

In  the  singular,  the  compound  forms  are  the  only  forms  that  are  used  in 
direct  reflexion,    iavrov  occurs  65  times,  airrov  31  times. 

In  the  plural,  the  simple  pronoun  is  used  9  times  as  a  direct  reflexive  (oipov 
4  times,  aiffiai  3  times,  and  a^  twice),  the  complex  pronoun  is  thus  used  94 
times,  and  the  compound  42  times  (iavrov,  etc.,  23  times,  airrov,  etc.,  19  times). 
This  does  not  include  the  use  of  the  possessive  genitive,  which  is  represented 
by  two  instances  of  ff^v,  one  of  <t^v  airrov,  and  103  (73  +  30)  of  iavrov  {airov). 

For  the  indirect  reflexive  use,  the  following  figures  indicate  the  frequency. 
The  simple  pronoun,  excluding  a^i^  (10  times),  occurs  439  times  (01 12,  ff^uv 
102,  oijfioi  239,  <T^df  86),  the  complex  pronoun  14  times,  iavrov  (airrov)  83  (48  + 
35)  times.  The  figures  include  the  instances  of  the  possessive  genitives: 
<T0ov  46,  iavrov  (airrov)  18  (13  +  5),  but  do  not  include  the  67  (39  +  28) 
instances  of  the  genitive  and  the  accusative  singular  of  iavrov  (airrov). 

In  the  case  of  the  compound  reflexive,  the  question  arises  as  to  whether 
ai;rov  or  avrov  is  the  correct  reading.  In  the  light  of  such  criteria  as  the  use 
of  the  complex  reflexive  in  corresponding  syntactical  groups,  the  correspond- 
ing use  of  the  simple  pronoun,  the  frequency  of  passages  containing  iavrov, 
the  aspiration  of  a  preceding  mute,  the  position  of  the  possessive  genitive 
between  the  article  and  the  substantive,  and  the  parallelism  of  pronouns  of 
the  1st  and  of  the  2d  person,  it  is  certain  that  the  compound  reflexive,  and  not 
airrov,  is  used  as  a  direct  reflexive.  It  is  also  regularly  used  as  an  indirect 
reflexive,  but  in  relative  participial  clauses,  the  pronoun  avrov  seems  to  be 


Digitized  by  VjOOQ IC 


RE  VIEWS  AND  BOOK  NOTICES.  221 

warranted  in  many  instances,  though  the  reflexive  is  not  uncommon.  So  too,  in 
most  clauses  introduced  by  conjunctions  and  in  dependent  sentences,  reflexive 
and  determinative  appear  side  by  side.  This  occurs  in  object  clauses  with 
5r<,  in  indirect  questions,  in  the  &OTe  w.  inf.  construction,  in  final  and  causal 
sentences,  and  in  clauses  forming  an  integral  part  of  an  infinitive  sentence. 
Of  the  other  dependent  clauses,  only  the  relative  sentences  introduced  by 
hao^  permit  the  reflexive,  the  rest  require  cojtov.  The  genitive  absolute  does 
not  admit  the  reflexive.  As  to  the  use  of  the  forms  iaurov  and  avrw^  it  is 
to  be  noted  from  the  figures  given  above  that  Thucydides  vastly  prefers  iavrdv 
to  arrrcw.  A  difference  in  meaning  between  iavrdv  and  avrw  does  not  exist, 
though  of  the  two  avroiv  seems  to  have  the  greater  emphasis. 

Thucydides  once  (i,  82,  i)  uses  the  plural  compound  reflexive  of  the  3d 
person  in  reference  to  the  ist  person. 

There  is  no  consistent  use  of  purely  reflexive  forms  of  the  ist  and  2d  person. 
In  the  plural,  the  simple  pronoun  is  used  as  a  direct  reflexive,  though  the 
complex  forms  are  also  found.  In  the  few  cases  of  the  direct  reflexive  use  of 
the  singular,  the  compound  form  is  regularly  used  except  in  the  formula  6ok& 
hv  ftoi  6,  38,  4.  For  the  indirect  reflexive  use,  the  simple  forms  are  the  rule 
for  both  singular  and  plural. 

The  pronominal  adjective  of  the  3d  person  is  found  only  in  the  form 
a^repoc,  which  occurs  90  times,  62  times  as  a  direct  reflexive.  Oi^repoc  airrdv 
occurs  13  times,  once  as  an  indirect  reflexive,  kfiavrov  and  ifMv  avrav  occur 
as  possessives,  and  e/i^,  a6c,  ^fitrepo^,  ifdrepo^  and  iifikrepo^  avruv  occur  as 
reflexives. 

Attic  Orators. 

Of  the  Attic  Orators,  Lycurgus,  Aeschines,  Dinarchus  and  Hyperides  do 
not  use  the  simple  pronoun  at  all ;  the  other  orators  use  only  ol  1 1  times,  ff0e<c 
twice,  a<^v  twice,  a^Uri  la  times  and  (T^dc  4  times.  Only  two  of  these  31 
occurrences  are  instances  of  the  direct  reflexive  use,  the  rest  are  indirect 
reflexives. 

The  complex  reflexive  is  used  only  in  the  plural  and  is  regularly  a  direct 
reflexive,  very  much  less  frequently  an  indirect  reflexive.  Like  the  simple 
pronoun,  it  disappears  towards  the  close  of  the  period  of  Attic  Oratory.  In 
Antiphon  and  Andocides,  it  predominates  over  the  corresponding  forms  of 
the  compound  reflexive,  the  plural  of  the  compound  pronoun  being  found 
only  in  the  genitive;  in  Lysias  and  Isocrates,  it  is  as  yet  pretty  frequent, 
especially  in  the  accusative;  in  Isaens,  the  rival  forms  are  about  equally 
divided;  in  Demosthenes  and  Hyperides  the  complex  forms  are  rare,  and 
they  are  entirely  wanting  in  Lycurgus,  Aeschines  and  Dinarchus.  The 
genitive  of  the  complex  pronoun  is  found  but  twice  as  a  possessive  genitive, 
the  genitive  of  the  compound  form  being  used  instead. 

The  form  avrov  is  everywhere  preferred  to  iavrov  except  in  Andocides  and 
Aeschines,  where  the  two  forms  balance.  In  Isocrates,  avrov  seems  to  have 
been  the  only  form  used.  In  this  connection  it  may  be  noted  that  aavrov 
likewise  is  more  common  than  aeavrov,  except  in  Andocides,  who  uses  only 
oeavrov,  and  in  Dinarchus,  who  has  four  instances  of  aeavrov  to  three  of  aavrov. 

The  possessive  adjective  is  found  only  in  the  form  atpirepoc  (airrotf).  It  is 
rare  where  the  simple  substantive  pronoun  is  rare  and  is  wanting  where  that 


Digitized  by  VjOOQ IC 


222  AMERICAN  JOURNAL   OF  PHILOLOGY. 

is  wanting.  The  ratios  of  the  direct  and  indirect  reflexive  use  of  cipirepoc  and 
a<^epoc  airrijv  are  for  a^repoq  8  :  6  and  for  afftirepoc  avruv  6o  :  4.  a^repoq  is 
never  anaphoric  in  the  Attic  Orators.  In  Antiphon,  Andocides  and  Lysias, 
the  frequency  of  the  adjective  forms  either  excels  or  nearly  equals  that  of  the 
corresponding  possessive  genitive  of  the  compound  reflexive,  whereas  in 
Ps.-Lys.,  Isoc,  Isae.,  and  Dem.,  the  possessive  genitive  far  outstrips  the 
adjective  forms  in  point  of  frequency. 

In  direct  reflexion,  except  in  the  formula  fioi  6oko,  which  occurs  10  times, 
the  compound  forms  of  the  pronoun  of  the  zst  and  of  the  2d  person  are  the 
rule  for  the  singular  and  the  complex  forms  are  the  rule  for  the  plural :  the 
simple  forms  are  rare.  In  indirect  reflexion  the  simple  pronoun  is  much  more 
common,  the  reflexive  forms  being  the  rule  only  when  the  indirect  reflexive 
use  borders  closely  upon  the  direct  reflexive  use. 

There  is  no  complex  form  of  the  possessive  adjective  of  the  ist  and  2d 
person  singular,  ifiavrov  and  aavrov  (aeavrov)  being  used  instead.  In  direct 
reflexion,  ifiavrov  and  aawov  {aeavrov)  are  very  much  preferred  to  kfid^  and  <i6c, 
whereas  in  indirect  reflexion,  the  forms  kfJiSc  and  fiov  are  used  by  preference. 
For  the  plural,  there  is  a  complex  possessive  adjective,  and  in  direct  reflexion 
this  is  very  much  more  common  than  the  simple  forms,  the  possessive  genitive 
^fiuv  (vfiav)  avTuv  being  very  rare.  In  indirect  reflexion,  the  complex  possessive 
adjective  is  quite  exceptional. 

There  is  no  instance  of  the  interchange  of  numbers.  Of  the  interchange  of 
persons  there  are  a  few  examples  in  the  plural,  a^v  avruv  is  once  used  for 
{;fia}v  avTov,  airrav  is  once  used  for  ifiav  avruv  and  8  times  for  viiov  avruv. 
These  instances  are  found  in  Andocides,  Lysias,  Demosthenes  and  Aeschines. 
In  the  singular  avrov  is  twice  used  for  hfiavrov  and  once  for  oavrm.  These 
examples  are  found  in  Antiphon  and  Andocides.  Other  instances  of  the  use 
of  iavTov  (avTov)  for  the  2d  person  are  found  in  the  MSS  of  Isocrates,  Aeschines, 
Dinarchus  and  Hyperides,  but  these  instances  are  very  uncertain,  inasmuch  as 
tradition  varies  and  iavrw  is  graphically  close  to  aavrov,  and,  besides,  the  MSS 
of  Lysias,  Isaeus,  Lycurgus  and  Demosthenes  furnish  no  such  examples. 

Plato. 

The  simple  pronoun  is  represented  by  the  forms  ov,  ol,  I,  a^v,  a^lai  and  a^c 
It  occurs  109  times,  and,  with  the  solitary  exception  of  at^^  in  Legg.  782  E, 
which  is  used  as  a  direct  reflexive,  its  function  is  that  of  an  indirect  reflexive. 

The  complex  reflexive  is  found  11  times  in  direct  and  6  times  in  indirect 
reflexion. 

The  compound  forms  are  used  2013  times,  1719  times  as  direct  reflexives 
and  294  times  as  indirect  reflexives.  The  form  airrov  is  preferred  to  iavrxnt, 
the  ratio  being  z  21 2  :  801.  aavrw  also  is  preferred  to  ffcavroi),  the  ratio  being 
no:  27. 

The  compound  pronoun  of  the  3d  person  plural  is  twice  used  in  Plato  and 
twice  in  Ps.-Plato  for  the  pronoun  of  the  1st  person,  and  it  is  once  used  in 
Ps.-Plato  for  the  pronoun  of  the  2d  person.  In  the  singular  the  reflexive  of 
the  3d  person  is  found  only  for  that  of  the  2d,  and  in  every  instance  aavrov 
may  be  readily  restored  for  iavrov  {avrov). 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES.  223 

With  but  few  exceptions,  the  singular  of  the  simple  pronoun  of  the  ist  and 
2d  person  is  used  as  a  direct  reflexive  only  in  the  formula  fioi  doKij  or  Soicij  fioi, 
and  even  in  this  formula  the  reflexive  is  used  when  a  contrast  is  involved.  In 
the  plural,  the  reflexive  form  is  likewise  the  rule  for  direct  reflexion.  In 
indirect  reflexion,  the  simple  pronoun  predominates,  though  the  compound 
and  complex  forms  are  also  used  except  in  the  genitive. 

airrd^  is  used  in  a  number  of  instances  both  before  and  after  non-reflexive 
(Tf,  etc.,  and  kue,  etc.,  without  forming  a  regular  complex  pronoun. 

Plato  is  the  only  one  of  the  Attic  prose-writers  that  uses  the  possessive  5c, 
and  he  uses  it  but  once,  and  that  as  an  indirect  reflexive  in  a  paraphrase  of 
the  Iliad  (Rpb.  394  A).  Even  the  plural  form  a^trepoq  {avriiv)  is  rare  in  direct 
reflexion  as  compared  with  the  possessive  genitive  eavrav.  In  indirect 
reflexion,  a^repo^  is  more  freely  used,  though  not  as  often  as  iavruv.  The 
ratio  of  iavrav  to  a^trepog  (avruv)  is  202  :  20.  c^repog  avruv  is  used  only  as  a 
direct  reflexive.  ^/liTepog  {yfihepoq)  avruv  and  simple  ^fiirepo^  (vfikrepoq)  are 
each  used  11  times  as  direct  reflexives,  while  possessive  ^fiuv  avrav  occurs  but 
twice.  Ifidf  (a6c)  is  used  30  times  in  direct  reflexion,  but  kfiavTov  {aavrov^ 
aeavToif)  occurs  58  times. 

The  simple  pronoun  is  found  in  but  six  of  the  admittedly  spurious  works, 
whereas  it  is  wanting  in  only  the  Critias,  the  Crito,  the  Meno  and  the 
Parmenides  of  the  genuine  works,  oh  and  i  are  used  only  in  the  Convivium 
and  in  the  Republic.  The  complex  reflexive  is  confined  to  four  of  the 
genuine  (Gorg.,  Politic,  Rpb.,  Legg.)  and  to  four  of  the  spurious  (Ale.  I,  Ale. 
II,  Eryx.,  Men  ex.)  dialogues.  The  form  avrov  predominates  in  most  of  the 
works,  but  iavroif  outnumbers  avrw  in  the  Apol.,  Parmen.,  Euthyd.,  Protag., 
and  in  seven  of  the  spurious  dialogues,  a^repo^  is  restricted  to  the  Rpb., 
Phaedr.,  Politic,  Tim.,  Legg.,  Soph.,  Euthyd.,  Menex.,  Eryx.  and  Epist.  III. 

Xenophon. 

Xenophon  uses  the  forms  ol,  (T^civ,  a^iai,  a^,  a^l^.  They  are  indirect 
reflexives  except  in  Cyr.  3,  2, 26,  where  o\  is  used  anaphorically.  The  complex 
reflexive  is  rare  in  both  direct  (5  times)  and  in  indirect  reflexion  (8  times),  the 
compound  form  being  the  prevailing  form,  except  in  the  dative  plural,  in 
which  a^ioi  is  the  most  common  form  for  indirect  reflexion,  iavrov  is  used 
679  times  (225  times  in  indirect  reflexion),  and  avrov  394  times  (203  times  in 
indirect  reflexion),  but  aeavrov  (16  times)  is  less  common  than  oavrov  (42  times) 

a^av  airrCw,  airruv  and  iavruv  are  used  once  each  for  the  reflexive  of  the  2d 
person  plural.  All  three  instances  occur  in  the  first  book  of  the  Hellenica. 
In  the  singular,  the  compound  reflexive  of  the  3d  person  is  found  for  that  of 
the  2d  person,  but  aavrov  should  everywhere  be  restored. 

The  simple  pronoun  of  the  ist  and  2d  person  is  used  as  a  direct  reflexive  when 
there  is  emphasis ;  elsewhere,  only  in  fioi  SoKa  and  Soku  fxot.  The  compound 
forms  are  used  in  direct  reflexion  103  times,  but  they  are  also  used  in  indirect 
reflexion.  The  simple  pronoun,  however,  is  exceedingly  frequent  in  indirect 
reflexion. 

The  use  of  the  possessive  forms  becomes  clear  from  the  following  figures : 
a^repoq  is  used  13  times  (once  in  direct  reflexion),  the  corresponding  posses- 
sive genitive  149  times  (12s  times  in  direct  reflexion),     a^repoc  avruv  does  not 


Digitized  by  VjOOQ IC 


224  AMERICAN  JOURNAL    OF  PHILOLOGY. 

occur.  In  direct  reflexion,  ^fttrepoq  avruv  is  used  once,  ifUrepoc  (vfihepo^y 
lo  times,  possessive  ifiw  avrw  once.  ifi6c  {06^)  is  used  32  times  as  a  direct 
reflexive,  possessive  ifiavrcv  (aavroi),  (Teavrov)  only  26  times. 

Anab.,  Cyr.,  and  Hell.  II  and  III  show  a  decided  preference  for  the  plural 
forms  of  the  simple  pronoun,  and  they  use  o<  rather  frequently.  All  but  three 
of  the  complex  reflexives  of  the  3d  person  are  found  in  the  Hellenica.  airrotf 
is  more  frequent  than  iavroif  in  only  the  Cyneg.  and  Hell.  I;  in  the  other 
works  iavTov  preponderates. 

C.  W.  E.  MiLLBR. 


Chrestomathie  fran^aise,  by  A.  Rambbau  and  J.  Passy.    Henry  Holt  &  Co. 
N.  Y.,  1 897.    Pp.  XXXV  +  250. 

The  phonetic  method  of  teaching  modem  languages,  while  it  has  as  yet 
scarcely  gained  a  foothold  in  this  country,  has  rapidly  won  favor  in  Germany 
and  Scandinavia,  and  is  gradually  coming  into  notice  in  France  and  England. 
For  the  slow  progress,  in  America,  of  a  system  that  undoubtedly  has  much  to 
commend  it,  there  are  at  least  two  potent  reasons;  in  the  first  place,  our 
educators  have  seen,  within  the  last  twenty  years,  the  rise  and  fall  of  so  many 
new  modes  of  linguistic  study,  each  one  loudly  proclaimed  as  infallible,  that 
they  are  inclined  to  look  with  distrust  upon  any  apparently  similar  innovation  ; 
and,  secondly,  as  a  result  of  much  experimenting,  our  ways  of  instruction,  in 
the  more  enlightened  regions,  are  really  less  antiquated  than  those  of  most 
other  countries,  and  the  need  of  a  change  is  correspondingly  less  urgent. 

The  *  phonetic'  or  'reform'  program  differs  from  nearly  all  other  methods  in 
that  it  is  based  on  really  scientific  principles  and  advocated  chiefly  by  men  of 
learning  and  successful  experience.  With  it  are  associated  especially  the 
names  of  Professor  Victor,  of  Marburg,  and  Dr.  Paul  Passy,  of  Paris.  A  large 
society  of  teachers,  the  Association  Phon^tique  Internationale,  is  devoted  to 
the  propagation  of  the  'reform'  creed.  It  has  two  thoroughly  reputable 
organs,  the  Mattre phon/Hqtu  in  France  and  the  Neueren  Sprachen  in  Germany. 
The  principal  articles  of  the  new  faith  are  these :  modem  language  instraction 
should  take  as  its  first  material  the  living,  spoken  tongue,  reserving  for  later 
study  the  more  or  less  obsolete  speech  of  literature  ;  pronunciation  should  be 
thoroughly,  accurately  and  scientifically  taught  from  the  very  outset.  For 
these  purposes  various  printed  aids  are  required :  charts  of  sounds,  with  well- 
chosen  key-words;  pictures  that  afford  topics  for  questions  and  answers; 
dialogues  and  simple  narratives  in  modem,  idiomatic  style  and  in  phonetic 
spelling.  This  latter  condition  is  indispensable;  for  the  advocates  of  the 
system  attach  the  greatest  importance  to  the  exclusive  use  of  a  phonetic 
notation  until  the  pupil  has  become  very  familiar  with  the  sounds  of  the 
language,  considered  both  as  artificially  isolated  phenomena  and  as  elements 
of  naturally  combined  phrases. 

Amid  the  surprisingly  copious  literature  that  the  new  method  has  called 
into  existence,  two  collections  of  phonetic  texts  have  merited  particular 
attention :  Sweet's  *  Elementarbuch  des  gesprochenen  Englisch,*  for  Germans 
who  are  acquiring  English,  and  the  '  Elementarbuch  des  gesprochenen  Fran- 
zOsisch,'  by  Franz  Beyer  and  Paul  Passy,  for  Germans  who  wish  to  leam 


Digitized  by  VjOOQ IC 


"anssHsmv^E:: 


REVIEWS  AND  BOOK  NOTICES,  22$ 

French.  To  these  is  now  added,  for  the  special  benefit  of  American  and 
English  students  of  the  French  language,  the  *  Chrestomathie  franQaise '  prepared 
by  Professor  A.  Rambeau,  of  Johns  Hopkins,  and  Jean  Passy,  a  brother  of  the 
distinguished  founder  of  the  Association  Phonetique.  Professor  Rambeau's 
linguistic  and  pedagogical  works  have  long  since  made  him  known  to  philolo- 
gists and  phoneticians;  and  Mr.  Passy  has  won  himself  a  reputation  as  a 
teacher  and  as  an  investigator  of  French  dialects.  The  present  volume  is, 
therefore,  the  product  of  men  expert  both  in  the  theory  and  in  the  practical 
side  of  their  science. 

The  Chrestomathie  is  not  meant  for  beginners,  but  is  intended  for  pupils 
who  have  already  used  some  more  elementary  work  of  a  similar  character ; 
hence  the  texts  are  given  in  two  forms — the  standard  spelling  and  the  phonetic 
transcription — on  opposite  pages.  The  book  begins  with  a  strong  defence  of 
the  *new  method';  then  follows,  condensed  into  less  than  twenty  pages,  a 
description  of  French  sounds  and  sound-groups.  The  rest  of  the  volume  is 
filled  by  the  texts  themselves ;  they  are  chosen  to  illustrate  all  sorts  of  styles 
in  prose  and  verse,  and  are  of  various  degrees  of  difficulty,  some  of  them 
being  very  hard,  and  none  particularly  easy.  The  figured  pronunciation  of 
the  poetry  conforms  to  Paul  Passy's  theory  of  accentuation.  The  phonetic 
alphabet  used  by  the  authors  is  that  of  the  Association  Phonetique ;  though 
rather  unsightly,  as  compared  with  Bell's  *  visible  speech '  or  Sweet's  '  broad 
romic,'  it  can  be  quickly  acquired  and  readily  deciphered.  The  print  is  clear 
and  sufficiently  large.  It  is  to  be  hoped  that  the  Chrestomathie,  which  repre- 
sents an  immense  amount  of  disinterested  labor,  will,  even  if  not  extensively 
used  in  America *for  years  to  come,  at  least  serve  fo  bring  home  to  many  of  our 
French  teachers  the  importance  of  phonetic  study. 

C.  H.  Grandgent. 


JoxntNAL  OF  Germanic  Philology.    Editor :  Gustaf  E.  Karsten,  University 
of  Indiana.    Vol.  I,  1897,  No.  I. 

The  first  number  of  the  Journal  of  Germanic  Philology  has  recently  appeared 
in  very  attractive  dress  on  heavy  paper ;  in  general  make-up  it  is  above  criti- 
cism, forming  a  pleasing  contrast  to  similar  journals  in  Europe.  But  not  merely 
its  exterior  reflects  credit  upon  Professor  Karsten ;  its  purpose  and  plan  are 
especially  deserving  of  the  highest  commendation.  The  problems  of  a  journal 
of  this  kind  in  America  are  not  only  distinctly  scientific,  but  are  also  decidedly 
practical.  It  ought  not  only  to  call  forth  and  foster  scientific  study  and 
scholarship  amongst  those  engaged  in  such  work  at  the  larger  institutions  of 
learning,  but  try  to  raise  the  general  average  of  scholarship  in  the  country 
by  disseminating  the  results  of  such  investigations  here  and  elsewhere  amongst 
the  larger  body  of  students  and  teachers ;  amongst  those  whose  time  is  so 
taken  up  by  their  routine  tasks  that  they  cannot  hope  to  follow  carefully  all 
the  latest  literature  in  their  lines  of  work,  but  who  are  forced  to  depend  upon 
abstracts  and  digests,  when  they  can  get  them,  or  who  are  not  near  libraries 
where  they  can  obtain  the  latest  literature,  particularly  such  as  is  to  be  found 
in  the  scientific  journals.  It  is  an  age  of  'Reviews  of  Reviews,'  and  such 
a  *  Review'  of  Germanic  studies  has  been  greatly  needed.     This  need  the 


Digitized  by  VjOOQ IC 


226  AMERICAN  JOURNAL    OF  PHILOLOGY, 

new  journal  intends  to  satisfy ;  in  the  first  number  it  has  made  an  excellent 
beginning  with  digests  of  the  contents  of  Anglia,  vol.  VI ;  Englische  Studien, 
vol.  XXII ;  a  general  discussion  of  the  purpose  and  aims  of  the  Euphorion  ; 
and  digests  of  the  first  three  volumes  of  Indogermanische  Forschungen.  If 
a  suggestion  might  be  allowed,  possibly  in  some  of  these  digests  a  little  more 
condensation  would  be  advisable.  Except  where  an  article  in  a  journal  is 
of  pretty  general  interest*  the  reader  can  hardly  expect  to  find  in  a  digest 
anything  but  the  main  points  that  he  may  be  interested  in  ;  if  he  desires  more 
detailed  information,  he  must  expect  to  go  to  the  original. 

The  body  of  this  first  number  of  the  Journal  provides  a  variety  of  well- 
selected  and  scientifically  interesting  studies,  hardly  needing  other  vouchers 
for  the  quality  of  their  contents  than  the  names  of  the  contributors.  Horatio 
S.  White  of  Cornell  contributes  the  first  article,  a  discussion  and  review  of  the 
various  theories  in  regard  to  the  home  of  Walther  von  der  Vogelweide,  which 
arrives  at  the  only  possible  conclusion  of  the  whole  matter,  that  it  is  still  incon- 
clusive. The  second  article,  by  George  Hempl  of  the  University  of  Michigan, 
is  on  Middle  English  -7&^-,  -wb-^  in  which,  after  a  careful  study  of  Chaucer's 
rimes,  he  establishes  a  new  rime-test  for  the  determining  of  Midland  and 
Southern  texts,  the  latter  riming  wd  with  gd  and  /g,  the  former  showing  the 
rimes  wd  :  do^  to.  The  investigation  further. traces  the  history  of  the  influence 
of  w  on  a  following  d,  establishing  definite  dates  for  the  change  of  g  to  ^  after 
w,  Edward  Payson  Morton,  of  the  University  of  Indiana,  in  the  next  article 
presents  the  results  of  a  study  of  Shakespeare's  popularity  in  the  seventeenth 
century,  as  evinced  by  the  number  of  different  Shakespearian  plays  put  on  the 
stage  during  the  century,  and  the  frequency  of  their  repetitions.  He  shows 
that  Shakespeare  was  popular,  notwithstanding  the  adverse  opinions  of  literary 
critics  of  the  times,  and,  at  least  as  far  as  representations  on  the  stage  are 
concerned,  was  as  popular  as  he  is  to-day,  judging  by  a  comparison  with 
statistics  from  the  Boston  theatres.  In  an  article  on  voiced  spirants  in  Gothic, 
George  A.  Hench,  of  the  University  of  Michigan,  establishes  by  a  careful 
investigation  of  all  cases,  first,  that  b  after  r  and  /  is  a  voiced  labial  spirant ; 
and,  secondly,  that  the  sandhi  theory  as  stated  by  Streitberg  (Gotisches 
Elementarbuch)  for  the  explanation  of  ^,  d  and  «,  where  /,/  and  s  would  be 
expected,  is  untenable,  as  are  likewise  the  theories  of  Kock  (Zfda.  XXV)  and 
Wrede  (Heyne's  Ulfilas,  gth  ed.).  The  forms  are  to  be  explained  rather  by 
leveling,  which  at  first  was  only  a  matter  of  spelling,  but  afterwards  *  prepared 
the  way  for  the  representation  of  the  real  voiced  spirant  in  sandhi,  which  is  to 
be  seen  in  the  first  eight  chapters  of  Luke,  perhaps  in  isolated  cases  else- 
where.* The  d  in  the  verbal  endings  is  due  to  *a  sound-change  in  East 
Gothic,  by  which  the  voiceless  spirant  became  voiced  in  an  unaccented 
syllable.'  In  the  fifth  article.  Otto  B.  Schlutter,  of  the  Hartford  High  School, 
offers  a  number  of  corrections  and  criticisms  of  Sweet's  edition  of  the  Oldest 
English  Texts.  H.  Schmidt- Wartenberg,  of  the  University  of  Chicago, 
follows  with  a  series  of  investigations  (illustrated)  made  with  the  Rousselot 
apparatus  on  r-sounds,  and  on  the  quantity  of  labials  in  Finnic  Swedish  as 
determined  with  Rosapelly's  lip  observer.  In  the  article  on  Teutonic  '  eleven ' 
and  '  twelve,'  F.  A.  Blackburn,  of  the  University  of  Chicago,  would  substitute 
for  the  derivation  and  explanation  of  the  ending  Uf  of  these  two  words,  as 


niuili  I  illi^  *    iOO^IP- 


.  II  ■  Will  I. 


REVIEWS  AND  BOOK  NOTICES,  22/ 

given  by  Kluge,  a  derivation  from  a  nominal  form  UH,  root  lip^  meaning 
*  addition.'  AinliH  would  then  mean  '  having  one  as  an  addition/  a  derivation 
which,  however,  fails  to  explain  the  Lithuanian  forms.  The  last  article,  On 
the  Hildebrandslied,  is  by  the  editor,  Professor  Karsten,  who  defends  the 
theory  that  the  original  text  was  OS.,  explains  the  HG.  forms  by  the  dialect 
of  the  first  scribe,  and  presents  emendations  for  verses  48  and  30. 

This  first  number  as  a  whole  fully  comes  up  to  the  high  expectations  which 
were  entertained  of  it,  and  augurs  well  for  the  future.  The  names  of  the 
co-editors,  Professors  Cook  of  Yale  for  English,  White  of  Cornell  for  the 
History  of  German  Literature,  and  Hench  of  Michigan  for  the  Historical 
Grammar  of  the  Germanic  Dialects,  together  with  Professor  Georg  Holz  of 
Leipzig,  and  a  large  number  of  European  scholars  who  have  promised 
co-operation,  guarantee  that  the  following  numbers  will  contain  thorough  and 
careful  work,  and  that  the  scholarly  character  of  the  journal  will  be  kept  up 
to  a  high  standard.  Its  continuance  is  provided  for  by  the  financial  support  of 
seven  gentlemen  in  Indianapolis,  to  whom  all  friends  of  Germanic  studies  in 
America  owe  a  debt  of  gratitude.  It  is  a  most  encouraging  sign  for  the  future 
of  learning  in  this  country  that  those  who  stand  outside  of  the  body  of  scholars, 
strictly  speaking,  should  so  munificently  show  their  interest  in  a  distinctly 
scientific  journal,  and  that  in  a  way  so  free  from  selfishness  or  ostentatious 
display.  Such  generosity  ought  to  call  forth  an  equally  generous  spirit  of 
support  in  the  community  of  scholars  and  students  more  directly  interested. 
Yale  UNXva«siTY.  GUSTAV  GrUENER. 


Digitized  by  VjOOQ IC 


REPORTS. 

Rbvue  de  Philologib.    Vol.  XX. 

No.  I. 

1.  Pp.  1-1 1.  Paul  Girard  discusses  two  passages  of  Aeschylus.  I.  Pers. 
527-31.  After  a  brief  examination  of  the  views  of  others,  M.  Girard  advances 
the  theory  that  these  verses  are  interpolated,  and  that  they  were  at  first 
inserted  after  v.  851. — II.  Theb.  961  ff.  He  examines  the  arguments  of  those 
that  reject  this  closing  scene,  and  finds  them  unsound. 

2.  Pp.  12-22.  Philippe  Fabia  investigates  the  conflicting  accounts  of  the 
adultery  of  Nero  and  Poppaea,  and  gives  the  preference  to  that  of  the  Annals 
of  Tacitus. 

3.  P.  22.    L.  Havet  proposes  *  ftiratrina'  in  Nonius,  p.  63  M. 

4.  Pp.  23-35.  Albert  Martin  publishes  an  article  left  nearly  complete  by 
Charles  Graux  on  some  unpublished  fragments  of  Lydus  irepl  duxrrffieiuv,  found 
in  the  Library  of  the  King  of  Spain. 

5.  Pp.  36-7.  C.  E.  Ruelle  discovers  that  the  fragment  of  *  Numenius  on 
Matter'  {^ovfiifviov  wepi  ih^c)  in  the  Escurial,  referred  to  by  some  writers,  is 
nothing  but  an  extract  from  Plotinus  (pp.  30S-22  ed.  princeps). 

6.  Pp.  3S-40.  Notes  on  some  MSS  of  Patmos,  by  J.  Bidez  and  L.  Parmen- 
tier.    I.  Fragments  of  Dio  Chrysostomus.    (To  be  continued.) 

7.  Pp.  41-2.  Louis  Duvau  reads,  Phaedr.  I  15. 1-2,  In  principatu  commu- 
tando  civium  |  nil  praeter  domin^j  inopgs  mutant  saepius.  Id.  Appendix  16,  7. 
for  *facinoris'  he  reads  *funeris'  =  cadaveris. 

8.  P.  42.     In  Babrius  LXI  (75),  id.  Tournier  proposes  ov  irapanaTo, 

9.  Pp.  43-52.  On  the  correspondence  of  Flavius  Abinnius,  by  Jules 
Nicoles.  Some  sixty  papyrus  MSS  found  at  Fayoum,  and  now  partly  in  the 
British  Museum,  partly  in  the  Library  of  Geneva,  furnish  an  outline  of  the 
life  of  Flavius  Abinnius  from  A.  D.  343  to  350.  Abinnius  was  commander  of 
an  ala  of  cavalry  {knapx<K  ^2^).  and  is  sometimes  called  also  irpaitrdotToc 
Kdarpoig,  All  the  documents  are  in  Greek  except  two  in  Latin.  Only  two  are 
written  by  him.  He  may  have  forgotten  to  send  these,  or  they  may  be  rough 
drafts  of  letters  sent.  Half  of  the  papers  are  official,  half  of  them  private. 
They  throw  important  light  on  several  questions.  Nicoles  publishes  the  text 
of  two :  the  first,  in  Latin,  dismissing  Abinnius  from  his  command  (A.  D.  344 ; 
in  346  he  is  found  reinstated) ;  the  second,  in  Greek,  an  instrument  conveying 
to  him  the  possession  of  two  cows,  for  which  he  has  paid  1200  talents  (in  the 
depreciated  currency  of  the  times).    One  of  the  cows  was  named  aoAc  . . . 


Digitized  by  VjOOQ IC 


f^m 


REPORTS,  22g 

(two  or  three  letters  obliterated),  the  other  oreeiaei.  Whatever  may  be  said  of 
the  former,  the  latter,  as  a  single  word,  belongs  to  none  of  the  languages  then 
used  in  Egypt— Latin,  Greek,  Egyptian.  Nicoles  suggests  that  the  explana- 
tion may  be  furnished  by  a  fact  which  M.  Rend  Bazin  records  in  his  Italians 
d*aufowrl*hui  (pp.  224  ff.),  that  in  various  parts  of  Southern  Italy  cattle  are 
called,  not  by  single  names,  but  by  short  phrases,  such  as  proverbs,  refrains  of 
popular  songs,  hucksters'  cries,  etc.  He  thinks  that  (yretuitt  may  be  the  begin- 
ning of  such  a  phrase,  6re  cl  act,  and  that  aake  .  .  .  may  be  aa^^ei  (jJ  vovf,  for 
instance).  [One  naturally  recalls  the  analogous  names  of  men  of  the  good 
old  Puritan  days,  such  as  If-God'had-nei-died-for-ihee'thoU'hadst'been'^amned 
Barebones^  The  name  of  Abinnius  is  written  *Apivatog,  'AjStwawf,  'Afiivatoiy 
'AfilwaioCt  *Afiiweioc,  'Afiiwto^,  *'Epiwtog,  The  use  of  fjt  seems  to  indicate  that  /? 
was  already  losing,  or  had  lost,  its  full  labial  character. 

10.  Pp.  53-6.  Geoi^es  Lafaye  defends  the  reading  of  the  editia  princeps 
(i.  e.  of  Cod.  Sangallensis)  in  Statins,  Silvae  I,  Preface,  1.  28  (Baehrens).  His 
defence  seems  conclusive. 

11.  Pp.  57-9.  Critical  notes  by  H.  van  Herwerden  on  seventeen  passages 
of  Callinicus,  Vita  S.  Hypatii. 

12.  Pp.  60-64.  Epigraphic  notes,  by  Jean  Negroponte.  Discussion  of  a 
bilingual  (Latin  and  Greek)  inscription  found  near  the  railway  station  of 
Deirmendjik,  and  published  (1895)  at  Athens ;  also  of  two  or  three  other  small 
inscriptions. 

13.  Pp.  65-7.  L.  Havet  explains  Lucilius  317  (Baehrens)  and  Phaedrus,  V 
7.  26.    Pascal  Monet  emends  Lucian,  Charon  15. 

14.  Pp.  68-72.  Book  Notices,  i)  Philo:  About  the  Contemplative  Life, 
or  The  fourth  book  of  the  Treatise  Concerning  Virtue,  critically  edited  with 
a  defence  of  its  genuineness,  by  F.  C.  Conybeare;  Oxford,  1895.  Joseph 
Viteau  gives  a  brief  description  of  this  work,  which  he  finds  full  of  valuable 
information,  but  objects  to  a  very  small  number  of  statements.  2)  J.  P.  Walt- 
zing, £tude  historique  sur  les  corporations  professionelles  chez  les  Romains, 
t.  I,  Louvain,  1895.  F.  C.  considers  this  a  much-needed  work,  and  predicts 
that,  when  completed,  it  will  add  much  to  our  knowledge.  3)  Bibliotheca 
scriptorum  Graecorum  et  Romanorum  Teubneriana.  Under  this  head  P.  C. 
gives  a  brief  and,  in  the  main,  favorable  account  of  Herondas  (2d  ed.)  by  O. 
Crusius,  the  Politica  of  Aristotle  by  Susemihl,  ApoUodori  Bibliotheca  by  R. 
Wagner,  Epicteti  Dissertationes  ab  Arriano  digestae  by  H.  Schenkl,  and 
barely  mentions  Dion  Cassius  by  Melber  and  Plut  Moralia,  vol.  6,  by  Bemar- 
dakis.  4)  P.  C.  commends  Goodwin  and  White's  Anabasis  and  White  and 
Morgan's  dictionary  to  the  Anabasis.  5)  P.  C.  pronounces  The  Hecuba  of 
Euripides,  by  W.  S.  Hadley,  Cambridge,  1894,  neither  the  best  nor  the  worst 
of  the  series  to  which  it  belongs. 

No.  2. 

I.  Pp.  73-83.  On  qu  in  liqtddus,  liquor^  Hquem^  aqua^  by  Louis  Havet. 
The  author  retracts  the  whole  of  his  article  on  this  subject  published  in  the 
Revue  de  Philologie,  1891,  pp.  8  ff.      He  now  denies  that  liqiUdus  (four 


Digitized  by  VjOOQ IC 


230  AMERICAN  JOURNAL    OF  PHILOLOGY. 

syllables),  dqUa  (three  syllables)  and  like  resolutions  occur  at  all.  Moreover, 
some  of  the  examples  of  hqu-  are  not  from  hqui,  but  from  liquere.  The 
length  of  the  syllable  is  simply  quantity  by  position  :■  liq-uidus.  The  same  is 
true  of  aqua  several  times.  Had  the  Pisistratidae  edited  Homer  with  digamma 
written  in  such  words  as  idfetaev,  aqua  would  have  been  as  common  Aspdiris, 
As  it  is,  only  Lucretius  and  Laevius  applied  the  principle  to  Latin  independ- 
ently of  Greek  models.  Even  liqumtia  flumina  in  Verg.  Aen.  IX  679  is  no 
exception,  although  liquenUa  here  cannot  come  from  liqui)  for,  as  Servius 
expressly  says,  this  is  a  proper  name  (in  adjective  form ;  cf.  stagna  Aufida  and 
the  like).  The  modem  name  is  Livenza^  though  it  seems  probable  that  the 
earliest  form  of  the  name  was  Liqu€tia,  and  had  nothing  to  do  either  with 
liqui  or  with  liquere.  The  insertion  of  n  was  due  to  analogy,  and  is  illustrated 
by  Vicenza,  which  was  ViceHa  in  ancient  times. 

2.  Pp.  84-8.  P.  Foucart,  by  means  of  two  Greek  inscriptions,  fixes  the 
reign  of  Tachos  between  360  and  357,  and  discusses  the  dates  of  events 
connected  with  the  contest  between  Samos  and  Priene,  especially  the  arbi* 
tration  of  the  Rhodians. 

3.  P.  88.  K.  D.  Mylonas  publishes  an  inscription  giving  the  name  of  a 
hitherto  unknown  sculptor,  fArpfo^  of  Pergamus. 

4.  Pp.  89-92.  Critical  notes  on  nine  passages  of  Aristot.  Poet.,  by  Mdderic 
Dufour. 

5.  Pp.  93-4.  Louis  Havet  proposes.  Plant.  Amphitruo  96,  Comoediai  dum 
huius  argumentum  eloquor,  and  shows  how  the  corruption  probably  arose  from 
V.  51. 

6.  Pp.  95-101.  Epigraphic  notes,  by  B.  Haussoulier.  Discussion  of  a  few 
inscriptions  from  the  neighborhood  of  Heronda.  These  establish  an  *kn6A}uuv 
Xitdavaaaeh^,  An  examination  of  avroiriK  shows  that,  contrary  to  what  some 
had  maintained,  it  has  its  ordinary  meaning  in  certain  inscriptions. 

7.  Pp.  101-2.  Louis  Havet  writes  an  interesting  note  on  C.  I.  L.  V  1939 
(Concordia). 

8.  Pp.  104-15.  On  the  first  two  Ptolemies  and  the  confederation  of  the 
Cyclades,  by  J.  Delamarre.  An  inscription  (containing  62  lines  of  about  35 
letters  each,  and  discovered  in  1893  on  the  little  island  of  TiiKovpyid  near 
Amorgos)  is  made  the  basis  of  an  instructive  investigation  of  the  origin  of  the 
confederation  of  the  Cyclades  and  its  relation  to  the  kingdom  of  Egypt.  The 
inscription  contains  other  valuable  information,  especially  concerning  the 
'isolympic'  games  celebrated  at  Alexandria. 

9.  Pp.  116-25.  Notes  on  some  MSS  of  Patmos  (continued),  by  J.  Bidez 
and  L.  Parmentier.  II.  First  a  critical  account  of  other  MSS  and  editions  of 
Evagrius'  Ecclesiastical  History  is  given,  then  a  Patmos  MS  is  described  and 
a  collation  of  many  important  passages  is  presented,  illustrating  the  value  of 
this  MS.  III.  The  same  MS  contains  also  the  Ecclesiastical  History  of 
Socrates,  which  is  briefly  described. 

10.  Pp.  126-8.  Book  Notices,  i)  J.  J.  Binder,  Laurion,  die  attischen 
Bergwerke  im  Alterthum;   Laibach,  1895;  unfavorably  criticised  by  £.  A. 


Digitized  by  VjOOQ IC 


REPORTS.  231 

2)  Carlo  Pascal,  II  culto  di  Apollo  in  Roma  nel  secolo  di  Augusto;  Roma, 
1895 ;  favorably  mentioned  by  Georges  Goyau.  3)  Carlo  Pascal,  Acca  Larentia 
e  il  mito  della  terra  Madre  a  proposito  di  un  passo  dei  Fasti  Prenestini; 
Roma,  1894.  A  work  of  31  pages,  considered  by  Georges  Goyau  a  useful 
collection  of  passages  relating  to  the  subject.  4)  Ettore  Ciccotti,  La  fine  del 
secondo  Triumvirato;  1895.  Georges  Goyau  gives  brief  summary.  It  is  a 
question  of  chronology.  5)  M.  Deloche,  Le  port  des  anneaux  dans  Tantiquit^ 
romaine  et  dans  les  premiers  si^cles  du  moyen-age;  Paris,  1895;  briefly  sum- 
marized, with  high  commendation,  by  Georges  Goyau.  6)  C.  Castellani,  Cata- 
logus  codicum  Graecorum  qui  in  bibliothecam  D.  Marci  Venetiarum  inde  ab 
anno  MDCCXL  ad  haec  usque  tempora  inlati  sunt.  Briefly  described  and 
pronounced  very  useful  by  C.  £.  R. 

No.  3. 

1.  Pp.  129-45.  Nero  and  the  Rhodians,  by  Philippe  Fabia.  I.  The  rela» 
tions  of  the  Rhodians  to  Rome  before  Nero.  II.  The  date  of  the  restoration 
of  their  autonomy.  Discussion  of  the  accounts  of  Tacitus  and  Suetonius, 
showing  that  the  preference  is  to  be  given  to  the  former,  and  that  the  date  was 
A.  D.  53.  III.  The  threat  of  Nero  to  escape  his  mother's  yoke  by  abdicating 
and  retiring  to  Rhodes,  and  his  reasons  for  selecting  that  place.  IV.  Discus- 
sion of  an  inscription,  recently  published  by  Hiller  von  Gaertringen,  relating 
to  an  embassy  from  the  Rhodians  to  Nero  in  the  first  year  of  his  reign. 
V.  The  escape  of  the  Rhodians  from  pillage  of  works  of  arts  at  the  hands  of 
Nero*s  agent,  Acratus. — An  interesting  and  instructive  article. 

2.  Pp.  146-8.  Louis  Havet  critically  discusses  Phaedrus,  IV  20,  V  i.  10, 
V  5.  18-19. 

3.  Pp.  149-50.  C.  E.  Ruelle  collates  two  pages  of  the  Epitome  prior  of  the 
CUmetUinae^  found  written  on  the  cover  of  a  MS  of  Ptolemy  (Paris,  Greek  MS 
1403). 

4.  Pp.  151-4.  A.  Cartault  declines  to  accept  the  conclusion  reached  by 
Louis  Havet  (Rev.  d.  Phil.  XII,  pp.  145  ff.)  and  approved  by  other  scholars, 
transposing  w.  616-20  of  Verg.  Aen.  VI  so  as  to  follow  v.  601.  He,  on  the 
contrary,  places  602-7  after  620,  shows  how  the  transposition  probably 
occurred,  and  that  the  proposed  arrangement  is  in  every  respect  satisfactory. 

5.  P.  155.    L.  Havet  proposes  sacerrume  in  Plant.  Trin.  540. 

6.  Pp.  156-8.  C.  E.  Ruelle  denies  the  correctness  of  avfi^uviag  and  avfju^ia 
in  the  disputed  passage  of  Arist.  Quintil.,  p.  26  (Meibom),  and  restores 
ofKXfuviag,  dfMM^ia.  The  use  of  avfu^ia  =  dfio^uvia  is  shown  to  be  inconsistent 
with  the  usage  of  Aristides  himself.  The  converse  change  of  aifjupuvoi  into 
6/i6^uvoi  occurs  in  all  the  MSS  of  Martianus  Capella,  De  Nupt.  Phil.  IX  947 
(Kopp). 

7.  Pp.  159-64.  Book  Notices,  i)  F.  Robiou,  L'etat  religieux  de  la  Grdce 
et  de  rOrient  au  sidcle  d' Alexandre.  II.  Les  regions  syro-babyloniennes  et 
r^ran ;  Paris,  1895  ;  unfavorably  mentioned  by  Ch.  Michel.  2)  Dionis  Prusae- 
ensis  quem  vocant  Chrysostomum,  quae  extant  omnia  edidit  etc.  J.  de  Arnim ; 
vol.  II,  Berlin,  1896;  described  and  commended  by  F.  C.     3)  F.  T.  Cooper, 


Digitized  byj 


232  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Word-formation  in  the  Roman  Sermo  Pkbeitu\  Boston,  1895.  T.,  in  a  notice 
of  some  length,  finds  that  this  work  exhibits  learning  and  diligence,  but 
otherwise  his  remarks  are  chiefly  unfavorable.  4)  P.  Terenti  Phormio,  with 
Notes  and  Introduction,  by  H.  C.  Elmer;  Boston.  1895.  Philippe  Fabia 
describes  this  work,  on  the  whole  favorably,  but  finds  that  in  the  Introduction 
the  special  study  of  the  play  is  too  much  sacrificed  to  generalities.  5)  The 
Adelphoe  of  Terence,  by  William  L.  Cowles;  Boston,  1896.  Pronounced 
"soign^  et  bien  imprime"  by  Philippe  Fabia,  though  some  slight  strictures  are 
made. 

No.  4. 

1.  Pp.  165-75.  P*  Couvreur  publishes  a  catalogue  of  the  papyrus  Greek 
MSS  discovered  in  recent  times.  The  names  of  authors,  whose  fragments  or 
works  are  contained  in  these  MSS,  are  given  in  chronological  order  in  two 
lists — one  for  poetry,  one  for  prose.  The  bibliography,  except  where  it  is 
very  voluminous,  as  in  the  case  of  Hero(n)das,  is  added ;  also  the  date  of  each 
MS.  Those  that  contain  anything  otherwise  unknown  are  marked  with  an 
asterisk.  The  author  requests  scholars  to  inform  him  of  any  omissions  he 
may  have  made.  This  catalogue  must  have  cost  much  labor,  and  Hellenists 
cannot  be  too  grateful  for  so  useful  a  work. 

2.  Pp.  175-7.  Paul  Tannery  proposes  *cacumen  perlibratum  cum  oculo^  in 
Vitruvius  Rufus,  §39. 

3.  Pp.  178-84.  Louis  Havet  critically  discusses  Phaedr.  Ill,  Prol.  38 
(II  Epil.  14) ;  III  15,  20 ;  III  Epil.  2 ;  V  5, 11-12  (and  I  29,  3);  Appendix  6, 6. 

4.  P.  185.     In  Ter.  Eun.  588,  A.  Mace  proposes  hiemem  for  homimm, 

5.  Pp.  186-7.  Otto  Keller  critically  discusses  Anecdota  Bernensia,  ed. 
Hagen,  p.  187 ;  Alexand.  Aphrodis.  2,  16;  Oros.  VII  9, 14. 

6.  Pp.  188-90.    J.  Chauvin  proposes  sucturrit  for  quaerit  in  Phaedr.  IV  9,  2. 

7.  Pp.  191  foil.  Book  Notices,  i)  Quelques  notes  sur  les  Silvae  de  Stace, 
premier  livre,  par  G.  Lafaye ;  Paris,  1896.  Jules  Chauvin  gives  numerous 
illustrations  of  the  great  value  of  this  work.  Of  special  importance  is  the 
happy  use  that  the  author  has  made  of  his  knowledge  of  archaeology. 
2)  Thucydides,  Book  III,  edited  with  Introduction  and  Notes,  by  A.  W. 
Spratt;  Cambridge,  1896.  E.  Chambry  reviews  this  work  at  considerable 
length.  Though  he  enumerates  some  details  which  he  cannot  approve,  he 
says  *^Non  ego  paucis  offendar^"*  and  pronounces  the  edition  an  excellent  one 
and  almost  as  exhaustive  as  it  is  possible  to  make  a  work  of  the  kind.  3)  De 
Flavii  Josephi  elocutione  observationes :  scripsit  Guilelmus  Schmidt ;  Leipzig, 
1893.  Briefly  and  favorably  mentioned  by  J.  Viteau.  4)  J.  J.  Hartmann,  De 
Terentio  et  Donato  commentatio ;  Leyden,  1895.  Philippe  Fabia,  after 
describing  this  book,  says  that,  of  its  four  chapters,  only  the  second  was  worth 
writing.  5)  P.  Cornelii  Taciti  Ab  excessu  divi  Augusti  quae  supersunt. 
Annales  de  Tacite,  texte  soigneusement  revu,  precede  d'une  introduction  et 
accompagne  de  notes  explicatives,  grammaticales  et  historiques.  par  MM. 
Leopold  Constans  et  Paul  Girbal ;  Paris,  1896.  Philippe  Fabia  does  not 
hesitate  to  pronounce  this  the  best  of  all  the  editions  of  the  Annals  that  have 


Digitized  by  VjOOQ IC 


REPORTS,  233 

ever  appeared  in  France.  He  finds  only  the  Introdaction  weak.  6)  Antho- 
logia  Latina,  pars  posterior,  Carmina  Epigraphica  conlegit  F.  Buecheler; 
fascic.  I,  Lipsiae,  1895.  Georges  Lafaye,  after  a  brief  history  of  other 
attempts  to  collect  poetical  inscriptions,  gives  an  account  of  the  origin  of  this 
valuable  work,  "worthy  of  the  eminent  master."  This  volume  contains 
inscriptions  in  the  Saturnian  verse,  iambics,  trochaics,  and  the  dactylic 
hexameter.  The  second  volume  will  contain  those  composed  in  the  elegiac 
form.  7)  Cassii  Dionis  Cocceiani  Historiarum  Romanarum  quae  supersunt 
edidit  Ursulus  Philippus  Boissevain  ;  vol.  I,  Berolini,  T895.  Briefly  described 
by  Dx.,  who  says  it  merits  the  thanks  of  philologians  and  especially  historians. 

The  Revue  des  Revues,  begun  in  a  previous  number,  is  finished  in  this 
number. 

Milton  W.  Humphreys. 


Englischb  Studibn.    Herausgegeben  von  Dr.  Eugen  KOlbing,  Leipzig. 

XXI.  Band,  1895. 

I. — F.  Graz,  Contributions  to  the  Textual  Criticism  of  the  so-called  Caed- 
monian  Poems.  In  his  article,  *  Die  Metrik  der  sogenannten  Cadmon'schen 
Dichtungen,'  in  Part  III  of  Studien  zum  Germanischen  Allitterationsvers, 
edited  by  Kaluza,  Graz  suggested  emendations  on  the  basis  of  the  metre. 
The  present  article  discusses  those  emendations  more  fully. 

Ph.  Aronstein,  John  Marston  as  a  Dramatist.  This  article  is  a  continuation 
of  a  study  begun  in  vol.  XX.  Part  II  is  devoted  to  the  literary  criticism  of 
the  poet's  work,  and  Part  III  is  a  brief  conclusion.  The  tragedies  and 
comedies  are  treated  separately.  In  the  first  group  are  Antonio  and  Mellida, 
Parts  I  and  II,  The  Malcontent,  Sophonisba,  The  Insatiate  Countess.  The 
comedies  are  What  You  Will,  The  Dutch  Courtezan,  Parasitaster  or  The 
Fawn.  The  order  of  discussion  in  each  case  is :  a  sketch  of  the  plot ;  the 
sources;  the  idea;  the  plot-treatment;  the  characters;  the  language  and 
style ;  final  estimate.  The  second  part  of  Antonio  and  Mellida,  called 
Antonio's  Revenge,  was  planned  as  a  satiric  comedy,  but  is  really  a  tragedy 
of  blood.  The  first  part  is  evidently  from  some  Italian  novel,  and  the  second 
follows  Thomas  Kyd*s  Spanish  Tragedy,  but  we  find  suggestions  of  Romeo 
and  Juliet  and  Much  Ado  About  Nothing.  The  poet  exercises  little  poetic 
justice.  In  diction  this  drama  best  illustrates  Marston's  excellencies  and 
faults.  He  is  a  reflective  lyrist.  Passages  of  tenderness,  such  as  IV.  i.  12, 
are  only  oases  in  a  wilderness  of  bombast.  Jonson  in  the  Poetaster  scores 
Marston  for  his  use  *of  wild,  outlandish  terms'  and  his  use  of  high-sounding 
diction  in  preference  to  simple  Anglo-Saxon.  The  whole  drama  shows  the 
need  of  a  discipline.  Of  the  Malcontent,  the  source  may  be  some  Italian 
novel,  or  it  may  have  been  constructed  by  Marston  himself  after  the  plan 
of  Antonio  and  Mellida.  Its  style  shows  Jonson's  influence.  Sophonisba 
is  a  historical  drama,  taken  directly  from  Livy,  bks.  27;  28;  29;  30,  §§i-i6. 
The  story  is  told  also  in  Appian's  history  of  Spain,  and  briefly  in  Polybius. 
The  subject  had  been  treated  by  Trissino  in  1524,  and  Marston  may  have 
made  some  use  of  that  treatment.     The  witch-scene  comes  directly  from 


Digitized  by  VjOOQ IC 


234  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Lucan,  Pharsalia  6.  488  ff.  With  a  few  exceptions,  the  style  is  bombastic  and 
repulsiye.  The  Insatiate  Countess  contains  two  poorly  joined  plots.  Its 
sources  are  the  fourth  and  fifteenth  novels  of  Bandello,  but  it  is  also  heavily 
marked  with  Shakespeare.  Of  the  comedies,  What  You  Will  is  drawn  directly 
from  the  Amphitruo  of  Plautus  or  through  the  Italian.  It  shows  some  skill  in 
detail,  but  is  without  unity.  Lampatho  Doria,  a  mad  scholar,  is,  Aronstein 
thinks,  a  caricature  of  Jonson,  while  Quadratus,  the  misanthrope,  is  Marston's 
self.  The  Dutch  Courtezan  is  one  of  the  best  of  the  Elizabethan  comedies. 
It  is  not  only  the  contrast  between  a  high  and  a  low  woman,  but  between  the 
ascetic  and  the  man  of  wide  experience.  Its  characters  and  diction  sire  the 
poet's  best.  Parasitaster  is  built  upon  a  device  of  the  Adelphi  of  Terence, 
which  appears  also  in  the  third  novel  of  the  third  day  of  the  Decameron.  It 
contains  enough  material  for  three  or  four  better  plays.  Marston  was  well 
acquainted  with  Latin  literature.  Seneca  was  his  inspirer.  Of  his  contem- 
poraries, he  follows  Jonson  more  closely  in  his  comedies  and  in  form,  but 
Shakespeare  is  his  help  in  ideas  and  motives,  and  in  the  tragedies.  He  is 
open  to  the  criticism  of  immoderation.  Plots  and  characters  are  in  the 
extreme,  though  their  range  is  small.  His  types  of  women  are  three :  the 
lover  and  heroine,  the  emancipated  woman,  and  the  low  woman.  Marston  is 
more  of  a  dilettant  than  a  poet,  but  the  friends  that  he  makes  are  faithful. 

E.  Nader  gives  an  interesting  report  of  the  Sixth  Summer-Meeting  for 
University  Extension  at  Oxford,  1894.  He  promises  the  historical  sketch  of 
the  movement  which  appears  later  in  the  volume.  • 

Under  Book  Notices  are  reviews  of  O.  Jespersen*s  Progress  in  Language 
with  special  reference  to  English,  P.  Cosijn's  Concise  Early  West  Saxon 
Grammar,  Hall's  Concise  Anglo-Saxon  Dictionary  for  Students,  the  second 
volume  of  Walker's  revision  of  Grein's  Library  of  Old  English  Poetry,  H.  A. 
Vance's  Late  Old  English  Sermo  in  Testis  S*«  Mariae  Virginis,  W.  H.  Hulme's 
Language  of  the  Old  English  Recension  of  Augustine's  Soliloquies,  C.  G. 
Child's  John  Lyly  and  Euphuism.  Under  the  continuation  from  vol.  XX  of 
reviews  of  the  latest  literature  on  the  Elizabethan  drama  are  Brandl's  Shaks- 
pere,  W.  Oechelhftuser's  Shakspeareana,  S.  von  Milletich's  The  Aesthetic 
Form  of  the  Conclusion  (abschliessenden  Ausgleiches)  in  the  Shakespearean 
Drama,  L.  Wurth's  The  Pun  in  Shakespeare,  Julius  Caesar,  translated  into 
German  by  A.  W.  von  Schlegel,  edited  by  A.  Englert.  Other  books  reviewed 
are  the  Manchester  Goethe  Society  Transactions,  E.  H.  Lewis'  History  of 
the  English  Paragraph,  E.  Hausknecht's  The  English  Student,  and  The 
English  Reader,  G.  KrQger's  Systematic  English-German  Vocabulary,  V. 
Olsvig's  Yes  and  No,  Dialogues  in  English  on  Holzer's  Charts. 

Jespersen's  book  is  an  enlargement  and  translation  into  English  of  his 
Studier  over  Engelske  Kasus.  He  rejects  the  theory  of  Schleicher  that  the 
order  of  linguistic  development  was  (i)  isolated  terms,  (2)  their  agglutination, 

(3)  inflection.  In  modern  English,  as  compared  with  ancient  speech,  he  finds 
that  (i)  its  forms  are  shorter,  (2)  there  are  fewer  forms,  (3)  fewer  irregularities, 

(4)  the  more  abstract  character  of  words  facilitates  expression.  Simplicity  was 
not  an  original  characteristic.  An  old  language  presents  with  simple  forms  a 
fixed  order,  and  a  fixed  order  is  *  the  highest,  finest,  and  accordingly  the  latest 


Digitized  by  VjOOQ IC 


-.  "JUW 


REPORTS,  235 

developed  expedient  of  speech.'  The  second  part  discusses  the  question  of 
the  English  plural  in  -f,  and  finds  that  its  uniformity  was  not  due  to  French 
influence.    Case-questions  of  less  interest  are  also  treated. 

Cosijn's  Grammar  would  serve  as  a  good  introduction  to  Sievers'.  The 
phonological  chapter  is  commended  by  Nader  for  the  abundance  of  corres- 
ponding Gothic  forms,  and  the  inflections  for  the  references  to  phonology. 

Hall's  Dictionary  will  be  used  by  the  learner,  where  the  specialist  will  use 
Bosworth-Toller. 

Volume  II  of  Grein's  Library  contains  reprints  of  poems  from  the  Vercelli 
Codex  and  the  Exeter  MS,  including  Andreas,  the  Fates  of  the  Apostles,  the 
Address  of  the  Soul  to  the  Body,  a  Homily  on  Ps.  28,  the  Dream  of  the  Rood, 
Elene  ;  in  the  second  part,  poems  from  the  so-called  Caedmon  MS  at  Oxford 
and  the  Corpus  Christi  MS,  the  Caedmon  Hymn,  and  the  lately  discovered 
inscription  on  the  Brussels  Cross.  The  concluding  volume  will  contain  the 
rest  of  the  Exeter  MS,  the  Metrical  Psalms,  Metres  of  Boethius,  Soloman  and 
Saturn,  and  several  minor  poems.  Gldde  gives  a  specimen  of  Winker's  work. 
Such  a  work  as  this  is  a  safeguard  against  mistakes  arising  from  a  scholar's 
confinement  to  a  narrow  circle  of  originals. 

Frankel,  in  his  review  of  the  late  literature  on  the  Elizabethan  drama, 
criticises  the  crowd  of  drivelers  or  demented  laymen  who  have  attempted  the 
biography  of  Shakespeare.  Shakespeare's  biographer  must  possess  both 
experience  and  scholarship.  The  object  of  Brandl's  book  is  to  show  the 
personality  of  Shakespeare  in  its  changing  phases,  and  the  apparatus  of 
Ikerary-historical  research  is  used  to  serve  this  purpose.  The  poet's  works 
fall  under  (i)  the  Falstaff  period,  (2)  the  Hamlet  period,  (3)  the  Lear  period, 
(4)  the  Romances.  -Then  follow  Cymbeline,  The  Winter's  Tale,  Tempest. 
Fr&nkel  inconsistently  criticizes  the  book's  obscure  style.  He  finds  some 
unwarranted  inferences.  The  book,  however,  is  a  precipitate  of  the  accumu- 
lated knowledge  of  Shakespeare,  and  marks  a  stage  in  the  advance  of  critical 
work. 

For  thirty  years  OechelhUuser  has  been  active  in  the  aesthetic  and  dramatic 
criticism  of  Shakespeare's  plays.  In  the  problems  of  their  staging  the  internal 
evidence  of  the  plays  must  be  considered.  His  book  contains  eight  essays, 
among  which  is  a  most  careful  analysis  of  Richard  III.  Most  of  the  work  is 
devoted  to  a  consideration  of  the  adaptability  of  the  plays  to  the  stage.  The 
author  looks  forward  to  an  advantageous  adaptation  of  the  plays  to  our  boards. 

Milletich,  from  the  standpoint  thaf  the  poet  must  in  his  conclusion  set  forth 
clearly  and  in  harmony  his  view  of  life,  treats  his  subject  with  much  help 
from  both  Zimmerman  and  Knauer  of  Vienna.  The  book  is  guilty  of  dilet- 
tantism and  some  inaccuracies.  Wurth  thinks  Shakespeare's  use  of  the  pun 
is  a  worthy  criterion  of  the  poet's  dramatic  art. 

Lewis'  treatment  of  the  paragraph  was  his  doctoral  thesis  at  Chicago. 
After  a  careful  historical  consideration,  beginning  with  the  oldest  MSS,  he 
concludes  that  Hunt's  definition, '  a  collection  of  sentences  unified  by  some 
common  idea,'  is  historically  the  most  accurate.  GlMe  especially  commends 
Lewis'  skill  of  selection. 

In  the  Miscellanea,  KSlbing  offers  emendations  to  the  text  of  William  of 
Shoreham's  most  interesting  though  incomplete   religious  poem,  *In  Holy 


Digitized  by  VjOOQ IC 


236  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Sauter  we  may  rede/  and  points  to  the  need  of  a  well-edited  edition  of  this 
poet's  complete  works.  Emendations  are  also  suggested  to  the  text  of  A.  W. 
Pollard's  English  Miracle  Plays,  Moralities,  and  Interludes,  Oxford,  1890. 
This  book  of  selections  goes  to  fill  a  great  want  in  a  rather  neglected  period. 
K51bing's  notes  on  Byron  explain  that  the  dedication  which  now  heads  Childe 
Harold,  *To  lanthe*  (Lady  Charlotte  Harley,  second  daughter  of  the  fifth 
Earl  of  Oxford),  appeared  first  in  the  seventh  edition,  1814.  The  song  *  Good 
Night,*  immediately  following  Canto  I.  xiii,  is  shown,  by  a  collection  of 
interesting  parallels,  to  be  an  imitation,  both  in  matter  and  spirit,  of  the 
Border  Minstrelsy,  edited  by  Scott,  and  of  Percy's  Reliques.  Other  notes  are 
by  Frankel,  on  the  Legend  of  the  Hermit  and  the  Angel ;  Wolfing,  on  the 
meaning  of  M.  E.  eroud\  Gnerlich,  on  the  etymology  oi  pedigree,  Gruber 
notes  the  discovery  in  Berlin  of  the  oldest  edition  of  Steele's  plays.  Its  date 
is  1723,  which  is  38  years  earlier  than  any  hitherto  known.  Ellinger  corrects 
Swaen  (Eng.  Stud.  XX  266  ff.)  in  a  note  on  the  verb  to  dare,  A.  SchrOer 
pays  tribute  to  the  service  of  Miss  Laura  Soames,  who  died  Jan.  24,  1895. 
Her  great  service  began  in  the  use  which  she  made  of  phonetics  as  a  means  of 
teaching  foreign  languages  to  children.  In  the  science  of  language-history 
the  phonology  of  the  living  tongue  grows  every  day  more  important.  Miss 
Soames'  work  is  most  valuable  for  its  conscientious  observation. 

II. — ^J.  H.  Hall  prints  three  short  religious  pieces  from  MS  Cotton  Galba  E. 
IX,  two  of  them  for  the  first  time. 

J.  Hoops,  Keats*  Youth  and  Early  Poems.  After  a  brief  review  of  the 
Georgian  poets  and  their  position,  the  author  says  that  the  two  whose  spirit 
has  stamped  the  Victorian  poetry  are  Wordsworth  and  Keats.  Both  are  little 
known  in  Germany :  Wordsworth  because  of  Anglo-Saxon  peculiarities;  Keats, 
who  is  more  universal,  through  lack  of  a  good  translation  and  a  stout  champion. 
The  translation  is  forthcoming  from  the  hands  of  Marie  Gothein.  The  present 
article  proposes  to  meet  the  translation  with  a  treatment  for  Germany  of  the 
biographical  and  literary  side.  It  contains  little  that  is  new,  and  makes  use 
of  much  second-hand  material.  The  following  sections  are  treated :  parentage 
and  early  childhood ;  school  at  Enfield ;  apprenticeship  at  Edmonton ;  study 
of  medicine  in  London ;  vacation  at  Margate ;  Keats  and  Leigh  Hunt ;  the 
winter  of  1816-17;  the  volume  of  1817  and  its  reception.  Naturally,  the 
article  deals  mostly  with  the  forces  which  entered  into  the  development  of 
the  poet*s  art.  On  the  evidence  of  some  remarks  by  Hunt  in  an  essay  entitled 
'Young  Poets,'  in  the  Examiner  of  Dec.  i,  1816,  Hoops  shows  that  Keats 
could  not  have  met  Hunt  until  shortly  before  this  date,  and  not  early  in  the 
year,  as  was  hitherto  believed.  Detailed  evidence  from  the  volume  of  1817 
shows  the  influence  upon  Keats  of  Chaucer,  Chapman,  Browne,  Milton,  and 
Moore.  Other  poets  whom  Keats  had  read  at  this  time,  but  whose  influence 
cannot  be  traced  in  detail,  are  Shakespeare,  Chatterton,  Byron,  and  Words- 
worth. The  early  poems  forecast  Keats'  wonderful  and  deep  familiarity  with 
nature,  as  well  as  his  inability  to  comprehend  human  passion  and  give  it 
poetic  expression. 

A.  Pakscher  describes  the  theory  and  working  of  the  Berlits  method  of 
instruction  in  foreign  languages. 


Digitized  by  VjOOQ IC 


REPORTS.  .      237 

In  the  Miscellanea  R.  R.  de  Jong  shows  that  the  distinction  between  -inde 
and  -ende  in  rimes  holds  not  only  in  Robert  of  Gloucester's  Chronicle,  as 
Balbring  (Eng.  Stud.  XX  149)  showed,  but  also  in  Sir  Beues  of  Hamtoun, 
probably  in  Guy  of  Warwick,  and  possibly  in  Sir  Ferumbras.  P.  Bellezza 
points  to  the  use  of  the  plowman  in  Macaulay  and  Tennyson.  Swaen  rejects 
the  old  derivation  of  Caliban  from  Cannibal,    It  may  be  from  Gipsy  kalo  = 


Digitized  by  VjOOQ IC 


238  AMERICAN  JOURNAL    OF  PHILOLOGY. 

being  a  noun.  This  seems  to  show  that  alliteration  is  a  comparatively  late 
embellishment  of  the  verse,  and  somewhat  external  to  it.  In  the  first  hemi- 
stich the  prevailing  alliterative  form  is  ajax^  where  a  is  the  alliterative  and  x 
the  non-alliterative  word.  This  analysis  reveals  the  great  flexibility  of  Old 
English  verse.  The  quiet  flow  of  normal  verses  might  be  broken  at  any 
moment  by  the  more  solemn  or  excited  expansion. 

The  object  of  F.  Maychrzak's  elaborate  study  of  Byron  as  a  translator  is 
twofold:  (i)  to  furnish  a  critical  treatment  of  Byron's  translations:  (2)  to 
show  their  relation  to  his  original  poetry.  The  article  falls  into  three  parts: 
(i)  Byron's  acquaintance  with  foreign  languages,  (2)  the  translations  and  their 
relations  to  the  originals,  (3)  the  relation  of  the  diction  in  his  translations  to 
his  diction  in  general.  (3)  is  to  be  treated  in  vol.  XXII.  In  school  Byron 
did  not  succeed  with  Latin,  Greek,  French,  and  German.  His  acquisition  of  a 
language  came  by  *  rote  and  ear  and  memory '  in  its  own  home.  Thus  it  was 
with  what  Spanish  and  Portuguese  he  knew.  His  modern  Greek  was  begun 
in  Albania.  In  Venice,  in  1816,  he  applied  himself  to  the  study  of  Armenian 
in  a  cloister  of  Armenian  monks.  His  knowledge  and  love  of  Italian  were 
most  important.  His  study  of  Bandello,  Dante,  the  tragedies  of  Alfieri  and 
Monti,  as  well  as  his  translation,  were  inspired  by  the  Countess  Guiccioli. 
The  translations  are  compared  line  for  line  with  the  originals,  and  separable 
amplifications,  which  generally  amount  to  one-third  or  one-half  the  length 
of  the  actual  translation,  are  collected  at  the  end.  In  his  treatment  of  the 
classics  Byron  merely  paraphrases,  though  some  passages,  like  the  Anacre- 
ontics, are  more  literal.  His  Morgante  Maggiore,  and  the  Francesca  da 
Rimini  from  Inferno  V,  are  much  finer  work.  The  mournful  ballad  on  the 
Siege  and  Conquest  of  Alhama  is  from  uncertain  Spanish  originals. 

E.  Nader  presents  a  short  but  interesting  historical  sketch  of  the  University 
Extension  movement,  especially  in  England. 

The  Miscellanea  contains  a  note  on  the  name  Ophelia  by  Sarrazin,  one  on 
Germanic  legends  in  England  by  Kluge,  a  note  on  pedigree  by  Skeat,  two  notes 
of  correction,  a  lately  discovered  letter  of  Charles  Dickens,  and  an  obituary 
notice  by  KOlbing  of  Julius  Zupitza,  who  died  July  6,  1895. 

Ophelia,  it  seems,  is  not  Greek,  but  Irish.  It  is  the  name  of  a  barony 
invaded  by  Essex  in  1599,  the  possible  date  of  Hamlet's  composition.  There 
are  evident  references  to  Essex  in  the  play.  It  is  probable  that  Ophelia 
merely  caught  the  poet's  ear.     Lord  Burleigh  may  have  suggested  Polonius. 

Julius  Zupitza  was  born  in  1844.  His  training  at  Breslau  and  Berlin,  under 
MOlIenhoff  and  Haupt,  was  most  thorough.  During  his  twenty-five  years  of 
teaching,  he  dealt  with  Gothic,  German,  Scandinavian,  English,  Old  French, 
and  Provencal.  The  first  part  of  this  time  was  spent  in  Vienna.  In  1872  he 
visited  England  to  do  comparative  work  on  Guy  of  Warwick.  In  1876  he 
was  called  to  Berlin  as  Professor  of  the  English  Language  and  Literature. 
He  was  most  successful  in  rousing  his  students  to  scholarly  efforts.  His  own 
great  work  was  done  in  textual  criticism.  Appended  to  the  notice  is  a 
complete  bibliography  of  his  publications. 

Yalb  UNivBasiTT.  Charles  Grosvbnor  Osgood,  Jr. 


Digitized  by  VjOOQ IC 


liEPORTS.  239 

Rhbinisches  Museum  fCr  Philologie,  Vol.  LII,  parts  i,  2. 

Pp.  1-12.  Der  prodigiorum  liber  des  lulius  Obsequens.  O.  Rossbach. 
The  author  of  the  liber  prodigiorum  was  probably  not  a  Christian,  and  the 
book  may  have  been  written  in  the  time  of  Hadrian  or  of  Antoninus  Pius. 
Textual  notes. 

Pp.  13-41.  Ueber  den  Cynegeticus  des  Xenophon.  II  (cf.  vol.  LI,  596- 
629;  A.J.  P.  XVIII  115).  L.  Radermacher.  The  use  of  the  word  yv^fui  in 
contrast  with  ^vofta  forbids  us  to  refer  the  closing  chapter  to  a  later  time  than 
the  others.  Chapters  2-13  must  be  ascribed  to  the  same  author.  The  Cyne- 
geticus cannot  be  a  youthful  essay  of  Xenophon,  and  it  is  not  like  his  later 
writings :  it  is  spurious.  The  sharp  distinction  between  ^iXdao^c  and  ao^ior^ 
suggests  that  the  author  was  influenced  by  Plato.  The  proem  was  probably 
written  by  a  later  hand»  not  earlier  than  the  third  century  B.  C:  it  is  mere 
rhetoric. 

Pp.  42-68.  Die  BegrUndung  des  Alexander-  und  Ptolemaeerkultes  in 
Aegypten.  J.  Kaerst.  Ptolemy  Soter  founded  Ptolemais  in  Upper  Egypt, 
and  was  worshipped  as  a  god  in  that  city.  This  was  probably  in  imitation  of 
the  worship  of  Alexander  at  Alexandria.  The  worship  of  the  Ptolemaic 
dynasty  extended  and  developed  its  external  ceremonial,  but  the  consecration 
gradually  became  a  simple  form,  and  the  title  of  *  god '  a  mere  title. 

Pp.  69-98.  Die  Ueberlieferung  von  *  Aeli  Donati  commentum  Terentii.' 
P.  Wessner.  It  is  probable  that  all  the  15th-century  MSS  are  derived  from 
two  recensions,  that  of  Mentz  and  that  of  Chartres.  The  former  is  the  more 
valuable. 

Pp.  99-104.  Die  Bukoliasten.  E.  Hoffmann.  The  various  traditional 
accounts  of  the  origin  of  pastoral  poetry  agree  in  making  it  the  product  of  a 
people  reduced  to  slavery  by  foreign  invaders.  The  propitiatory  sacrifice  to 
Artemis  took  the  form  of  a  symbolic  restitutio  in  integrum,  and  on  that  day 
the  slaves  seem  to  have  enjoyed  some  such  freedom  of  speech  and  action  as 
the  Roman  slaves  enjoyed  during  the  Saturnalia. 

Pp.  105-25.  Delphische  Beilagen.  (S.  Band  LI,  S.  580.)  III.  Die  Th&tig- 
keit  der  Alkmeoniden  in  Delphi.     H.  Pomtow. 

Miscellen. — Pp.  126-9.  O.  Immisch.  Vergiliana.  I.  The  writer  would 
transpose  verses  40  and  41  of  the  fourth  book  of  the  Aeneid.  II.  The  con- 
ception of  the  Helena  taedifera  of  Aen.  VI  518  is  probably  derived  from 
Stesichorus.  There  may  be  lurking  in  it  something  of  an  old  popular  super- 
stition. If  the  **  fratres  Helenae,  lucida  sidera  *'  brought  safety  to  the  mariner, 
the  flame  of  Helen,  ^*iXivavg^  i^vSpog,  iXiirroXtg/^  may  have  indicated  disaster. 
— Pp.  129-31.  M.  Ihm.  Zum  Carmen  de  bello  Actiaco.  The  poem  contains 
many  reminiscences  of  Vergil  and  Ovid. — Pp.  131-5.  M.  Manitius.  Hand- 
schriftliches  zix  Germanicus'  und  Ciceros  Aratea. — Pp.  135-7.  H.  Sch5ne. 
Sechzehnsilbige  Normalzeile  bei  Galen. — Pp.  137-40.  C.  Wachsmuth.  Ein 
neues  Fragment  aus  Lydus*  Schrift  de  ostentis. — Pp.  140-43.  C.  Wachsmuth. 
Das  Heroon  des  Themistokles  in  Magnesia  am  Maiandros. — P.  143.  M.  Ihm. 
Zu  den  graeco-syrischen  PhilosophensprUchen  ttber  die  Seele.    (Cf.  vol.  LI, 


Digitized  by  VjOOQ IC 


240  AMERICAN  JOURNAL   OF  PHILOLOGY, 

p.  529.)  A  parallel  from  Xen.  Cyrop.  V  i. — P.  144.  R.  FOrster.  Cyriacus 
von  Ancona  zu  Strabon.  Nachtrag  zu  LI,  S.  490. — P.  144.  R.  WUnsch.  Zu 
Band  LI,  S.  148. 

Pp.  145-67.  Studien  zu  Ciceros  Briefen  an  Atticus  (IX,  X).  O.  E.  Schmidt. 
Textual  notes  on  forty  passages. 

Pp.  168-76.  Zu  attischen  Dionysos-Festen.  A.  KOrte.  i.  £iiovixJta  rd  iiri 
ArjvaXtfi.  DOrpfeld,  Das  griechische  Theater,  p.  9,  has  accepted  Gilbert's  view 
that  the  Lenaea  was  the  last  day  of  the  Anthesteria.  It  is  clear  from  C.  I.  A. 
II  834  b  that  they  were  separate  festivals.  The  official  name  in  the  fifth  and 
fourth  centuries  was  not  A^mx,  but  Aiovvaia  rd  em  AffvaUf>.  This  name  seems 
to  have  been  retained  long  after  the  place  of  celebration  was  changed.  2.  Der 
Agon  der  komischen  Schauspieler.  The  first  hypothesis  to  Aristophanes' 
Peace  speaks  of  a  competition  between  comic  actors  at  the  *City'  Dionysia  in 
B.  C.  421.  The  earliest  competition  of  this  sort  mentioned  by  the  inscriptions 
occurred  at  the  Lenaean  festival  in  B.  C.  354.  Possibly  the  first  hypothesis 
has  confounded  the  Peace  with  the  other  Peace  which  Aristophanes  brought 
out  at  the  Lenaea.  3.  Der  KitharOde  Nikokles.  The  inscription  C.  I.  A.  II 
1367  cannot  be  earlier  than  the  third  century  B.  C.  The  Isthmian  contest  in 
music,  in  which  Nicocles  was  the  first  victor,  must  have  been  introduced  in 
the  third  century,  not  in  the  fourth,  and  the  dithyramb,  which  was  unknown 
to  the  Lenaea  in  the  time  of  Demosthenes  (XXI  10)  and  Aristotle  (ABtjv,  liok, 
57),  was  not  added  to  this  festival  until  the  Hellenistic  period. 

Pp.  177-86.  Anecdoton  Fulgentianum.  R.  Helm.  This  is  an  allegorical 
explanation  of  the  story  of  Thebes,  with  grotesque  etymologies  of  the  proper 
names,  found  in  a  13th-century  MS,  Paris.  3012.  The  author  is  a  Christian 
writer,  who  quotes  from  Vergil,  Horace,  Ovid,  Lucan  and  the  New  Testament. 
It  is  probably  the  work  of  Fulgentius. 

Pp.  187-204.  Buphonien.  H.  von  Prott.  A  study  of  the  various  legends 
as  to  the  origin  and  significance  of  the  ^ov(f>6vta,  the  sacrifice  of  a  bull  to  Zeus 
Polieus.  It  is  possible  that  this  represents  an  earlier  human  sacrifice.  Cf. 
Ailianos,  Hist.  An.  XII  34 ;  Porphyrios,  De  Abst.  II  55  ;  Athenaios,  X  456  C. 

Pp.  205-12.  Zu  lateinischen  Dichtem.  M.  Ihm.  i.  Vespae  indicium  coci 
et  pistoris  iudice  Vulcano.  This  comic  epyllion  cannot  be  a  carmen  infimae 
Latinitatis.  2.  Das  carmen  contra  Flavianum  (Cod.  Paris.  8084).  A  list  of 
the  Vergiliana  in  the  poem.  The  author  seems  to  have  made  use  of  Petronius 
and  of  the  eclogues  of  Nemesianus,  and  to  have  read  some  of  the  epigrams  of 
Damasus.  3.  Ein  verschoUenes  Gedicht  des  Damasus?  An  anonymous 
glossary  contained  in  Cod.  Paris.  Lat.  7598  (saec.  XIII  or  XIV)  refers  to  a 
poem  of  Damasus,  "Prophetatio  Nicei  (Nicaenif)  Concilii." 

Pp.  213-36.  Beitrage  zur  Quellenkunde  des  Orients  im  Alterthum.  L.  Jeep. 
A  study  of  the  epitome  of  the  church  history  of  Philostorgios,  III  4-1 1.  The 
relation  of  Philostorgios  to  Agatharcides  and  Artemidoros. 

Pp.  237-85.  Zu  den  Assyriaka  des  Ktesias.  (Cf.  vol.  L,  pp.  205-40.)  P. 
Krumbholz.  5.  Inferences  to  be  drawn  from  Justinus,  Diodoros  and  Kepha- 
lion  as  to  the  statements  of  Ktesias.    6.  The  relation  of  Ktesias  to  earlier 


Digitized  by  VjOOQ IC 


REPORTS.  241 

historians  (Herodotus,  etc.)  and  to  later  vrriters.    7.  Diodoros  and  Ktesias. 
Ktesias  represents  a  Persian  anti-Assyrian  tradition. 

Pp.  286-92.  Varia.  W.  Kroll.  Textual  notes  to  Porphyrins,  Stobaeus, 
Damascius,  Galen. 

Miscellen. — Pp.  293-4.  J.  Ziehen.  Zwei  Vermuthungen  zur  griechischen 
Kunstgeschichte.  i.  The  Vienna  bronze  statuette  (Sacken,  Bronzen  in  Wien, 
I  44 ;  Dilthey,  Taf.  IX  f.)  probably  represents  Menelaus,  not  Ares  or  Achilles. 
2.  The  **  Alexander  et  Philippus  in  quadrigis"  of  Euphranor  (Plin.  N.  H. 
XXXIV  77)  was  probably  made  at  Alexander's  command,  after  the  death  of 
Philip. — Pp.  294-6.  O.  Hirschfeld.  Der  Brand  von  Lugudunum.  The  burn- 
ing of  Lugudunum  referred  to  by  Seneca,  E.  M.  91,  probably  took  place  in 
65  A.  D.  That  Seneca  does  not  directly  mention  the  great  fire  at  Rome  in  64 
may  be  due  to  the  popular  belief  that  it  was  caused  by  Nero. — P.  296.  A. 
Wilhelm.  Zum  Carmen  de  hello  Actiaco.  The  epithet "  pars  imperii,"  III 
25,  recurs  in  Propertius,  I  6,  34. — Pp.  296-8.  R.  F5rster.  Expletur  lacuna  in 
Libanii  declamatione  quae  inscribitur  fwyov  icanryopia. — Pp.  298-9.  R.  FOrster. 
Zur  Ueberlieferung  der  Physiognomik  des  Adamantios. — Pp.  299-302.  C. 
Heldmann.  Ein  neuentdecktes  Priscianbruchstttck.  A  new  fragment  of  the 
Instit.  Gram.  (XIV  33/34),  apparently  written  in  the  8th  century. — P.  302. 
C.  We3rman.  Zur  Anthologia  Latina  Epigraphica.  The  '  sinergima'  of  Carm. 
lAt.  Epigr.  1356,  19  B  is  not  for  awipyrffia.  The  s  belongs  to  the  preceding 
word.  For  *inergima'  cf.  Prud.  Apoth.  400  f.— Pp.  302-3.  F.  B.  Carmen 
Epigraphicum.  A  short  poem  from  a  stone  recently  discovered  at  Cologne. — 
Pp.  303-4.  E.  Lommatzsch.  Carpus.  The  name  of  Trimalchio's  carver 
(Petron.  36)  appears  frequently  in  Latin  and  Greek  inscriptions.  The  Greek 
name,  KdpTroc,  is  derived  not  from  Kapndc  *  fruit,'  but  from  Kapirdg  *  hand.'  It  is 
a  name  which  denotes  dexterity.  There  is  no  direct  evidence  of  a  word 
carpus  in  Latin,  but  the  word  catpo  *htind'  exists  in  Italian ;  whence  the  word 
carpone  (K6rting,  Latein.-roman.  W5rterbuch,  1688).  Carpere  is  the  technical 
expression  for  'carving'  (Friedlander  on  Mart.  Ill  13,  i);  Carpus^  which  is 
formed  from  the  same  stem,  corresponds  to  the  carptor  of  Juv.  IX  no.  Carpere 
is  for  an  older  ^scarpere^  which  was  retained  in  popular  speech  (LOwe,  Coni. 
Plant.,  p.  209 ;  Stowasser,  Archiv,  I,  p.  287 ;  cf.  Usener  on  coruscus,  *scoruscus, 
Rh.  Mns.  XLIX,  p.  463).  The  initial  s  shows  that  carpere  has  nothing  to  do 
with  jcopTrdf  *  fruit,'  but  belongs  rather  to  Kopirdg  'hand.'  The  proper  name 
Scarpus  is  rare,  but  some  coins  of  a  certain  Pinarius  Scarpus  show  the  device 
of  an  open  hand  (Cohen,  Med.  Imp.  I',  p.  136).  We  may  thus  assume  the  loss 
of  an  initial  j*  from  Kapirdg  *  hand,  wrist.' 

Havxrpobd  C0U.B6B.  Wilfred  P.  Mustard. 


Digitized  by  VjOOQ IC 


BRIEF  MENTION. 

Professor  Max  Schnkidewin  has  presented  the  world  with  a  bulky  volume 
of  558  pp.  entitled  Die  aniike  Humamt&t  (Weidmann),  in  which  he  has  brought 
together,  without  any  attempt  at  literary  finish,  many  facts  and  reflexions  in 
regard  to  a  theme  of  permanent  and  universal  interest.  The  author  does  not 
profess  to  have  ransacked  every  nook  and  corner  of  antiquity  for  documents, 
and  the  draughts  he  has  drawn  on  Cicero,  whom  he  sets  up  as  the  accepted 
type  of  antique  '  humanity,'  are  so  considerable  that  this  book  may  be  regarded 
as  a  companion-piece  to  the  slighter  performance  of  the  same  writer  published 
in  1890, '  Die  Horazische  Lebensweisheit.'  No  wonder,  then,  that  the  work 
revives  for  the  reader  the  charm  of  Cicero  and  Cicero's  circle,  which  is  not 
less  real  because  it  is  exotic,  which,  like  the  charm  of  the  winter  palaces  of 
Russia,  is  only  heightened  by  the  rigor  of  the  atmosphere  without.  When  we 
are  with  Cicero  we  are  in  good  society,  society  that  is  redolent  of  Scipionic 
traditions,  and  it  would  be  rude  to  scratch  the  skin  of  this  and  that  Roman 
grandee  and  compare  the  fine  Greek  sentiments  with  the  merciless  downright- 
ness  of  Italian  action.  Doubtless  Cicero,  the  novus  homo^  and  Horace,  liherHno 
patre  naius^  were  saturated  with  Greek  *  humanity,'  but  the  Greek  must  have 
the  credit  of  it  all,  directly  or  indirectly,  and  there  is  evidence  enough  that 
the  Hellene  or  Hellenist,  Greek  or  Greekling,  whichever  you  choose,  was 
fully  alive  to  the  essential  hardness  of  the  Roman  character  and  was  fully 
aware  of  his  own  success  and  his  own  failure  in  the  emollient  process. 


But  there  are  other  sides  to  Cicero  than  the  Greek  side,  the  ethical,  the 
philosophical,  the  humane  side.  He  was  much  more  than  a  translator  of 
Panaetius,  though  the  de  Officiis  has  proved  itself  a  potent  book ;  much  more 
than  a  clever  lawyer,  though  the  French  Revolution  is  said  to  have  been  the 
work  of  lawyers ;  and  in  an  essay  which  takes  the  form  of  a  discourse  in 
celebration  of  the  second  millennium  of  Cicero's  birth.  Professor  Zielinski 
has  produced  a  sketch  of  Cicero's  influence  on  the  ages  which  forms  a  striking 
contrast  to  the  work  just  mentioned,  both  in  bulk  and,  if  it  must  be  said,  in 
brilliancy.  With  such  a  champion  as  Professor  ZisuNSKI  is,  the  friends  of 
Cicero  may  well  take  heart,  for,  as  one  reads  this  masterly  summary  of 
Cicero's  after-life,  Cicero  im  Wandel  der  Jahrhnnderte  (Teubner),  Drumann's 
savagery  and  Mommsen's  sarcasm,  the  bludgeon  of  the  one  and  the  rapier  of 
the  other,  lose  weight  and  point.  The  salient  features  are  tipped  with  light, 
and  the  test -question,  'What  thinkest  thou  of  Cicero?,'  is  most  effectively  put 
to  the  leaders  of  human  thought  and  action.  Cicero's  immense  influence 
on  style  is  generally  recognized  after  a  vague  fashion,  though  perhaps  few 
are  aware  that  every  penny-a-liner  on  the  daily  press  is  swayed  by  his 


Digitized  by  VjOOQ IC 


BRIEF  MENTION.  243 

example  and  his  precepts ;  but  his  influence  on  the  coarse  of  history  at  its 
critical  points  is  a  matter  that  only  such  a  cross-section  as  ZiEUNSKl  has 
given  us  can  bring  to  the  consciousness.  What  Cicero  did  for  Christianity, 
what  for  the  Renascence,  for  the  Reformation,  for  the  French  Revolution, — 
how  he  affected  the  leaders  of  those  great  transitional  periods,  this  is  the 
theme  of  an  essay  which  combines  the  rhetorical  swing  of  the  panegyrist  with 
the  sober  merits  of  historical  research.  That  Augustin  was  converted  by 
reading  Cicero  is  a  familiar  story,  and  no  one  that  has  once  read  is  likely  to 
forget  the  passage  in  Luther's  Table-talk  in  which  he  extols  the  man  who  has 
wrought  and  suffered  above  that  *ass  of  leisure,  Aristotle' — *weit  ttberlegen,' 
he  says,  'dem  mttssigen  Esel  Aristoteli';  but  the  influence  of  Cicero  the 
humanitarian  on  Voltaire,  of  Cicero  the  orator  on  Mirabeau,  of  Cicero  the 
republican  statesman  on  the  leaders  of  the  French  Revolution  is  not  always 
present  to  the  average  mind.  Vergniaud  was  the  Cicero  of  the  Gironde  and 
denounced  Robespierre  in  phrases  borrowed  from  the  Catilinariae,  and  Robes- 
pierre defended  his  cause  and  prolonged  his  power  by  a  telling  use  of  passages 
taken  from  the  Oratiopro  P,  Sulla,  With  the  close  of  the  French  Revolution 
ZiELiNSKi  bids  the  procession  stop  and  contents  himself  with  citing  Taine  to 
show  the  estimate  in  which  Cicero  is  held  by  that  penetrating  student  of 
history  and  literature,  and  with  reinforcing  in  a  brief  summary  the  important 
lesson  that  with  every  advancing  stage  of  culture  the  vision  for  the  antique 
becomes  wider  and  deeper  and  that  the  value  of  the  antique  is  enhanced 
from  stage  to  stage. 


All  who  admire  the  scholarship,  the  precision,  the  balance  of  M.  Henri 
Weil  will  be  glad  to  have  in  a  convenient  volume  the  collection  of  his  papers 
entitled  Etudes  sur  le  drame  antique  (Hachette).  Nearly  all  these  studies 
belong  to  a  recent  period.  One,  it  is  true,  goes  back  to  the  remote  date  1847, 
one  to  1864,  but  of  the  remaining  eight  there  is  none  older  than  1886,  and  the 
eighth  deals  with  the  important  work  of  M.  Masqubray,  Les  formes  lyriques 
de  la  tragSdie  grecque^  which  was  published  as  late  as  1895  and  is  still  awaiting 
the  notice  it  deserves  in  this  Journal.  It  is  a  book  which  M.  Weil  justly 
praises  for  the  exhaustive  command  of  the  literature,  its  wide  scope,  its  fine 
appreciation  of  the  rfio^  of  the  lyric  measures  of  tragedy.  M.  Weil's  admi- 
ration of  Wilatnowit^s  Herakles^  the  subject  of  another  chapter,  is  frankly 
expressed,  while  he  preserves  the  independence  of  his  judgment  in  details,  a 
hard  thing  to  do,  if  one  yields  at  all  to  the  rush  of  that  fervid  genius.  ZieliiC- 
ski*s  ayliv  with  all  its  minute  subdivisions  M.  Weil  cannot  bring  himself  to 
accept,  but  he  recognizes,  as  some  have  refused  to  do  (A.  J.  P.  X  383),  the 
popular  element  that  lies  at  the  basis  of  the  comic  debate,  and  compares  the 
quarrel  between  tanner  and  sausage-seller  in  the  Knights  with  the  altercation 
of  the  modem  carnival.  ''On  pense,"  he  says,  *'i  notre  camaval:  deux 
masques  se  provoquent,  se  criblent  de  lazzi ;  on  fait  cercle  autour  d'eux,  on 
les  encourage,  on  les  excite,  comme  fait  le  choeur  de  Tantique  com^die.  De 
pareilles  scenes  nVtaient  sans  doute  pas  rares  dans  les  joyeux  ebats  des 
Dionysiaques."  In  another  article  M.  Weil  takes  up  M.  Decharme's  book, 
Euripide  et  V esprit  de  son  th^dtre,     M.  Decharme  is  especially  emphatic  on  the 


Digitized  by  VjOOQ IC 


244  AMERICAN  JOURNAL    OF  PHILOLOGY. 

atheism  and  rationalism  of  Euripides,  and  here,  as  elsewhere,  M.  Weil  has  a 
wise  word  of  caution.  True,  every  scholar  knows  that  atheism  does  not  mean 
the  same  thing  in  Greek  as  it  does  in  English  (A.  J.  P.  XVII  362),  but  it  was 
well  worth  the  while  to  say  (p.  105):  **Si  Ton  dit  que  Ic  thfeatre  d*Euripide 
agit  comme  un  dissolvant  sur  les  vieilles  fables  et  les  croyances  populaires,  on 
dit  vrai,  .mais  on  ne  dit  pas  tout.  Euripide  n'a  pas  seulement  ebranle  les 
opinions  revues,  il  a  puisamment  contribu^  i  repandre  une  conception  plus 
haute  du  divin,  qui  devait  6tre  celle  de  I'avenir."  In  the  same  paper  DOrp- 
feld's  theory  of  the  stage  comes  up.  M.  Weil  minimizes  the  difference 
between  the  old  view  and  the  new,  but  holds  after  all  to  the  raised  wooden 
stage,  and  the  words  hrcX  r^g  oiofv^g  are  to  him  a  stone  of  stumbling,  as  they 
have  been  to  many  philologians  (A.  J.  P.  XVIII  iig).  "II  faut  vraiment,"  he 
says,  '*  beaucoup  de  bonne  volonte  pour  traduire  [ces  mots]  par  prh  de  la  scene 
plutOt  que  par  sur  la  sehte^*  and  after  the  appearance  of  D6rpfeld  and  Reisch's 
book  he  adds :  **  Tout  le  monde  ne  se  persuadera  pas  non  plus  que  les  acteurs 
sont  appel^s  ol  and  aiofv^g  parcequ'ils  sortaient  de  la  aiap^*^ 


Mr.  M ARCHANT  has  added  BooJk  VI  to  the  three  books  of  Thukydides  he  has 
already  edited,  II,  III,  VII  (Macmillan).  The  text  is  based  on  Hude's,  but 
the  editor  shows  his  wonted  independence  in  minor  matters.  There  is  a 
chapter  of  new  explanations  headed  'Some  Cruces*  which  will  be  read  with 
interest  by  Thukydidean  scholars.  An  adjutant  and  admirer  of  Dr.  Ruther- 
ford's, Mr.  Marchant  has  learned  from  his  master  the  importance  of  a  sharp 
formulation  of  Attic  usage,  and  his  work  shows  advancing  appreciation  of 
syntactical  phenomena.  As  he  has  referred  to  this  Journal  (XIII  7,^<^yii propos 
of  the  negative  in  c.  81,  5,  it  may  be  as  well  to  say  that  I  cannot  see  any  call 
for  *  mobility '  in  order  to  understand  so  simple  a  case  as  rriv  irpbg  ifiag  ix^pav  fu^ 
av  ppaxelaif  yevdfievjpf.  The  article  with  the  participle  gives,  as  it  often  does, 
the  impulse  to  the  negative  ///;,  and  the  resolution  is  not  what  Mr.  Marchant 
has,  ^  oifK  &u  ppaxeia  yevoiro,  but  f/  fir)  ap  Ppaxela  ykvotro^  the  so-called 
characteristic  relative  {q  —  nrig)  taking  fi^.  See  A.  J.  P.  I  54,  56,  and  for 
rel.,  fi^  dv,  opt.  comp.  Dem.  19,  313;  20,161;  21,  203;  Plato,  Phileb.  20  A; 
Legg.  839  A,  872  D.  For  a  parallel  use  of  (iti  av  c.  partic.  ^ee  Dem.  54,  40 
6  ftndht  av  ofiSaaCt  with  Sandys*  note.  This  is  one  of  the  many  points  that 
show  the  importance  of  an  historical  survey  for  the  appreciation  of  syntactical 
phenomena.  It  was  only  when  the  participle  was  consciously  employed  as 
the  shorthand  of  a  hypotactic  sentence  that  the  neg.  fii^  could  be  used  with  it. 
Pindar*s  6  fiij  awulq  (N.  4,  31)  is  a  distinct  advance  not  only  on  Homer,  but 
also  on  Hesiod,  whose  /3odf  .  . .  ^  reroKviTK  (O.  et  D.  591)  is  under  the  domi- 
nation of  the  imperative  opt.  tlij. 


Dr.  Rutherford's  Introduction  to  his  Scholia  Aristophanica  (Macmillan)  is 
a  prolonged  growl  over  the  uncongenial  work  that  has  cost  the  leisure  of  no 
less  than  seven  years.  AieiCcj  Ai&rtfwv,  The  subject-matter  of  the  scholia  of 
the  Codex  Ravennas,  he  says,  would  not  have  tempted  him  to  edit  them.    In 


Digitized  by  VjOOQ IC 


mammtmmmmB/m 


BRIEF  MENTION,  245 

fact,  "  the  direct  value  of  any  corpus  of  scholia  as  a  commentary  upon  the  text 
to  which  it  belongs  is  in  no  degree  commensurate  with  its  indirect  value  as 
evidence,  on  the  one  hand,  of  the  manner  in  which  classical  texts  have  been 
manipulated  at  different  periods  in  the  history  of  learning,  and,  on  the  other 
hand,  of  the  kind  of  corruption  and  interpolation  to  which  they  have  been 
exposed."  We  know  the  note  from  the  Introduction  to  his  Fourth  Book  of 
Thukydides.  Still,  a  man  might  be  worse  employed  than  in  laboring  over  the 
Greek  scholia.  It  is  higher  work  than  the  preparation  of  an  index,  and  the 
preparation  of  an  index  is  better  than  making  canons  of  Greek  usage  on  the 
basis  of  imperfect  induction.  It  is  something  to  have  to  one's  credit  two 
such  stately  volumes  as  these.  The  third  volume  is  still  due,  the  volume  that 
is  to  contain  Dr.  Rutherford's  conclusions,  drawn  from  his  seven  years'  study 
of  the  scholia ;  and  while  we  are  grateful  for  all  that  these  two  volumes  hold, 
it  is  the  third  volume  in  which  we  shall  behold  the  flower  of  the  Scottish  thistle. 


The  most  striking  characteristic  of  Professor  Tyrrell's  edition  of  the 
TrocLdes  (Macmillan)  is  the  sympathetic  discernment  with  which  he  has 
brought  out  the  poetic  vein  of  Euripides.  In  so  doing  he  has  made  free  use 
of  translation — now  an  apt  rendering  of  his  own,  now  an  extract  from  Mr. 
Way*s  brilliant  version.  The  book  is  meant  for  boys,  and,  as  Professor 
Tyrrell  justly  remarks,  *  a  boy  should  not  be  encouraged  to  think  that  the 
Greek  poets  were  bald  and  frigid.'  How  soon  the  attention  of  the  student 
should  be  called  to  the  dissonances  of  Euripidean  style,  designed  or  not,  is 
another  matter.  Dr.  Verrall's  'Euripides  the  Rationalist*  would  not  be  a 
good  book  to  put  in  the  hands  of  a  beginner  in  Euripides,  and  the  young 
student  would  be  rather  puzzled  than  edified  by  a  demonstration  of  the 
contrarieties  of  the  diction  and  the  syntax  of  Euripides,  the  matching  of  cloth 
of  gold  with  cloth  of  frize.  The  metres  are  not  neglected,  as  in  so  many 
English  editions,  but  it  is  to  be  regretted  that  Professor  Jebb's  example  has 
not  been  followed  and  that  Schmidt's  schemes  have  not  been  reproduced.  It 
seems  rather  late  in  the  day  to  cite  Dr.  Kennedy's  views  in  the  matter  of 
Greek  metres. 


Dr.  Sandys*  edition  of  the  First  Philippic  and  the  Olynthiaes  of  Demosthenes 
(Macmillan)  is  marked  by  his  unfailing  adequacy.  Every  side  of  his  author  is 
treated  with  sound  judgment,  excellent  taste  and  rare  command  of  the  litera- 
ture. The  proof-reading  is  good.  An  odd  mistake  occurs  p.  36,  §25  (critical 
note),  where  read  *^suu5  locus  est  infinitivo  supra  §12,  Bl.'  By  the  way,  if 
Blass  means  to  differentiate  between  participle  and  inf.  in  the  two  passages, 
he  sees  too  much.  §12  reads:  ri  rh  KxSkvov  if  avrihf  iarai  Padi^eiif  bvroi 
pob^£T2i/  §25:  rig  avrdv  Ko^vaei  devpo  padi^ovra;  As  padi^ovra  is  condi- 
tional, =  eav  Padii^%  the  difference  is  naught.  In  conditional  relations  inf. 
and  part,  often  meet,  aiaxwoifiifv  av  avriXiyuw  (X.  Mem.  2,  6,  37)  =  el  avTikk- 
yoifu  =  avTiXtyeiv,    See  Hertlein  (1853)  on  X.  Cyr.  3,  2,  16. 


Digitized  by  VjOOQ IC 


246  AMERICAN  JOURNAL    OF  PHILOLOGY. 

An  esteemed  correspondent  sends  to  the  Journal  the  following  note  on 
FOgner's  Lexicon  Livianum^  Fasciculus  III^  s,  v.  ady  cum  gerundio  vel  gerun^ 
divoy  which  seems  to  belong  to  the  black  list  of  Brief  Mention  : 

**  The  following  incorrect  references  have  been  noticed :  28,  9,  i  for  28,  29, 
I,  p.  432.  8 ;  44. 19,  4  for  41,  19.  4.  p.  44M  ;  10.  55,  4  for  10,  35,  4.  p.  447,  16 ; 
25,  35.  4  for  25,  36.  4,  p.  448,  24 ;  31,  47.  2  for  31,  46.  2,  p.  448.  38 ;  23.  34,  9 
for  29,  34,  9,  p.  457,  23.  In  a  few  instances  the  Lex.  fails  as  a  guide  for  the 
Weissenbom  ed.:  4, 11,  5  triumviri  ad  colon iam  Ardeam  deducendam  is  not 
given  p.  428,  2  (creo),  nor  p.  457, 40  (triumviri).  40, 24, 5  ad  quod  celebrandum 
is  not  given  p.  434,  39.  42,  10,  8  ad  quam  pestem  frugum  tollendam  . . . 
missus,  ingenti  agmine  hominum  ad  colligendas  eas  coacto.  The  first  gerun- 
dive is  not  given  s.  v.  mitto ;  the  second  is  not  given  p.  426, 19,  where  is  given 
9,  21,  3  magno  exercitu  coacto  ad  eximendos  obsidione  socios." 


Brief  Mention  has  received  the  following  note  from  Dr.  J.  KEELHOFFt  of 
Antwerp:  **Sur  Texpression  tl  fi^  6ia  cf.  Rost,  Griech.  Gram.,  7te  Aufl.,  p.  641, 
note:  *Zu  erg&nzen  (Plat.  Gorg.  516  E)  wk  iviireoev^  also  der  reine  Gegensatz 
des  im  Hauptsatx  enthaltenen  Praedikates,  wie  immer  bei  dieser  fVendung* 
Votre  explication  [A.  J.  P.  X  124,  XVI  396.  XVII  128]  se  rencontre  done  avcc 
celle  de  Rost,  ce  qui  augmente  encore  les  chances  de  probability.  On  trouve 
de  bien  bonnes  choses  dans  cette  syntaxe  qu*on  ne  consulte  plus  gu^re."  To 
my  mind  the  explanation  is  so  evident  that  it  only  needs  to  be  stated,  and  I 
am  not  surprised  that  so  sensible  a  grammarian  as  Rost  was  had  reached  the 
same  formula,  which,  however,  does  not  occur  in  the  earlier  editions,  to  which 
alone  I  had  access. 


Digitized  by  VjOOQ IC 


NECROLOGY. 

The  double  blow  that  has  fallen  on  Harvard  University  in  the  last  few 
weeks  will  be  felt  throughout  the  scholarly  world,  will  be  felt  with  peculiar 
poignancy  by  those  who  were  privileged  to  know  personally  the  two  masters 
whose  names  are  henceforth  to  be  a  memory.  The  forces  they  set  in  motion 
will  never  die,  but  their  living  presence  is  to  inspire  and  to  guide  us  no  more. 
The  thirtieth  of  June  closed  the  career  of  America's  greatest  Latinist,  George 
Martin  Lane.  The  end  was  not  unexpected,  yet  when  it  came,  it  came  with 
a  sudden  pang  to  those  who  had  watched  the  bulletins  of  his  failing  health. 
It  seemed  a  hard  fate  that  he  was  not  to  bestow  on  the  world  with  his  own 
hands  the  summary  of  his  long  life  of  keen  observation,  of  loving  study.  And 
yet  to  those  who  can  sympathize  with  the  temper  of  the  man,  who  understand 
as  he  did  the  inexorableness  of  the  ideal,  his  life  as  a  rebuke  to  pretentious 
ignorance,  to  hasty  performance,  to  rash  generalization,  has  served  a  high 
purpose.  The  best  text-books  must  pass  away,  but  the  lessons  of  a  great 
teacher  become  incarnate  in  generations  of  living  men.  Lane  faded  out  of 
life.  Five  short  weeks  afterwards,  Aug.  4,  his  dear  friend,  Frederic  De 
Forest  Allen,  fell  without  warning,  struck  down  in  a  moment,  snatched 
rudely  from  the  midst  of  an  active  career,  at  an  age  when  the  intellectual 
faculties  are  in  their  happiest  balance  and  most  successful  play.  Born  to  a 
time  when  American  classical  scholarship  was  ripe  for  advanced  work  along 
the  whole  line,  Allen  had  taken  his  place  at  once  among  the  leaders  in 
university  study,  and  what  he  wrought  for  his  wide  domain  as  teacher  and  as 
author  showed  the  mind  and  the  will  of  a  true  master.  There  is  no  space  in 
this  number  of  the  Journal  to  set  forth  the  work  and  the  character  of  these 
departed  scholars.  In  the  next  issue  a  more  fitting  tribute  will  be  paid  to 
their  great  services.  Standing,  as  I  do,  between  the  two  in  years,  the  one 
who  was  intimately  associated  with  my  own  student  life  in  the  dear  GOttingen 
days,  the  other  for  whom  I  foresaw  the  accomplishment  of  ever  greater  work 
for  classical  philology  in  its  widest,  highest,  noblest  sense,  I  look  backward 
and  forward  with  a  sense  of  bereavement  which  all  the  teaching  of  old 
experience  will  not  school  into  resignation. 

Basil  L.  Gilderslbbve. 


Digitized  by  VjOOQ IC 


RECENT  PUBLICATIONS. 

Thanks  are  due  to  Messrs.  Lemcke  &  Buechner,  8x2  Broadway,  New 
York  City,  for  material  furnished. 

AMERICAN. 

Cicero  (Marcus  TuUius).  The  Correspondence  of;  arranged  according 
to  its  chronological  order ;  with  a  revision  of  the  text,  a  commentary,  and 
introductory  essays,  by  Rob.  Yelverton  Tyrrell  and  L.  Claude  Purser.  In 
7  V.  V.  5.  (Dublin  University  Press  Series.)  New  York,  Longmans, 
Green  &*  Co.,  1897.     75  +  4^2  pp.     Svo,  cl.,  net  $4. 

The  Fourth  Verrine;  ed.  for  schools  by  F.  W.  Hall.     New  York, 

The  Macmillan  Co,,  1897.     60+  187  pp.     i6mo,  cl.,  net  90  cts. 

Collar  (  W.  C.)  Via  Latina.  Boston,  Ginn  &*  Co,,  1897.  c.  6+  203  pp. 
Map,  i2mo,  cl.,  85  cts. 

Comba  (T.  £.)  Compendium  of  Italian  Pronunciation.  New  York, 
Truslove  &*  Comba  [1897].     2  +  47  pp.     Sq.  i2mo,  cl.f  75  cts. 

Hall  (J.  R.  Clark).  A  Concise  Anglo-Saxon  Dictionary.  New  cheaper 
ed.  New  York,  TAe  Macmillan  Co,,  1897.  369  pp.  4to,  cl.,  reduced  to 
net  11.75. 

Jackson  (E.  Hilton).  Law  Latiiu  Washington,  D.  C.,/.  Byrne  &*  Co., 
1897.     c     14-I-219PP.     X2mo,  shp.,  |2. 

Thucydides.  Book  6 ;  ed.  by  E.  C.  Marchant.  New  York,  The  Macmil- 
lan Co,,  1897.     54  +  299  PP*     x6mo,  cl.,  net  |x.io. 

Xenophon.  The  Fifth  Book  of  Xenophon's  Anabasis ;  ed.  by  Alfred  G. 
Rolfe.    'Boston,  Ginn  &*  Co,,  i^*j,    c.    5+ x  15  pp.    Map,  x6mo,  cl.,  45  cts. 

ENGLISH. 

Barrire  (A.)  and  Leland  (C.  G.)  Dictionary  of  Slang,  Jargon  and  Cant. 
2  vols.     London,  1897.     Svo,  950  pp.     15s. 

Evans  (D.  L.)  Welsh  Dictionary.  Geiriadur  Cymraeg.  Pt  4.  D-Dd. 
London,  1897.    8vo,  564  pp.     S.  1893,  ^r*  1062.     14s. 

Kohut  (G.  A.)  Semitic  Studies  in  Memory  of  A.  Kohut.  London,  1897. 
Svo.     £1  4s. 

MacCauley  (C.)  Japanese.  An  Introductory  Course.  London,  1897. 
Svo.     los.  6d. 

Morfill  (  W.  R.)  A  Short  Grammar  of  the  Bulgarian  Language.  London, 
1S97.     Svo.     5s. 

Murray  (J.  A.  H.)  A  New  English  Dictionary.  Vol.  IIL  Distrustfully- 
Doom.     Vol.  IV.  Flexuosity-Foister.    London,  1S97.    4to.     2s.  6d. 

Northall  (G.  F.)  A  Warwickshire  Word-book.  London,  1897.  Svo, 
300  pp.     15s. 


Digitized  by  VjOOQ IC 


RECENT  PUBLICATIONS,  249 

Papyri,  Greek.  Series  IL  New  classical  fragments  and  other  Greek 
and  Latin  papyri.  Edited  by  6.  F.  Grenfell  and  A.  S.  Hunt.  London, 
1897.    4to.     I2S.  6d. 

Pegge  (S.)  Two  Collections  of  Words  and  Phrases  peculiar  to  the 
Natives  and  Inhabitants  of  the  County  of  Derby.  London,  1897.  8vo, 
162  pp.     I2S.  6d. 

Petrie  Papyri.  The  Hieratic  Papyri  from  Kahun  and  Gurob.  Edited  by 
F.  L.  Griffith.  Autotypes,  Kahun,  plates  1-37,  Gurob,  plates  38-40.  Part  I. 
London,  1897.    4to,  x8  pp.    Compl.,  £2  12s.  6d. 

Temple  (G.)  Glossary  of  Indian  Terms  relating  to  Religion,  Customs, 
Government,  Land,  etc.     London,  1897.     8vo,  332  pp.     7s.  6d. 

Thimm  (C.  A.)  Egyptian  Self-Taught  (Arabic).  London,  1897.  8vo, 
71  pp.     2S. 

Wyatt  (A.  J.)  An  Elementary  Old  English  Grammar  (Early  West 
Saxon).     London,  1897.    8vo,  170  pp.     4s.  6d. 

FRENCH. 

Harlez  (de).  Le  Yi-King,  traduit  d'apr^s  les  interpr^tes  chinois  avec  la 
version  Mandchoue.     In-8.     E,  Leroux.     7  fr.  50. 

Henry  (V.)  La  relation  locative  dans  les  Ungues  italiennes.  In-8. 
Maisonneuve.     i  fr.  50. 

Houdas  (O.)     Precis  de  grammaire  arabe.     In-8.    J,  Andr/ et  Cie.    6  fr. 

Lexique  de  la  langue  de  Moli^re.  T.  Ill  et  dernier.  In-8.  II.  Welter. 
Les  3  vol.,  45  fr. 

McCarthy  (R.  P.  Laurent).  Capucins.  Grammaire  Hindoustani-fran- 
^aise.     In-8.     i  vol.     l^i^ixoWt^^  SoeiA/ beige  de  Librairie.    4  fr.  50. 

Meillet  (A.)  Recherches  sur  I'emploi  du  genitif  accusatif  en  vieuz  slave. 
Gr.  in-8.     i  vol.    E,  Boullon,     6  fr. 

Nouveau  Larousse  illustr^.  Dictionnaire  encyclop^dique  aniversel  en 
six  volumes.     In-4.     Hollier- Larousse  et  Cie.     Prix  de  souscription,  150  fr. 

Passy  (J.)  et  Rambeau  (A.)  Chrestomathie  frangaise.  In-8.  H,  Le 
Soudier.     5  f  r. 

Rousselot  (Abbe).  Principes  de  phonetique  ezp^rimeqtale.  ire  partie, 
avec  fig.     In-8.     II.  Welter.     15  fr.  en  souscrivant. 

GERMAN. 

Aristophanis  equites,  recensuit  Adph.  v.  Velsen.  Ed.  II,  quam  curavit 
Konr.  Zacher.     gr.  8.     xxii,  109  S.     L.,  B.  G.  Teubner.    m.  3. 

Beitiage,  Berliner,  zur  getmanischen  u.  romanischen  Philologie.  Ver- 
Offentl.  V.  Emil  Ebering.  XIII.  Romanische  Abtlg.  Nr.  7.  gr.  8.  B., 
E,  Ebering. — 7.  Schayer  (Siegb.)  Zur  Lehre  vom  Gebrauch  des  unbe- 
stimmten  Artikels  u.  des  Teilungsartikels  im  AltfranzOsischen  n.  im  Neu- 
franzOsischen.     viii,  152  S.     m.  4. 

^■^  Milnchner,  zur  romanischen  u.  englischen  Philologie.  Hrsg.  v.  H. 
Breymann  u.  J.  Schick.  XII.  Hft.  gr.  8.  L.,  A.  Deichert  Naehf.  '  m.  2.80. 
—XII.  Klein  (Frdr.)  Der  Chor  in  den  wichtigsten  TragOdien  der  franz<5- 
sischen  Renaissance,    xx,  144  S.    m.  2.80. 


Digitized  by  VjOOQ IC 


250  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Bibliothek,  assyriologische,  hrsg.  v.  Frdr.  Delitzsch  u.  Paul  Haapt. 
XIII.  Bd.  2.  Hft.  gr.  4.  L.,  /.  C.  Hinrichs'  VerU^z.  Craig  (James  A,) 
Assyrian  and  Babylonian  Religious  Texts,  etc.  Vol.  II.  With  corrections 
to  vol.  I.    xi  S.  tt.  21  Bl.     m.  7. 

romanische,   brsg.   v.    Wendelin    Foerster.     XIV.    8.     Halle,  M. 

Niemeyer, — XIV.  Estoria,  La,  de  los  quatro  dotores  de  la  santa  eglesia. 
Hrsg.  y.  Frdr.  Laucbert.    xiv,  443  S.    m.  12. 

Brock  (Artb.)  Quaestionum  grammaticarum  capita  duo.  gr.  8.  184  S. 
Jurievi  (Dorpati).     l^^  JC,  F»  Koekler  Sert,     m.  3. 

Brugmann  (Karl)  u.  DelbrUck  (Berthold).  Grundriss  der  ver^eichenden 
Grammatik  der  indogermanischen  Spracben.  i.  Bd.  Einleitung  u.  Laut- 
lebre.  i.  Halfte.  Von  Karl  Brugmann.  2.  Bearbeitg.  i.  Bd.  Einleitung 
u.  Lautlebre.  x.  HSlfte  (§1  bis  694).  gr.  8.  xlvii»  622  S.  Strassburg, 
X.  J.  TrUbner  Verl.     m.  16. 

4.  Bd.  Vergleichende  Syntax  der  indogerman.  Spracben  v.  B. 

DelbrQck.  2.  Tbl.  gr.  8.  xvii,  560  S.  Strassburg,  AT.  /.  TrUbner  Verl. 
m.  15. 

Callimacbi  bymni  et  epigrammata.  Iterum  ed.  Udalr.  de  Wilamowitz- 
Moellendorff.     8.    66  S.     B.,  IVeidmann,    m.  —-80. 

Commentaria  in  Aristotelem  graeca.  Edita  consilio  et  auctoritate  acade- 
miae  litterarum  regiae  borussicae.  Vol.  XIV,  pars  2.  gr.  8.  B.,  G, 
Reimer, — XIV  2.  loannis  Philoponi  in  Aristotelis  libros  de  generatione  et 
corraptione  commentaria.     Ed.  Hieron.  Vitelli.     x,  356  S.     m.  14. 

— ^  Vol.  XV,  gr.  8.— XV.  loannis  Philoponi  in  Aristotelis  de  anima 
libros  commentaria.     Edidit  Mich.  Hayduck.     xix,  670  S.     m.  27. 

Corpus  inscriptionum  graecarum  Graeciae  septentrionalis.  Vol.  III. 
Fasc.  I.  Fol.  B.,  G,  Reimer.  Kart. — III  x.  Inscriptiones  graecae  Phocidis, 
Locridis,  Aetoliae,  Acarnaniae,  insularum  maris  lonii.  Ed.  Guil.  Ditten- 
berger.    vii,  212  S.     m.  22.50. 

Denssen  (Paul).  Sechzig  Upanishad's  des  Veda.  gr.  8.  xxvi,  920  S. 
L.,  F.  A,  Brockhaus,     m.  20. 

D6rwald  (Paul).  Die  Formenbildungsgesetze  des  Hebraischen.  gr.  8. 
vii,  X23  S.     B.,  Mayer  A*  Mailer,     m.  2.40. 

Eranos.  Acta  philologica  suecana.  Ed.  Vilelm.  LundstrOm.  Vol.  II. 
1897.  4  fascc  gr.  8.  i.  fasc.  48  S.  Upsaliae.  Leipzig,  O,  Harrasscwitt 
in  Komtn.     m.  6. 

Gassner  (Armin.)  Das  altspanische  Verbum.  gr.  8.  viii,  208  S.  Halle, 
M.  Niemeyer.    m.  5. 

Grdber  (Gust.)  Grundriss  der  romanischen  Philologie.  II.  Bd.  3.  Abtlg. 
2.  Lfg.     gr.  8.     S.  X  29-256.     Strassburg,  K.J,  TrUbner  Verl.    m.  2. 

Grundriss  der  indo-arischen  Philologie  u.  Altertumskunde.  Hrsg.  v. 
Geo.  BQhler.  IIL  Bd.  i.  Hft.  A.  u.  2.  Hft.  gr.  8.  Strassburg,  K.  J. 
TrUbner  Verl.—lU  I.  Macdonell  (A.  A.)  Vedic  Mythology.  176  S.  m. 
9. — 2.  Hillebrandt  (Alfr.)  Ritual-Litteratur.  Vedische  Opfer  u.  Zauber. 
189  S.     m.  9.50. 

Handbuch  der  klassischen  Altertums-Wissenschaft,  hrsg.  v.  Iwan  v. 
Mttller.      IIL  Bd.   3.  Abtlg.    i.  Halfte.     gr.  8.     MUnchen,  C.  H.  Beck 


Digitized  by  VjOOQ IC 


sgn 


RECENT  PUBLICA  TIONS,  2$  I 

III  3,  I.  Jung  (Jul.)  Grundriss  der  Geographie  v.  Italien  u.  dem  Orbis 
romanus.     2.  Aafl.     Mit  alphabet  Register,    viii,  178  S.     m.  3.50. 

Heusler  (Andr.)  Zwei  Islfinder-Gescbichten,  die  H^nsna-JxSres  a.  die 
Bandamanna  saga,  m.  Einleitg.  n.  Glossar  hrsg.  gr.  8.  iv,  Ixii,  164  S. 
B.,  Weidmann,     m.  4.50. 

Homeri  Odysseae  carmina  cum  apparatn  critico  ediderunt  J.  van  Leeuwen 
J.  F.  et  M.  B.  Mendes  da  Costa.  Ed.  II.  Accedunt  tabulae  III.  Pars  I. 
Carm.  I-XIL     gr.  8.     xxvii,  292  S.     Leiden,  A.  W,  Sijthoff,     m.  3. 

Jacob]  (Herm.)  Compositum  u.  Nebensatz.  Studien  Ub.  die  indoger- 
man.  Sprachentwicklg.     gr.  8.    z,  127  S.     Bonn,  F,  Cohen,     m.  3. 

Koschwitz  (Ed.)  Anleitung  zum  Stodium  der  franzOsischen  Philologie. 
gr.  8.    viii,  148  S.     Marburg,  N.  G.  Elwerfs  Verl,     m.  2.50. 

Lamer  (Joa.)  De  choriambicis  Graecorum  poetarum  versibus.  Diss, 
gr.  8.     142  S.     L.,  E,  Gr&fe,     m.  2. 

Lexer  (Matthias).  Mittelhochdeutsches  TaschenwSrterbuch.  5.  Aufl. 
8.    vii,  4x3  s.     L.,  5".  Hirul.    m.  5. 

Litteraturdenkmaler,  lateinische,  des  XV.  u.  XVI.  Jahrh.  Hrsg.  v.  Max 
Herrmann.  13.  8.  B.,  Weidmann, — 13.  Macropedius  (van  Langveldt) 
(Georgius).     Rebelles  u.  Aluta.     Hrsg.  v.  Jobs.  Bolte.    xlii,  104  S.     m.  3. 

Martin  (E.)  u.  Lienhart  (H.)  W6rterbuch  der  elsassischen  Mundarten. 
Im  Auftrage  der  Landesverwaltg.  v.  Elsass-Lothringen.  (In  ca.  6  Lfgn.) 
I.  Lfg.     gr.  8.     xvi,  x6o  S.     Strassburg,  K.  J.  Triibner  Verl,     m.  4. 

Neue  (Frdr.)  Formenlehre  der  lateinischen  Sprache.  3.  Bd.  Das  Verbum. 
3.  Aufl.  V.  C.  Wagener.  10.  u.  11.  (Schluss-)Lfg.  gr.  8.  ii  o.  S.  577-664. 
B.,  S,  Calvary  &*  Co.     Subskr.-Pr.,  m.  2.40;  3.  Bd.  kplt.,  m.  21. 

Parmenides*  Lehrgedicht.  Griechisch  u.  deutsch  v.  Herm.  Diels.  Mit 
e.  Anh.  ab.  griech.  ThQren  n.  Schl5sser.  gr.  8.  164  S.  m.  49  Fig.  B., 
G,  Reimer,     m.  5. 

Pedersen  (Holger).  Aspirationen  i  Irsk,  en  sproghistorisk  unders^gelse. 
I.  del.  Med  et  tillaeg  :  Thesen  til  den  indoevropaeisk  sproghistorie.  gr.  8. 
viii,  200  S.     L.,  M,  Spirgatis,    m.  4.50. 

Philonis  Alexandrini  opera  quae  supersunt,  edd.  Leop.  Cohn  et  Paul. 
Wen^lland.  Vol.  H.  Edidit  Paul.  Wendland.  gr.  8.  xxxiv,  315S.  B., 
G.  Reimer,     m,  9. 

Ed.  minor.     Vol.  IL     Recognovit  W.    8.    xiii,  306  S.    Berlin,  G. 

Reimer,    m.  2, 

Rerum  normannicarum  fontes  arabici,  edidit  Alex.  Seippel.  Fasc  I, 
textum  continens.  Progr.  gr.  4.  148  S.  m.  4  farb.  Karten.  Christiania, 
UniversitdtS'Bibliothek,     m.  22.20. 

Simon  (Joh.  Adph.)  Exoterische  Studien  zur  antiken  Poesie.  i.  Tl.  Zur 
Anordng.  der  Oden,  Epoden  u.  Satiren  des  Horaz.  Lex. -8.  iv,  80  S. 
Koln,  Kolner  Verlags-Anstalt  u,  Druckerei,    m.  2. 

Studien,  romanische,  ver6ffentl.  v.  Emil  Ebering.  i.  u.  2.  Hft.  B.,  E, 
Ebering, — i.  Pillet  (Alfr.)  Die  neuprovenzalischen  SprichwOrter  der  jttn- 
geren  Cheltenhamer  Liederhandschrift.  gr.  8.  131  S.  m.  3.60. — 2.  Erd- 
mannsdOrffer  (E.)     ReimwOrterbuch  der  Trobadors.    8.    vii,  199  S.    m.  5. 

Testamenti,  Veteri8,prophetarum  interpretatioistro-croaticasaeculi  XVI. 


Digitized  by  VjOOQ IC 


252  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Ed.  V.  Jagic.  Lex. -8.  vii,  316  S.  Wien,  A,  Holthausetu  B.,  Weidmann, 
m.  10. 

Trombetti  (Alfredo).  Indogermanische  u.  semitische  Forschungen.  gr.  8. 
vii,  78  S.     Bologna,  Fratelli  Treves,     m.  3.20. 

Winckler  (Hugo).  Altorientalische  Forschungen.  V.  Zur  babylonisch- 
assyr.  Geschichte.  Zur  phOnicisch-karthag.  Geschichte.  Zur  Geschichte 
des  alten  Arabien.     gr.  8.     iii  u.  S.  371-468.     L.,  E,  Pfeiffer,     m.  6. 

Zupitza  (Jul.)  Alt-  u.  mittelenglisches  Uebungsbuch,  m.  e.  Wdrterbuche. 
5.  Aufl.  V.  J.  Schipper.  gr.  8.  viii,  311  S.  Wien,  W.  Braumaller,  m.  6; 
geb.  in  Leinw.,  m.  6.40. 

ITALIAN. 

Angiolini  (Francesco).  Vocabolario  milanese-italiano.  Torino.  8vo, 
109  X  pp.     L.  6.50. 

Borsari  (L.)  Topografia  di  Roma  antica.  Milano,  1897.  i6mo,  442  pp. 
Con  7  tav.     L.  4.50. 

Caruselli  (G.)  Sulle  origini  dei  popoli  italic!.  I.  Italianismi  della  lingua 
greca.     Palermo.     i6mo.     L.  i. 

Decia  (G.)  Florilegio  greco.  Parte  II-III.  2a  impressione.  Firenze. 
x6mo,  pp.  275-797.    L.  3. 

De  Sanctis  (Fr.)  La  letteratura  italiana  nel  secolo  XIX.  Napoli.  i6mo, 
619  pp.     L.  5. 

Librandi  (V.)  Grammatica  albanese.  Milano,  Man.  Hoepli,  i6mo, 
214  pp.     L.  3. 

Monaci  (Ernesto).  Crestomazia  italiana  dei  primi  secoli.  Citt&  di 
Castello.    8vo.     Fasc.  II.    L.  10. 


Digitized  by  VjOOQ IC 


Digitized  by  VjOOQ IC 


254  AMERICAN  JOURNAL    OF  PHILOLOGY, 

Hill  (G.  F.)  Sources  for  Greek  History  between  the  Persian  and  Pelo- 
ponnesian  Wars.     Oxford,  Clarendon  Press^  1897. 

Huizinga  (J.)  Die  Vidflsaka  in  het  indisch  toonell.  (Proefschrift.) 
Groningen,  P,  Noordhoff^  1897. 

Journal  of  the  American  Oriental  Society.  Vol.  18,  second  half.  New 
Haven,  1897. 

Jung  (Julius).  Grundriss  der  Geographie  von  Italien  a.  dem  Orbis 
Romanus.  2te  umgearb.  u.  vermehrte  Aufl.  (Handbuch  der  klass.  Alter- 
tumswissenschaft  herausg.  von  I.  v.  Miiller.  Dritter  Bd.  3.  Abt.  i.  Halfte.) 
Mttnchen,  Oskar  Beck^  1897.     3  m.  50  pf. 

Kuhns  (L.  Oscar).  The  Treatment  of  Nature  in  Dante's  Divina  Corn- 
media.     New  York  and  London,  Edwin  Arnold^  1897.     I1.50. 

Macdonell(A.)  Vedic  Mythology.  (Grundriss  der  indo-arischen  Philo- 
logie  u.  Altertumskunde.  HI.  Bd.  x.  Heft.  A.)  Strassburg,^./.  TrUhner^ 
1897.    9  m.     Subscript*  6.50. 

Merrill  (Elmer  T.)  Fragments  of  Roman  Satire  from  Ennius  to  Apuleius. 
New  York,  Cincinnati,  Chicago,  American  Book  Co,     75  cts. 

Mooney  (W.  D.)  A  Brief  Latin  Grammar.  New  York,  Cincinnati, 
Chicago,  American  Book  Co,^  1897.     75  cts. 

Muret-Sanders.  Encyclopadisches  Wdrterbuch  der  englischen  u.  deut- 
schen  Sprache.  Teil  H  (Deutsch-Englisch.)  Lfg.  2 :  Anbau-anpragcn. 
Berlin,  Langenscheidtsche  Verlagsbuchkandlung,  New  York,  The  Interna^ 
iional  News  Co,     24  Lfgn.  ii  i  m.  50  pf. 

Neuc  (Friedrich).  Formcnlehre  der  lateinischen  Sprache.  Dritter  Band. 
3te  Aufl.  von  C.  Wagener.  lote  u.  iite  Lief.  Berlin,  Calvary  6r»  Co,^  1897. 
2.40  m. 

A  New  English  Dictionary  on  Historical  Principles.  Edited  by  James 
A.  H.  Murray.  Doom-Dziggetai.  (Vol.  IH.)  Oxford,  Clarendon  Press, 
Ss. 

Rambeau  (A.)  et  Passy  (J.)  Chrestomathie  fran^aise  avec  prononciation 
figuree.     New  York,  Henry  Holt  et  Cie,^  1897. 

Ramsay  (W.  M.)  Impressions  of  Turkey  during  Twelve  Years'  Wan- 
derings. New  York,  G,  P.  Putnam^s  Sons,  London,  Hodder  6r*  Stoughton, 
1897. 

Schmid  (Wilhelm).  Der  Atticismus.  Registerband.  Stuttgart,  W, 
Kohlhammer^  1897.     6  m. 

Spencer  (Frederic).  Chapters  on  the  Aims  and  Practice  of  Teaching. 
Edited  by  F.  S.  Cambridge,  University  Press,  New  York,  Macmillan  dr» 
Co.    I1.75. 

Talmud,  Babylonian,  New  edition  of.  By  Michael  L.  Rodkinson.  Vol.  IV. 
New  York,  New  Talmud  Publishing  Co, 

Thucydides.  Book  VL  Edited  by  E.  C.  Marchant.  London,  Macmillan 
&*  Co.,  1897. 

Toepffer  (Johannes).  Beitrage  zur  griech.  Altertumswissenschaft.  Ber- 
lin, Weidmannsche  Buchhandlung^  1897.     10  m. 

Williams  (Ralph  Olmsted).  Some  Questions  of  Good  English.  Henry 
Holt  6r*  Co,,  1897. 


Digitized  by  VjOOQ IC 


AD  VER  TtSEMBNTS. 


LeMCKE  &  BUECHNER, 

Formerly  B.WMtMiiuinn  A  Co., 

Publishers,  Booksellers 
and  importers, 

812  Broadway,  New  York, 

With  Offices  at  London^  Leipzig  and  Paris. 


Just  Published — the  American  Edition  of 

FLUEQEL-SCHMIDT-TANQER'S 

School-  and  Hand-Lexicon  of  the  Qerman  and  English  Languages. 
2  vols.,  large  8vo,  half  leather,  $4.50. 
The  German-English  part  separately,  I2.60. 

Destined  to  replace  the  venerable  Adler  and,  more  or  less,  all  other 
German-English  Dictionaries  now  used  in  Colleges  and  High  Schools.  It 
is  JQSt  the  book  which  Professors  of  German  have  been  asking  us  to 
supply  for  many  years.  When  it  was  not  promptly  forthcoming,  there 
was  a  project  considered  by  these  gentlemen  for  a  time  to  compile  a 
dictionary  here.  Happily  Fluegel  now  is  ready,  and  orders  for  Schools 
can  be  filled.     Professors  will  please  communicate  with  us  direct. 

Special  Notice. 


The  stock  of  complete  sets  in  the  hands  of  the  management  of  The 

American  Journal  of  Philology  was  exhausted  some  time  ago,  but  a 

few  sets  have  been  made  up  of  late  by  purchase  and  exchange.    These  sets 

will  be  sold  at  the  regular  price,  $51  for  the  seventeen  volumes,  cash  to 

accompany  the  order.     Of  single  volumes  (I3)  Volumes  IV  and  VI  cannot 

be  furnished.     Of  single  numbers  (|i)  13,  15,  21,  22,  23,  49,  50,  52  are 

wanting.     No.  40,  which  contains  the  Index  to  Vols.  I-X,  is  not  furnished 

separately,  but  a  small  edition  of  the  Index  itself  is  for  sale  at  |i.oo  a 

copy.    Address, 

B.  L.  GiLDBRSLEBVK,  Baltimore,  Md. 


Digitized  by  VjOOQ IC 


AMERICAN  JOURNAL  OF  PHILOLOGY. 


The  Macmillan  Company's 

New  Books  in  Foreign  Languages,  etc. 

An  Aid  of  Great  Importance  to  Students  of  Early  English^ 

Student's  Dictionary  of  Anglo-Saxon  (The).    By  Henry  Swket,  M.  A., 
Ph.  D.,  LL.D.,  Corresponding  Member  of  the   Munich  Academy  of 
Sciences.     Small  4to,  cloth,  pp.  xvi  +  217.     Price,  |i.7S  net. 
The  Anglo-Saxon  Primer  and  Reader  by  Henry  Sweet  are  so  well  known 

that   this  needs  no  introduction;    they  rank  at   once  among  the   books 

necessary  to  every  student  of  Anglo-Saxon. 

FRENCH. 

BIART. — Quand  J'etais  Petit.  Histoire  d'un  Enfant  racont^e  par  un 
Homme.  By  Lucien  Biart.  Adapted  for  Use  in  Schools,  with  Notes 
and  Vocabulary  by  James  BoIelle,  B.  A.  Part  I.  i6mo,  cloth.  Pitt 
Press  Series.     Price,  50  cents  net. 

MOLIERE — L'Avare.  Par  J.-B.  P.  MoLiiRE.  Edited  with  Introduc- 
tion and  Notjcs  by  E.  G.  W.  Braunholtz,  M.A.,  Ph.D.,  University 
Lecturer  in  French.  i6mo,  cloth,  pp.  xlvii-l-245.  Pitt  Press  Series. 
Price,  70  cents  net. 

GERMAN. 

MACMILLAN*S  FOREIGN  SCHOOL  CLASSICS.  Edited  by  G. 
Eugene  Fasnacht.  LESSINQ. — Minna  von  Bamhelm.  Edited 
with  Introduction  and  Notes  by  Rev.  Charles  Merk,  M.  A.,  Ph.  D., 
Leipzig.     i2mo,  cloth,  pp.  lxviii-{-  224.     Price,  75  cents  net. 

AMERICANA  GERMANICA.  A  Quarterly  devoted  to  the  Comparative 
Study  of  the  Literary,  Linguistic  and  other  Cultural  Relations  of 
Germany  and  America.  Editor,  Professor  M.  D.  Learned,  of  the 
University  of  Pennsylvania.  Vol.  I,  No.  i.  Single  number,  75  cents. 
Annual  subscription,  $2.00. 

ENGLISH. 

BARRERE  and  LELAND.^A  Dictionary  of  Slang,  Jargon,  and  Cant. 
Embracing  English,  American  and  Anglo-Indian  Slang,  Pidgin  English, 
Gypsies'  Jargon,  and  other  irregular  phraseology.  Compiled  and 
edited  by  Albert  Barr^re,  Officier  de  IMnstruction  publique;  Pro- 
fessor R.  M.  a.  Woolwich,  author  of  'Argot  and  Slang,*  etc,  etc.; 
and  Charles  G.  Leland,  M.  A.,  Hon.  F.  R.  S.  L.,  author  of  *The 
Breitmann  Ballads,'  *The  English  Gypsies  and  their  Language,'  etc.* 
etc.  In  2  vols.  Vol.  I,  A-K,  pp.  xix-f-soo;  vol.  II,  L-Z,  pp.  415. 
Crown  8vo.     Price,  $4.00. 

In  Preparation. 
GERMAN. 

HEINE.— Selections  from  Heine.  Edited  with  Notes  and  Introduction 
by  C.  A.  BucHHEiM,  Professor  of  German  Literature  in  King's  College. 
Uniform  with  Deutsche  Lyrik,  Golden  Treasury  Series. 


The  Macmillan  Company,  66  Fifth  Ave.,  New  York. 


Digitized  by  VjOOQ IC 


AD  VER  TISEMENTS, 


THE  FRIEDENWALD 

COMPANY, 


Printers, 
Lithographers, 
Book  Binders, 


BALTIMORE,  EUTAW  AND  GERMAN  STREETS, 

BALTIMORE,  MD. 


THE  LEADING  HOUSE, 


Printers  of  the  American  Journal  of  Philology,  American 
Journal  of  Mathematics,  Memoirs  from  the  Biolog- 
ical Laboratory,  Studies  in  Historical  and 
Political  Science,  issued  by  the  Johns 
Hopkins  University. 


ESTIMATES  CHEERFULLY  SUBMITTED. 


Digitized  by  VjOOQIC 


AMERICAN  JOURNAL  OF  PHILOLOGY. 


INDISPENSABLE  TO  EVERY  LATIN  TEACHER. 
OILDBRSLEEVE'S  LATIN  ORAMMAR. 

THIRD  EDITION,  REVISED  AND  ENLARGED. 

By  B.  L.  GiLDBRSLSEVE,  Ph.  D.,  Professor  of  Greek  in  the  Johns  Hopkins 
University,  Editor  of  the  American  Joarnal  of  Philology,  and  Gonzalez 
Lodge,  Ph.  D.,  Associate  Professor  of  Latin  in  Bryn  Mawr  College,  Pa. 
550  pp.     Cloth,  9x.ao. 

For  school  and  college  use^  the  new  edition  of  Gildenleeve's  Latin  Grammar,  in  point  of 
fulness,  accuracy  and  convenience,  rivals,  if  it  does  not  surpass,  all  Latin  grammars  with 
which  I  am  acquainted.— J.  P.  Postgatb,  Fellow  of  Trinity  College,  Cambridge,  England. 

It  is  a  superfluous  compliment  to  praise  any  work  of  Prof.  Gildersleeve,  who  never  does 
any  which  is  not  praiseworthy.— W.  W.  Goodwin,  Ph.  D.,  Eliot  Prof,  of  Greek  Literature, 
Harvard  University,  Cambridge,  Mass. 

CAESAR'S  CIVIL  WAR, 

With  Maps  and  Vocabularies.  Edited  for  the  use  of  Schools,  with  refer- 
ences to  the  Latin  Grammars  of  Gildersleeve,  Allen  and  Greenough, 
and  Harkness.  By  B.  Perrin,  Ph.  D.,  Professor  of  Greek  in  Yale 
University.     i2mo,  348  pp.     Price,  Cloth,  $z.oo. 

Prof.  Tract  Pbck,  Yale  University.  New  Haven,  says:  "It  is  matter  for  congratulation 
that  at  last,  and  so  auspiciously,  the  'Civil  War'  has  been  made  available  and  attractive  for 
American  schools." 

Prof.  W.  A.  Packard,  College  of  New  Jersey,  Princeton.  N.  J.,  says:  "Perrin's  'Civil 
War'  is  thoroughly  and  skillfully  edited,  and  is  a  welcome  addition  to  our  preparatory  text- 
books in  Latin." 

UNIVERSITY  PUBLISHING  COMPANY, 

Boston,  352  Wathington  St.         43,  45,  47  E.  Tenth  St.,  NEW  YORK. 
Nkw  Orleans,  714-716  Canal  St,  [4T. 

ESSAYS   AND   STUDIES 

BY 

BASIL  LANNEAU  GILDERSLEEVE 
J20  pages,  small  quarto,  bound  in  cloth.    Price,  tj-SO  («^0« 


Educational  Essays: — i.  Limits  of  Culture.  2.  Classics  and 
Colleges.  3.  University  Work  in  America.  4.  Grammar 
and  Aesthetics. 

Literary  and  Historical  Studies: — i.  Legend  of  Venus. 
2.  Xanthippe  and  Socrates.  3.  ApoUonius  of  Tyana. 
4.  Lucian.  5.  The  Emperor  Julian.  6.  Platen's  Poems. 
7.  Maximilian ;  his  Travels  and  his  Tragedy.  8.  Occa- 
sional Addresses. 

N.  MURRAY, 
The  Johns  Hopkins  Press, 

Oct.,  1890.  Baltimore,  Md. 


Digitized  by  VjOOQ IC 


AD  VER  TISEMENTS, 


HELPS  IN  CLASSICAL  STUDY. 


HARPER'S  CLASSICAL  DICTIONARY. 

Harper's  Dictionary  of  ClasAical  Literature  and  Antiquities.  Edited  by  Hakkt  Thur- 
STON  Pbck,  M.  a..  Ph.  D.p  Professor  of  the  Latin  Languages  and  Literature  in  Columbia 
University,  City  of  New  York.  With  the  cooperation  of  many  special  contributors. 
Pp.  xvi,  1 701.  with  about  1500  illustrations.  Royal  8to,  cloth,  $6.00;  half  leather,  $8.00. 
Tvro-Tolume  edition,  cloth,  $7.00. 
Unquestionably,  in  many  respects,  the  most  authoritative  work  of  the  kind  that  has  vet 

been  published.  ...    In  short,  this  is  what  the  editor  claims  for  it— a  classical  encyclopaedia. 

There  is  no  one  volume  on  the  market  at  once  so  complete  and  so  succinct,  and  whose  vast 

collection  of  information  is  so  readily  and  immediately  accessible. — N,  V,  Hgr€Ud, 
The  student  of  classical  antiquity,  whether  in  art,  life,  religion,  or  literature,  will  find  the 

volume  a  whole  library  of  information,  admirably  arranged  for  practical  use.  . .  .    The  text  is 

clear,  terse,  and  pointed. — Chicago  Inttr^Octan, 

MODERN  GREEK  MASTERY. 

A  Short  Road  to  Ancient  Greek.    By  Thomas  L.  Stbdman,  A.  M.,  M.  D.    Pp.  viii,  380. 

I^Mt  8vo,  cloth,  $1.50. 
Dr.  Stedman  has  filled  a  very  actual  want. ...  A  work  by  means  of  which  the  mastery  of 
the  so-called  modern  Greek  language  will  be  very  greatly  facilitated  and  that  of  the  ancient 
will  be  made  more  easjr. .  .  .  The  parts  dealing  with  the  conjugation  of  the  verb,  especially 
the  sections  on  the  aorist  and  future,  are  written  with  the  greatest  clearness,  and  are  readily 
comprehensible.  . . .  There  never  has  been  published  in  the  English  language  such  a  practical 
method  for  the  mastery  of  the  Greek  tongue;  and  our  opinion  is  that  this  important  and,  as 
far  as  that  is  possible,  perfect  work  of  Dr.  Stedman  is  a  production  of  great  weight.— ^//aM/tr, 
New  York. 

AN  ENQLISH-QREEK  DICTIONARY. 

A  Concise  Dictionary  of  the  English  and  Modern  Greek  Languages,  as  Actually  Written 

and  Spoken.    By  A.  M.  Jannasis,  Ph.  D.    English*Greek.    Pp.  xvi,  436.    Post  8vo,  cloth, 

la.50. 

The  author  has  endeavored  to  render  this  work  compact  and  concise  by  employing  a  judicious 

system  of  abbreviation  and  by  omitting  all  references  and  quotations.    The  Dook  is  of  interest 

to  the  student  of  Classic  Greek  who  possesses  any  tendency  towards  philology.    The  extinc- 

tion  of  the  present  infinitive,  of  the  optative,  and  of  part  of  the  imperative  will  be  noticed  in 

examining  the  verb.    Likewise  the  replacing  of  the  dative  by  the  accusative,  the  disappear- 

ance  of  rho  final,  and  the  abandonment  of  the  reduplication  in  nouns  and  participles.    Archaic 

or  learned,  colloquial,  literary  and  ecclesiastical  expressions  are  all  given,  and  are  marked 

with  different  siens.    Careful  comparison  reveals  the  Roman,  Byzantine,  or  Turkish  influence 

in  some  of  the  older  forms,  and  the  tendency  towards  learned  Greek  in  the  present  Restoration 

Period  of  the  languages. 

PROFESSOR  GILDERSLEEVE*S  PINDAR. 

Pindar:  The  Olympian  and  Pythian  Odes.   With  Introductory  Essay.  Notes,  and  Indexes. 
By  Basil  L.  Gildbkslbbvb,  Professor  of  Greek  in  the  Johns  Hopkins  University. 
Pp.  cxvi,  396.    xamo,  cloth,  ft. 50. 
Prepared  with  great  exzctneMt.—Brookfyn  Union. 

Never  before  did  the  appearance  of  an  American  edition  of  a  Greek  poet  so  nearly  approach 
an  'event'  in  scholastic  circles  as  the  publication  of  Gildersleeve's  ' Pindar.'— A^tfW  York 
Times, 

Pour  ceux  qui  commencent  I'^tude  de  Pindare  rien  ne  sauraic  remplacer  ce  livre.— i?/vtM  tU 
Philologie, 

PROFESSOR  QILDERSLEEVE*S  PERSIUS. 

The  Satires  of  A.  Persius  Flaccus.    Edited  by  Basil  L.  Gildbrslbbvb,  Ph.  D.  (G6t> 
tingen),  LL.  D.,  Professor  of  Greek  in  the  Johns  Hopkins  University.    Pp.  73a.    ismo, 
cloth.  90  cents.    By  mail,  fz.oo. 
I  discover  in  it  at  once  ample  evidence  of  the  accurate  and  thorough  scholarship  which  I 
expected  to  find. ...    I  have  myself  found  instruction  and  recreation  in  it.    The  'Introduc- 
tion' and  'Notes'  are  as  good  a  whetstone  as  the  'Satires'  themselves. — W.  S.  Tylbs, 
Professor  of  Greek,  Amherst  College. 

JUSTIN  MARTYR. 

The  Apologies  of  Justin  Martyr.  To  which  is  appended  the  Epistle  to  Diognetus.  With 
an  Introduction  and  Notes  by  Basil  L.  Gildbrslbxvb,  Ph.  D.  (Gfitt.).  LL.  D.,  Professor 
of  Greek  in  the  Johns  Hopkins  University.    Pp.  xlii,  390.    zsmo,  cloth,  ^i.jo.    By  mail. 

Published  by  HARPER  &  BROTHERS,  New  York. 

Th*  above  works  art  for  sale  by  all  booksellers,  or  will  be  sent  by  the  Publishers  on  receipt 
of  Price,    1/ ordered  sent  by  mail^  lo/er  cent,  must  be  added  to  cover  cost  0/ Postage,       [at. 


Digitized  by^ 


THE  JOHNS  HOPKINS  PRESS, 


I.  American  Journal  of  Mathematics. 

Edited  by  T.  Craig,  with  tlie  co-operation  of  S.  Nswcomb.     Quarterly. 
4to.     Volume  XIX  in  progress.     $5  per  volume. 

II.  American  Chemical  Journal. 

I.  KsMSBN,  Editor.   Ten  numbers  a  year.     8vo.      Volume  XIX  in  progress. 
$4  per  volume. 

III.  American  Journal  of  Philology. 

B.  L.  GiLDSRSLEEVE,  Editor.     Quarterly.    8vo.     Volume  XVII  complete. 
$3  per  volume.     Index  to  Vols.  I-X,  $1, 

IV.  Studies  from  the  Biological  Laboratory.    8vo.     Volume  V  complete. 

$5  per  volume. 

V.  Studies  in  Historical  and  Political  Science. 

H.  B.  Adams,  Editor.     Monthly.    8vo.     Volume  XV  in  progress.     $3  per 
volume. 

VI.  Johns  Hopkins  University  Circulars. 

Containing  reports  of  scientific  and  literary  work  in  progress  in  Baltimore. 
4to.    Vol.  XVI  in  progress.    $1  per  year. 

VII.  Johns  Hopkins  Hospital  Bulletin.    4to.    Monthly.    Volume  VIII  in 
progress.    $1  per  year. 

VIII.  Johns  Hopkins  Hospital  Reports.    4to.    Volume  VI  in  progress.     ^5 
per  volume. 

IX.  Contributions  to  Assyriolog^  and  Comparative  Semitic  Philology. 
(Beitrsige  zur  Assyriologie,  etc.)    Vol.  Ill  in  progress. 

X.  Memoirs  from  the  Biological  Laboratory. 

W.  K.  Brooks,  Editor.     Volume  III  complete.     $7.50  per  volume. 

XI.  Annual  Report. 

Presented  by  the  President  to  the  Board  of  Trustees,  reviewing  the  opera- 
tions of  the  University  during  the  past  academic  year. 

XII.  Annual  Register. 

Giving  the  list  of  officers  and  students,  and  stating  the  regulations,  etc,  of 
the  University.     Published  at  the  close  of  the  academic  year. 


In  addition  to  the  serials  above  named,  copies  may  be  obtained  of  the  works 
mentioned  below : 
The  Critical  Edition  of  the  Hebrew  Text  of  the  Old  Testament. 

Edited  by  Professor  Paul  Haupt.     Prospectus  on  application. 
Rowland's  Photograph  of  the  Normal  Solar  Spectrum.     10  plates. 

$20.00. 
Description  of  the  Johns  Hopkins  Hospital.     J.  S.  Billings,  Editor. 

116  pp.  4to.     56  plates.     $7.50,  cloth. 
The  Teaching  of  the  Apostles.    Complete  facsimile  edition.    Edited  by 

J.  Rendel  Harris,     no  pp.  4to.     10  plates.     $5.00. 
Reproduction  in  Phototype  of  a  Syriac  MS  with  the  Antilbgombna 

Epistles.     Edited  by  I.  H.  Hall.    I3. 
The  Oyster.     By  W.  K.  Brooks.     240  pp.  i2mo.     14  plates.     $1.00. 
Studies  in  Logic     By  members  of  the  Johns  Hopkins  University.     C.  S. 

Pierce,  Editor.     123  pp.  i2mo.     $2.00,  cloth. 
The  Constitution  of  Japan,  with  Speeches,  etc,  illustrating  its  significance. 

48  pp.  i6mo.     50  cents. 
BiBLiOGRAPHiA  HoPKiNSiENSis.   Part  I.  Philology.   Parts  Il-in.  Chemistry, 

Mineralogy   and    Geology.      Parts   IV-V-VI.    Physics,   Astronomy   and 

Mathematics.     8vo.     30  cents  per  part. 

A  full  list  of  publications  will  be  sent  on  application. 


Communications  in  respect  to  exchanges  and  remittances  may  be 
tent  to  The  Johns  Hopkins  Press,  Baltimore,  Maryland. 


Digitized  by  VjOOQ IC 


AMERICAN 
JOURNAL    OF    PHILOLOGY 

Vol.  XVIII,  3.  Whole  No.  71. 


I.— THE  ETHICS  AND  AMENITIES  OF  GREEK 
HISTORIOGRAPHY.^ 

The  great  Bishop  of  the  fourth  century,  the  Christian  Hero- 
dotus, ''making  it  his  object  to  show  that  heathen  lore  was 
generally  false  and  foolish,"  quotes  first  from  Clement  of  Alex- 
andria proofs  that  Greek  literature  was  full  of  plagiarism  (Praepar. 
Evang.  X  2).  He  then  more  gladly  draws  his  proofs  from 
Porphyry,  the  great  antagonist  of  Christianity  in  the  third 
century.  Porphyry  describes  a  symposium  to  commemorate 
the  name  of  Plato,  held  in  Athens  at  the  home  of  his  master,  the 
great  Platonist,  Longinus.  The  literjury  chat  at  this  symposium 
may  well  give"u§*"th«>^best  Athenian  opinion  of  the  third  century 
of  our  era.  As  the  symposium  advances  a  lover  of  Ephorus  and 
a  lover  of  TheopQmpus  get  by  the  ears  in  advocacy  of  their 
favorite  historians.  The  heated  debate  proves  that  both  Ephorus 
and  Theopompus  were  wholesale  thieves,  Theopompus  particu- 
larly. He  stole  a  good  story  about  Pythagoras  from  Andron  of 
Ephesus,  merely  changing  name  and  place.  He  stole  large 
material  from  Xenophon's  HelUnica,  changing  only  for  the 
worse.  "Yes,**  said  ApoUonius  the  grammarian,  and  he  surely 
.knew,  "Ephorus  and  Theopompus,  being  lazy  men,  stole  as  a 
matter  of  course.  But  catalogues  have  been  made  of  the  thefts  of 
such  men  as  Sophocles  and  Menander.  Hyperides  stole  from 
Demosthenes,  Hellanicus  took  chapters  of  barbarian  customs 
from  Herodotus,  as  Herodotus  had  taken  much  from  Hecataeus 

^  President's  Address  at  the  28th  annual  session  of  the  Am.  Phil.  Ass.,  Bryn 
Mawr,  Pa.,  July  6,  1897. 


Digitized  by  VjOOQ IC 


256  AMERICAN  JOURNAL    OF  PHILOLOGY. 

with  trifling  changes.  Hesiod's  primal  testimony — *  Nothing 
better  than  a  good  woman,  nothing  worse  than  a  bad  one,'  is 
stolen  by  Simonides  and  Euripides.  Nay,"  said  Apollonius, 
**  there  is  a  library  of  writers  on  literary  theft — "  a  library  of 
klopediaSy  "that  I  may  not  be  found  out  to  be  a  thief  myself 
on  theft.  There  is  Lysimachus — two  books  on  the  thievery  of 
Ephorus,  Polio — a  book  on  the  thievery  of  Herodotus,"  etc. 

Then  up  spake  the  banqueter  Prosenes,  "Ye  have  shown  all  the 
rest  to  be  thieves,"  said  he,  "but  I  declare  that  even  the  divine 
Plato  himself,  whose  memory  we  honor  in  this  banquet,  made 
great  use  of  his  predecessors,  not  to  call  it  thievery."  "What 
sayest  thou ! "  cried  one  aghast.  "  I  not  only  say,"  replied  the 
sacrilegious  Prosenes,  "  but  I  can  prove  what  I  say,  though  Plato's 
predecessors  wrote  few  books."  And  he  does  so  by  many 
instances. 

To  a  company  of  Platonists,  then,  gathered  at  a  banquet  in 
Athens  towards  the  middle  of  the  third  century  of  our  era,  the 
most  salient  feature  of  Greek  literature  in  general,  and  of  Greek 
historiography  in  particular,  was  its  klepHcistn.  And  this  opinion 
of  Athenians  of  the  third  century  is  urged  by  Christian  writers  of 
the  fourth  century  in  their  crusade  against  pagan  literature  as  such. 

Lucian  will  represent  for  us  the  opinion  of  a  versatile  and  gifted 
cosmopolitan  Greek  of  the  second  century.  His  criticisms  apply 
mainly,  it  is  true,  to  the  historiography  of  his  day.  But  his  his- 
torians are  Greek,  with  Greek  models  and  literary  inheritances. 
And  Lucian*s  attitude  toward  the  great  models  of  his  victims  is 
seen  incidentally.  He  brings  very  serious  charges  against  the 
historians  of  his  own  day  (Quomodo  Hist.  Conscr.  passim). 

First,  servile  flattery  of  generals  and  leaders,  encomium  instead 
of  history,  whereas — and  here  sounds  the  oft-recurring  Thucy- 
didean  tone  in  Lucian — posterity  should  be  the  approvers  of  a 
historical  work,  not  contemporaries. 

Second,  poetical  instead  of  prose  canons;,  one  law,  viz.  the 
pleasure  of  the  author  and  his  audience,  and  in  obedience  to  this 
law  every  species  of  poetical  embellishment;  epic  invocations  to 
muses;  such  dangerous  Homeric  comparisons  as  that  of  the 
Roman  general  to  Achilles,  and  his  Parthian  antagonist  to  Ther- 
sites ;  tedious  epic  descriptions. 

Third,  self-praise,  exaltation  of  the  historian's  native  city,  and 
of  his  opportunities  as  a  historian. 

Fourth,  indiscriminate  appropriation  of  the  successful  compo* 


Digitized^yVjOOQlC 


GREEK  HISTORIOGRAPHY,  257 

sitions  of  predecessors ;  Herodotean  and  Thucydidean  exordia  \ 
the  funeral  oration  of  Pericles  adapted  to  a  new  funeral. 

Fifth,  invention  of  extravagant  exploits;  false  statistics  of 
armies  engaged  and  of  losses  in  battle,  altogether  regardless 
of  official  reports. 

Plagiarism  and  falsehood,  then,  are  prominent  in  Lucian's 
charges  against  the  historians  of  his  day.  Theopompus  he  calls 
a  general  viliiier,  Ctesias  a  general  liar.  Ctesias  is  blamed,  how- 
ever, not  for  lying,  but  for  supposing  that  people  would  not  know 
that  he  was  lying  (Vera  Hist.  4). 

Josephus,  the  learned  Hebrew  warrior,  statesman  and  historian 
of  the  first  century,  whom  Jerome  calls  the  Greek  Livy,  whom 
we  may  call  the  Hebrew  Polybius — ^Josephus  had  occasion  to 
review  critically  the  historiography  of  the  Greeks  (Contra 
Apion.).  Their  lack  of  agreement  on  the  same  subject  is  what 
he  specially  emphasizes,  and  the  acrimonious  correction  of  each 
by  his  successor.  Hesiod  is  accused  of  falsehood  and  corrected 
by  Acusilaus,  he  says,  Acusilaus  by  Hellanicus.  Then  Ephorus 
shows  that  Hellanicus  lies  in  most  that  he  says,  Ephorus  in  turn 
is  attacked  and  corrected  by  Timaeus,  Timaeus  by  his  successors, 
and  Herodotus  by  everybody.  Not  plagiarism  or  klepticism^ 
then,  is  the  burden  of  the  arraignment  which  Josephus  makes  of 
the  Greek  historians,  but  falsehood.  They  are  not  only  xXcirrat, 
but  ^vtrrai,  and  Herodotus  is  the  arch-liar.  The  Atthides^ 
Josephus  says,  all  differ  from  each  other,  so  do  the  Argolic  annals, 
and  the  histories  called  ktIqw.  Above  all,  in  the  histories  of  the 
Persian  wars  the  most  famous  historians  are  widely  at  variance. 
Even  Thucydides,  he  says,  is  accused  by  many  of  lying. 

This  sweeping  charge  of  Josephus  is  only  an  echo  of  charges 
made  by  the  Greeks  themselves.  Strabo  complains  (p.  341)  that 
the  ancient  writers,  like  Hecataeus,  "say  much  that  is  false,'' 
because  they  were  reared  on  £adsehood  in  their  mythography. 
Therefore  they  do  not  agree  with  one  another  on  the  same 
theme.  This  is  significant  from  a  man  who  regarded  Homer 
as  an  epitome  of  all  knowledge.  But  we  find  no  better  agree- 
ment among  the  later  writers  whom  Strabo  uses.  They  not 
only  do  not  agree  with  one  another,  but  each  in  turn  accuses  his 
predecessor  of  falsehood.  Starting  with  Timaeus,  the  arch- 
censor,  Timaeus 'Epitimaeus, — Timaeus  says  that  the  greatest 
fault  in  history,  by  which  he  means  the  Greek  historians,  is  want 
of  truth.    And  he  accordingly  advises  all  whom  he  has  convicted 


Digitized  by  VjOOQ IC 


258  AMERICAN  JOURNAL    OF  PHILOLOGY. 

of  making  false  statements  in  their  writings,  to  find  some  other 
name  for  their  ,books,  and  to  call  them  anything  they  like  except 
history  (Polyb.  XII  12).  Now  Timaeus  convicted  all  his  pre- 
decessors of  falsehood,  to  say  nothing  of  other  faults,  and  seems 
to  have  spared  no  contemporary.  "  He  makes  such  a  parade  of 
minute  accuracy,"  says  Polybius,  "and  inveighs  so  bitterly  when 
refuting  others,  that  people  come  to  imagine  that  all  other  histo- 
rians have  been  mere  dreamers,  and  have  spoken  at  random  in 
describing  the  world"  (Polyb.  XII  26,  Shuckburgh's  translation). 
But  this  arch-censor  found  his  own  censor,  judge  and  executioner, 
all  in  one.  What  historical  sin  is  not  charged  upon  Timaeus  in 
that  fearful  book  which  Polybius  devotes  especially  to  him? 
Malevolent,  deliberate  falsehood ;  mendacious  omissions ;  coarse, 
partisan  abuse;  simulated  veracity;  manufactured  evidence; 
childish  ignorance;  neglect  of  personal  inquiry;  incomplete 
study  of  his  subjects ;  great  professions  and  scant  performance. 
He  is  inaccurate  and  untrustworthy  even  when  reporting  the 
evidence  of  his  own  eyes ;  he  is  ill-informed  and  easily  misled 
about  localities  where  he  was  born  and  bred;  he  is  uncritical, 
undignified,  swayed  by  personal  jealousies  and  animosities;  he 
is  a  carping,  false,  impudent,  unphilosophical,  paradoxical,  obsti- 
nate, vituperative  word-juggler. 

Polybius  may  have  found  a  severe  judge  in  his  successor,  Posi- 
donius,  and  Posidonius  in  Strabo.  Plutarch  may  say  of  Duris  of 
Samos  (Pericles,  28)  that  he  lied  even  when  he  had  no  conceivable 
motive  for  lying,  and  much  more  when  he  had  personal  griev- 
ances. But  surely  the  grand  climax  in  the  critical  depreciation 
of  a  predecessor  is  reached  in  this  twelfth  book  of  Polybius.  All 
else  of  this  nature  in  the  long  line  of  further  historical  transmission 
has  the  diminishing  intensity  of  a  dramatic  exodos. 

It  is  the  culmination  of  a  long  practice,  so  long  as  to  have 
become  a  literary  tradition.  To  the  Hesiodic  poet,  the  Homeric 
poet  is  a  liar.  He  sings  of  what  never  was.  Didactic  poetry 
must  teach  men  the  truth.  But  the  truth  is  just  what  Acusilaus 
and  the  earliest  logographers  miss  in  Hesiod  and  didactic  poetry. 
They  will  therefore  tell  the  truth  in  prose.  But  their  effort  seems 
to  have  been  vain,  for  Hecataeus  begins  his  genealogies  with  a 
scornful  reference  to  their  lack  of  veracity.  "These  things  I 
write,  as  I  think  them  to  be  true.  For  the  stories  of  the  Hellenes 
are  many  and  ridiculous."  By  "the  Hellenes"  he  means  Acusi- 
laus particularly,  his  predecessor  and   main  source,  a  literary 


Digitized  by  VjOOQ IC 


GREEK  HISTORIOGRAPHY,  259 

mannerism  which  can  be  traced  from  Hecataeus  through  Hero- 
dotus to  Pausanias. 

But  Hecataeus  in  his  turn  is  found  ridiculous  by  his  great  suc- 
cessor, Herodotus,  for  whom  he  served  as  an  important  source. 
How  the  Father  of  History  makes  merry  at  the  expense  of  the 
Father  of  Chartography.  and  of  his  absurd  theory  of  ocean  cur- 
rents! "It  makes  me  laugh,"  he  says  (IV  36),  "to  see  so  many 
people  drawing  maps  of  the  world,  and  not  a  man  of  them 
describing  it  sensibly.  They  draw  Oceanus  flowing  round  the 
earth,  which  is  a  perfect  circle,  and  they  make  Asia  and  Europe 
of  the  same  size."  It  is  well  that  Herodotus  had  his  laugh  before 
his  successors  took  hold  of  him.  For  everybody  corrects  and 
ridicules  Herodotus,  as  Josephus  says,  and  "everybody,"  for  us, 
includes  Professor  Sayce. 

And  who  does  not  know  the  magnificent  scorn  which  Thucy- 
dides  has  for  Herodotus?  The  great  trait  of  the  jcr^/ia  h  ati 
should  be  its  truth,  even  though  it  lacked  the  charm  of  the  mythic 
falsehood.  But  neither  Thucydides,  the  father  of  what  the 
modern  world  calls  history,  who  invented  the  art  of  determining 
from  what  is  handed  down  what  really  was,  nor  his  great  con- 
temporary Hellanicus,  the  founder  of  scientific  chronology,  escape 
the  charge  of  falsehood  from  their  successors,  Ephorus  and 
Theopompus,  who,  in  their  turn,  as  we  have  seen,  come  under 
the  sweeping  censures  of  Timaeus-Epitimaeus.  And  of  him 
Polybius  says  (XII  25),  "Those  who  are  most  ready  at  finding 
fault  with  others  are  most  prone  to  error  in  their  own  life." 
This  general  principle  may  safely  be  applied  to  the  long  line  of 
celebrities  in  Greek  historiography,  especially  since  we  find 
Polybius  sharply  criticising  Fabius  Pictor,  Philinus  and  Phylar- 
chus  for  the  very  defects  which  are  most  conspicuous  in  himself. 

To  the  early  Christian  scholars,  then,  to  the  cosmopolitan 
Lucian,  to  Josephus,  Greek  historians  are  thieves  and  liars,  and 
the  successive  Greek  historians  themselves  have  this  same  opinion 
of  their  predecessors.  On  the  ethical  side,  therefore,  Greek 
historiography,  judging  from  the  opinions  of  those  who  knew  it 
best,  left  much  to  be  desired.  The  studies  of  our  own  later  times 
lengthen  this  list  of  its  ethical  shortcomings,  and  find  in  it, 
besides,  a  surprising  lack  of  the  fine  amenities  which  characterize 
modern  historical  transmission,  even  when  it  is  polemical  or  con- 
troversial. 

This  constant  depreciation,  this  contempt  for  a  predecessor,  is 
always  most  pronounced  when  that  predecessor  is  most  generously 


Digitized  by 


Gangtr- 


260  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Utilized.  Hecataeus  treated  with  bold  and  harsh  scepticism  the 
oral  and  literary  tradition  on  which  he  based  his  genealogies. 
Herodotus  could  hardly  have  written  his  second  and  fourth  books 
had  it  not  been  for  the  pioneer  work  of  Hecataeus,  and  yet  he 
mentions  Hecataeus  only  once  by  name,  and  that  to  mock  at 
his  aristocratic  lineage.  He  refers  to  him  often  by  vague  plurals 
or  gentile  adjectives,  but  then  only  to  correct,  complete,  or  ridicule. 
The  note  of  obligation  is  never  sounded.  The  tone  of  con- 
temptuous superiority  is  never  absent. 

So  Thucydides  makes  his  stately  introduction  a  labored  depre- 
ciation of  the  great  work  of  Herodotus,  by  which,  more  than  by 
anything  else,  his  survey  of  the  history  prior  to  his  own  point  of 
departure  is  made  possible.  That  great  work  was  the  tale  of  a 
chronicler  who  sought  to  please  the  ear  rather  than  to  speak  the 
truth  (I  20).  Thucydides  will  not  so  much  as  write  the  name  of 
Herodotus,  but  denies  the  greatness  of  his  theme,  and  refuses  to 
admit  the  essential  truth  of  his  narrative.  He  protests  also  against 
the  seductive  charm  of  his  manner.  He  does  once  break  this 
contemptuous  silence  about  his  literary  sources,  and  scornfully 
accuses  Hellanicus  of  insufficiency  and  inaccuracy,  but  at  just  the 
point  where  he  is  obliged  to  use  Hellanicus  most  freely  (I  97,  2). 
Ephorus  and  Theopompus  in  their  day  and  generation  despise 
both  Herodotus,  Thucydides  and  Hellanicus,  and  yet  they  often 
merely  rewrite  the  material  furnished  them  by  these  predecessors. 
And  so  on  down  the  line.  Even  Polybius  devotes  a  book  to  the 
crushing  of  Timaeus-Epitimaeus  at  just  the  point  where  his  obli- 
gations to  Timaeus  become  greatest.  Polybius  can  rise  to  a  much 
higher  level,  it  is  true.  He  can  even  approach  the  best  modern 
•  sentiment  of  obligation  toward  a  predecessor.  "Judging  from  their 
lives  and  principles,"  he  says  of  his  sources  Philinus  and  Fabius 
(I  xiv),  "  I  do  not  suppose  that  these  writers  have  intentionally 
stated  what  was  false ;  but  I  think  that  they  are  much  in  the  same 
state  of  mind  as  men  in  love.  Partisanship  and  complete  pre- 
possession made  Philinus  think  that  all  the  actions  of  the  Cartha- 
ginians were  characterized  by  wisdom,  honor,  and  courage; 
those  of  the  Romans  by  the  reverse.  Fabius  thought  the  exact 
opposite."  "One  must  not  shrink,  however,  either  from  blaming 
one's  friends  or  praising  one's  enemies ;  nor  be  afraid  of  finding 
fault  with  and  commending  the  same  persons  at  different  times." 
"  For  it  is  impossible  that  men  engaged  in  public  affairs  should 
always  be  right,  and  unlikely  that  they  should  always  be  wrong." 


Digitized  by  VjOOQ IC 


GREEK  HISTORIOGRAPHY,  261 

And  yet  this  same  Polybius  speaks  in  his  polemic  against  Timaeus 
as  though  the  profit  one  got  from  one's  predecessors  was  the 
malicious  joy  of  detecting  their  errors.  "Study  of  documents," 
he  says,  "involves  no  danger  or  fatigue,  if  one  takes  care  to  lodge 
in  a  city  rich  in  such  records,  or  to  have  a  library  in  one's  neigh- 
borhood. You  may  then  investigate  any  question  while  reclining 
on  your  couch,  and  compare  the  mistakes  of  former  historians 
without  any  pain  to  yourself"  (XII  27).  But  it  needs  no  further 
citation  to  show  that  the  fine  sentiment  of  obligation  to  pioneer 
predecessors,  a  sentiment  which  permeates  the  best  modern 
writing,  and  which  we  find  also  in  Pliny,  Jerome,  and  Plutarch,  is 
wholly  lacking  in  Greek  historiography.  The  note  of  obligation 
is  never  sounded ;  the  tone  of  contemptuous  superiority  is  seldom 
absent. 

Aggressive  ingratitude,  like  that  now  described,  is  of  course 
more  striking,  but  perhaps  no  more  objectionable  to  the  best 
modern  spirit  than  the  ingratitude  of  silence.  In  Greek  historio- 
graphy, before  as  well  as  after  the  days  of  compilations  and  com- 
pends,  there  is  steadfast  reluctance  to  name  a  source  at  all.  This 
is  one  thing  that  makes  source-criticism  at  once  so  fascinating  and, 
in  the  main,  so  fruitless.  The  feeling  is  even  more  than  mere 
reluctance  to  name,  it  is  desire  to  conceal  a  source.  And  the 
charge  must  not  be  laid  at  the  doors  of  Herodotus  or  Pausanias,  or 
other  great  scape-goats  of  criticism  alone.  It  is  equally  true  of 
Aristotle.  '*  Seinen  Autor  zu  verschweigen,"  says  Diels, "  das  ist  ja 
antike  Sitte."  Herodotus  treats  Hecataeus  in  this  matter  exactly 
as  Aristotle  treats  Herodotus.  The  great  chronographers  built 
further  on  the  system  of  Eratosthenes,  giving  him  either  no 
thanks  at  all,  or  blame.  The  practice  has  been  found  to  be  so 
fixed  and  constant,  that  what  Ernst  Curtius  said  of  Pausanias 
early  in  the  century  must  now  be  exactly  reversed,  till  it  takes  on 
paradoxical  form.  The  fact  that  Pausanias  does  not  mention 
certain  authors,  instead  of  proving  that  he  did  not  use  them, 
proves  rather  that  he  did. 

Having  thus  noticed  the  tendency  in  the  Greek  historians  to 
depreciate  or  conceal  the  literary  sources  to  which  they  are  most 
largely  indebted,  it  is  natural  to  refer  next  to  the  tendency  to 
parade  special  sources,  oral,  written  or  monumental,  which  do 
not  exist,  perhaps  never  did  exist,  or  at  least  did  not  exist  when 
the  author  claims  to  have  used  them.  Fictitious  witnesses  are 
quoted,  impossible  documents  are  cited,  long-vanished  monu" 


Digitized  by 


Go6gIe 


262  AMERICAN  JOURNAL    OF  PHILOLOGY. 

ments  are  restored  by  the  magic  formula  In  Ka&  ^fias — "down  to 
my  day."  Instances  abound  from  Herodotus  to  Plutarch. 
Cephalion  in  Hadrian's  time  cites  a  letter  from  Priam  to  the 
Assyrian  king  Teutamus,  begging  for  succor  now  that  Hector 
was  slain  (Eusebius,  Chron.  p.  41,  apud  Miiller,  F.  H.  G.  HI,  p. 
626).  This  we  may  perhaps  explain  as  a  bold  extension  of  the 
demegoric  principle  which  developed  the  speeches  of  Thucydides 
and  Livy.  But  Dio  Chrysostom,  writing  at  Rome  toward  the  end 
of  the  first  century  (Or.  XI  148  f.),  tells  us  that  he  heard  a  Mede 
say  that  the  Persians  do  not  agree  at  all  with  the  Hellenic  accounts 
of  the  Persian  wars.  The  Persians  say  that  Darius  sent  Datis  and 
Artaphernes  against  Naxos  and  Eretria,  and  they  took  those  cities 
and  retiurned  straightway  to  the  king.  As  they  lay  at  anchor  near 
Euboea,  a  few  ships— not  above  twenty  in  all — straggled  off  toward 
Attica.  Their  crews  had  a  skirmish  with  the  people  who  lived 
on  shore.  Such  was  the  Persians'  Marathon.  After  this,  they 
say,  Xerxes  made  an  expedition  against  Hellas,  conquered  the 
Spartans  at  Thermopylae  and  slew  their  king,  then  took  Athens, 
destroyed  it,  and  sold  its  inhabitants  into  slavery.  Then  he 
imposed  tribute  on  the  Hellenes,  and  returned  to  Asia.  For  the 
Persians,  then,  there  was  no  Salamis  at  all. 

Now  who  was  this  Mede,  the  informant  of  Chrysostom  ?  An 
imaginary  person,  and  his  version  of  the  Persian  wars  is  the 
version  Dio  imagines  that  a  Persian  might  give.  But  here  is  a 
rather  startling  parade  of  oral  source. 

So  when  Plutarch  tries  to  solve  that  great  puzzle  of  Athenian 
wisdom,  from  what  teacher  did  Themistocles  get  his  wonderful 
training?  he  takes  refuge  from  two  dubious  answers  which  he 
finds  proposed — Anaxagoras,  Melissus — with  "those  who  say 
that  Themistocles  was  a  disciple  of  Mnesiphilus,  who  was  a 
sort  of  practical,  natural  statesman,  inheriting  from  Solon  and 
bequeathing  to  the  Sophists"  (Them.  II  4).  Now  who  are  these 
who  say  this  of  Mnesiphilus,  that  malicious  invention  of  Athenian 
local  tradition  to  rob  Themistocles  of  the  glory  of  Salamis,  whom 
Herodotus  all  too  willingly  adopts  ?  None  say  this  save  Plutarch. 
It  is  his  own  delightful  yvii\i.i\  which  he  here  parades  before  us 
under  the  cloak  of  a  vague  "  they  say." 

Akin  to  this  parade  of  fictitious  authorities  is  the  ostentation  of 
entire  originality,  like  that  so  suspiciously  claimed  by  the  Father 
of  History.  "Others  have  written  of  this,"  he  says  (VI  55),  "I 
will  therefore  omit  it.     But  what  others  have  not  touched  upon, 


Digitized  by  VjOOQ IC 


■    ^^^  ,i^  < 


GREEK  HISTORIOGRAPHY.  263 

of  this  I  will  make  mention."  Had  not  others  "touched  upon*' 
Egypt  and  Scythia  and  Libya,  how  would  the  Father's  great 
geographical  excursus  have  looked?  In  just  this  way,  if  we  may 
trust  Clement,  Acusilaus  translated  Hesiod  into  prose,  Kai  va  tdta 
cgiyw^Kev,  "and  published  it  as  his  own"  (Strom.  VI,  p.  629  A). 
Here  belongs  also  in  our  list  of  literary  sins  the  constantly  recur- 
ring claim  of  autopsy  by  the  historian,  where  we  know  that 
autopsy  was  impossible.  Here,  too,  the  claims  of  special  advan- 
tages never  enjoyed,  of  extensive  travels  never  undertaken,  of 
long  periods  of  time  never  spent  upon  the  work.  A  typical 
example  will  suffice.  Diodorus  begins  his  immense  bibliotheca 
with  a  long  elaboration  of  the  theme  of  the  great  lack  of  general 
histories  and  universal  compends,  in  spite  of  their  great  useful- 
ness. He  claims  to  have  spent  thirty  years  of  unremitting  toil 
upon  the  work,  and  to  have  journeyed  over  Asia  and  Europe,  pro- 
tected by  the  aegis  of  the  Roman  Empire,  in  order  to  secure  the 
advantages  of  autopsy.  But  thirty  years  is  a  fabulously  long  time 
for  the  preparation  of  such  a  work  as  that  of  Diodorus  is  seen  to 
be,  there  is  no  trace  of  autopsy  in  the  work,  and  no  claim  to  it 
elsewhere,  except  in  the  borrowed  words  of  his  sources,  and  the 
whole  is  a  mechanical  copy  of  a  famous  passage  of  Polybius 
(III  59).  "Ctesias,"  says  Lucian  (V.  H.  I  3),  "wrote  many 
things  about  India  and  its  peoples  which  he  had  never  seen  him- 
self, nor  heard  another  tell."  But  Ctesias  claimed  both  to  have 
seen  himself  and  to  have  heard  from  others. 

When  quotations  and  citations  are  made  by  the  Greek  histo- 
rians, how  unsatisfactory  to  modern  ethical  standards,  which  are 
perhaps  never  too  high,  the  procedure  is !  The  practice  of  citing  at 
second  or  third  hand  confuses  all  our  science.  Even  the  encyclo- 
paedic Aristotle  takes  quotations  from  Plato,  or  at  least  alludes 
to  them  because  Plato  had.  He  mentions  Herodotus  only  once 
in  the  'KBi\vaUv  iroUrtla,  but  either  uses  him  freely  elsewhere 
without  citing  him,  or  quotes  him  here  only  because  his  imme- 
diate source  so  quotes  him.  Citation  is  made,  as  we  have  seen, 
for  unimportant  details,  or  else  for  controversial  minutiae,  when 
the  whole  context  as  well  is  borrowed.  Specific  authors  are 
hidden  behind  nebulous  plurals  or  gentile  adjectives.  Real 
sources  are  concealed,  and  others  suggested.  Eldest  sources  of 
late  compends  are  cited  from  the  compends,  with  archaistic  effect, 
ignoring  the  compends.  This  may  be  a  valuable  practice  for  us, 
and  the  fruit  of  a  learned  and  scholastic  spirit,  but  it  is  contrary 


Digitized  by  VjOOQ IC 


264  AMERICAN  JOURNAL    OF  PHILOLOGY, 

to  our  ethics.  And  when  at  last  we  come  to  the  compilers,  and 
to  Diodorus  their  chief,  how  the  ethics  and  amenities  stagger 
us !  Large  inheritances  of  questionable  practice  from  original 
sources  are  here  employed  without  attempt  at  concealment. 
''The  only  privilege  of  the  original  man,"  says  Lowell,  "is  that, 
like  other  sovereign  princes,  he  has  the  right  to  call  in  the  current 
coin  and  reissue  it  stamped  with  his  own  image/'  In  this  sense 
Herodotus  is  original.  He  restamps  before  reissuing.  So  do 
Ephorus,  Timaeus,  and  Poly  bins,  to  great  extent.  But  the  com- 
pilers simply  reissue.  They  reissue  with  the  personal  judgments, 
local  colors,  political  combinations,  and  architectural  monuments 
of  bygone  centuries  still  intact  and  unchanged.  And  over  all  the 
gaping  seams  they  try  to  cast  the  mantle  of  some  master's  grand 
manner. 

On  either  side  of  this  long  list  of  what,  from  the  modern  point 
of  view,  seem  to  be  literary  malpractices,  let  us  hang  the  two 
portraits  of  the  true  historian  drawn  by  Poly  bins  and  Lucian. 
The  true  historian,  says  Polybius  (XH  passim)^  will  be  a  man 
of  action,  versed  in  political  and  military  affairs.  He  will  not  con- 
fine himself  to  the  study  of  documents  and  monuments  merely, 
although  he  will  not  neglect  these.  He  will  study  carefully  and 
in  person  the  topography  of  the  actions  he  describes.  He  will 
ask  questions  of  as  many  people  as  possible  who  are  connected 
in  any  way  with  the  events  or  places  which  he  is  describing,  and 
he  will  believe  those  most  worthy  of  credit,  and  show  critical 
sagacity  in  judging  all  their  reports.  He  will  be  a  man  of  dignity 
and  good  sense.  When  he  resolves  to  retaliate  upon  a  personal 
enemy,  he  will  think  first,  not  of  what  that  enemy  deserves,  but 
of  what  it  is  becoming  in  himself  to  do  to  that  enemy ;  what  his 
self-respect  will  allow  him  to  say  of  that  enemy. 

The  true  historian,  says  Lucian  (Quomodo  Hist.  Conscr.  pas- 
sim) ^  will  be  a  man  of  natural  gifts,  with  an  interpretative  power 
cultivated  after  the  best  models.  He  will  have  acquaintance  with 
politics  and  armies.  He  will  be  fxiyvvnyf  rather  than  rroiTnJff,  trans- 
mitter rather  than  producer.  He  will  not  try  to  determine  what 
it  was  that  happened,  but  how  it  happened,  inasmuch  as  it  has 
already  happened  (Wfrpajcrai  yap  ^diy).  He  will  aim  to  be  under- 
stood by  the  many,  and  will  therefore  call  figs  figs  and  spades 
spades^  but  he  will  aim  to  be  praised  by  the  cultured.  The  best 
preparation  for  his  work  will  be  persona]  experience;  but  if  this  is 
impossible,  he  will  follow  the  best  and  most  incorruptible  narra- 


Digitized  by  VjOOQ IC 


GREEK  HISTORIOGRAPHY,  265 

tors.  Above  all  things  he  will  be  iXtvdfpog  in  spirit — a  freeman, 
knowing  no  fear  and  courting  no  favor.  He  will  convert  the  v\rj 
of  historical  material  into  the  fair  shape  of  history,  as  Phidias 
his  gold  and  ivory  and  wood  into  the  Athena  Parthenos. 

Of  the  Polybian  standard  Josephus  may  have  thought  when  he 
said  of  his  own  qualifications  for  writing  a  history  of  the  Jewish 
war,  "  Of  many  things  I  was  myself  the  cause,  of  most  the  eye- 
witness, and  I  am  ignorant  of  nothing  said  or  done."  Before 
Lucian  the  great  work  of  Thucydides  seems  to  have  stood  as 
model.  And  at  the  very  opening  of  scientific  Greek  historio- 
graphy we  have  that  great  master's  severe  conception  of  the  work 
and  calling  of  the  true  historian.  "  If  he  who  desires  to  have 
before  his  eyes  a  true  picture  of  the  events  which  have  happened, 
and  of  the  like  events  which  may  be  expected  to  happen  here- 
after in  the  order  of  human  things,  shall  pronounce  what  I  have 
written  to  be  useful,  then  I  shall  be  satisfied.  My  history  is  an 
everlasting  possession,  not  a  prize  composition  which  is  heard 
and  forgotten"  (I  22).  This  conception  surely  was  in  the  mind 
of  Mommsen  when  he  closed  the  introduction  to  the  fifth  volume 
of  his  great  history  with  the  words :  "  mit  Entsagung  ist  dies 
Buch  geschrieben,  und  mit  Entsagung  mochte  es  gelesen  sein." 

Greek  historiography,  then,  constantly  alleges  against  itself 
falsehood  and  theft  while  professing  truth  and  originality.  It 
uniformly  depreciates  the  predecessor  and  main  literary  source. 
It  labors  to  conceal  or  openly  denies  obligation  to  predecessors 
when  obligation  is  the  strongest.  It  assumes  qualifications, 
advantages,  originality,  sources  and  methods  which  are  fictitious. 
And  yet  early  in  its  development  it  presents,  and  it  constantly 
reverts  to,  the  loftiest  ideals.     How  can  this  anomaly  be  explained  ? 

No  one  can  claim  that  the  sense  of  truth  and  the  instinct  of 
obligation  for  benefits  received  were  as  keen  in  the  best  ancients 
as  they  are  in  the  best  moderns.  With  the  increased  richness  of 
content  in  modern  life  and  thought  has  come  also  increased 
sensitiveness  in  matters  regarding  truth  and  honor.  We  have 
plagiarisms  still :  whole  sermons  in  the  New  York  pulpit,  whole 
books  in  the  New  York  publishers'  lists,  thoughts  and  expressions 
consciously  and  unconsciously  appropriated  everywhere.  But 
the  practice  is  not  tolerated,  much  less  is  it  canonized,  and  the 
practitioner  is  pilloried  when  detected.  We  have  historical 
frauds  still,  as  all  will  admit  who  remember  the  elaborate  paper 
on  "The  Secession  of  Jones  Co.,  Missouri,"  in  the  Magazine  of 


Digitized  by  VjOOQ IC 


266  AMERICAN  JOURNAL    OF  PHILOLOGY, 

American  History  for  October,  1886  (pp.  387-90),  exposed  in  the 
New  York  Nation  six  years  later  (March  24,  1892,  March  31, 
1892).  But  such  frauds  are  not  common,  and  are  certainly  not 
encouraged.  They  do  not  add  to  the  reputation  of  a  historian 
who  perpetrates  them.  But  it  would  be  gross  injustice  to  attribute 
the  superiority  in  the  ethics  and  amenities  of  modern  historiography 
over  the  ancient  Greek'  merely  and  wholly  to  the  higher  ethical 
tone  of  modern  life  and  its  increase  in  justice  and  politeness* 
There  were  certain  special  causes,  no  longer  existing,  which  pro- 
duced in  ancient  Greek  historiography  features  which  seem  to  us 
so  reprehensible. 

In  the  first  place,  Greek  historiography  was  rooted  and  grounded 
in  poetry  and  mythography,  as  Strabo  says  in  a  passage  already 
quoted  (p.  341).  Between  the  epos  and  Thucydides  came 
logography,  which  to  us  seems  hardly  more  than  a  transfer  of 
poetical  to  prose  narration.  To  the  logographers  themselves, 
however,  it  was  evidently  much  more.  It  was  truth  as  opposed 
to  fiction,  fact  rescued  from  overwhelming  fancy.  Indeed,  that 
development  of  epic  poetry  which  we  call  Hesiodic  was  itself  an 
appeal  from  the  exuberant  fancies  of  the  Homeric  poems  to  a 

soberer  truth.      Not  -^M^a  voWa  \iy€iv  €TVfioiaiif  Sfuna,   but   oKrfBta 

Xeyciv  was  its  aim.  This  desire  to  speak  the  truth  must  have  been 
the  leading  motive  in  the  early  logographers,  taking  the  place  to 
a  great  extent  of  the  poet's  desire  to  please,  or,  better,  they 
sought  to  please  by  telling  the  truth  rather  than  by  telling  fanci- 
ful inventions.  Hence  they  could  abandon  the  poetic  form  for 
simple  prose.  Their  reliance  was  upon  other  charms  than  those 
of  rhythm.  But  Dionysius  of  Halicarnassus  tells  us  (de  Thuc. 
jud.  V  4)  that  even  their  simple  prose  had  a  "nameless  grace.'* 
The  stern  appeal  to  pleasing  truth  as  opposed  to  pleasing  fiction, 
made  already  in  the  Hesiodic  poems  as  they  competed  for  favor 
with  the  Homeric,  made  by  Acusilaus  and  Hecataeus  as  they 
attempted  to  improve  upon  both — this  reliance  upon  the  charm 
and  value  of  facts  as  opposed  to  the  charms  of  fancy,  coupled  with 
annalistic  form  and  neglect  of  dramatic  evolutions,  culminates  in 
Thucydides.  His  mechanical  "summer  and  winter"  remind  us 
of  the  Hesiodic  principle  of  composition  by  annexation.  But  the 
"nameless  grace"  of  early  logographic  prose  was  brought  into 
the  service  of  the  Homeric  spirit  by  Herodotus.  Upon  vast  stores 
of  fact  and  fancy  gathered  alike  from  the  speech  and  the  writings 
of  men  he  imposed  the  Homeric  principles  of  artistic  narration. 


Digitized  by  VjOOQ IC 


GREEK  HISTORIOGRAPHY,  267 

The  form  of  narration  became  again  in  him  the  chief  thing,  but 
the  truth — "I  am  obliged,"  he  says  (VII  152),  **to  say  what  is 
said  to  me,  but  I  am  not  obliged  to  believe  it  in  all  cases,  and  I 
wish  this  remark  to  apply  to  my  whole  work." 

Gradually,  then,  and  very  slowly,  Greek  historiography  emerged 
from  poetic,  fictitious  narrative  in  prose  and  verse — from  myth- 
ography.  What  corresponded  to  our  Saxon  Annals  began  with 
the  mythical  foundings  of  cities.  Myth  was  the  soil  of  history, 
and  exhalations  from  this  soil  cover  the  initial  periods  of  history. 
It  was  impossible  for  any  generation  before  Aristotle  to  draw  the 
line  between  myths  and  facts.  It  is  impossible  for  us.  "The 
better  Greek  history  is  known,"  says  Wilamowitz,  "the  later  it 
begins."  Each  generation  accused  its  predecessor  of  credulity 
and  blindness,  each  claimed  to  know  and  speak  the  truth.  Heca- 
taeus  was  a  radical  sceptic.  Diels  calls  him  the  "enlightener  of 
the  sixth  century."  But  to  Herodotus,  Hecataeus  is  a  laughable 
old  pedant,  and  to  Thucydides  merely  a  credulous  logograph. 
To  him  as  well  as  to  Herodotus  and  the  whole  class  he  applies 
the  same  scornful  characterization — omission  of  research,  accept- 
ance of  what  comes  ready  to  hand,  admixture  of  the  fabulous, 
aim  at  effect  first  and  truth  afterward.  Scepticism  and  enlighten- 
ment thus  go  hand  in  hand  with  the  severest  depreciation  of  the 
predecessor,  till  such  depreciation  becomes  norm  and  sign  of 
progress.  Two  great  quests  also  become  dominant  in  historio- 
graphy, the  quest  of  form,  with  Homeric  traditions,  and  the  quest 
of  truth,  with  Hesiodic  traditions.  Each  quest  becomes  dominant 
and  classic  in  the  work  of  a  master,  the  form-quest  in  Herodotus, 
the  truth-quest  in  Thucydides.  In  neither  is  either  quest  entirely 
exclusive  of  the  other,  though  in  each  the  one  quest  is  supreme. 

The  gradual  emergence  of  Greek  historiography  from  myth- 
ography  in  verse  and  prose,  together  with  the  lack  of  the  finer 
sentiments  of  obligation  to  a  predecessor  whose  standpoint  may 
have  been  outgrown,  will  account  for  much  of  the  falsehood 
charged  upon  Greek  historians  down  to  Herodotus  and  Thucy- 
dides. The  charges  of  plagiarism  in  the  same  period  may  be 
accounted  for  in  the  same  way.  For,  through  most  of  the  period, 
it  is  not  new  facts  or  fancies  which  are  stated  by  successive  his- 
torians, rather  the  same  facts  or  fancies  are  stated  in  a  different 
way.  Whether  the  daughters  of  Proetus  were  afHicted  with 
madness  because  they  refused  to  honor  Dionysus,  as  Hesiod 
says,  or  because  they  offended  Hera,  as  Acusilaus  says,  is  imma- 


Digitized  by  VjOOQ IC 


268  AMERICAN  JO URNAL    OF  PHIL OLOGY. 

terial.  They  were  punished,  if  at  all,  for  impiety.  But  just  such 
items  of  mythography  led  to  charges  both  of  mendacity  and 
plagiarism. 

A  second  special  reason  for  many  of  the  peculiar  phenomena 
we  have  noted  in  Greek  historiography  is  the  fact  that  earliest 
Greek  tradition  was  almost  wholly  oral,  in  the  time  of  Herodotus 
and  Thucydides  was  both  oral  and  written,  and  never  became  so 
'  capable  of  literary  and  documentary  control  as  in  modern  times. 
Josephus  contrasts  scornfully  the  age  of  the  Hebrew  sacred 
records  with  the  comparative  youth  and  dearth  of  Greek  records. 
Thucydides  and  Polybius  lay  far  more  stress  on  oral  testimony, 
especially  the  oral  testimony  of  eye-witnesses,  than  the  modern 
historian  will  consent  to  do.  For  the  Homeric  and  Hesiodic 
poets,  tradition  was  originally  wholly  oral,  and  the  persistence 
of  oral  renditions  and  oral  publication  of  literature  even  long 
after  the  introduction  of  writing,  together  with  the  late  and  slow 
growth  of  reading,  a  reading  class,  and  libraries,  kept  the  air  and 
mannerisms  of  oral  tradition  natural  and  even  necessary  in  an 
attractive  literary  style.  The  great  dramatic  literature  was  long 
meant  for  public  oral  rendition  rather  than  for  reading.  At  some 
early  period  of  necessarily  oral  tradition,  the  first  reducer  to 
writing  might  have  expressed  himself  toward  his  various  and 
indefinite  oral  sources  much  as  Hecataeus  in  his  genealogies 
expresses  himself  toward  Acusilaus  and  Hesiod.  Demosthenes 
and  the  orators  luar  the  history  they  quote.  Aelian,  in  the  Varia 
Historia  (XII  43)  hears  that  Darius  the  son  of  Hystaspes  was  a 
quiver-bearer  of  Cyrus.  The  remove  from  this  literary  mannerism, 
this  reminiscence  of  oral  tradition,  to  Dio  Chrysostom's  Median 
informant,  is  not  a  violent  one.  The  freedom  of  transmission 
which  characterized  oral  tradition  imparted  itself  to  literary 
tradition.  Plato's  citations  of  Homer  often  have  the  ^\x  oivw& 
voce  reminiscence.  Even  Polybius  recasts  the  spirited  reply  of 
the  Greeks  to  Gelo  (Hdt.  VII  161  =  Polyb.  XII  26).  They  do 
not  say  as  Herodotus  makes  them  say — *' Hellas  sent  us  to  thee 
for  troops,  not  for  a  leader,"  but  rather — *'  Come  to  our  aid  and 
the  logic  of  events  will  give  the  command  to  the  bravest." 

We  find  proof  of  the  awakening  of  a  sense  of  accountability  for 
romantic  oral  testimony  as  early  as  the  12th  book  of  the  Odyssey 
(389-90).  The  conversation  between  Helius  and  Zeus  about  the 
slain  cattle  of  the  sun-god  Odysseus  claims  to  have  heard  from 
Calypso,  who  heard  it  frOm  Hermes,  who  was  presumably  present 


Digitized  by  VjOOQ IC 


GREEK  HISTORIOGRAPHY.  269 

in  person.  And  far  down  into  the  age  of  literary  tradition  we  can 
trace  the  feeling  that  oral  tradition  \vas  really  superior.  The 
ideal  historian  of  Polybius  will  ask  questions  of  as  many  people 
as  possible,  will  believe  those  worthy  of  credit,  and  will  show 
critical  sagacity  in  judging  their  reports.  A  phraseology  of 
original  oral  transmission,  kept  natural  and  effective  by  oral 
renditions  and  reminiscences  even  in  the  literary  period,  explains 
much  apparent  falsity  of  claim.  Continuity  of  method  and 
apparatus  is  eagerly  sought  by  the  historian  under  the  impulse  of 
a  keen  love  of  form.  Old  formulae  are  superimposed  upon  new 
material,  until  we  can  have  a  Pausanias  collecting  oral  and  ocular 
evidence  like  a  Polybius,  amassing  literary  evidence  like  a 
Timaeus,  and  clothing  the  combination  with  Herodotean  manner- 
isms. 

Thirdly,  the  agonistic  spirit  in  Greek  literary  production 
accounts  for  many  of  the  amenities  which  we  find  so  questionable. 
Greek  life  had  no  word  of  sympathy  for  the  vanquished.  Suc- 
cess was  not  so  much  the  success  of  positive  achievement  as  the 
glory  of  comparative  personal  triumph  over  rivals.  The  court 
bard  of  Odysseus  (a  351)  found  that  his  audience  liked  the 
newest  things  best,  and  the  standing  of  rival  bards  must  have 
depended  largely  on  their  ability  to  present  novelties  in  matter 
or  form.  And  must  not  Homer  and  Hesiod,  Pindar  and  Corinna, 
vie  with  one  another  in  prize  contests  ?  Did  not  Aeschylus  and 
Sophocles,  Sophocles  and  Euripides,  Cratinus  and  Aristophanes 
contend  with  each  other  for  a  prize?  So  the  Xoyon-oior  competed 
with  the  tnonoMs,  truth-quest  with  form-quest,  Thucydides  with 
Ephorus,  for  popular  acceptance.  As  enlightenment  spread  and 
knowledge  increased,  they  led  to  contempt  of  previous  stadia,  as 
the  victor  and  his  admirers  despised  the  vanquished.  ''All  the 
Greek  historians,"  says  Josephus,  "seek  not  truth,  but  Xoyw  dvvafiiv 
-—powerful  expression^  and  each  writer  applies  himself  to  eclipse  in 
fame  his  predecessor."  "  In  rhetoric,"  he  says,  "all  must  yield  to 
the  Greeks,  but  not  in  truth  about  antiquity."  "The  reason  for 
their  ignorance  is  lack  of  travel  and  intercourse.  The  reason  for 
their  lying  is  their  desire  to  get  the  reputation  of  telling  more 
history  than  their  rivals."  "The  abuse  of  their  predecessors  is 
due  to  the  repute  of  their  victims,  to  jealousy  of  that  repute,  and 
desire  to  eclipse  it"  (contra  Apion.  passitn).  The  scorn  which 
Thucydides  feels  for  Herodotus  was  due  not  only  to  a  natural 
dislike  of  the  broad  and  flowing  Ionian  narrative,  but  also  to  the 


Digitized  by  VjOOQ IC 


270  AMERICAN  JOURNAL    OF  PHILOLOGY. 

influence  of  a  literary  tradition  of  long  standing.  If  Thucydides 
yielded  to  this  influence,  much  more  all  following  historians,  who, 
able  to  correct  details,  proclaimed  their  superiority  to  their  great 
source.  And  so  Herodotus  is  criticised  by  everybody  because 
used  by  everybody,  and  the  degree  to  which  he  is  criticised  is  a 
fairly  safe  criterion  of  the  extent  to  which  he  has  been  used. 
The  sense  of  obligation  toward  even  the  imperfect  achievements 
of  the  past  is  one  of  the  choicest  flowers  of  modern  life. 

Fourthly,  with  Herodotus  and  Thucydides,  and  for  all  histo- 
rians after  them,  whether  given  chiefly  to  the  quest  of  form  or  the 
quest  of  truth,  certain  literary  conventions  were  established. 
Dramatic  conventions  must  be  understood  before  we  can  mount 
a  play.  They  become  bolder  and  more  startling  as  we  go  back  in 
the  history  of  the  drama.  "  What's  bad  unless  it  seem  so  to  the 
spectator  ?  "  asks  Dionysus  in  the  Frogs,  In  dramatics  certainly, 
if  not  in  ethics,  the  pleasure  of  the  spectator  is  the  commonest 
standard  of  right  and  wrong.  So  in  all  literary  compositions, 
but  particularly  in  the  historical,  the  pleasure  of  the  hearer  or 
reader  is  to  a  great  degree  the  writer's  standard  of  right  and 
wrong.  This  attitude  of  superiority  over  his  predecessor  on  the 
part  of  each  succeeding  ancient  Greek  historian;  this  correction 
of  error  under  the  form  of  accusation  of  falsehood,  while  merely 
recasting  a  material  already  furnished,  became  literary  conven- 
tions. In  the  presentation  of  new  historical  material  also,  the 
accumulations  of  the  advancing  centuries,  other  great  literary 
conventions  were  tacitly  assumed.  They  are  all  more  or  less 
distinctly  traceable  to  the  domination  of  the  old  form-quest.  For 
the  Greek  after  all  was  more  concerned  for  the  manner  than  the 
matter.  Herodotus,  according  to  Dionysius  of  Halicarnassus 
(Epist  ad  Cn.  Pomp.  Ill  7),  did  not  avoid  the  same  material  as 
Charon  and  Hellanicus,  but  he  believed  he  could  produce  some- 
thing better  than  they  had,  and  he  did  so.  Ephorus  and  Theo- 
pompus  recast  the  material  of  Herodotus  and  Thucydides,  not  in 
the  interest  of  truth  but  of  form.  Like  Curtius  Rufus,  they  used 
their  historical  basis  as  a  corpus  vile  on  which  to  practice  their 
rhetorical  art.  They  could  say  the  same  things  in  a  better  way. 
And  since  the  Greek  public  cared  not  so  much  for  correct  report 
as  for  free  and  effective  description  of  the  writer's  personal  impres- 
sions, the  rhetorical  age  impressed  upon  their  work  all  the  con- 
ceits of  reflned  rhetorical  craft.  So  epic  and  heroic  influences 
had  triumphed  in  Herodotus,  political  and  diplomatic  oratory  in 


Digitized  by  VjOOQ IC 


GREEK  HISTORIOGRAPHY.  2/1 

Thucydides,  as  later  the  Alexandrian  imperial  sentiment  dominated 
in  Callisthenes,  the  Roman  imperial  sentiment  in  Polybius — all  at 
the  sacrifice  of  the  strictest  truth.  The  testimony  of  hearsay  was 
less  sure  and  true  than  that  of  an  eye-witness,  but  often  pleasanter, 
according  to  Polybius  (XII  27).  And  just  as  Timaeus  chose  the 
pleasanter  testimony,  so  the  successive  generations  of  Greek  his- 
torians strove  to  please.  The  great  protests  of  the  Hesiodic  poems 
and  of  Thucydides  did  not  prevail.  The  form-quest  as  exemplified 
in  Herodotus  carried  the  day,  till  even  truth-quest  became  largely 
matter  of  form,  always  excepting  Polybius,  and  the  severer  truth- 
seekers,  on  whom  Herodotus  and  Livy  both  relied,  were  relegated 
to  oblivion  by  the  very  triumphs  of  these  great  artists. 

Other  literary  conventions  might  be  enumerated  which  grew 
up  in  consequence  of  the  mythical  beginnings  of  Greek  history, 
its  originally  oral  tradition,  its  agonistic  spirit,  and  its  quest,  in 
the  main,  of  form  rather  than  of  truth. 

Foremost  of  all,  and  most  prolific  in  minor  devices  of  a  similar 
nature,  is  the  literary  embellishment  of  the  set  speech — demegoria. 
It  is  already  a  great  feature  of  epic  poetry,  we  find  traces  of  it 
in  the  fragments  of  the  logographers,  it  is  dramatically  developed 
by  Herodotus,  adapted  to  the  political  spirit  of  his  time  by  Thucy- 
dides, and  it  becomes  a  historical  mannerism  ever  after,  reflecting 
the  dominant  influences  of  each  succeeding  age.  But  it  tends  even 
in  its  simplest  form  to  distort  history,  as  most  reflection  distorts, 
and  it  is  a  breeder  of  inferior  devices  subversive  of  the  truth  of 
tradition,  such  as  the  citation  of  impossible  letters  and  documents. 
No  sooner  therefore  do  genuine  historical  documents  begin  to 
abound  in  the  period  of  Alexander,  than  spurious  documents  yet 
more  abound.  The  modern  feeling  toward  this  literary  device  of 
the  set  speech  is  well  shown  in  a  citation  from  Voltaire's  preface 
to  his  History  of  Russia,  made  by  Professor  Jebb.  "  If  one  put 
into  the  mouth  of  a  prince  a  speech  which  he  had  never  made, 
the  historian  would  be  regarded  as  a  rhetorician.  Set  speeches 
are  a  sort  0/  oratorical  lie  ** 

"A  lady  once  told  me,"  wrote  the  late  Master  of  Balliol,  "that 
Lord  Weslbury  was  an  esprit  faux^  but  I  do  not  think  that  this 
was  true,  although,  like  Plato,  he  could  invent  Egyptians,  or  any- 
thing else."  The  great  convention  of  the  set  speech,  itself 
designed  to  promote  the  sense  of  verisimilitude,  led,  as  has  been 
said,  to  numberless  other  devices  of  a  similar  nature.  But  the 
writers  who  used  them  were  not  necessarily  esprits faux.    They 


Digitized  by  VjOOQ IC 


272  AMERICAN  JOURNAL    OF  PHILOLOGY. 

are  not  deceiving,  are  not  trying  to  deceive  their  readers.  And 
it  is  only  fair  to  say  that  charges  of  deceit  on  these  scores  come 
mainly  from  modern  critics,  for  whom  such  literary  conventions 
no  longer  exist.  The  professions  of  superior  truthfulness,  or 
superior  advantages ;  claims  of  autopsy,  travel,  study,  discovery  ; 
the  use  of  fixed  schemes  for  descriptions  of  battles,  scenery,  or 
speeches ;  the  adaptations  of  great  models  of  historical  achieve- 
ment to  new  surroundings — all  these  become  conventional. 
"Former  historians,"  says  Polybius  (XII  25),  "showed  their 
sense  of  the  necessity  of  making  professions  of  personal  expe- 
rience in  matters  about  which  they  wrote.  When  the  subject 
was  political  they  were  careful  to  state  that  the  writer  had  of 
course  been  engaged  in  politics,  and  so  had  had  experience  in 
matters  of  this  sort."  A  historian  who,  as  Lucian  knew,  had 
never  set  foot  out  of  Corinth,  began  his  history  of  the  Parthian 
wars  with  the  sententious  phrase  which  had  already  done  duty 
for  Heracleitus  and  Herodotus — "  Eyes  are  more  trusty  than  ears  ; 
I  shall  write  therefore  only  what  I  have  seen,  not  what  I  have 
heard"  (Quomodo  Hist.  Conscr.  38).  And  Lucian  ridiculed 
merely  for  barrenness  of  invention  another  historian  who  borrowed 
Thucydides*  description  of  the  plague  at  Athens  in  order  to  tell 
about  the  disease  which  afflicted  Nisibis  in  Mesopotamia  for 
closing  its  gates  to  the  Roman  armies.  The  historian  omitted 
the  Pelargikofif  and  the  Long  Walls  of  Thucydides,  but  his  plague 
started  in  Aethiopia,  then  descended  upon  Egypt  and  Persia, 
exacdy  as  Thucydides  had  made  his  plague  behave,  and  Lucian 
left  this  poor  historian  "  burying  Athenians  in  Nisibis,"  and  knew 
by  heart  all  the  rest  of  his  description. 

The  step  from  the  invention  of  ornamental  details  to  the  inven- 
tion of  vital  facts  is  a  natural  one — in  the  course  of  human 
nature  an  inevitable  one.  The  oldest  of  the  younger  school  of 
Roman  annalists,  we  are  told  (Wachsmuth,  Alte  Geschichte,  pp. 
622  ff),  was  true  to  matters  of  fact,  did  not  invent  his  facts  out- 
right, but  did  invent  speeches,  documents  and  letters,  not  to 
deceive,  but  to  enliven.  Later  annalists,  however,  invented  their 
facts  outright,  invented  statistics  corroborative  of  their  invented 
facts,  and  reported  the  actual  words  of  legendary  characters. 
They  deceived  no  one,  it  is  true,  u  e.  no  one  for  whom  they  wrote. 
But  it  is  hard  for  us  moderns  to  differentiate  them  from  Cato's 
Ligurians,  who  were  "uneducated  liars,  with  no  power  of  recol- 
lecting facts."     Duris  of  Samos,  whom  Plutarch  never  tires  of 


Digitized  by  VjOOQ IC 


GREEK  HISTORIOGRAPHY.  273 

discrediting,  criticises  his  predecessors,  Ephorus  and  Theopompus, 
in  words  which  seem  to  our  modern  ethics  rather  praise  than 
blame.  They  fell  far  behind  their  predecessors,  he  says,  **for 
they  did  not  cultivate  the  art  of  dramatic  imitation  and  of  pleasing 
invention,  but  devoted  themselves  to  the  mere  recital  of  what  had 
happened"  (Photius,  176,  as  cited  and  emended  in  Schaefer's 
Quellenkunde,  I,  p.  83). 

Another,  and  the  last  literary  convention  to  be  here  mentioned, 
is  the  borrowing  of  famous  mots  or  phrases,  and  weaving  them 
into  new  narrative,  mutatis  mutandis^  for  fresh  conquests.  The 
practice  is  as  early  certainly  as  Herodotus.  Some  maladjustment 
of  context,  or  lack  of  perfect  application  to  the  new  situation, 
reveals  the  device.  How  ill  the  exquisite  Periclean  funeral  meta- 
phor of  war's  robbery  of  the  spring  from  the  year  is  put  by  Hero- 
dotus in  the  mouth  of  Gelon  of  Syracuse  when  he  declines  to 
reinforce  the  allied  Greeks  (VH  162)!  How  imperfectly  the 
capital  parable  of  the  assorted  evils  of  mankind  is  applied  by  the 
same  historian  to  the  malevolent  charges  of  Argive  Medism  (VH 
152)!  One  is  even  tempted  to  believe  that  his  famous  story  of 
the  wife  of  Intaphernes  and  her  choice  is  skilfully  adapted  to 
Persian  context  from  the  Antigone  of  Sophocles,  instead  of  vice 
versd  as  is  generally  held.  But  the  audience  were  well  aware  of 
the  real  ownership  in  all  these  cases.  There  was  no  deception, 
rather  a  confident  and  flattering  appeal  to  the  literary  culture  of 
the  hearer,  such  as  those  which  so  amaze  us  in  Aristophanes. 
Polybius  finds  this  same  literary  device  in  Timaeus  (XII  26). 
Timaeus  puts  into  a  great  speech  of  his  Hermocrates  such  well 
known  literary  gems  as — "  In  peace  sleepers  are  waked  by  cocks, 
in  war  by  trumpets" — "In  peace  the  old  are  buried  by  the 
young,  as  nature  directs,  in  war  the  young  are  buried  by  the  old." 
Such  arguments,  Polybius  says,  would  have  been  employed  by  a 
youth  who  had  devoted  himself  to  scholastic  exercises  and  studies 
in  history.  Precisely  so.  They  are  borrowed  literary  finery. 
But  every  one  knew  their  owner,  even  in  an  age  when  the  sense  of 
literary  ownership  was  not  keen.  And  we  should  remember 
when  we  consider  the  wholesale  charges  of  plagiarism  among  the 
ancients,  that  literary  productions  are  the  last  form  of  property 
for  which  the  modern  world  is  devising  adequate  safeguards. 

A  word  should  be  said  of  the  great  safety-valve  for  modern 
historiography  afforded  by  the  distinctive  literary  form  of 
romance.     Romance  existed  in  Greek  poetry  and  historiography 


Digitized  by  VjOOQ IC 


274  AMERICAN  JOURNAL    OF  PHILOLOGY. 

long  before  it  became  a  distinct  species  of  composition.  Even 
after  mythography  came  to  be  dearly  distinguished  from  history, 
the  romantic  tendencies  which  had  produced  an  epos  and 
a  Sage  remained  imbedded  in  historical  composition.  Both 
romantic  history  and  historical  romance  were  classed  as  history. 
Imagine  the  safety-valve  of  modem  fiction  closed.  What  would 
become  of  our  historiography  ? 

Another  word  should  be  said  of  the  immense  relief  to  modern 
historiography  afforded  by  the  footnote.  The  ancient  artist 
knew  not  this  form-conserving  device,  and  yet  most  ancient 
artists  had  their  hearts  set  mainly  on  form.  Imagine  a  Busolt's 
History  of  Greece  with  footnotes  artistically  blended  with  the 
main  text ! 

The  oppressiveness  of  accumulated  historical  material  has  been 
felt  by  the  world  many  times  before  this  age  of  compends  and 
encyclopedias  and  handbooks  and  "Libraries  of  the  World's 
Best  Literature."  The  great  Alexandrian  scholars  felt  it.  Marius 
the  Epicurean  felt  it  under  Marcus  Aurelius.  The  great  Byzan- 
tine compilers  felt  it  in  the  tenth  century.  Periodical  attempts 
will  always  be  made  to  strip  off  from  historical  tradition  the  accre- 
tions due  to  fancy  and  the  desire  for  pleasant  form.  But  each  age 
must  do  its  own  work  here.  It  will  not  be  satisfied  with  the  work 
of  any  previous  age.  Even  the  ultimate  facts  of  history  will  be 
constantly  restated.  If  this  were  not  so,  where  were  the  charm 
of  historical  research  ?  What  would  become  of  source-criticism  ? 
Toward  historical  truth,  as  toward  the  higher  truth  which  Lessing 
so  bravely  sought,  we  may  feel  as  he  felt — "  If  God  held  all  truth 
shut  in  his  right  hand,  and  in  his  left  nothing  but  the  ever  restless 
instinct  for  truth,  though  with  the  condition  of  for  ever  and  for 
ever  erring,  and  should  say  to  me,  *  Choose!'  I  should  bow 
humbly  to  his  left  hand  and  say,  'Father,  give!  Pure  truth  is  for 
Thee  alone/  *' 

B.  Perrin. 


Digitized  by  VjOOQ IC 


II.— THE  SUBJUNCTIVE  IN  INDEPENDENT  SEN- 
TENCES IN  PLAUTUS. 

II.— Forces. 

The  presentation  of  the  facts  of  usage  in  the  preceding  part  of 
this  paper  should  make  upon  any  reader  who  may  have  patience 
to  follow  it  the  same  impression  that  is  made  by  the  facts  them- 
selves, the  impression  that,  in  the  last  analysis,  the  subjunctive  use 
consists  of  a  great  variety  of  more  or  less  specialized  usages,  differ- 
ing often  but  slightly,  yet  distinct  and  deserving  separate  treatment. 
Some  are  narrow  in  range  and  clearly  defined,  like  hand  dicam 
dolo,  egone  austm,  ueniat  uelim ;  others  are  broader  and  of  more 
general  application,  like  the  use  of  the  2d  singular  of  the  present, 
and  have  scarcely  crystallized  into  definite  meaning.  These 
varieties  of  usage  are,  it  is  true,  connected  not  only  by  the  unity 
of  the  mode,  but  also  by  resemblances  which  justify  somewhat 
more  specific  terms:  jussive,  hortatory,  deliberative.  But  it 
must,  I  think,  be  acknowledged  that  the  study  of  resemblances, 
the  grouping  and  re-grouping  of  usages  which  are  for  the  most 
part  ill-defined,  has  not  greatly  advanced  our  knowledge  of 
syntax  during  the  last  quarter  of  a  century.  The  search  for  a 
GrundbegriffhAS  not  been  fruitless  in  suggestion,  but  it  has  been 
less  productive  of  real  advance  than  might  have  been  hoped. 
What  we  call  *the  subjunctive'  is  not  an  entity;  it  has  no  objec- 
tive existence.  Only  the  individual  subjunctive  forms — sim^ 
dicam^^existt  and  the  term  *the  subjunctive'  is  the  result  of  a 
generalization  by  which  we  endeavor  to  include  in  a  single  idea 
all  that  is  common  to  the  many  individual  forms  or  the  narrow 
and  clearly  defined  usages.  No  single  form  or  usage  exactly 
corresponds  to  the  type ;  each  falls  short  of  the  type  in  some 
particulars  and  in  other  particulars  each  may  have  acquired 
suggestions  of  meaning  which  are  not  found  in  the  type.  By  a 
process  analogous  to  natural  selection,  a  process  pardy  of  exclu- 
sion, partly  of  acquisition,  indiscriminate  and  therefore  shifting 
applications  of  the  modal  forms  have  crystallized  into  definite 
and  therefore    expressive   usage.     The  understanding  of  this 


Digitized  by  VjOOQ IC 


276  AMERICAN  JOURNAL    OF  PHILOLOGY. 

process— and  this  is  the  true  work  of  the  student  of  historical 
syntax — must  begin  with  detailed  and  precise  description  and 
definition  of  usage,  such  as  I  have  attempted  to  give  above ;  the 
second  step  is  the  discussion  of  the  forces  which  have  been  at 
work  to  produce  such  usages. 

These  forces  are  of  two  kinds  :—yfrj/,  each  subjunctive  has 
person,  number,  tense,  voice,  and  the  stem  of  the  verb  has  its 
own  proper  meaning.  Secandy  each  subjunctive  stands  in  a 
particular  setting,  in  a  sentence  of  a  particular  kind,  interrogative 
or  not,  with  particles  or  adverbs  which  limit  it  and  at  the  same 
time  suggest  additional  meanings,  often  in  a  paratactic  relation 
to  another  verb,  sometimes  with  a  negative,  preceded  by  other 
sentences  and  conditioned  by  all  the  preceding  course  of  thought, 
and,  in  the  spoken  language,  made  definite  by  the  circumstances, 
by  inflection  and  by  gesture.  Of  these  two  kinds  of  forces,  the 
first  are  of  course  inseparable  from  the  form ;  the  influence  of 
person  and  number  upon  the  modal  meaning,  whatever  it  may 
be,  must  always  have  accompanied  and  modified  the  subjunctive. 
In  part  this  is  true  also  of  the  forces  of  the  second  kind,  but  the 
additions  to  the  subjunctive,  especially  the  paratactic  verbs,  are 
later  and  are  at  the  same  time  definitions  of  the  subjunctive  and 
evidence  of  the  need  of  definition. 

In  the  study  of  these  forces  I  have  two  objects  in  view.  Firsts 
a  conscious  and  deliberate  weighing  of  the  various  elements 
which  make  up  what  we  call  the  *  context'  ought  to  give  greater 
precision  to  our  interpretation.  Second^  in  so  far  as  it  can  be 
shown  that  the  usages  described  above  are  the  result  of  known 
forces,  working  along  traceable  lines,  the  necessity  and,  indeed, 
the  scientific  propriety  of  referring  such  usages  to  supposed  I.E. 
functions  falls  away.  If,  for  example,  any  other  forces  can  be 
found  which  of  themselves  are  sufficient  to  limit  ^///^/^r^/ufn/ to 
a  curse  or  saluos  sis  to  a  greeting,  then  it  is  not  permissible  to 
explain  them  as  survivals  of  an  I.E.  optative  function  or  to 
attempt  to  show  how  the  optative  function  extended  to  the 
subjunctive  forms  di  ie  perdant  and  ualeas. 

The  discussion  of  these  forces  is  not  in  any  case  complete,  but 
I  have  hoped  that  it  might  be  suggestive  of  a  point  of  view  some- 
what different  from  the  usual  one.  Only  the  more  salient  points 
are  brought  out  under  each  head,  and  no  attempt  has  been  made 
to  trace  the  influence  of  each  force  upon  every  kind  of  subjunctive 
usage.    Thus,  under  person  and  number  the  hypothetical  cases 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  2/7 

and  the  questions  are  not  touched,  because  they  are  chiefly 
influenced  by  other  forces. 

I.  Person  and  Number. 

Four  acts  are  connected  in  thought  with  every  subjunctive: 
speaking,  willing,  hearing  and  performing  the  action  expressed 
by  the  verb.  These  four  acts  may  be  done  by  one,  two,  three  or, 
rarely,  by  four  persons. 

In  the  ist  person  singular  ihese  acts  may  be  performed  by  one 
person ;  the  speaker  may  express  his  own  will  to  himself  about 
an  act  which  he  is  to  do.  These  cases,  however,  are  rare,  partly 
because  the  occasion  is  rare,  partly  because  other  expressions 
like  the  periphrastic  are  more  precise.  Where  such  forms  do 
occur,  it  is  evident  that  the  function  of  the  mode  is  much 
restricted.  The  will  of  the  speaker  cannot  take  the  form  of 
command  or  entreaty;  it  is  limited  to  ideas  of  propriety  or 
necessity  or  determination.  And  even  for  these  ideas  the  bare 
subjunctive  form  is  an  inadequate  expression,  needing  to  be 
supplemented  by  a  paratactic  addition  like  opiumumsty  necessest. 
In  general,  either  the  speaker  divides  himself  and  by  a  Action,  a 
dramatic  doubling  of  his  personality,  addresses  himself  in  the 
forms  of  the  2d  person,  which  can  express  command  or  entreaty, 
or  else  the  sentence  is  addressed  to  another  person.  In  the  latter 
case  the  verb  is  not  really  an  expression  of  will,  but  a  statement 
to  the  other  person  of  the  speaker's  determination — that  is,  it  is  a 
future.  This  is,  of  course,  a  very  common  situation,  and  most  of 
the  3d-conjugation  forms  in  Plautus  show  by  their  context  that 
they  are  of  this  kind.  The  line  of  division,  however,  is  faint. 
Asin.  605,  sermoni  iam  finem  face  tuo :  huius  sermonem  accipiam, 
is  like  Trin.  11 36,  quid  ego  cesso  hos  conloqui?  sed  maneam 
etiam,  opinor,  in  expressing  desire  rather  than  determination,  but 
it  shades  toward  the  future  because  it  is  half  addressed  to  the 
other  person.  So  also  in  Asin.  719  ecastor  ambae  sunt  bonae.  || 
sciam,  ubi  boni  quid  dederint. 

In  the  large  majority  of  cases  in  ist  sing,  the  wilier  is  the 
person  addressed,  the  speaker  and  actor  being  one  person. 
These  cases  are  in  questions  and  will  be  treated  below. 

The  very  fact,  then,  that  a  verb  is  used  in  ist  sing,  of  the 
subjunctive  restricts  the  possible  meaning  of  the  mode  within 
narrow  limits  and  tends  to  produce  a  well-marked  though  infre- 
quent usage.    So  strong,  in   fact,  is  the  force  of  person  and 


Digitized  by  VjOOQ IC 


278  AMERICAN  JOURNAL    OF  PHILOLOGY. 

number  in  this  case  that  it,  as  it  were,  compels  the  subjunctive  to 
express  ideas  of  necessity  and  propriety  which  are  more  naturally 
expressed  by  other  forms. 

In  the  2d  pers.  sing,  the  speaker  and  wilier  is  one  person,  the 
hearer  and  actor  another  person,  who  is  present.  The  relation  of 
wilier  and  actor  is  the  most  direct  and  simple  possible.  It  needs 
no  definition ;  its  definiteness  even  helps  to  define  the  nature  of 
the  will.  For  the  kind  of  will  exerted,  though  it  is  not  defined 
by  the  verb,  is  of  necessity  fixed  by  the  relation  of  the  persons — 
father  to  son,  master  to  slave,  friend  to  friend — and  by  the  circum- 
stances, which  are,  so  to  speak,  visible  to  both  and  unconsciously 
taken  for  granted  in  the  selection  of  expressions.  Definiteness  of 
language  is  unnecessary  where  the  hearer  must  understand  suffi- 
ciently the  sense  in  which  the  speaker,  standing  before  him,  wills 
the  action.  On  the  one  hand,  therefore,  the  use  in  2d  sing,  is 
limited  to  the  more  direct  expressions  of  will,  and  on  the  other 
hand  there  is  no  need  of  distinct  forms  to  differentiate  advice 
from  command  or  entreaty  from  demand,  because  this  distinction 
is  implied  in  the  circumstances.  The  wish  without  uiinam  is 
almost  unknown ;  the  two  cases,  in  the  marriage  song  in  Cas.  and 
in  the  curse  in  Trin.  351  (if  malum  be  read,  with  GSO>  are  only 
half-wishes.^ 

In  the  3d  sing,  three  persons  are  involved,  the  speaker  and 
wilier,  the  bearer,  and  the  actor,  not  present  or  treated  as  not 
present.  The  bare  subjunctive  (without  paratactic  additions)  is 
therefore  indefinite  in  two  respects.  First,  in  all  cases  where  the 
subject  of  the  verb  is  not  a  definite  person,  and  occasionally  even 
when  the  actor  is  definite,  there  is  nothing  in  the  relation  of  the 
persons  to  define  the  mode,  either  by  the  exclusion  of  many 
kinds  of  will,  as  in  ist  sing.,  or  by  a  direct  and  evident  relation 
of  wilier  to  actor,  as  in  2d  singular.  Second,  the  separation  of 
hearer  from  actor  leaves  the  part  which  the  former  is  to  play 
undefined.  For  the  details  I  will  refer  to  what  has  been  said 
above  in  describing  the  uses  of  the  3d  sing,  present.     The  vary- 

^  It  may  be  noted  in  passing  that  the  gap  in  meaning  between  ist  sing.  fut. 
and  1st  sing.  pres.  subj.  is  much  less  than  in  2d  sing.  In  1st  sing,  determin- 
ation runs  into  futurity  (English  wili  into  shall\  and  many  futures  like  ibo 
express 'determination.  But  in  the  2d  pers.  the  subjunctive  expresses  clear 
forms  of  will,  while  the  future  expresses  futurity  with  little  trace  of  will. 
Therefore,  while  dicam  serves  for  both  modes,  the  2d  person  selects  the  two 
differing  forms  dUas  and  dices. 


Digitized  by  VjOOQIC^ 


THE  SUBJUNCTIVE  IN  PLAUTUS,  279 

ing  relations  of  the  hearer  to  the  action  are  of  course  not  expressed 
in  the  mode;  they  are  indeed  not  expressed  at  all  in  language 
except  in  the  paratactic  uses,  but  the  fact  that  they  exist  and  are 
in  the  mind  of  both  speaker  and  hearer  tends,  by  a  common  law 
of  language,  to  associate  them  with  the  form.  In  other  words, 
the  3d  sing,  vaguely  suggests  to  the  hearer  that  he  also  is  to  act. 
\{  ferat  alone  is  used  in  the  sense  oiiubeferat^  it  comes  to  mean 
iubeferai.  And  this  comes  out  more  plainly  when  the  subject  is 
a  thing  and  the  verb  is  passive;  by  necessary  implication  effe- 
rantur  omnia  means /^^  efferantur} 

Further  evidence  of  the  way  in  which  the  meaning  of  the  mode 
seems  to  be  shifted  by  causes  which  lie  outside  the  mode  may  be 
had  from  the  cases  of  3d  pers.  where  the  subject  is  not  a  definite 
person.  With  an  ideal  or  typical  person  (Amph.  960  proinde  eri 
ut  sint,  ipse  [the  slave]  item  sit;  Pers.  125  cynicum  esse  egentem 
oportet  parasitum  probe :  pallium,  marsuppium  habeat)  the  sub- 
junctive expresses  only  propriety,  the  direct  forms  of  will  being 
excluded.  So  when  persons  of  a  class  are  the  subject  {regesy 
inimici^,  especially  if  they  are  in  the  audience  and  are  addressed 
indirectly  (matronae  tacitae  spectent,  tacitae  rideant,  Poen.  prol. 
32).  When  the  subject  is  a  thing  and  the  verb  is  passive,  the 
hearer  becomes  the  real  actor,  and  the  direct  forms  of  will 
reappear,  as  in  the  2d  person. 

In  one  class  the  3d  pers.,  sing,  as  well  as  plur.,  is  rather 
narrowly  defined.  The  wishes  containing  the  word  di  or  the 
name  of  a  god  are  definite  in  respect  to  the  nature  of  the  desire 
and  to  the  actors*  The  nature  of  the  desire  is  partly  fixed  by 
the  meaning  of  the  verb,  which  will  be  discussed  later,  but  the 
fact  that  the  gods  are  to  be  the  actors,  not  only  in  wishes  like  di 
me  ament,  but  also  in  the  impersonal  expressions  male  sit  tibi^ 
quod  bene  uoriaty  excludes  advice  or  command  and  confines  the 
mode  to  the  expression  of  that  kind  of  desire  whose  attainment 
is  beyond  human  power.  The  fact,  also,  that  the  gods  are  not 
directly  addressed  excludes  prayer  and  entreaty. 

^  It  is  perhaps  not  speculating  too  freely  to  hazard  the  guess  that  it  is  the 
vagueness  of  the  3d  pers.  which  rendered  possible  the  wide  extension  of  usage 
in  the  subordinate  clause  in  narrative  Latin.  I  am  very  sure  that  in  tracing 
the  origin  of  subordinate  clauses  careful  attention  should  be  paid  to  the  person 
of  the  verb.  Clauses  in  the  ist  or  ad  pers.  are  more  likely  to  be  idiomatic 
and  to  be  directly  connected  with  independent  uses,  while  it  is  the  3d  pers. 
which  swings  away  most  widely  from  independent  uses  and  develops  special 
subordinate  functions. 


Digitized  by  VjOOQ IC 


28o  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Of  all  the  varieties  of  subjunctive,  none  is  more  clearly  defined 
than  the  ist  plural.  In  it  the  speaker  and  wilier  is  an  actor  and 
the  person  addressed  is  also  an  actor ;  it  combines,  as  has  often 
been  remarked,  the  ist  person  with  the  2d,  and  can  therefore  be 
used  only  in  those  senses  which  are  common  to  both.  This 
excludes  almost  entirely  all  uses  like  command,  permission, 
entreaty,  advice,  and  leaves  only  the  special  sense  which  we  call 
exhortation.  No  name  in  all  our  imperfect  grammatical  nomen- 
clature is  more  suitable  and  more  precise  than  Hortatory,  if  it  be 
confined,  as  I  think  it  should  be,  to  the  ist  pers.  plural. 

In  a  few  cases  in  ist  plur.  the  subjunctive  has  somewhat  the 
effect  of  a  command,  where  tu  (True.  840  eamus  tu  in  ius)  or  a 
vocative  (Poen.  1342  leno,  eamus  in  ius)  is  expressed.  That  is, 
where  the  2d  person  is  brought  forward  into  prominence,  the 
direct  shades  of  will  which  are  associated  with  the  2d  person  also 
appear. 

The  forms  of  the  sigmatic  aorist  afford  a  striking  proof  of  the 
effect  of  person  upon  mode.  All  cases  in  ist  sing,  are  hypothet- 
ical, all  in  2d  pers.  are  prohibitive,  all  but  one  in  3d  pers.  are 
optative.  It  seems  certain  that  the  force  which  produced  this 
curious  difference  \i^\.\^^txifaxim,faxis  and  faxii  can  have  been 
nothing  else  than  the  person.  It  is  true  that  the  ist  sing,  in 
general  is  not  always  hypothetical,  but  it  leans  more  strongly  that 
way  than  the  2d  or  the  3d,  while  the  2d  pers.  in  general  leans 
toward  the  direct  forms  of  command  and  the  3d  pers.  contains  by 
far  the  largest  number  of  wishes.  These  general  tendencies, 
working  upon  forms  which  were  becoming  obsolete  and  were 
therefore  preserved  only  in  idiomatic  uses,  brought  about  an 
absolute  uniformity  of  usage  in  the  sigmatic  aorist,  which  would 
be  impossible  with  forms  which  were  in  free  and  general  use. 

These  illustrations  will,  I  hope,  show  sufficiently  how  greatly 
the  modal  meaning  is  influenced  by  the  force  of  person  and 
number,  and  how  necessary  a  recognition  of  this  force  is  to 
precise  interpretation.  It  is  in  the  main  an  exclusive  or  selective 
forced  Of  the  whole  range  of  possible  applications,  certain  ones 
are  excluded  when  the  speaker  addresses  himself  about  his  own 
action,  certain  others  when  the  hearer  is  to  be  the  actor.  But  it 
is  also  capable  of  leading  to  an  expansion  of  modal  meaning  by 
association  and  suggestion.  The  idea  of  propriety  or  conformity 
to  an  ideal  implied  in  some  of  the  3d-pers.  uses  is  such  an  exten- 
sion, and  in  the  forms  of  the  sigmatic  aorist  these  added  concep- 
tions have  become  permanently  associated  with  the  mode. 


Digitized  by 


Google 


P«M! 


THE  SUBJUNCTIVE  IN  PLAUTUS,  28 1 

2.  Tense. 

The  influence  of  tense-force  upon  the  modal  meaning  is  slight. 
Tense-force  in  the  subjunctive  is  in  general  less  clearly  marked 
than  in  the  indicative,  partly  because  the  very  nature  of  the 
subjunctive  removes  it  somewhat  from  time  limitations  (e.  g.  in 
the  hypothetical  uses,  which  are  often  timeless),  partly  because 
in  a  considerable  number  of  cases  the  will  and  the  act  belong  to 
different  times.  So  Ps.  131  ostium  lenonis  crepuit.  ||  crura  mauel- 
lem  modo.  Cure.  512  tacuisse  mauellem ;  in  neither  case  is  mauel- 
lent  parallel  to  malebam.  Pers.  710  eras  ires  potius  is  an  extreme 
case;  it  means  'You  ought  (some  time  ago)  to  have  decided  to 
go  to-morrow,'  i.  e.  the  obligation  is  past,  the  act  future. 

In  one  respect,  however,  the  tense-force  is  well  marked  and 
strongly  affects  the  meaning  of  the  mode.  The  impf.  2d  sing, 
(and  in  some  cases  in  other  persons)  expresses  an  obligation 
which  should  have  been  felt  in  the  past ;  e.  g.  Rud.  842  caperes 
aut  fustem  aut  lapidem  'you  should  have  taken  . .  .'  The  same 
thing  is  true  of  several  questions  in  the  ist  sing.,  the  connection 
of  which  with  2d  sing,  is  close.  Obligation  is  one  of  the  many 
meanings  of  the  subjunctive  in  the  present,  though  it  is  some- 
what infrequent,  and  it  is  the  only  one  which  has,  so  to  speak, 
survived  the  transfer  from  present  time  to  past.  The  other  shades 
of  will  or  desire  cannot  be  used  of  a  past  feeling.  Command, 
entreaty,  advice,  permission,  determination  are  excluded;  only 
obligation  remains.  The  influence  of  tense,  therefore,  though  it 
is  not  wide,  is  especially  clear  and,  when  put  into  comparison 
with  other  influences,  especially  instructive. 

The  use  of  nan  with  these  cases  will  be  considered  later. 

It  is  worth  while  to  note,  also,  in  connection  with  the  question 
of  tense,  how  infrequent  the  subjunctive  of  the  past  is  in  Plautus. 
There  are  77  instances  of  the  impf.  and  7  of  the  plupf.,  against 
1366  of  the  present.  The  subjunctive  in  the  spoken  language  is 
a  direct  and  simple  expression  of  desire,  dealing  with  the  present 
or  the  immediate  future.  It  is  only  in  a  complicated  style,  in  the 
complexity  of  the  conditional  sentence  or  the  subordinate  clause, 
that  the  plupf.  flnds  a  place. 

3.  Voice. 

The  simplicity  and  directness  of  the  feeling  which  lies  behind 
the  subjunctive  is  also  shown  by  the  infrequency  of  passives. 
They  occur  as  follows:  pres.  ist  sing.  4,  2d  sing,  i,  3d  sing.  17, 


Digitized  by  VjOOQ IC 


282  AMERICAN  JOURNAL    OF  PHILOLOGY. 

3d  plur.  5,  impf.  3d  sing,  i,  3d  plur.  i.  Of  these,  6  are  hypo- 
thetical, 4  are  in  parataxis,  4  are  in  the  statement  of  a  plan  (a 
kind  of  half-paratactic  use),  4  are  in  the  peculiar  phrase  mos 
geratur,  two  or  three  are  without  context  or  are  doubtful  in  text, 
and  not  more  than  half  a  dozen  are  in  ordinary  expressions  of 
desire.  It  is  impossible  to  tell  without  more  trouble  than  the 
point  is  worth  whether  this  proportion  (29  out  of  1600,  less  than 
2  per  cent.)  is  smaller  than  in  the  indie,  but  it  seems  probable 
that  Plautus  found  the  mode  to  some  extent  inconsistent  with  the 
passive  voice,  as  it  was  inconsistent  (though  to  a  less  degree) 
with  the  past  tenses.  It  was  a  living  expression  of  desire,  not  a 
mere  symbol,  as  it  became,  for  example,  in  the  cum  clause.  Of 
these  29  cases,  all  but  5  are  in  the  3d  pers.,  where  the  personal 
relations  are  least  direct.  And  where  the  passive  uses  are  nearest 
to  the  directness  of  the  active,  it  is  because  a  2d-pers.  use  really 
lies  behind  the  3d  pers.,  2&facproferantur  is  almost  the  same  as 
proferas. 

4.  The  Meaning  of  the  Verb. 

In  Amph.  928  ualeas,  tibi  habeas  res  tuas,  reddas  meas,  the 
three  subjunctives  are  evidendy,  so  far  as  the  mode  is  concerned, 
identical;  the  close  connection  makes  any  other  interpretation 
impossible.  But  if  ualecLS  stood  alone,  as  it  often  does,  it  would 
be  called  a  wish,  while  habeas  in  10  out  of  the  15  passages  where 
it  occurs  expresses  a  permission,  as  here,  and  reddas  is  a  demand. 
In  modal  force  the  three  verbs  are  alike ;  the  difference  in  effect 
is  due  to  the  difference  in  the  meaning  of  the  verbs. 

The  same  thing  appears  in  a  few  other  cases  where  two  verbs 
are  used  together:  Cas.  611  missurun  es  ad  me  uxorem  . . .?  || 
ducas  (permission)  easque  in  maxumam  malam  crucem  (curse); 
Pers.  352  inimici  famam  non  ita  ut  natast  ferunt.  ||  ferant  eantque 
in  maxumam  malam  crucem;  Ps.  1015  quid  nunc?  ||  argentum 
des,  abducas  mulierem,  a  demand  or  expression  of  will  and  a 
permission  or  expression  of  willingness.  A  few  other  cases  occur 
here  and  there,  the  difference  growing  slighter  as  the  two  verbs 
approach  in  meaning. 

The  classification  of  acts  of  will  according  to  the  nature  of  the 
action  willed  appears  to  rest  upon  a  good  psychological  basis. 
The  attitude  of  the  mind  in  willing  that  another  person  shall 
retain  an  object  which  belongs  to  him  is  different  from  the 
attitude  of  willing  that  he  shall  give  up  what  is  not  his.  Through 
all  the  long  succession  of  choices  and  determinations,  the  mind  is 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  283 

constantly  changing  its  tone  by  slight  and  almost  imperceptible 
degrees.  Language,  it  is  true,  does  not  find  distinct  expression 
for  each  of  these  varying  shades  of  will,  but  they  exist  in  the 
mind  of  the  speaker  and  are  felt,  by  suggestion,  by  the  hearer, 
and  therefore  become  a  restricting  and  modifying  force,  shaping 
the  modal  meaning. 

The  same  conclusion  follows  from  the  modal  behavior  of  single 
verbs,  mereo  {-or)  is  used  5  times,  all  in  the  limited  sense  nan 
{quid')  merear  (mereat),  like  the  English  *I  would  not  do  it  for 
the  world.'  The  meaning  of  the  verb  inclines  it,  though  it  does 
not  necessarily  confine  it,  to  expressions  of  hypothetical  determin- 
ation, excluding  the  ordinary  jussive  uses,  patiar  is  found  only 
in  1st  sing.,  in  repudiating  sentence  questions ;  Asin.  810  egone 
haec  patiar  aut  taceam  ?,  Men.  559  egone  hie  me  patiar  esse  in 
matrimonio  . .  .  ?  In  these  cases  it  is  an  insertion,  like  uin  or 
like  dicam  in  quis  questions.  Its  sense  does  not  exclude  it  from 
other  uses  {perpetiare  is  used  once  in  advice),  but  it  makes  it 
peculiarly  fitted  for  a  single  use.  habeas  is  usually  an  ironical 
permission,  though  it  is  also  used  in  other  ways,  dicam  occurs 
33  times,  but  only  twice  in  expressions  of  will,  both  in  repudiating 
sentence  questions.  It  is  inserted  in  quis  questions  16  times,  is 
used  in  parataxis  8  times  and  7  times  in  hypothetical  (or  perhaps 
here  I  may  say  potential)  uses,  dicai  is  used  once  and  dicant  4 
times  in  hypothetical  senses ;  that  is,  out  of  55  cases  of  dicere  in 
the  subjunctive,  12  are  hypothetical,  much  beyond  the  ordinary 
proportion,  which  is  about  i  :  50.  This  tendency  toward  the 
hypothetical  may  be  like  the  English  'I  should  think';  but  I  am 
not  concerned  here  with  the  explanation ;  whatever  that  may  be, 
the  fact  of  the  tendency  of  dicere  away  from  the  ordinary  uses  is 
plain.  The  verbs  used  in  the  ist  person  plural  are  all  verbs  of 
activity ;  there  are  no  cases  of  esse^  scire,  dicere,  habere,  uelle. 
The  marked  subjunctive  force  of  this  person  almost  excludes 
verbs  which  are  strictly  verba  sentiendi  et  declarandi.  The  verb 
esse  might  be  expected  to  show  an  equal  distribution  over  the 
whole  range  of  subjunctive  use.  It  is,  however,  limited  in  three 
directions : — a)  It  is  essentially  passive  in  meaning  and  therefore 
unfitted  for  the  more  direct  and  vigorous  expressions  of  will. 
Among  the  1 1  cases  where  it  is  used  of  desire,  there  is  no  case  of 
demand  or  command.  The  two  cases  in  2d  sing,  are  the  mildest 
kind  of  permission ;  the  seven  cases  in  3d  sing,  are  either  conces- 
sions {sii  per  me  quidem)  or  expressions  of  propriety,  of  the 


Digitized  by^ 


284  AMERICAN  JOURNAL    OF  PHILOLOGY. 

character  that  a  type  should  have  (improbus  sit  cum  improHs). 
b)  Its  general  and  vague  sense  calls  especially  for  paratactic 
additions,  which  are  used  in  29  of  the  92  cases  and  by  which 
greater  directness  and  force  can  be  given  to  the  verb  (J'ac  sis 
frugi).  Other  defining  words,  adverbs  and  particles,  are  also 
unusually  frequent  wiih  it.  c)  The  fact  that  esse  combines  with 
a  predicate  shades  its  sense  in  various  ways  and  gives  rise  to 
some  narrowly  restricted  idioms,  especially  in  wishes  like  saluos 
sis  J  male  sit  tibi. 

No  single  usage  of  the  subjunctive  is  more  distinctly  marked 
than  the  wishes  which  contain  di  or  the  name  of  a  god :  di  te 
perdanty  luppiier  te  perduit,  iia  me  di  amenty  etc.;  they  have 
been  given  above  in  some  detail.  One  of  the  forces  which  bring 
about  this  distinctness  has  been  spoken  of  above,  under  person 
and  number ;  the  fact  that  the  gods  are  to  be  the  actors  shuts  out 
many  kinds  of  will.  To  this  restricting  agency  another  is  added 
by  the  meaning  of  the  verbs.  The  situation  in  which  it  would  be 
natural  tcTexiiCic  or  command  or  entreat  a  person  to  destroy 
himself  would  be  very  rare,  and  the  natural  result  of  the  working 
of  two  restricting  forces  is  to  produce  a  group  of  sharply  defined 
idioms. 

One  verb,  uelUy  is  so  remarkable  in  its  modal  behavior  as  to 
call  for  special  notice,  although  the  uses  are  not  due  so  much  to 
its  limiting  force  as  to  its  peculiar  adaptability  to  the  subjunctive 
mode.  It  is  used  in  the  present  tense  78  times,  of  which  73  (or 
74)  are  uelim  and  compounds.  All  other  verbs  together  are 
used  in  ist  sing.  pres.  of  desire  or  will  not  more  than  20  or  30 
times.  In  the  impf.  ist  sing,  uellem  and  compounds  are  used  17 
times,  while  other  verbs  are  not  used  at  all  except  with  utinam  or 
in  questions  or  parataxis.  That  is,  uelle  is  used  three  or  four 
times  as  often  as  all  other  verbs  together  in  expressions  of  will  in 
the  ist  person  singular.  The  cases  are  classified  above  according 
to  their  syntactical  relations,  and  the  meaning  of  each  group 
is  given,  uellem  differs  from  uelim  only  in  tense,  expressing  a 
present  wish  about  a  past  act;  it  is  never  the  subjunctive  of 
uolebam. 

As  to  the  force  of  uelim,  the  following  points  deserve  notice : 

i)  uelim  is  not  in  Plautus  a  subjunctive  of  'mild  assertion'  or 
of  *  modesty.*  The  proof  of  this  is  in  the  cases  themselves,  of 
which  full  lists  are  given  above.  In  most  cases  any  such  meaning 
is  absolutely  excluded,  for  example,  in  the  cases  which  express  a 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS,  285 

curse  {pulmoneum  edepol  nimis  uelim  uomitum  nomas;  perii,  || 
uerum  sit  uelim;  in  anj^inam  ego  nunc  me  uelim  uorW),  In  a 
few  cases,  taken  alone,  such  an  interpretation  is  not  impossible, 
though  it  is  not  anywhere  necessary.  As  all  these  cases  must  be 
essentially  alike,  we  are  bound  to  adopt  that  interpretation  which 
is  possible  for  all,  not  that  which  is  possible  for  only  a  part  of  the 
cases.  The  instances  which  seem  most  like  a  mild  assertion  are 
those  with  malim,  which  are  really  influenced  by  the  comparative 
deg^ree  (cf.  potius). 

2)  There  is  no  stronger  evidence  that  uelim  is  potential.  The 
closest  parallel  should  he  faximj  which  is  always  hypothetical, 
huifaxim  regularly  has  a  protasis  either  in  the  same  sentence  or 
in  the  immediate  context,  while  uelim  has  a  protasis  in  only  5 
cases  (these  are  not  included  in  the  lists  above),  4  with  sipossim 
(possit).  In  the  large  majority  of  cases  the  sense  excludes  a 
potential  meaning  just  as  clearly  as  it  excludes  the  idea  of 
modesty  or  politeness. 

3)  uelim  is  in  Plautus  a  sign  of  a  wish,  an  optation,  parallel  in 
the  main  to  uiinam.  This  usage  is  mentioned  in  some  of  the 
grammars,^  as  a  secondary  use  of  uelim.  It  may  have  been 
secondary  in  classical  Latin,  but  in  Plautus  it  is  the  first  and  the 
prevailing  use.  I  should  explain  it  as  having  arisen  by  attraction 
through  parataxis.  The  simple  subjunctive,  ueniat  for  instance, 
is  frequently  so  indefinite  as  to  call  for  a  defining  addition  (see 
below,  on  parataxis).  Alone  ueniat  might  mean  *tell  him  to 
come,*  Met  him  come/  'make  him  come,*  *I  desire  (will)  that  he 
should  come*  or  *I  wish  he  would  come.'  Of  the  various  para- 
tactic  additions  which  supply  the  needed  definition,  uolo  empha- 
sizes the  will,  while  uelim,  repeating  the  mode  of  ueniat,  empha- 
sizes the  modal  force,  the  optative,  and  differentiates  {uold)  ueniat 
•  I  will  that  he  should  come,'  from  {uelim)  ueniat '  I  wish  that  he 
would  come.'  In  the  same  way  paiiar  is  inserted  into  a  repu- 
diating sentence  question,  assuming  the  mode  of  the  verb,  which 
then  becomes  an  infin.,  and  dicam  is  inserted  into  quis  questions. 
OS,  Verg.  Aen.  IV  24  sed  mihi  uel  tellus  optem  prius  ima  dehiscat, 
X  443  cuperem  ipse  parens  spectator  adesset.  In  this  way  uelim 
became,  like  utinam,  a  sign  of  the  wish.  In  contrast  with  uolo  it 
is  a  milder  word,  and  it  became  finally  almost  a  separate  verb, 
meaning  '  I  wish,*  and  is  used  in  this  sense  increasingly  in  later 
Latin  with  scarcely  any  feeling  of  its  original  optative  force.     But 

*  A.  and  G.,  267,  c ;  Gildersleeve,  261,  546,  2. 


Digitized  by  VjOOQ IC 


286  AMERICAN  JOURNAL    OF  PHILOLOGY, 

the  continuance  of  its  paratactic  use  in  familiar  style  (uelim  exis- 
times,  Cic.  ad  Fam.  I  9,  24 ;  uelim  uerum  sit,  ad  Att.  XV  4,  4)  is 
a  reminder  of  its  origin. 

If  this  hypothesis  as  to  the  origin  and  use  of  iulim  is  correct, 
the  meaning  of  lulle  determines  its  modal  use,  not,  as  in  the  case 
of  some  verbs,  by  its  unfitness  for  certain  uses,  but  by  its  peculiar 
adaptability  to  other  (the  wishing  and  willing)  uses. 

A  further  illustration  of  the  effect  of  the  meaning  of  the  verb 
upon  the  modal  force  may  be  found  by  comparing  the  similar 
influence  of  verb-meaning  upon  other  constructions.  Blase, 
Geschichte  des  Plusquamperfekts  im  Lat.,  pp.  9  f.,  35  f.,  has 
shown  how  the  shifted  sense  of  the  plupf.  tense  extends  along  the 
line  of  verbs  of  obligation,  oportuerat^  debuerat,  aequam  erat 
Foth,  Verschiebung  lateinischer  Tempora,  in  Boehmer's  Roman. 
Stud.,  1876,  pp.  243  ff.,  shows  that  the  peculiar  present  sense  of 
the  perfects  fui '  I  am  no  more,'  habui  *  I  have  no  longer,'  is  the 
result  of  verb-meaning.  A  still  better,  because  more  distinctly 
independent,  support  may  be  found  in  the  discussion  of  Greek 
tenses  by  Hultsch,  Die  erzahlenden  Zeitformen  bei  Poly  bios, 
which  I  know  only  through  the  review  in  A.  J.  P.  XVI  2  (62), 
by  C.  W.  E.  Miller.  Exactly  as  the  inherent  temporal  force  of 
Greek  verbs  modifies  and  limits  the  meaning  of  tenses,  so  that 
the  aorist  of  one  verb  is  not  the  same  as  the  aorist  of  another,  so 
does  the  different  modal  meaning  of  verbs  modify  and  limit  the 
modal  force. 

From  these  suggestions,  imperfect  and  superficial  as  I  am 
conscious  that  they  are,  I  draw  two  conclusions.  In  the  first 
place,  it  is  obvious  that  a  consideration  of  the  effect  of  the  verb- 
meaning  upon  the  mode,  intensifying  or  lessening  or  shifting  the 
modal  force,  may  be  of  considerable  service  in  accurate  interpre- 
tation. In  the  second  place,  though  the  verb- meaning  alone  may 
not  be  the  cause  of  idioms  or  of  restricted  and  precise  usages,  it 
may  combine  with  other  forces  to  produce  such  usages.  The 
most  marked  illustration  is  in  the  wishes  like  saluos  sis,  di  ie 
perdanL  And  when  several  forces  are  thus  combined,  a  very 
distinct  influence  may  be  exerted  upon  a  great  number  of 
subjunctive  usages. 

The  influences  thus  far  considered,  from  person,  number,  tense, 
voice  and  verb-meaning,  are  inherent  in  the  form.  No  subjunctive 
can  exist  without  feeling  some  or  all  of  them.  The  forces  which 
remain  to  be  considered  are  exerted  upon  the  subjunctive  by 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  287 

Other  words  in  the  sentence.  These  are  apart  from  the  subjunc- 
tive form,  but,  as  a  bare  subjunctive,  without  other  words,  is  of 
rare  occurrence,  and  even  then  the  preceding  sentences  influence 
the  mode,  we  may  say  that  the  form  of  the  sentence,  which  is  to 
be  treated  next,  is  almost  as  necessary  and  inherent  an  influence 
as  person  and  number. 

5.  The  Interrogative  Sentence. 

The  first  and  most  striking  fact  is  the  difl*erence  in  extent  of 
usage  between  the  subjunctive  in  questions  and  the  subjunctive 
in  non-interrogative  sentences.  Omitting  uelim  and  cases  with 
uiinam,  but  including  the  hypothetical  uses  and  the  indefinite  2d 
pers.,  the  statistics  stand  thus  : 


Non-interrog. 

Interrog. 

I  St  sing., 

34 

212 

2d  sing., 

177 

24 

3d  sing., 

179 

20 

These  figures,  with  the  ratio  of  i  :  6  in  ist  sing,  more  than 
reversed  in  the  2d  and  3d  sing.,  are  in  themselves  proof  that  the 
relation  of  the  subjunctive  to  the  sentence  diflers  with  the  form 
of  the  sentence,  or,  in  other  words,  that  the  interrogative  form  is 
a  force  which  favors  the  use  of  the  ist  sing.,  while  it  partially 
excludes  the  2d  and  3d  sing.  The  diflerence  is  in  the  changed 
relation  of  speaker  and  wilier  in  the  question.  In  the  non-inter- 
rogative sentence  in  ist  sing,  the  speaker,  the  wilier  and  the  actor 
are  one  person ;  in  the  question  the  speaker  asks  (or  exclaims) 
about  the  will  of  the  other  person  in  regard  to  his  (the  speaker's) 
action.  The  speaker  and  the  wilier  are  two  difl*erent  persons. 
sed  maneam  etiam  means,  in  full,  *  I  will  (judge,  decide)  that  I 
should  remain';  maneam f  means  'do  you  will  that  I  should 
remain  ?  *  The  first  situation  is  rare,  the  second  is  very  common. 
We  must  therefore  note  that  in  questions  in  ist  sing,  we  are 
dealing  with  a  subjunctive  which  seems  to  be  the  same  as  the 
ist  sing,  in  non-interrogative  sentences,  but  which  is  in  reality 
different  in  the  very  important  point  of  the  relation  of  speaker 
and  actor  to  wilier.  The  situation  in  questions  in  ist  sing,  in  fact 
corresponds  (with  some  exceptions,  which  will  be  noticed)  rather 
to  the  2d  sing,  in  non-interrogative  sentences.  And,  on  the  other 
hand,  the  question  in  2d  sing,  corresponds  to  the  non-interrogative 
ist  sing.,  in  that  the  wilier  and  actor  are  one,  though  the  speaker 


288  AMERICAN  JOURNAL    OF  PHILOLOGY. 

is  a  different  person.  That  is,  abeam?  may  be  said  to  be  the 
interrogative  form  of  abeas,  and  abeasf  *you  want  to  go  away  ?* 
the  interrogative  oi  abeam  *I  want  to  go  away.* 

But  beside  this  general  effect  produced  by  the  interrogative 
form,  the  particular  kind  of  question  also  affects  the  mode  in 
narrower  but  equally  distinct  ways. 

The  sentence  question  with  simple  subjunctive  (without  para- 
taxis) is  largely  exclamatory.  Of  the  77  sentence  questions,  all 
but  six  or  eight  are  repudiating.  The  repudiation  is  not  a  matter 
of  mode;  questions  of  the  same  form,  with  egone^  etc.,  or  exclam- 
atory repetitions  without  a  particle  are  always  repudiating,  even 
with  the  indicative.  When  the  previous  sentence  contains  a 
statement,  the  statement  is  rejected  ;  when  it  contains  an  expres- 
sion or  implication  of  will,  the  repudiation  is  directed  upon  the 
will.  Occasionally  special  emphasis  leads  to  a  separation  of  the 
will  and  the  repudiation,  as  in  Poen.  149  egone  istuc  ausim  facere  ?, 
where  faciam  is  expanded  into  ausim  facere^  in  order  to  make 
the  repudiation  stronger,  as '  I  should  not  wish  (venture)  to  do 
that'  is  stronger  than  '  I  would  not  do  it.' 

In  these  cases  the  subjunctive  is  not  dubitative  or  deliberative. 
It  is  a  simple  subjunctive  of  will  or  desire,  repeated  with  the 
necessary  change  of  person  and  corresponding  in  general  to  the 
2d  sing,  non-interrogative.*  There  are,  however,  a  few  cases 
which  correspond  to  the  ist  sing,  in  soliloquy  in  non-interrogative 
sentences.  They  are  the  disjunctive  questions  in  Cist.  641,  Cure. 
589,  Merc.  128  (but  this  is  better  taken  as  an  indirect  question) ; 
in  form  Pers.  26  deisne  . . .  aduorser?  cum  eis  belligerem  . . .  ?  is 
deliberative  and  in  soliloquy,  but  the  sense  is  repudiating.  In  a 
few  cases,  where  quid  agam  {Jaciam)?  precedes,  there  is  a 
slight  pretence  of  deliberation.  But  the  whole  number  is  small^ 
certainly  less  than  one-twelfth  of  the  sentence  questions. 

It  has  been  said  that  the  repudiating  exclamations  reject  the 
expressed  or  implied  desire  of  the  other  person.  But  will  can  be 
repudiated  only  by  will.  In  the  brief  and  typical  form  abi,  || 
abeam  f  the  exclamation  means  *  You  want  me  to  go  away !,'  but 
it  also  implies  *  I  don't  want  to  go  away.'  The  will  of  the  person 
addressed  is  repeated  in  the  mode,  the  will  of  the  speaker  is 
indicated  by  the  form  of  the  sentence.  Now,  in  many  cases  the 
previous  implication  of  will  is  not  strong,  while  the  repudiation  is 

^  Cf.  the  latest  and  best  discussion  of  these  questions  by  Wilh.  Gathmann, 
Ueber  eine  Art  unwilliger  Fragen  im  Lat.,  NUrnberg.  1891. 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  289 

definite  and  positive.  In  such  cases  the  force  of  the  mode  is 
weakened  to  a  claim  or  an  expression  of  obligation,  so  that  it 
might,  at  first  sight,  be  overlooked  entirely,  and  the  sentence 
might  be  called  deliberative  because  only  the  will  of  the  speaker 
is  apparent.  E.  g.,  St.  297  nunc  ultro  id  deportem  ?  is  in  a 
soliloquy ;  the  idea  has  occurred  to  the  speaker,  as  if  it  were  a 
suggestion  from  without,  that  he  might  offer  his  good  news 
unasked.  But  he  at  once  rejects  the  idea  by  the  form  of  the 
exclamation,  and  makes  the  rejection  plainer  by  the  next  words, 
hau  placet  neque  id  uiri  ofBcium  arbitror.  It  is  only  when  the 
suggestion  is  faint  and  there  is  no  rejection  in  the  form  of  the 
sentence,  that  the  question  can  properly  be  called  deliberative. 
There  are  few  cases  as  near  the  line  as  St.  297 ;  usually  a  fairly 
careful  examination  of  the  context  makes  the  case  plain.  The 
will  of  the  speaker,  in  all  such  exclamations,  is  suggested  by  the 
form  of  the  sentence  and  has  nothing  whatever  to  do  with  the 
mode. 

None  of  these  forms  is  a  true  question,  asked  in  order  to  have 
a  reply.  The  subjunctive  alone  was  too  vague,  too  liable  to 
confusion  with  repudiation  or  deliberation.  For  a  true  question, 
clearly  expressed,  the  will  must  be  separated  from  the  act  willed. 
A  question  in  regard  to  the  act  would  be  either  a  simple  form  of 
sentence  question  (with  -ne  appended  to  the  verb)  or  a  quis 
question ;  a  question  in  regard  to  the  will  was  expressed  by  the 
insertion  of  uin  in  parataxis,  as  in  Men.  606  uin  hunc  rogem  ?, 
Poen.  990  uin  appellem  hunc  Punice  ?  This  form  of  question 
occurs  26  times  (lists  above).  It  must  of  course  be  supposed 
that  there  was  a  time  when  the  bare  subjunctive,  appellem  or 
perhaps  appellemne^  was  capable  of  expressinjg^  this  sense,  but  the 
form  uin  appellem  expressed  it  so  much  more  clearly  that  appellem 
alone  fell  back  into  the  more  restricted  function  of  repudiation, 
and  the  intermediate  forms  {appellemne^  fell  out  of  use. 

With  the  quis  question  the  matter  is  somewhat  more  compli- 
cated, since  the  variety  in  the  form  of  the  question  is  greater.  In 
general,  there  are  here  also  two  lines  of  usage,  differing  according 
as  they  relate  to  the  will  of  the  speaker  or  to  the  will  of  the 
person  addressed.  They  are  well  illustrated  in  the  quidfaciamf, 
quid  agamf  forms,  which  are  given  in  some  detail  in  the  lists. 
quidfaciam  is  usually  a  question  for  advice  or  direction,  following 
an  impv.  or  its  equivalent,  and  answered,  if  at  all,  by  an  impv.  or 
equivalent.     At  the  other  extreme  quid  ego  nunc  faciamf  is 


Digitized  by  VjOOQ IC 


290  AMERICAN  JOURNAL    OF  PHILOLOGY. 

usually  in  soliloquy  and  deliberative,  and  quid  ego  nunc  agam  f 
is  invariably  so,  with  no  reference  to  the  will  of  another  person. 
Between  these  extreme  uses  quid  ego  faciam  {agam)f  and  quid 
nunc  faciam  {agam)f  are  used  either  way,  perhaps  leaning  a 
little  more  toward  the  deliberative.  There  is  but  one  explanation 
of  these  facts.  They  illustrate  the  gradual  advance  of  language, 
by  the  formation  of  special  idioms,  from  a  single  widely  inclusive 
expression  to  more  specialized  and  precise  expressions,  quid 
faciam  f  was  once  used  for  both  functions,  to  ask  for  advice  or 
to  express  deliberation.  But  as  the  difference  between  these 
functions  was  felt,  quid  faciam  f  was  expanded  into  quid  ego 
nunc  faciam  t  for  the  deliberative,  and  quid  faciamf  retained 
only  the  more  direct  function.  For  the  most  precise  expression 
of  a  question  in  regard  to  the  will  of  another  person  it  was 
expanded,  as  the  sentence  question  was,  by  the  insertion  of  uis 
into  quid  uis  faciamf  or,  as  in  Most.  556,  into  quid  nunc  faci- 
undum  censes  f  That  these  are  all  one  and  the  same  'subjunctive' 
seems  to  me  beyond  question,  and  that  subjunctive  use  is  the 
same  that  is  found  in  all  exclamatory  repetitions  of  an  expression 
of  will,  i.  e.  is  itself  a  subjunctive  of  will.  It  is  modified,  first,  by 
the  fact  that  it  is  in  a  question  and  therefore  is  concerned  with  the 
will  of  the  person  addressed,  and,  second,  by  the  use  of  a  question 
that  is  originally  meant  for  dialogue  in  monologue,  where  the 
speaker  addressed  the  question  to  himself,  quid  faciamf  'What 
do  you  want  me  to  do  ? '  then  becomes  '  What  do  I  want  myself 
to  do?,' and,  as  in  soliloquy  in  non-interrogative  sentences,  the 
self-address  eonfines  the  will  to  narrow  limits,  to  ideas  of  deter- 
mination or  choice  or  propriety,  as  in  sed  maneam  etiam  or 
taceam  opiumvmsi.  Further,  such  a  question,  'What  am  I  to 
do?,'  spoken  in  soliloquy,  suggests  ideas  like  'What  can  I  do?,' 
which  have  sometimes  led  to  the  use  of  the  term  potential  for 
this  kind  of  subjunctive ;  not  improperly,  if  the  fact  is  kept  in 
mind  that  it  is  directly  connected  with  the  subjunctive  of  will. 
The  name  deliberative  also  applies  well  to  these  questions, 
though  not  so  well  to  the  mode.  The  deliberation  is  in  the 
question  more  than  in  the  subjunctive,  and  is  expressed  even  in 
questions  which  have  the  indicative  (jquid  agof)^  where  we  do 
not  speak  of  the  '  deliberative  indicative.' 

The  same  general  distinction  between  questions  of  will  addressed 
to  another  person  and  questions  of  deliberation  in  soliloquy  runs 
through  all  the  quis  questions.    The  forms  are  not  so  well  marked 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS,  29 1 

as  in  the  idiomatic  quidfaciamf,  nor  can  the  connecting  forms  be 
pointed  out  so  distinctly.  In  questions  like  quern  ie  diucm  auiem 
nominemf,  guam  ob  rem  ego  argentum  enumerem  forasf  the 
force  of  the  mode  is  obscured  by  various  other  ideas  and  is  not 
so  easily  felt  as  in  an  empty  form  like  quid  faciamf  The 
number  of  questions  that  can  properly  be  called  deliberative  is 
small,  but  the  distinction  is  a  real  one.  It  is  supported,  too,  by 
the  usage  in  indirect  quis  questions.  These,  omitting  the  cases 
where  a  question  is  not  asked,  but  the  statement  is  made  that  a 
question  was  asked  (Becker  in  Studemund's  Stud.  I,  p.  211  f.), 
depend  either  upon  an  inserted  uis  or  upon  nescio  or  its  equiv- 
alent (Aul.  730,  nunc  mi  incertumst  quid  agam.  abeam  an 
maneam  an  adeam  an  fugiam :  quid  agam  edepol  nescio. 

Beside  these  general  characteristics,  common  to  all  quis 
questions,  there  are  certain  peculiarities  which  are  due  to  the 
form  of  quis.  These  have  been  given  in  the  lists,  and  I  note  here 
only  the  more  striking. 

quo  modo  is  in  all  cases  but  one  an  appeal  implying  nulla  modo^ 
e.  g.  M.  G.  1206  quo  modo  ego  uiuam  sine  te?,  and  this  runs 
easily  into  a  translation  by  can,  as  in  some  cases  oiquid  fiuiamf 
Cf.  Ps.  236  quonam  uincere  pacto  possim  animum  ? 

Questions  with  quid 'why '  and  quor  regularly  imply  a  negative 
answer.  A  question  in  regard  to  the  motive  or  reason  for  acting 
is  of  necessity  argumentative,  and  the  argumentative  tone  excludes 
command,  so  that  the  subjunctive  expresses  only  a  vague  sense 
of  obligation,  imposed  by  the  person  addressed  and  repudiated 
by  the  speaker.  Cf.  the  same  tendency  in  repudiating  sentence 
questions. 

In  qui  ego  isiuc  credamf  (the  only  use  of  ^i  *how*)  there  is 
the  same  argumentative  and  rejecting  sense,  but  it  is  here 
modified  by  the  meaning  of  the  verb  credere.  Belief  is  not 
under  the  control  of  the  will ;  credos  would  inevitably  slip  into 
*  You  should  (ought  to)  believe*;  even  crede  mihi  is  an  appeal. 
And  qui .  . .  credamf  rejects  the  implied  obligation  to  believe  by 
the  implied  answer  '  I  cannot.' 

On  the  other  hand,  the  forms  quern,  quo  *  whither,'  quid  as 
object,  do  not  modify  the  force  of  the  subjunctive,  except  in  the 
general  ways  noted  above  for  all  questions. 

From  what  has  been  said  I  hope  it  may  be  clear  that  the 
interrogative  form  of  sentence  is  a  most  potent  force  in  influencing 
the  meaning  of  the  mode.    That  mode  is  the  same,  in  origin  and 


Digitized  by  VjOOQ IC 


292  AMERICAN  JOURNAL    OF  PHILOLOGY, 

in  essential  meaning,  in  questions  as  in  non-interrogative  sen- 
tences, but  the  question  form  excludes  or  greatly  restricts  the 
more  direct  expressions  of  will,  introduces  the  reference  to  the 
will  of  another  person,  and  implies  by  the  repudiating  form  an 
exertion  of  the  speaker's  will.  In  certain  idioms  with  specialized 
forms  of  quis^  the  idea  of  will  is  diminished  to  a  sense  of  obliga- 
tion, and  by  association  and  suggestion  the  subjunctive  so  far 
approaches  the  potential  meaning  that  it  may  be  translated  by 
can  and  negatived  by  non. 

With  the  following  sections  I  come  to  consider  the  meaning 
and  effect  of  added  words,  not  at  all  a  part  of  the  mode,  which 
help  out  the  mode  where  it  lacks  clearness  and  add  related 
meanings. 

Looked  at  historically,  the  sentence  is  the  result  of  gradual 
accretions  gathering  about  a  nucleus.  The  single  cries  or  words 
which  were  the  primitive  signs  of  emotion  and  thought  corres- 
ponded in  their  vagueness  and  inclusiveness — in  their  applica- 
bility to  a  wide  range  of  different  occasions  and  objects — to  the 
vagueness  of  primitive  thought.  The  words  which  gradually 
added  themselves  to  the  nucleus  were  the  signs  of  the  gradual 
rise  into  consciousness  of  one  or  another  modifying  or  defining 
aspect  of  the  general  idea.  Such  new  features  of  the  thought 
and  the  new  words  which  represented  them  were  on  the  one 
hand  related  to  the  original  germ  and  had  elements  of  likeness 
to  it,  and  on  the  other  hand  they  added  something  which  was 
not  contained  in  the  original  sentence.  This  process  of  sentence- 
growth,  which  must  have  been  infinitely  complex,  still  continues 
in  language  as  long  as  the  language  is  in  a  formative  stage, 
repeating  itself  on  a  small  scale  in  the  growth  of  idiomatic 
phrases.  The  expansion  of  quid  faciamf  to  quid  ego  nunc 
faciamf  is  of  this  sort.  And  along  the  line  of  this  general 
principle  is  to  be  found  the  explanation  of  nearly  all  subjunctive 
parataxis  in  Plautus. 

6.  Parataxis, 

There  are  some  315  or  320  cases  of  paratactic  use  of  the 
subjunctive,  occurring  as  set  down  in  the  accompanying  table. 
The  group  at  the  end,  mainly  of  verbs  of  saying  and  causing, 
comprises  the  cases  of  indirect  quotation,  in  which  the  leading 
verb  states  that  an  expression  of  will  has  been  used.  The 
remaining  cases,  about  300  in  number,  are  all  of  one  kind.     The 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS. 


293 


leading  verb  is  rarely  modified  by  adverbs  or  phrases ;  it  usually 
stands  next  to  the  subjunctive,  either  before  or  after  it,  and  it 
frequently  comes  in  the  middle  of  the  sentence.  An  examination 
of  a  few  of  these  cases,  of  which  full  lists  have  been  given,  will 
make  it  entirely  clear  that  what  is  called  the  leading  verb, 
syntactically,  is  not  the  leading  verb  in  thought,  but  an  addition, 


Present, 
3  X 


sine,  18       I     14 

fac,  facito,  facite,  9    20     10 


3 
I 
8 


iabe, 

3 

2 

uide, 

2 

I 

roga, 

I 

cedo, 

2 

da.  date, 

I 

I 

mane, 

I 

caue. 

I 

10 

I 

I 

I 

uolo. 

8 

7 

2 

3 

I 

malo, 

I 

I 

nolo, 

5 

3 

I 

opsecro,  etc.. 

3 

credo,  etc., 

I 

nil  interdico. 

I 

faxo. 

8 

4 

2 

faciam. 

2 

I 

uelim. 

4 

7 

malim, 

I 

I 

uellem. 

mallem, 

<si)  exoptem. 

I 

faxim. 

I 

I 

5 

<utinam)  faxint. 

I 

optumumst, 

6 

I 

necessest, 

I 

decretumst, 

I 

certumst, 

5 

nil  opust. 

I 

licet. 

4 

2 

decet, 

I 

censen. 

I 

(quid)  uis, 

8 

uin,  uis. 

26 

I 

2 

credin. 

I 

potin, 

2 

Other  indie. 

4 

2 

I 

8^ 

6^ 

61 

7 

10 

28 

Pcrf.  2d  sing.,  I. 


Perf.  2d  sing.,  16 ;  3d  sing.,  I ;  Sigm.  aor.  2d  sing.,  16. 


Lists  entirely  incomplete. 

Incomplete.    Impf.  2d  sing.,  2;  3d  sing.,  i. 

Perf.  2d  sing.,  2 ;  3d  sing.,  3. 

Perf.  3d  sing.,  2 ;  3d  plur.,  2. 

Impf.  3d  plur.,  I;  Plupf.  3d  sing.,  I ;  3d  plur.,  I. 
Impf.  3d  sing.,  I. 


Scattering,  8. 

Present,  259 ;  other  tenses,  58.     Total,  317. 


Digitized  by  VjOOQ IC 


294  AMEKJCAN  JOURNAL    OF  PHILOLOGY. 

an  insertion,  into  a  sentence  already  formed.  The  germ  of  the 
sentence  is  the  subjunctive  verb ;  to  it  almost  all  the  modifying 
words — subject,  object,  adverbs — belong,  and  the  sentence  would 
be  intelligible,  though  not  equally  precise,  without  the  indicative 
verb.  In  many  cases  sentences  very  similar  or  even  absolutely 
identical  occur  without  the  added  verb.  With  the  few  exceptions 
spoken  of  above — and  I  think  it  could  be  shown  that  these  are 
not  really  exceptions — all  cases  of  paratactic  subjunctive  in 
Plautus  are  of  this  kind. 

It  has  been  said  above  of  the  sentence  in  general  that  the 
additions  to  it  have  always  something  in  common  with  the  germ 
out  of  which  they  come,  some  element  which  is  a  repetition  of  an 
element  in  the  central  idea,  and  that  they  also  bring  something 
new,  which  was  not  before  in  the  sentence.  This  is  true  also  of 
the  added  paratactic  verbs ;  they  both  repeat  and  amplify. 

The  idea  which  is  taken  up  from  the  subjunctive  verb  and 
repeated  may  appear  in  the  form  of  the  added  verb  or  in  its 
meaning  or  in  both.  It  appears  in  the  form  when  an  impv.,  sine^ 
f<Uy  uide,  is  used  with  a  subjunctive  in  ist  or  2d  pers.  Thus 
amem  means  *  I  want  to  love  you,'  and  sine  repeats  and  empha- 
sizes the  expressed  desire;  uideas  expresses  a  command  which 
the  impv. /a^  repeats;  ^nd  so  \n  fac  sciam,  fac  sis  frugt, prae- 
cepia  sobrie  adcures  face,  linguam  conpescas  face,  and  many 
more.  Any  of  these  subjunctives  might  stand  alone,  but  the 
jussive  force  would  be  less  clearly  expressed  than  it  is  when  it  is 
strengthened  by  the  impv.  verb. 

The  meaning  of  the  subjunctive  is  repeated  in  the  meaning  of 
the  leading  verb  by  uolo,  maio,  nolo  (together  about  one- tenth  of 
all  the  cases),  by  opsecro,  quaeso,  oro  (the  lists  of  which,  I  regret 
to  say,  are  entirely  incomplete),  used  with  the  2d  and  3d  pers. 
chiefly,  where  the  speaker  is  emphasizing  or  defining  his  own 
will,  and  by  licet,  nil  opusi,  decet,  which  define  the  nature  of  the 
desire.  Thus  animum  aduortas  may  stand  alone  as  an  expression 
of  will,  but  the  will  of  the  speaker  is  doubled  when  uolo  is  added. 
So  taceas  malo  quam  . .  .,  habeas  licet,  hoc  agas  uolo,  iuris 
iurandi  uolo  gratiam  facias,  and  many  more.  The  phrases  in 
the  impersonal  group,  optumumst,  necessest,  decretumst,  certumst, 
are  used  almost  exclusively  with  ist  sing,  and  repeat  those 
modified  kinds  of  desire  which  are  latent  in  these  infrequent 
forms,  sed  maneam  etiam  might  perfectly  well  have  been 
explained  by  optumumst^  as  sed  taceam  (Epid.  59)  would  have 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  295 

been  intelligible  without  optumumst\  or  quod  perdundumsi  pro- 
per em  perdere  might  be  glossed  by  necessesi.  These  all  belong 
to  the  subjunctive  of  will.  The  subjunctive  of  view  or  opinion  is 
occasionally  repeated  in  censeo,  credo,  scio,  though  these  verbs, 
used  with  the  subjunctive,  may  express  an  opinion  as  to  what  is 
best,  as  in  sed  maneam  eHam,  opinor,  where  opinor  is  nearly  the 
same  as  optumumst  As  the  more  direct  expressions  of  will 
imply  a  determination  on  the  part  of  the  speaker  to  see  that  the 
command  is  obeyed,  they  are  emphasized  \>y  faxo.faciam. 
.  In  a  few  cases  the  idea  of  the  mode  is  repeated  both  in  the 
form  and  the  meaning  of  the  added  verb.  So  especially  sine, 
and  also  roga,  iubey  though  these  have  more  to  do  with  defining 
the  relation  of  persons.  The  group  of  subjunctives,  uelimy  maitm, 
ueUetHy  mallemy  also  belong  here,  repeating  by  their  meaning  the 
will-force  of  the  mode  and  by  their  form  defining  the  will  as  a 
wish,  faxim  is  used  only  in  apodosis,  where  the  subjunctive 
verb  was  or  would  have  been  the  apodosis  M  faxim  had  not  been 
thrust  in;  e.  g.  Pers.  73  si  id  fiat,  faxim  nusquam  appareant, 
Amph.  511  si  sciat .  . .,  ego  faxim  ted  Amphitruonem  esse  malis. 
The  hypothetical  idea  is  doubled.  The  one  case  oi  faxint  is 
perhaps  clearer;  Amph.  632  utinam  di  faxint  infecta  dicta  re 
eueniant  tua,  where  the  added  thought,  di  faxint^  is  a  wish 
because  utinam  eueniant  alone  would  have  been  a  wish.  These 
cases  afford,  I  think,  some  independent  support  to  the  explanation 
of  uelim  given  above.  The  use  of  uis  in  quis  questions  and  of 
uin  in  sentence  questions  may  also  be  mentioned  here.  The  verb 
by  its  meaning  repeats  the  meaning  of  the  mode  and  by  its 
combination  with  quis  or  -ne  repeats  the  question. 

Repetition,  however,  is  not  the  function  of  these  added  verbs ; 
it  is  only  the  condition  which  makes  their  close  union  with  the 
subjunctive  possible.  Their  function  is  to  define,  to  bring  out 
more  clearly  the  particular  kind  of  will  or  desire  which  is 
expressed  too  vaguely  in  the  mode,  or  to  express  with  precision 
something  in  the  relation  of  the  persons  involved  which  the 
subjunctive  merely  suggests. 

As  to  the  meaning  of  the  mode,  to  illustrate  the  definition  of  it 
by  examples  would  be  to  repeat  the  lists  already  given.  Will  in 
its  more  direct  forms  is  defined  by  uolo,  uis,  uin ;  wish  (optation) 
by  uelim  \  preference  by  maio\  determination  and  choice  by 
faxOf  decretumst,  certumst ;  entreaty  by  opsecro,  oro ;  permission 
by  licet \  propriety  by  decet\  necessity  by  necessest\  decision  and 


Digitized  by  VjOOQ IC 


296  AMERICAN  JOURNAL    OF  PHILOLOGY, 

sense  of  obligation  by  opiumumst ;  belief,  opinion  by  credo ^  censeo^ 
scio.  The  expressions  of  desire  which  appear  infrequently  or  not 
at  all  in  parataxis  are  those  which  are  in  themselves  most  explicit, 
the  ist  plur.  and  the  2d  sing,  of  command :  the  latter,  however, 
is  frequently  intensified  by  fac  or  uolo.  The  variety  and  extent 
of  these  uses,  defying  precise  classification,  indicate  the  variety  of 
application  of  which  the  subjunctive  was  capable. 

It  is  chiefiy  in  the  3d  person  that  the  relation  of  the  persons 
involved  needs  definition,  because  here  the  hearer  may  be  con- 
cerned with  the  action,  though  his  part  is  left  to  suggestion.  In 
many  cases  this  makes  no  difference.  When  the  subject  is  in  the 
plural  or  is  one  of  a  class,  and,  generally,  when  the  hearer  is 
merely  a  bystander,  other  forces,  chiefly  due  to  person  and 
number,  limit  the  range  of  the  mode  so  that  further  definition  is 
unnecessary.  But  when  both  hearer  and  actor  are  definite 
persons,  it  is  often  necessary  for  the  sake  of  clearness  that  the 
hearer's  part  should  not  be  left  to  suggestion.  The  varieties  of 
usage  and  the  corresponding  paratactic  forms  have  been  suffici- 
ently illustrated  above.  Am  ph.  951  euocate  hue  Sosiam :  ... 
Blepharonem  arcessat,  suggests  the  same  idea  which  is  definitely 
expressed  by  the  addition  of  iube^  in  Most.  930  die  me  aduenisse 
fillo  .  . .  iube  in  urbem  ueniat.  Most.  920  hodie  accipiat  implies 
that  the  slave  is  to  attend  to  the  matter,  and  in  Pers.  445  facito 
mulier  ad  me  transeat,  this  idea  is  important  enough  to  find 
expression  va  facito.  But,  as  has  been  said,  many  verbs  with/ac 
are  passives  and  have  few  or  no  parallels  outside  of  parataxis. 
Compare  also  M.  G.  iioo  aurum  habeat  sibi  .  . . :  sumat,  habeat, 
auferat,  with  M.  G.  1244  sine  ultro  ueniat,  quaeritet,  desideret. 
In  all  these  cases  it  is  the  need  of  more  precise  expression  of  the 
hearer's  part  in  the  action  which  has  led  to  the  addition  of  the 
imperative. 

Some  of  the  sporadic  cases  of  defining  parataxis  are  especially 
interesting.  Of  the  nine  cases  in  ist  sing,  with  fac^  all  are  with 
verbs  of  knowing,  noscam^  uidcam^  sciam ;  a  phrase  meaning  '  I 
desire  to  know '  and  addressed  directly  to  another  person  is  in 
effect  an  appeal  for  information.    The  phrases  da  cLbsorbeam 

^Occasionally  iube  loses  its  proper  sense,  as  in  Ter.  Adelph.  914  f.  iube 
nunciam  dinumeret  ille  Babylo  uiginti  minas;  see  SpengePs  note.  So  in 
Most.  426  iube  uenire  nunciam,  like  Engl,  slang  *now  bring  on  your  man,* 
and  this  is  the  sense  once  with  a  paratactic  subjunctive,  Rud.  708  iube  modo 
accedat  prope. 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  297 

(Cure.  313),  cedo  bibam  (True.  367,  Most.  373,  where  the  MSS 
give  ut  bibam^  which  may  be  right),  mane  sis  uideam  (Most.  849), 
and  perhaps  concede  inspiciam  (Cure.  427)  illustrate  a  eonnection 
between  the  added  verb  and  the  subjunctive  in  which  the  amount, 
so  to  speak,  of  addition  greatly  exceeds  the  repetition.  That  is, 
mane  uideam  is  a  brief  expression  for  mane  et  sine  uideam,  and 
cedo  bibam  is  for  cedo  ei  fac  bibam.  Cf.  Most.  344  da  illi  quod 
bibat,  where  the  thought  is  somewhat  more  expanded,  and  Verg. 
Aen.  IV  683  f.  date  uolnera  lymphis  Abluam  et  extremus  siquis 
super  halitus  errat  Ore  legam.  Capt.  961  quod  ego  fatear,  credin 
pudeat  quom  autumes  ?  is  an  expansion  by  the  insertion  of  credin^ 
as  uin  is  inserted  with  a  slightly  different  meaning ;  cf.  also  M.  G. 
614  quodne  nobis  placeat,  displiceat  mihi?  In  the  same  way 
potin  is  prefixed  to  abeas  in  Pers.  297,  Cas.  731,  a  phrase  which 
later  expands  into  potin  ut  abeas.  In  Most.  679  f.  euocadum 
aliquem  oeius,  roga  circumdueat,  roga  is  substituted  for  the  more 
common  iube  because  the  aliquis  was  not  the  speaker's  slave. 

The  group  of  cases  in  which  the  leading  verb  is  an  indicative 
(other  than  uolo,  credo^  f<^X0y  etc.),  in  all  about  a  dozen  cases, 
differ  somewhat  from  the  ordinary  parataxis  and  have  been 
passed  over  in  the  foregoing  remarks.  They  approach  more 
nearly  the  usual  conception  of  parataxis  as  the  joining  in  thought 
of  two  sentences,  each  of  which  is  complete  in  itself.  Yet  in 
most  eases  a  distinct  relationship  to  complementary  parataxis  can 
be  traced.  Thus  in  Amph.  257,  uelatis  manibus  orant  ignoseamus 
peccatum  suom,  the  mode  is  repeated  in  the  meaning  of  orant. 
In  M.  G.  54,  at  peditasteUi  quia  erant,  siui  uiuerent,  there  is 
combined  a  quotation  of  a  past  thought  (''I  said  to  myself 
^uiuanf')  and  a  repetition  of  the  subjunctive  in  sino.  So  in 
nearly  all  the  verbs  which  quote  a  subjunctive,  uoltis^  suadeSy 
impetraui,  rogarat,  coniurauimus,  accuratum  habuit,  there  is  an 
element  of  meaning  which  harmonizes  with  the  mode.  In  Stich. 
624,  dixi  in  careerem  ires,  there  is  no  repetition  of  the  mode, 
such  as  would  be  expressed  by  iussi,  but  simple  quotation.  The 
greatest  expansion  of  the  prefixed  verb,  so  great  that  it  would 
perhaps  be  correct  to  speak  of  it  as  an  independent  sentence, 
appears  in  St.  177  hoe  nomen  repperi  eo  quia  paupertas  fecit 
ridieulus  forem,  and  Rud.  681  quae  uis  (the  noun)  uim  mi 
adferam  ipsa  adigit. 

I  have  said  nothing  thus  far  of  the  negative  verbs,  nolo,  nil 
inierdico  (jnterduo)^  nil  opust,  or  of  caue.    The  process  of  accre- 


Digitized  by  VjOOQ IC 


298  AMERICAN  JOURNAL    OF  PHILOLOGY, 

tion  cannot  have  produced  the  negative  sentence  out  of  the 
affirmative ;  every  negative  sentence  must  have  been  negative  in 
thought  from  the  beginning  of  its  conception.  Such  a  phrase, 
therefore,  as  nolo  antes  (Cas.  233)  does  not  begin  with  antes  and 
then  prefix  nolo\  it  begins  with  the  prohibition,  ne  ames,  and 
expands  that  by  the  insertion  of  uoio  into  ne-uolo  antes.  With 
caue  the  matter  is  more  complicated.  It  is  used  47  times;  10 
times  with  pres.  2d  sing.,  16  with  perf.  2d  sing.«  16  with  the 
sigmatic  aorist  2d  sing.,  the  rest  scattering.  These  are  the  forms 
which  are  also  largely  used  with  ne  in  prohibitions,  and  in  many 
uses  caue  and  ne  are  exactly  parallel.  Thus  Capt.  439  caue 
fidem  fluxam  feras  (^geras  GS.),  and  443  infidelior  mihi  ne  fuas 
quam  ego  sum  tibi;  Capt.  431  horunc  uerborum  causa  caue  tu 
mi  iratus  fuas,  and  Amph.  924  ignosce,  irata  ne  sies ;  Pers.  816 
caue  sis  me  attigas,  ne  tibi . .  .  malum  magnum  dem,  and  True. 
276  ne  attigas  me.  So  ne  feceris  and  caue  feceris^  ne  dixeris 
and  caue  dixeris^  ne  istuc  dixis  and  caue  tu  istuc  dixis.  On  the 
other  hand,  there  are  no  cases  oi  caue  parallel  to  the  frequent  ne 
posiules  {censeaSyfrustra  sis)  and  none  like  ntolestus  ne  sis,  which 
is  common.  The  sentences  with  caue  are  generally  longer  and 
are  more  frequently  accompanied  by  sis  (si  uis).  It  would 
appear  that  while  caue  has  in  a  considerable  number  of  cases 
sunk  to  a  mere  negative,  not  to  be  distinguished  from  ne,  it  has 
also  retained  enough  of  its  proper  verbal  force  to  prevent  its  use 
in  certain  forms  of  prohibition.  It  is  not  likely  that  the  sinking 
to  a  negative  force  can  have  occurred  except  through  the  use  in 
combination  with  another  verb,  for  a  prohibition  implies  an  action 
to  be  prohibited.  As  long  as  it  was  used  alone,  caue  would  mean 
*  take  care ! '  or  '  beware !  *  It  could  mean  *  don't*  only  when  some 
definite  action  was  proposed.  The  prohibition  with  ne  must  be 
older  than  the  prohibition  with  caue,  and  that  in  its  turn  is 
probably  older  than  caue  ne,  which  is  rare  in  early  Latin  (only 
half  a  dozen  times  in  Plautus).  All  this  would  be  explained  if  we 
suppose  that  caue  was  prefixed  to  the  subjunctive,  chiefly  in  2d 
sing.,  on  the  analogy  of  other  impvv.,  sine,fac,  iube,  uide,  as  a 
periphrasis  of  the  ne  prohibition,  but  with  a  slightly  different 
force,  emphasizing  the  watchfulness  and  the  activity  of  the  person 
addressed,  in  accordance  with  the  proper  meaning  of  cauere. 
This  would  be  analogous  to  the  prefixing  oiuelint  to  differentiate 
the  wish  from  the  expression  of  will  by  means  of  uoio.  The 
natural  sphere  of  such  a  use  would  be  in  the  more  elaborate  and 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS,  299 

formal  prohibitions,  but  it  would  tend  to  degenerate  into  an 
equivalent  of  ne^  though  never  so  far  as  to  be  used  at  all 
frequently  in  blunt  prohibitions  like  molesius  ne  sis} 

The  term  parataxis,  which  has  been  used  above  of  the  prefixing 
of  a  verb  to  a  subjunctive,  is  commonly  employed  to  designate 
the  dependence  of  a  sentence  complete  in  itself  upon  another 
complete  sentence,  without  any  sign  of  the  subordination.  It  is 
in  this  sense  that  the  word  is  used  by  Draeger,  §§368-75,  and  by 
Stolz-Schmalz,  §208,  and  this  is  sometimes  regarded  as  the  only 
proper  use  of  the  term.'  But  this  kind  of  parataxis  is  only  the 
most  obvious  form  of  dependence  without  subordinating  sign, 
which  has  been  accepted  as  representative  of  parataxis  in  general. 
It  needs  to  be  broken  up,  to  be  analyzed,  so  that  the  different 
varieties  of  the  process  which  has  produced  all  subordinating 
words  may  be  more  accurately  understood.  The  program  of 
Weissenhom  (Parataxis  Plautina,  Burghausen,  1884)  and  the 
dissertation  of  Weninger  (de  parataxis  in  Ter.  fab.  uestigiis, 
Erlangen,  1888)  follow  the  lines  of  Draeger,  and  the  fuller  work 
which  seemed  to  be  promised  by  Becker's  program  (Beiordnende 
und  unterordnende  Satzverbindung,  Metz,  1888)  has  not  yet 
appeared.  Lindskog  (de  parataxi  et  hypotaxi  ap.  prise,  lat., 
Lundae,  1896)  deals  chiefly,  though  not  wholly,  with  what  may 
be  called  correlative  parataxis,  where  the  two  sentences  are 
balanced  by  the  repetition  of  like  words  or  by  some  other  simi- 
larity in  the  structure  (Tac.  Ann.  I  28  tarda  sunt,  quae  in  com- 
mune expostulantur :  priuatam  gratiam  statim  mereare,  statim 
recipias).  This  variety  of  parataxis  must  be  fully  studied  in 
order  to  reach  an  understanding  of  the  relative  or  of  protasis  and 
apodosis.  In  what  has  been  said  above  I  have  dwelt  upon  the 
facts  at  greater  length  because  the  prefixing  of  the  verb  seems  to 
be  a  different  and  hitherto  little  noticed  variety  of  the  subordin- 

>  It  is  possible  that  caue  ne  may  have  an  independent  origin  from  caue:  ne 
facias^  but  it  is  also  possible  (and  to  me  it  seems  more  probable)  that  it  is  an 
expansion  oicaue  facias  by  the  insertion  of  n^,  after  ne  had  acquired  conjunc- 
tional force.  So  I  should  explain  some  phrases  with  ut  (e.  %^fac  utualeas  as 
an  expansion  of  fac  ualeas),  though  doubtless  in  many  cases  both  ne  and  ut 
were  a  part  of  the  subordinated  sentence.  Another  explanation  of,  e.  g.,  caue 
eatlas  as  meaning  originally  *look  out !  you  may  fair  is  difficult  because  codas 
alone  in  Plautus  would  never  mean  *  you  may  fall/  and,  if  it  had  that  meaning, 
would  not  take  a  prefixed  imperative. 

'So  Earle,  Proc.  Amer.  Phil.  Assoc,  Dec.  1894,  p.  50.  The  prefixing  of 
Pui'^t,  deXeig,  which  is  like  the  insertion  of  uis,  uin^  he  calls  verbal  preposition. 


Digitized  by  VjOOQ IC 


300  AMERICAN  JOURNAL    OF  PHILOLOGY, 

ating  process,  and  to  be  of  some  value  for  the  explanation  of 
certain  subordinate  clauses.  It  will  explain,  in  part,  the  limitation 
of  the  ut  clause  to  certain  kinds  of  leading  verbs,  the  presence  of 
the  negative  in  the  leading  clause  upon  which  a  quin  clause 
depends,  and  it  has  been  largely  instrumental  in  producing  the 
indirect  question.  Any  process  through  which  subordinating 
force  is  acquired  may  fairly  be  called  parataxis;  this  variety 
might  be  called  defining  or  complementary  parataxis. 

Nothing  could  show  more  clearly  than  these  constructions  the 
wide  applicability  of  subjunctive  forms,  or — if  we  attempt  to 
group  them  and  to  speak  of  the  meaning  of  the  subjunctive — 
the  inherent  vagueness  of  the  mode.  In  spite  of  the  effect  of 
person  and  number,  of  tense,  voice,  form  of  sentence,  verb- 
meaning,  all  working  more  or  less  effectively  toward  precision, 
a  paratactic  addition  is  needed  in  about  one-fifth  of  the  cases  in 
Plautus  to  bring  out  clearly  the  latent  meaning.  And  the  more 
one  examines  the  cases,  the  more  will  the  variety  and  beauty  of 
this  means  of  attaining  to  precise  expression  be  apparent.  They 
are  a  running  commentary  on  the  meaning  of  the  mode,  showing 
both  by  what  they  repeat  and  by  what  they  add  how  the  mode  is 
to  be  interpreted.  They  show  where  aijd  how  the  mode  seemed 
inadequate  to  those  who  used  it,  and  for  correct  interpretation,  at 
least  in  colloquial  Latin,  they  are  of  more  service  than  any  other 
means  at  our  disposal. 

Of  other  words,  adverbs  and  particles,  which  limit  and  define 
the  meaning  of  the  mode  it  is  not  necessary  to  speak  in  detail. 
Most  of  them  are  well  known,  and  for  the  present  purpose  it  is 
necessary  only  to  allude  to  the  need  of  distinguishing  between 
the  function  of  the  mode  and  that  of  the  particle.  In  Pers.  542, 
uideam  modo  mercimonium.  ||  aequa  dicis,  the  restrictive  force  is 
not  in  the  mode;  uideam  alone  might  have  any  one  of  the 
meanings  possible  in  the  ist  sing.,  'I  want  to,'  'I  had  better,'  'I 
must,'  but  modo  adds  a  restrictive  idea  which  is  contained,  it  is 
true,  in  the  mode,  but  not  expressed  by  it.  So  ita  in  ita  dime 
ament  may  be  taken  to  be  a  sign  of  the  asseveration.  With 
utinam  the  precise  steps  by  which  it  became  a  sign  of  the  wish 
are  unknown,  but  its  use  shows  the  need  and  function  of  such 
defining  words.  With  the  most  distinct  and  vehement  forms  of 
wish  it  is  used  very  rarely  or  not  at  all.  Thus  in  the  3d  plur.  it 
is  used  three  times,  but  not  at  all  with  the  118  cases  which 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  3OI 

contain  the  word  di.  With  these  «/,  a  less  distinct  sign  of  the 
wish,  is  used  four  times  and  at,  guin,  qui  still  more  frequently. 
In  the  3d  sing,  the  forms  quae  res  bene  uortat,  bene  {maW)  sit, 
luppiter  te  perdat,  never  have  utinam^  though  ut  ie  Mercurius 
perdat  occurs  once.  The  function  of  utinam  is  to  distinguish 
wishes  of  a  general  character,  not  already  specialized  by  verb- 
meaning  or  by  direct  mention  of  the  gods,  from  other  uses  of  the 
subjunctive  with  which  they  might  be  confused.  The  reference 
of  proin,  proinde  to  the  preceding  thought  gives  it  a  certain 
argumentative  force,  which  makes  it  unfitted  for  use  with  the 
more  intense  expressions  of  will.  It  is  found  especially  with  the 
2d  sing.,  though  somewhat  with  other  persons,  in  advice  or 
direction,  and  may  be  said  to  be  to  some  extent  a  mark  of  such 
uses  of  the  subjunctive.  In  the  same  way  other  words,  especially 
comparatives,  are  associated  with  the  potential  uses,  which  will 
be  discussed  below. 

E.  P.  Morris. 


Digitized  by  VjOOQ IC 


III.— CAECILIUS  OF  CALACTE. 

A  Contribution  to  the  History  of  Greek  Literary 
Criticism. 

In  the  time  of  Augustus  the  two  leading  critics  in  the  literary 
world  of  Greece  and  Rome  were  Dionysius  of  Halicarnassus  and 
Caecilius  of  Calacte.  Of  the  merits  of  the  former  it  is  easy  to 
judge  from  his  extant  works.  The  task  is  more  difficult  in  the 
case  of  the  latter,  whose  remains  are  few  and  fragmentary.  Yet 
the  figure  of  Caecilius  is  so  interesting,  and  in  some  ways  so 
significant,  that  it  seems  worth  while  to  review  the  scattered' 
notices  of  his  life,  and  to  form  such  general  notions  as  we  can  of 
the  nature  of  his  writings.  In  this  work  of  reconstruction  not  a 
litde  help  may  be  obtained  from  a  careful  examination  of  the 
Treatise  on  the  Sublime,  a  book  attributed,  by  a  tradition  long 
since  challenged,  to  Longinus,  the  minister  of  Queen  Zenobia. 

Suidas,  our  principal  authority  with  regard  to  the  life  of 
Caecilius,  tells  us  that  he  was  a  Sicilian  rhetorician  who  practised 
at  Rome  in  the  time  of  Augustus  Caesar,  that  he  was  according 
to  some  accounts  of  servile  birth,  that  his  original  name  was 
Archagathus,  and  that  he  was  *in  faith  a  Jew.*^  Suidas,  it  will  be 
seen  from  the  extract  given  below,  adds  (if  the  words  are  to  be 
regarded  as  genuine)  the  surprising  statement  that  his  life 
extended  till  the  advent  of  Hadrian,  whose  reign  began  more 
than  a  century  after  the  death  of  Augustus.  This  inexactitude 
has  led  Blass  to  assume  that  Caecilius,  the  rhetorician,  has  here 
been  confused  with  Q.  Caecilius  Niger,  the  quaestor  of  Verres, 
about    whom   Plutarch    makes    statements    similar  to  those  of 

'Suidas,  s.  v.  Ka/xi7tof  Y.cukxKlo^  {KtKihjjo^  codd.)  ItLKthLrri^  KaXavriavdc^ 
KdXavTig  6e  irdXtg  2</ceX/ac,  Mr^Pt  ooi^iOTEhaaq  h  'Vufii)  kirl  tov  le^currov  Kaiffopoc 
KOI  Eug  *A6piavov^  Kai  aTro  doh^uv,  wf  riveg  laropffKcuJi,  /cat  irpdrepov  fuv  Kokohfievog 
'Apx&yadoi,  T^  6^  66§av  'lovScuoc.  There  seems  little  doubt  (cp.  Athen.  VI 
272,/;  XI  466,  a)  that  Ka?MKTivoc  and  KaMtcnj  should  be  read  for  KaXavriavdc 
and  KdXavTic.  Archagathus^  it  may  be  added,  seems  to  have  been  a  specially 
Sicilian  name:  see  G.  Kaibel,  Inscriptiones  Graecae  Siciliae,  210,  211,  212, 
330  (conjecturally),  376. 


Digitized  by  VjOOQIC 


CAECILIUS  OF  CALACTE.  303 

Suidas.^  It  has  led  an  earlier  writer  to  go  further  still,  and  to 
assume  the  identity  of  the  rhetorician  and  the  quaestor.'  But 
however  much  or  however  little  truth  there  may  be  in  these 
hypotheses,  or  in  C.  Miiller's  conjecture  (F.  H.  G.  Ill  331  a) 
that  his  ancestors  had  been  brought  as  slaves  from  Syria  to 
Sicily,  it  is  not  disputed  that  Caecilius  Calactinus  taught  rhetoric 
at  Rome,  wherein  he  resembled  Dionysius,  of  whom  he  was  in 
fact  an  inlimate  friend.' 

The  biographical  notes  thus  given  by  Suidas  reappear,  almost 
without  variation,  in  the  *i»vid  (  Violarium,  bed  of  violets)  attributed 
to  Eudocia,  who  flourished  in  the  eleventh  century,  and  was  suc- 
cessively the  wife  of  the  emperors  Constantine  XI  (Ducas)  and 
Romanus  IV  (Diogenes).  There  is  a  like  correspondence  also 
in  the  lists  of  Caecilius'  writings  as  supplied  by  Suidas  and  by 
Eudocia.  The  same  works  are  mentioned,  and  in  the  same 
order.^  Departing  from  this  order  for  the  sake  of  convenience, 
we  may  classify  the  productions  of  Caecilius  under  the  two  heads 
of  history  and  literary  criticism. 

Of  the  historical  writings  of  Caecilius  we  know  little.  But  the 
fact  itself  that  he  attempted  history  is  not  without  an  interest  of 
its  own,  quite  apart  from  the  further  point  of  contact  which  it 
affords  between  him  and  Dionysius.  Athenaeus,  who  is  here  our 
principal  informant,  says  that  a  history  of  the  Servile  Wars  in 
Sicily  was  brought  out  by  Caecilius  the  rhetorician  of  Calacte.* 
When  we  remember  that  Caecilius  was  himself,  according  to  the 
story,  of  servile  origin,  and  when  we  remember,  further,  that  his 
town  of  Calacte  had  been  founded  by  the  rebel  leader  of  an 
earlier  era,  Ducetius,  we  can  imagine  that  he  would  recount  the 
exploits  of  Spartacus  with  peculiar  zest.  Athenaeus  also  refers 
to  a  treatise  of  his  on  history,  which  contained  an  anecdote  of 

*  Plut.  Cic.  VII :  ai:tkj£vBtpuu^  ivOpumo^,  ivox^  ^V  tovdiUl^eiVy  bvofia  KeiuXto(, — 
Friedrich  Blass,  Die  griechische  Beredsamkett  in  dem  Zettraum  von  Alexander 
bis  auf  Augustus,  p.  174.  But  cp.  Th.  Reinach,  Revue  des  £tudes  Juives, 
XXVI  36. 

*G.  Buchenau,  De  scriptore  libri  irepl  v^»ovf,  pp.  41,  42. 

»  Dionys.  Hal..  Epist.  ad  Cn.  Pompeium,  p.  777  (ed.  Reiske) :  tfioi  fUvroi  luu 
r^  ^i7jTdTi,i  KcuKtXiif)  doKci  ra  kvdv/i^fiaTa  avrov  (sc.  OovKvdidov)  /tdXurrd  yt  nal 

*On  the  spuriousness  of  the  'lufid  see  Pulch,  Hermes,  XVII  177 ;  A.  J.  P. 
Ill  489.  IV  109.  V  114  f..  VII  104. 
^Athenaeus,  VI  272,  /:  a<ryypafifjid  re  kKSidcMu  irepl  tuv  dovXiKctv  nuXkfuiv 


Digitized  by  VjOOQ IC 


304  AMERICAN  JOURNAL    OF  PHILOLOGY. 

the  Sicilian  tyrant  Agathocles.*  Whether  the  work  in  question 
was  theoretical  or  practical  we  cannot  say.  It  may  have  been 
identical  with  one  given  under  a  somewhat  different  title  in 
Suidas'  list.' 

And  now  we  approach  Caecilius  in  his  special  rdle,  that  of  a 
literary  critic.  And  first  of  all  it  is  worth  notice  that  his  more 
purely  literary,  or  aesthetic,  judgments  rested  on  a  sufficiently 
solid  foundation  of  verbal  scholarship.  He  was  not  wanting  on 
the  technical,  philological,  grammatical  side.  The  key-note  of 
his  literary  activity  was  sounded  in  his  «cora  ^pvyw^  which  was,  as 
its  title  indicates,  a  kind  of  Antibarbams^  His  energies  were 
chiefly  spent  in  waging  war  against  the  licence  of  the  Asiatic 
school,  and  in  inculcating  a  pure  Attic  style.  As  a  means  to  the 
same  end  he  prepared  that  'select  glossary'  of  Attic  phrases  to 
which  Suidas  refers,  and  also  a  lexicon  of  rhetorical  terms  (Xf(iic6v 
pfrropiK6v).  Both  of  these  were  destined  to  have  many  subsequent 
imitators,  the  latter  being  the  prototype  of  Harpocration,  from 
one  of  whose  articles,  indeed,  its  existence  is  ehiefly  inferred. 
Caecilius  seems  also  to  have  written  an  ar/  of  rhetoric^  and  a  work 
on  figures.*'  The  latter  is  frequently  quoted  by  other  Greek 
rhetoricians  and  by  Quintilian.* 

*  Athenaeus,  XI  466,  a :  KaMciXiOf  cT  b  {iifrup^  6  airb  Ka?.^c  fl«^.  fv  r^p  nept 
icTopiac  ^AyaBoKXia  ^rial  tov  ripawov^  eKw^fiara  XP^^^  ImdeiKvhifTa,  roi^  iraipotg 
if>AaKeiv,  e§  aw  tKEpdpevae  KartOKtvaKkvai  ravra. 

'  Suidas,  1.  c,  nepl  rwv  Koff  laropiav  fj  irap*  laropiav  tipijukvuv  toI^  jji/ropatv. 
This  title  would  seem  to  show  that  Caecilius  was  well  aware  that  the  rheto- 
rician did  not  always  make  an  ideal  historian. 

'Suidas'  enumeration  of  Caecilius*  writings  may  be  conveniently  given  here 
in  full  from  Imm.  Bekker's  edition,  p.  555 :  pipXia  6k  avTov  wo^2a,  Kara  ^puyuv 
fi''  iari  6e  Kara  aroix^lov  airddet^/g  tov  elpyaOat  naaav  /.i^iv  Ka^XipprfpoavvrfC  eon 
6e  fxAoy^  Xi^euv  Kara  (rroixf^lov  •  avyKpiat^  AfffioaOevov^  Kal  KiKipovo^  •  rivt  diat^tpei 
6  'Xttiko^  ^tl"^  TOV  'Aaiatfov  •  nepi  tov  x^P^^^VP^  ''wv  /  prjrdpuv '  ffiyKptat^  ^rifio- 
aBkvovq  Kai  AitJx'tvov '  irtpl  ArfpoaBivovg^  noiot  avTov  yvijoiot  Adyoi  koI  iroioi  v6fhi  • 
irepl  TCiv  Koff  laropiav  [^  nap*  itnopiav,  codd.  et  G.  Bernhardius]  Elprjptvuv  Toi^ 
piiTopat,  Ka\  &X?ja  ^r^Zcrra.— >In  the  section  at  present  under  consideration  Blass 
suggests:  koto,  ♦pvywv  p''  Iqti  6i  anddei^i^  tov  (delv)  eip^aOai  naaav  M^iv  ev 
Ka7Juppr]uoaivi)'  Iti  de  c/cAo)-^  A/fe&w  Kord  OToixeiov.  M.  H.  E.  Meier  (Opuscula 
Academica,  1 131)  proposed  fUTa  KdX?dpprfpo<rit'r;g,  By  /ca?.?./pp^/«M76w7  *  elegance 
of  diction '  seems  to  be  meant. 

*Tixvv  priTopiKij  and  "rrtpl  (T;t7?/idrwv.  For  the  former  cp.  Quintil.,  Inst.  Orat, 
Hit,  16. 

*By  the  rhetoricians  Alexander  mpX  ffxVf^aTiJv  (Spengel,  Rhetores  Graeci. 
Ill  7-40),  Phoebammon  ax^^a  nepi  axn/^oTuv  /jrfTopiKov  (ibid.  Ill  41-56),  and 
Tiberius  nepl  axffpaTuv  (ibid.  Ill  57-82).     By  Quintilian,  IX  3,  89:   haec 


Digitized  by  VjOOQ IC 


CAECILIUS  OF  CALACTE.  3OS 

Next  come  two  works  of  a  more  distinctly  literary  cast,  one 
illustrating  the  differences  of  the  Attic  and  the  Asiatic  style,  and 
the  other  characterising  the  ten  (Attic)  orators.*  In  both  these 
writings  Caecilius  was  on  ground  which  he  had  made  specially 
his  own.  Atticism  as  opposed  to  Asiaticism  was  his  great  pre- 
occupation, and  it  is  in  him  that  we  find  the  first  specific  reference 
to  the  so-called  canon  of  the  ten  Attic  orators.  That  he  was  the 
first  to  frame  the  canon,  we  are  hardly  entitled  to  assert,  though 
many  good  authorities  have  held  that  view.  The  probability, 
rather,  seems  to  be  that  it  should  be  referred  neither  to  Rome  and 
Caecilius,  nor  yet  (as  the  traditional  opinion  among  scholars  since 
Ruhnken's  time  has  been)  to  Alexandria,  but  to  Pergamus  and  the 
end  of  the  second  century  B.  C*  Pergamus,  like  Alexandria,  was 
a  notable  centre  of  learning.  Rhetoric  in  particular  flourished 
greatly  there,  as  in  Asia  Minor  generally,  whereas  at  Alexandria 
it  was  but  little  studied.  Over  and  above  this  general  treatise  on 
the  distinctive  features  of  the  ten  Attic  orators,  Cae(:ilius  wrote 
separately  on  the  authenticity  of  the  speeches  of  Demosthenes, 
on  Antiphon,  and  on  Lysias.' 

omnia  (viz.  the  whole  question  of  rhetorical  figures)  copiosius  sunt  exsecuti, 
qui  non  ut  partem  operis  transcurrerunt,  sed  proprie  libros  huic  operi  dedica- 
verunt,  sicut  Caecilius,  Dionysius,  Ruttlius,  Cornificius,  Visellius  aliique  non 
pauci,  sed  non  minor  erit  eorum  qui  vivunt  gloria.  The  same  treatise  is 
probably  indicated  in  Quintilian,  V  10,  7  ;  IX  3,  38,  46,  91,  97. 

*  SuidaSi  1.  c:  rivi  SuujtlpeL  6  'Arr</cdf  C^^^f  ^<w  'Aatavov  and  Trepl  tov  ;t'a^a\r^/>of 
ruv  i  prrrdpuv,  Liddell  and  Scott  interpret  C^Aof  of  Asiatic  extravagance.  But 
the  title  of  Caecilius*  book  seems  to  suggest  a  more  general  meaning,  such  as 
emulation^  imitatum^  manner.  At  the  same  time  the  word  appears  to  be  used 
specially  of  the  Asiatic  school.  Cp.  Plut.,  Anton.  Vit.  2 :  exPV^o  ^^  tu  /caPjw- 
fUWf)  fiEv  *Aaiav(f)  'C.ifh^  tCjv  X6yuv  aiSmnm  fiakuna  Kof  SKeivcv  Tbv  XP^^^t  kxovri  6i 
TTo?^^  6fjioi6T7/Ta  TTpbg  TOV  P'tov  avTov  KOfiir66rf  Kai  ^pvayfiariav  dvra  Kal  kevw 
yavpidfMTo^  Kal  (U7joTifiiaq  avQfjLah}v  fuOT&v,  and  Strabo,  648  :  '^yrjaiag  h  inirup^  bg 
yp^e  /idAiara  tov  'Aaiavoif  Xey'O/xivov  C^Aov,  ^^rapaKpdeipac  rd  KaOeaTTfKog  Idoq  to  'Attikov. 

'See  Brzoska's  learned  and  ingenious  dissertation,  De  Canone  Decem 
Oratorum  Quaestiones,  Vratislaviae,  1883.  H.  Usener,  however,  declares  for 
Alexandria:  Dionysii  Hal.  librorum  de  imitatione  reliquiae,  p.  132. 

'Suidas,  1.  c:  irepl  AfifioadivovCf  iroioi  airrov  yv^moi  Tidyoi  xal  ttoIoi  v66ot, 
[Plutarch],  X  Oratorum  Vitae,  832  E:  Km«A<of  ev  r^  wept  avToif  (sc.  'Avnqcjv 
Toc)  owTdyfMTi,  [Longinus].  ^rtpi  'rVwuf,  XXXII  8:  6  KeKiTuoQ  kv  To'ig  vnkp 
Xvaiov  avyypdfifiaaiv.  In  this  last  passage  the  plural  and  the  preposition  are 
to  be  noted.  Caecilius,  it  seems  to  be  implied,  often  dealt  with  Lysias,  and  in 
the  spirit  of  an  advocate  rather  than  a  judge.  Cp.  Baudat,  iStude  sur  Denys 
d'Halicamasse  et  le  Traite  de  la  Disposition  des  Mots,  p.  16. 


Digitized  by  VjOOQ IC 


306  AMERICAN  JOURNAL    OF  PHILOLOGY. 

[Photius  has  preserved  a  passage  from  the  book  on  Antiphon  in 
w&ich  Caecilius  remarks  that  that  orator  seldom  uses  the  'figures 
of  thought/  and  only  where  nature  herself  is  his  prompter.^  By 
'figures  of  thought'  (a;rif"»T"  diapoias)  Caecilius  denoted  irony, 
rhetorical  question,  and  the  like,  as  distinguished  from  'figures  of 
language'  (trxil^'^*^  Xf^cor),  viz.  assonance,  balance  of  clauses,  and 
so  onr/The  whole  passage  is  interesting,  and  it  forms  the  longest 
fragment  we  possess  of  Caecilius.' 

The  consideration  of  Caecilius'  attitude  towards  another  Attic 
orator,  Lysias,  brings  us,  as  we  have  already  noted,  into  direct 
contact  with  the  De  Sublimiiate,  In  the  thirty-second  chapter  of 
that  treatise  we  read :  '  Fastening  on  such  defects  [as  have  previ- 
ously been  mentioned],  Caecilius,  in  his  writings  in  praise  of 
Lysias,  ventured  to  make  the  assertion  that  Lysias  was  altogether 
superior  to  Plato.  In  so  doing  he  gave  way  to  his  feelings, 
unlike  a  true  critic,  in  two  respects.  Loving  Lysias  better  than 
his  own  person,  he  nevertheless  hates  Plato  more  perfectly  than 
he  loves  Lysias.  He  is  carried  away  by  the  spirit  of  contention, 
and  even  his  premises  are  not,  as  he  thought,  admitted.  For  he 
prefers  the  orator,  as  faultless  and  immaculate,  to  Plato  as  one 
who  has  often  made  mistakes.  But  the  facts  are  not  of  this 
nature,  nor  anything  like  it.' " 

If  we  accept  this  passage  without  qualification,  we  shall  certainly 
feel  bound  to  form  a  poor  opinion  of  Caecilius  as  a  judge  of  great 
literature.  But  we  must  not  forget  that  the  De  Sublimiiaie  is  a 
polemical  treatise.  As  its  opening  words  show,  it  is  directed 
against  those  shortcomings  of  Caecilius  which  suggested  its 
preparation.  Moreover,  the  author  of  the  treatise  cannot,  surely, 
himself  wish  to  imply  more,  at  the  most,  than  that  Caecilius 
compared  the  two  as  writers  solely,  and  not  as  thinkers  and 
artists.  The  author  (whoever  he  was)  and  Caecilius  were,  both 
of  them,  opposed  to  Asianism ;  and  Caecilius  may  well  have  held 

1  Phot.,  Cod.  259.  p.  485  B.  29,  Bekker. 

« Cp.  R.  C.  Jebb,  Attic  Orators',  I  28 ;  F.  Blass.  Die  attiichc  Beredsamkeit*, 
I  118. — The  following  definition  of  a  figure  is  attributed  by  Phoebammon 
(Spengel,  Rhett.  Gr.  Ill  34)  to  Caecilius:  ILamOuo^  Si  6  KaAoxr/rjTf  (codd. 
Ka?MifdiTfjg)  (jpiaaro  ovtw  oxVf^  ^cri  Tfxmf^  fif  to  fiij  Kara  ^vaiv  rd  nyf  diavoiac  kqI 
"Xk^ELtq,  Is  this  to  be  harmonised  with  the  passage  in  Photius  on  Shakespeare*s 
principle  that  it  is  sometimes  natural  to  be  unnatural  ?  '  This  is  an  art  |  Which 
does  mend  nature,  change  it  rather,  but  |  The  art  itself  is  nature,'  Winter's 
Tale.  IV  4. 

»7rcpi  t)V«wf,  XXXII8. 


Digitized  by  VjOOQ IC 


CAECILIUS  OF  CALACTE,  307 

that,  from  this  and  other  points  of  view,  Lysias  was  a  safer  model 
for  the  young  student  of  composition  than  Plato.  So  to  hold 
would  simply  be  to  recognise  that  average  humanity  should 
choose  more  modest  standards  than  those  presented  by  the 
loftiest  and  most  daringly  original  genius.  But,  however  inter- 
preted, the  critic's  words  can  hardly  be  acquitted  of  the  charge 
of  exaggeration ;  they  seem  to  show  that  he  was  himself  'carried 
away  by  the  spirit  of  contention.'  At  all  events,  we  know  from 
another  source  that  Caecilius  was  no  such  blind  and  uncritical 
admirer  of  Lysias  as  is  here  suggested.  On  the  contrary,  he 
found  fault  with  him  on  the  ground  that  he  was  less  skilful  in  the 
arrangement  of  arguments  than  in  invention.* 

It  will  be  well  here  to  quote,  without  abridgment,  from  the  De 
Sublimitate  its  opening  sentence,  for  in  the  original  it  is  all  one 
sentence,  though  rather  a  long  one:  'You  will  remember,  my 
dear  Postumius  Terentianus,  that  when  we  examined  together 
the  treatise  of  Caecilius  on  the  Sublime,  we  found  that  it  fell 
below  the  dignity  of  the  whole  subject,  while  it  failed  signally  to 
grasp  the  essential  points,  and  in  consequence  conveyed  to  its 
readers  but  little  of  that  practical  help  which  it  should  be  a 
writer's  principal  aim  to  give.  In  every  systematic  treatise  two 
things  are  required.  The  first  is,  to  indicate  what  the  subject  is. 
The  second  in  order  ranks  higher  in  importance.  It  concerns 
the  means  and  methods  by  which  we  may  attain  our  end.  Now, 
Caecilius  essays  to  show  the  nature  of  the  sublime  by  countless 
instances,  as  though  our  ignorance  demanded  it,  but  the  consid- 
eration of  the  means  whereby  we  can  avail  to  bring  our  own 
natures  to  a  certain  pitch  of  elevation  he  has,  in  some  strange 
way,  passed  by  as  unnecessary.  However,  it  may  be  the  man 
ought  not  so  much  to  be  blamed  for  his  omissions  as  praised  for 
his  happy  thought  and  his  enthusiasm.' 

It  is  clear  from  this  preface  that  Caecilius  had  written  a  treatise 
— ^apparently  a  short  one — of  which  the  subject,  and  probably  the 
title,  was  ircpi  v^rovr.  This  treatise  his  successor  in  the  same  field, 
a  writer  who  is  now  generally  supposed  to  have  belonged  to  the 
first  century  of  our  era  rather  than  to  the  third,  had  examined  in 
conjunction  with  his  young  Roman  friend  Postumius  Terentianus.' 

^Phot.,  Cod.  262,  p.  489  B.  13. 

'It  should  be  noted  that  the  MSS,  in  this  passage,  gire  the  name  as 
IloaTobfue  ^Xupevriavi.  I  hope  to  discuss  this  reading  elsewhere,  in  cob- 
nexion  with  the  general  subject  of  the  authorship  of  the  nepl  v^fnv^. 


Digitized  by  VjOOQ IC 


308  AMERICAN  JOURNAL    OF  PHILOLOGY, 

They  had  found  in  it  much  to  desire,  but  due  credit  is  given  to 
its  author  for  originality  in  his  choice  of  theme.* 

The  work  of  Caecilius  on  the  Sublime  has  been  lost  entirely, 
and  that  of  his  successor  exists  only  in  a  mutilated  form,  about 
one-third  of  it  having  disappeared.  It  is  therefore  impossible  to 
speak  with  any  certainty  about  the  two  books  and  their  relation 
to  one  another.  But  it  seems  open  to  question  whether  the  later 
treatise  was  not  guilty,  to  some  extent,  of  the  omission  which,  in 
its  proemium,  it  imputes  to  the  earlier.  In  any  case,  it  seems  to 
have  trodden  closely  in  the  footsteps  of  Caecilius,  especially 
when  treating  oi  figures  and  tropes.  The  references  to  Caecilius 
are  either  direct  or  indirect.  The  direct  references,  besides  those 
already  mentioned,  are  the  following.  In  the  eighth  chapter  we 
are  told  that  he  had  omitted  some  of  the  five  sources  of  elevated 
speech,  passion  {ytoBoi)  being  specially  mentioned.  Towards  the 
end  of  the  chapter  we  have  the  same  criticism  driven  home  thus : 
*  If,  however,  Caecilius  considered  that  passion  never  contributes 
at  all  to  sublimity,  and  if  it  was  for  this  reason  that  he  did  not 
deem  it  worthy  of  mention,  he  is  altogether  deceived.  I  would 
affirm  with  confidence  that  there  is  no  tone  so  lofty  as  that  of 
genuine  passion,  in  its  right  place,  when  it  bursts  forth  in  a  wild 
gust  of  mad  enthusiasm,  and  fills  the  words,  so  to  say,  with 
frenzy.'  Inc.  XXXI  Caecilius  is  again  taken  to  task:  'In  this 
way,  too,  that  original  expression  of  Theopompus  merits  praise. 
Owing  to  the  correspondence  between  word  and  thing,  it  seems 
to  me  to  be  very  expressive ;  and  yet  Caecilius,  for  some  unex- 
plained reason,  finds  fault  with  it.  "Philip,"  says  Theopompus, 
"had  a  genius  for  stomaching  {dvayKo<l>ay^trat)  things."  Now,  a 
homely  expression  is  sometimes  much  more  telling  than  eloquent 
language,  for  it  is  understood  at  once,  since  it  is  drawn  from 

^  There  seems  no  valid  reason  for  questioning  this  originality,  though  from 
the  nature  of  the  case  we  cannot  demonstrate  it.  At  first  sight  it  might  seem 
likely  in  itself  that  a  man  with  Hebrew  inclinations  should  conceive  the  idea. 
But  we  do  not  know  precisely  what  was  Caecilius'  attitude  to  his  theme.  It 
would,  however,  appear  probable,  from  the  character  of  his  own  fragments 
and  from  his  known  regard  for  Lysias,  that  he  favoured  a  plain  rather  than  a 
heightened  style.  But  we  suffer  everywhere  from  want  of  information.  For 
instance,  we  cannot  tell  whether  he  confined  (as  he  might  almost  seem  to  have 
done)  his  investigations  to  prose- writers,  and  excluded  the  poets,  who  figure 
so  largely  in  the  De  SubUmitau,  Nor  yet  can  we  assert  that  he  did,  or  did 
not,  agree  with  so  many  of  his  Greek  and  Roman  contemporaries  and 
successors  in  associating  literary  criticism  with  art-criticism. 


Digitized  by  VjOOQ IC 


CAECILIUS  OF  CALACTE.  3O9 

common  life,  and  the  fact  that  it  is  familiar  only  makes  it  the 
more  convincing.  So  the  words  'stomaching  things'  are  used 
most  strikingly  of  a  man  who,  for  the  sake  of  attaining  his  own 
ends,  patiently  and  with  cheerfulness  endures  things  shameful 
and  vile.'  In  the  next  chapter  it  is  mentioned,  apparently  in  an 
approving  rather  than  in  a  merely  critical  spirit,  that  'with  regard 
to  the  number  of  metaphors  to  be  employed,  Caecilius  seems  to 
assent  to  the  view  of  those  who  lay  it  down  that  two,  or  at  the 
most  three,  should  be  ranged  together  in  the  same  passage.' 
Finally,  when  in  c.  IV  the  author  is  illustrating  the  vice  of 
frigidity  from  the  writings  of  the  historian  Timaeus,  he  excuses 
himself  from  a  lengthy  enumeration  of  examples  on  the  ground 
that  'most  of  them  have  already  been  quoted  by  Caecilius.' 

Thus  the  direct  references  are,  as  usually  happens  when  a  new 
writer  is  treating  a  subject  previously  handled  by  some  one  else, 
of  a  rather  controversial  nature.  But  this  is  not  all.  The  general 
contents  of  the  treatise,  and  its  sequence,  or  want  of  sequence, 
seem  to  be  influenced  by  the  fact  that  the  author  had  the  book  of 
Caecilius  before  him,  and  assumed  the  same  of  his  reader  or 
readers.  This  is  probably  also  the  explanation  of  the  rather 
abrupt  way  in  which  some  of  the  literary  illustrations  make  their 
appearance.  And  we  may  possibly  include  among  indirect  allu- 
sions to  Caecilius  such  expressions  as  rhv  ypa^vra  in  c.  XXXVI  3, 
where  the  passage  runs:  'In  reply,  however,  to  the  writer  who 
maintains  that  the  faulty  Colossus  is  not  superior  to  the  Dory- 
phorus  of  Polycleitus,  it  may  be  readily  said,  among  many  other 
things,  that,'  etc.;  and  the  words  6  rols  xpi^^'^ofiaBovaiif  imrifi&p  in  c. 
II  3,  where  the  complete  sentence  is:  'If,  I  say,  the  critic  of 
those  who  desire  to  learn  were  to  turn  these  matters  over  in  his 
mind,  he  would  no  longer,  it  seems  to  me,  regard  the  investigation 
of  the  subject  as  unnecessary  and  useless.'  It  has  also  been 
maintained  that  in  c.  II  i  the  word  (^i/cri  should  be  understood  of 
Caecilius,  but  this  does  not  seem  altogether  probable.  There  is  a 
more  likely  instance  in  XXIX  i.* 

^On  the  whole  question  see  M.  Rothstein  in  Hermes,  XXIII  1-20;  L. 
Martens,  De  Libello  Uepi  "TV'ovf,  Bonnae,  1877;  Morawski,  Qaaestiones 
Quintilianeae,  Posnaniae,  1874,  and  De  Dionysii  et  Caecilii  Studiis  Rhetoricis 
in  Rheinisches  Museum,  XXXIV,  pp.  370  seqq.;  Burckhardt,  Caecili  Rhetoris 
Fragmenta,  Basileae,  1863;  Weise,  Quaestiones  Caecilianae,  Berolini,  1888; 
F.  CaccialanzE,  Cecilio  da  Calatte  e  V Ellen ismo  a  Roma  nel  secolo  di  Augusto 
(Rivista  di  Filologia,  XVIII  1-73). 


Digitized  by  VjOOQ IC 


3IO  AMERICAN  JOURNAL    OF  PHILOLOGY, 

Suidas,  it  will  have  been  noticed,  ascribes  to  Caecilius  a  com- 
parison between  Demosthenes  and  Aeschines,  and  another  between 
Demosthenes  and  Cicero.  On  the  subject  of  the  latter  compari- 
son, Plutarch  in  his  Life  of  Demosthenes  has  some  caustic 
remarks.  We  will  forego,  says  he,  the  task  of  contrasting  the 
two  orators,  and  of  pronouncing  upon  their  superiority  in  charm 
or  intensity.  We  must  remember,  he  continiyes,  the  proverb 
about  a  fish  out  of  water ^  which  'the  all- accomplished  Caecilius 
overlooked  when  he  had  the  hardihood  to  publish  a  comparison 
between  Demosthenes  and  Cicero.  However,  it  may  well  be  that 
if  the  saying  Know  thyself  were  always  present  to  everybody's 
mind,  it  would  not  have  been  thought  a  divine  behest.'  ^ 

These  severe  strictures  upon  Caecilius  recall  the  equally  severe 
remark,  already  quoted  from  the  De  Subltmiiaie,  that  'although 
he  loved  Lysias  better  than  his  own  person,  he  nevertheless  hated 
Plato  more  than  he  loved  Lysias.*  Clearly,  Caecilius  was  unpop- 
ular. He  may  have  been  one  of  those  men  who  are  described, 
by  those  who  like  them,  as  original  and  versatile,  and  by  those 
who  like  them  not,  as  self-confident  and  audacious.  We  cannot 
disregard  the  mild  irony  of  Plutarch,  whose  attitude  is  usually 
kindly ;  yet  neither  can  we  fail  to  recognise  in  Caecilius  a  pioneer 
in  the  fruitful  region  of  comparative  literature.  And  he  must 
have  been  doubly  a  pioneer  if  there  is  truth  in  the  suggestion  that 
the  author  of  the  De  SublimiicUe  owes  to  him,  as  being  '  in  faith  a 
Jew,*  his  celebrated  reference  to  the  *  legislator  of  the  Jews '  and 
to  a  passage  'at  the  very  commencement  of  his  Book  of  Laws.'* 
There  must  have  been  originality,  and  true  scientific  instinct,  in 
the  man  who,  probably  for  the  first  time,  compared,  in  however 
rudimentary  a  way,  three  several  literatures.  The  De  Sublimiiaie 
itself  has  a  comparison — ^this  too,  very  possibly,  suggested  by 
Caecilius — between  Demosthenes  and  Cicero,  in  which  the  author 
likens  the  former  to  a  thunderbolt,  the  latter  to  a  conflagration. 
But  it  is  there  prefaced  by  an  apology:  if  we  too,  as  Greeks,  are 
permitted  to  form  an  opinion  upon  the  point?  This,  seemingly,  is 
also  Plutarch*s  doubt.    If  the  doubt  were  caused  by  the  con- 

*  Pint.,  Vit.  Demosth.,  c.  Ill :  rh  61  rohc  Myovc  avre^erd^etv  koI  awo^iveoSat, 
x&repoi  ^iuv  ^  detv&repo^  eiirelv,  k&oofJLtv,  Kaxel  ydp,  6c  ^tv  6  'luv,  AeX^lvoc 
kv  x^P^V  /5*flf  ^  ^  rrepirrb^  ev  airaot  Kexlhoc  dyvofyjac  eveavteitaaro  aijyKpioiv 
Tov  AfffJUJoBivovg  KaX  Kudpuvog  k^tveyKtiv,  oAAa  yap  lauc,  ei  iravrbc  ijv  rd  Tv£)6i. 
ffeavrbv  Ix^iv  irp6xetpov,  ovk  hv  eddKei  wpdarayfia  Btiov  tivai, 

«DeSubl.  1X9.  "Ibid.  XII  4. 


Digitized  by  VjOOQ IC 


CAECILIUS  OF  CALACTE.  3II 

sciousness  of  imperfect  knowledge  of  Latin,  it  was  creditable  to 
those  who  felt  and  owned  it ;  but  if  it  were  due  to  any  awe  of  the 
Roman  conqueror,  it  would  be  spurned  by  Caecilius,  who  hailed 
from  Sicily,  the  birthplace  of  rhetoric,  and  who  had  chronicled 
the  intrepid  resistance  there  offered  to  the  Roman  power  by 
Spartacus.^ 

It  is  in  the  indtependence  of  mind  which  led  Caecilius,  if  we  are 
right,  to  bring  three  literatures  into  comparison,  that  we  seem  to 
detect  his  true  significance  and  originality.  The  historian  of 
ideas— especially  of  rhetorical  or  literary  ideas — must  always 
speak  with  due  diffidence.  He  is  not  entitled  to  affirm  more  than 
that,  as  far  as  his  researches  have  gone,  this  or  that  thinker  was 
the  parent  of  this  or  that  idea.  With  this  substantial  reservation, 
it  may  be  claimed  that  Caecilius  inaugurates  the  era  of  compar- 
ative literary  criticism.  And  this,  if  it  stood  alone,  would  be 
enough  to  make  him  the  man  of  mark  he  clearly  was  among  his 
contemporaries  and  successors. 

But  he  was  also,  together  with  his  friend  Dionysius,  in  the 
thick  of  a  great  movement  for  the  purification  of  literary  taste, 
the  movement  comprehensively  known  as  Atticism,  He  was  one 
of  the  leaders  in  the  revolt  against  the  tendency  to  prefer  the 
florid  writers  (broadly  termed  Asiatic)  of  the  age  between 
Demosthenes  and  Cicero  to  the  Attic  writers  of  an  earlier  and  a 
better  time.  In  this  controversy  some  originality,  as  well  as 
much  vigour,  may  safely  be  attributed  to  him.  It  is  possible,  as 
Wilamowitz  von  MoellendorfT  maintains,  that  the  Atticist  revival 
began  with  Apollodorus  of  Pergamus,  who  was  the  Jeacher  of 
Octavianus  and  probably  of  Caecilius.'    But  it  would  appear  that 

^  It  would  be  interesting  to  determine,  if  we  could,  what  knowledge  of  Latin 
was  possessed  by  Caecilius,  Dionysius  Halic,  Plutarch,  and  the  author  of  the 
De  SubtimitaU,  Egger's  essay  De  T^tude  de  la  langue  latine  chez  les  Grecs 
dans  Tantiquite  (contained  in  his  Memoires  d*histoire  ancienne  et  de  philo- 
logie)  may  be  consulted  in  the  matter. — I  do  not  think  it  has  previously  been 
remarked  that  Vancher's  elaborately  developed  theory  that  Plutarch  was  the 
author  of  the  De  Subiimiiate  seems  to  break  down  (even  if  there  were  no  other 
objections  to  it)  in  the  presence  of  the  set  comparison  which  the  treatise 
contains  between  the  oratory  of  Demosthenes  and  that  of  Cicero.  At  the 
same  time,  the  evidently  close  relation  in  which  the  treatise  stands  to 
Caecilius  is  one  of  the  chief  reasons  for  rejecting  the  tradition  of  its  third- 
century  authorship. 

*  Hermes,  XII  333;  but  see,  on  the  other  hand,  Rohde  in  Rheinisches 
Museum,  XLI  176.— Suet.,  Aug.  89,  and  Quint.  IX  i,  12. 


Digitized  by  VjOOQ IC 


312  AMERICAN  JOURNAL    OF  PHILOLOGY. 

ApoUodorus  approached  the  whole  question  in  a  somewhat  narrow 
scholastic  spirit.  That  is,  at  all  events,  the  impression  we  form  of 
the  schools  headed  by  ApoUodorus  and  his  rival  Theodonis  of 
Gadara.^  They  lost  themselves  in  rhetorical  rules  and  subtleties.' 
Dionysius  and  Caecilius  seem  to  have  stood  on  an  altogether 
higher  plane.  They  were  true  men  of  letters,  not  mere  masters 
of  technic.  Their  view  of  literary  criticism  was  not  mechanical, 
but  aesthetic.  They  had  something  of  the  wide  outlook  and 
sympathy  possessed  by  the  best  Roman  writers,  such  as  Cicero, 
for  whom  the  adoption  of  a  pure  Attic  standard  had  a  living,  and 
not  simply  an  antiquarian  interest. 

W.  Rhys  Roberts. 

Univbrsity  Collbgb  op  Nohth  Walbs«  Banco*. 

*  Theodoras  was  the  teacher  of  Tiberius,  of  whom  he  gave  the  famous 
definition  Tr^^of  ai^ri,  irefvpafiivog  (Suet.,  Tiber.  57);  and  probably  also  of  the 
author  of  the  Treatise  on  the  Sublime  (cf.  the  use  of  tKd'Xei  in  the  reference  to 
him  in  De  SubL  III  5). 

'  Strabo,  p.  625 ;  koI  'Airo^^^apog  6  p^up  6  rdf  Tixvn(:  avyypdrJMi^  xai  n)v 
'A7roXh)66p€ioif  aipeaiv  irapayay^v,  ^ig  ttot'  kari  •  TroAAd  yap  kireKpaTei,  fuH^ova  6e 
f/  Koff  rjfiaQ  Ix^"^^  "^^  Kftiaiif,  uv  ion  xal  ^  * A7ro?2/}dC)peiog  alpeai^  koi  //  Qeo66peioc. 
Tacitus,  Dialogus  de  Oratoribus,  XIX :  iam  vero  longa  principiorum  praepa- 
ratio  et  narrationis  alte  repetita  series  et  multarum  divisionum  ostentatio  et 
mille  argumentoram  gradus  et  quidquid  aliud  aridissimis  Hermagorae  et 
Apollodori  libris  praecipitur,  in  honore  erat. 


Digitized  by  VjOOQ IC 


IV.— ARE  THE  LETTERS  OF  HORACE  SATIRES? 

Students  of  Horace  have  doubtless  often  observed  that  his 
allusions  to  his  own  satirical  muse  seem  to  give  it  a  character  of 
violence  and  acerbity  which  in  fact  it  does  not  reveal.  This  is 
most  conspicuous  in  those  satires  which  deal  more  or  less  directly 
with  questions  relative  to  his  own  literary  work,  notably  the  fourth 
of  the  first  book  and  the  first  of  the  second.  The  real  nature  of 
this  inconsistency  is  not  far  to  seek.  It  does  not  represent  so 
much  the  momentary  mood  of  the  writer  as  it  does  the  difference 
between  the  narrow  generalizations  of  the  literary  criticism  of 
antiquity  and  the  wider  facts  of  literary  practice.*  Thus,  from 
whatever  source,  whether  from  suggestions  of  Lucilius  himself 
or  from  the  criticism,  childishly  imitative  of  Greek  models,  of 
scholars  soon  after  his  time,  we  find  in  Varro  (whether  in  these 
words  or  not)  the  fixed  formulation  of  satire  as  a  camun  male- 
dicum  . .  .  archaeae  camoediae  charactere  campositum  (Diom.,  p. 
485).'  Horace  passes  the  coin  on  without  change  in  his  allusion 
to  Lucilius  in  the  beginning  of  Serm.  I  4,  but  in  H  i  he  shows 
that  he  has  a  very  genuine  appreciation  of  other  and  better 
qualities  in  the  earlier  satirist  than  those  in  which  the  critics 
had  comprehended  his  genius.  Yet  in  the  same  poem,  for  the 
purposes  of  the  situation  he  has  created,  he  plays  with  the  tradi- 
tional doctrine  of  the  acerbity  of  satire,  and  threatens  vengeance 
on  his  enemies  with  a  vehemence  which  is  un-Horatian  in  all  but 
its  sly  fun. 

The  facility  with  which  literary  judgments  became  fixed  and 
then,  regardless  of  correspondence  with  fact,  went  on  to  affect 
subsequent  practice,  is  one  of  the  most  remarkable  features  of 
ancient   literary  history.      That   only  .the  smallest    proportion 

'  It  is  my  purpose  at  another  time  to  consider  in  some  detail  the  attitude  of 
Horace  toward  ^the  literary  criticism  of  his  time  relative  to  satire. 

'  In  this  identification  criticism  may  have  been  influenced  not  only  by  the 
ovofxaoTi  KUfi^6eiv  of  Lucilius,  like  that  of  the  old  comedy,  but  also  by  the  fact 
that  Lucilius  criticised  and  ridiculed  the  work  of  the  chief  contemporary 
tragedian,  Accius,  in  a  manner  analogous  to  the  parody  of  Euripides  by 
Aristophanes.    Cf.  Horace,  Ser.  I  10,  53  and  scholia  ad  loc. 


Digitized  by  VjOOQ IC 


314  AMERICAN  JOURNAL    OF  PHILOLOGY, 

of  the  satires  of  Horace  betray  even  remotely  a  character 
of  personal  attack  did  not  afiect  or  modify  the  theory  of  satire, 
and,  living  up  to  it  as  the  times  would  allow,  Persius  "broke 
his  milk  teeth  on  Alcibiades  and  Dama,"^  and  Juvenal  would 
cause  the  chills  to  course  through  the  conscious  vitals  of  the 
evil-doer.  Though  Persius  has  given  us  a  true  and  admirable 
characterization  of  Horace,  he  still  pays  his  tribute  to  the  tradition 
of  satire  (I  107  ff.).  Juvenal  is  not  so  discriminating,  and  confuses 
the  tradition  with  the  poet;  for  when,  after  setting  forth  the 
wrongs  that  burn  in  his  heart  and  feed  his  wrath  (I  45),  he 
continues:  haec  ego  nan  credam  Venusma  dig7ia  lucerna,  haec 
ego  nan  agitemf  (vs.  51),  he  is  of  course  thinking  rather  of  the 
stereotyped  character  and  function  of  satire  than  of  the  practice 
of  the  bard  of  Venusia.  In  Horace  the  theory,  though  frankly 
expressed  in  various  places,  had  least  effect  on  his  practice,  and 
he  gradually  worked  farther  and  farther  away  from  it  in  the 
development  of  that  mild  philosophical  humor  and  playful  wit 
which  culminated  in  the  perfect  urbanity  and  charm  of  his 
Letters.  These  are  so  far  removed  from  the  tradition  of  the 
censorious  nature  of  satire  that  it  seems  to  have  been  forgotten, 
for  the  most  part,  that  they  were  ever  looked  upon  as  represen- 
tatives of  this  department  of  Roman  poetry.  Some  scholars  will 
perhaps  recall  that  Casaubon  protested  with  energetic  emphasis 
against  the  habit  of  his  time  of  considering  the  Letters  a  form  of 
poetry  separate  from  satire,  but  his  words  were  not  supported  by 
sufficient  evidence  to  carry  conviction,  and  apparently  the  pre- 
vailing belief  to-day  is  not  different  from  that  which  he  attacked 
three  hundred  years  ago.*  The  question  which  I  have  put  in  my 
title  is  therefore  not  a  new  one,  and  a  warning  is  due  to  the 
reader  of  much-abused  benevolence,  that  it  would  not  be  sum- 
moned into  court  again  if  the  writer  did  not  hope  that  the  use 
of  new  evidence  and  a  different  presentation  of  some  older  obser- 
vations might  lead  us  nearer  to  a  settlement  of  the  matter. 

While  editors  are  not  yet  agreed  whether  the  first  two  books  of 
Horace's  poems  in  hexameters  are  to  be  called  saiirae  or  sermones, 

^  Gildersleeve,  Persius,  Int.,  p.  xxii,  q.  v.  See  also  article  *  Satire'  in  John- 
son^s  Cyclopaedia  (new  edition). 

'Casaubon,  De  Satyrica  Graecorum  et  Rom.  Satira,  ed.  Rambach,  p.  229: 
Ferendi  non  sunt  qui  epistolanim  libros  satirarum  appellatione  ac  numero 
censerint  excludendos.  Casaubon  does  not  support  his  view  by  appeal  to 
ancient  evidence. 


Digitized  by  VjOOQ IC 


ARE  THE  LETTERS  OF  HORACE  SATIRES?  315 

it  does  not  appear  that  there  is  any  disposition  to  designate  the 
last  two  books  otherwise  than  episiuUu.  To  my  knowledge, 
indeed,  only  one  attempt  has  been  madein  recent  years  to  claim 
anothe£^^signation  for  the  Letters  or  a  part  df  them,  viz.  by'  O. 
Miiller,^wlio  endeavored,  with  most  futile  arguments,  to  prove 
that  Epp.  I  '^o  was  an  epilogue  to  three  books  of  satires  or 
sermakes.  Before  him,  Heinrich,  the  celebrated  editor  of  Juvenal, 
was  wont  to  claim  in  his  lectures  and  in  some  published  utterances 
that  the  general  title  Sermones  was  the  only  correct  designation 
of  the  poems  of  Horace  in  hexameters.*  But,  though  there  has 
been  agreement  in  practice,  it  is  doubtless  generally  known  that 
the  oldest  evidence  on  this  point  has  been  thought  to  render 
-uncertain  the  designation  of  the  Letters  at  the  time  of  their 
publication  and  in  the  century  or  more  succeeding  the  author's 
death.  In  consequence,  editors  have  very  wisely  been  content  to 
follow  tradition  as  embodied  in  the  superscription  of  the  MSS. 
The  status  of  the  question  is  given  clearly  by  Ribbeck  in  his 
introduction  to  the  Letters,  and  what  there  is  of  older  discussion 
is  alluded  to  and  reported  by  Diintzer.' 

For  the  name  episiulae  no  internal  evidence  is  available,  and  if 
there  are  any  allusions  by  Horace  to  his  Letters,  they  are  included 
by  him  with  his  earlier  works  under  the  designation  sermaneSj 
chosen  to  characterize  their  style,  approximating  to  that  of  prose. 
Of  external  evidence,  the  earliest  hitherto  used  includes  Sermones 
and  Epistulae  under  the  term  saiirae.  Thus  Suetonius,  in  his 
life  of  the  poet  (early  in  the  second  century),  confirms  his  state- 
ment that  Horace  was  short  and  fat  by  the  words  ut  a  semei  ipso 
in  saiiris  describiiur^  where  his  data  are  derived  from  Epp.  I  4, 
15;  I  20,  24,  and  Sat.  \\  3,309.  The  same  view  seems  to  be 
shared  by  Quintilian,  who,  while  alluding  to  Horace  as  a  writer 
of  lyric,  iambic  and  satirical  poetry,  does  not  make  separate 
mention  of  the  Letters.  Are  we  to  infer  that  he  meant  to  include 
the  Letters  in  his  allusion  to  satire,  or  did  he  pass  over  them 
either  thoughtlessly  or  intentionally,  as  irrelevant  to  his  purpose  ?  ^ 
Ribbeck  and  others  have  thought  that  they  are  included  in  the 

^  Ein  Begleitschreiben  des  Horaz  zu  seinen  Sennonen.    Prog.    Berlin,  1876. 

'  DQntzer,  Kritik  und  Erkl.  d.  Hor.,  vol.  Ill,  p.  70  and  note. 

'Des  Q.  Hor.  Flaccus  Episteln,  ed.  O.  Ribbeck,  Berlin,  1869,  p.  79. 
Dantzer,  1.  1. 

*  Orelli-Hirschfelder,  Prolegomena,  p.  xxxiv:  Quintilianns  cum  carmina 
Horatiana  recenset  satiras  iambos  lyrica  affert,  epistulas  omittit. 


Digitized  by  VjOOQ IC 


3l6  AMERICAN  JOURNAL    OF  PHILOLOGY. 

mention  of  satire ;  to  whom  Schiitz  has  made  answer,  not  without 
appearance  of  probability,  that  Quintilian  is  not  professing  to  give 
complete  lists  and  might  well  have  passed  over  the  Letters,  as  he 
has  omitted  mention  of  the  Bucolics  and  Georgics  of  Virgil. 
While  it  will  be  granted  that  this  point  is  incapable  of  positive 
demonstration,  I  think  that  a  consideration  of  the  matter  will 
show  that  Quintilian  very  probably  meant  to  include  the  Letters 
in  his  mention  of  Roman  satire.  In  the  first  place,  he  enumer- 
ates with  needless  fulness,  from  the  whole  range  of  classical 
poetry,  all  those  writers  who  will  be  of  service  to  the  orator  in 
the  formation  of  style.  Now,  it  is  obvious  that  for  his  purpose 
the  Letters  of  Horace  were  quite  as  valuable  as  his  Satires,  nor 
is  it  likely  that,  as  an  avowed  lover  of  the  poet  {nisi  fallor  amore 
eius),  he  would  have  neglected  to  mention  them  through  over- 
sight. That  he  recognized  in  them  a  particular  character  different, 
let  us  say,  from  the  Sermones,  which  rendered  them  unfit  for  his 
purpose,  no  one  would  contend.  As  for  Schiitz's  observation, 
that  with  equal  reason  we  might  argue  analogously  from  the 
omission  of  the  Bucolics  and  Georgics,  let  us  see.  Quintilian, 
following  an  Alexandrine  canon  of  scripiares  classici  more  or 
less  completely,  does  not  admit  as  serviceable  for  imitation  all 
whom  he  mentions  and  characterizes.  Thus  he  alludes  to  The- 
ocritus in  these  words:  Admirabilis  ifi  suo  genere  Theocritus^ 
sed  musa  ilia  tustica  el  pastor alis  nan  forum  modo,  verum  ipsam 
etiam  urbem  re/ormidat  (X  i,  55).  Now,  that  Quintilian  does 
not  refer  to  the  Georgics  and  Bucolics  (musa  rustica  et  pastor  alis)  ^ 
which  must  have  been  dismissed  in  almost  the  same  words,  will 
not  cause  surprise.  Of  important  cla<;sical  Latin  poetry,  only  the 
minor  works  of  Virgil  and,  presumably,  the  Letters  of  Horace 
are  not  mentioned.  The  reason  for  the  omission  of  the  former 
has  been  shown :  that  the  latter  were  also  omitted  is  a  priori 
improbable ;  nor  is  their  omission  susceptible  of  any  explanation 
analogous  to  the  case  of  the  Bucolics  and  Georgics.  We  shall 
consider  it  therefore  as  highly  probable  that  Quintilian  included 
the  Letters  of  Horace  in  his  treatment  of  Roman  satire^  From 
late  antiquity  there  is  finally  the  harmonious  evidence  of  two 
witnesses,  the  scholiast  Porphyrio  (third  or  fourth  century)  and 
the  Gallic  writer  Sidonius  Apollinaris  (middle  of  the  fifth  century). 
Porphyrio  ad  Serm.  I  i,  i :  quamvis  saturam  esse  opus  hoc  suum 
Horatius  ipse  confiteatur,  cum  ait  'Sunt  quibus  in  satira  videdr 
nimis  acer  et  ultra  legem  tendere  opus^  tamen  proprios  titulos 


Digitized  by  VjOOQ IC 


ARE  THE  LETTERS  OF  HORACE  SATIRES?  317 

voluit  ei  accamodare.  nam  has  priores  duos  libros  Sermonum, 
posteriores  Epistularum  inscripsit  Sidonius  carm.  IX  221  if.: 
non  quod  per  saHras^  epistularum  sermonumque  sales, . .  .  voluit 
sonare  Flaccus. 

A  word  in  explanation  of  this  evidence  before  going  further. 
If  the  error  had  not  been  made  by  many  scholars,  it  would 
scarcely  seem  necessary  to  warn  the  reader  that,  because  certain 
works  were  looked  upon  as  satires,  they  did  not  therefore  neces- 
sarily bear  this  title.  Now,  none  of  this  evidence  shows  that  any 
one  of  these  four  witnesses  believed  that  the  Letters  of  Horace 
bore  the  title  Satirae.  It  only  shows  that  they  considered  that 
the  Letters  were  satires.  Quintilian  touches  on  satire  as  a  form 
of  poetry.  He  does  not  mention  separate  works,  but  only  writers. 
He  does  not,  therefore,  concern  himself  as  to  whether  various 
collections  of  satires  were  called  sermones,  epistulae^  Menippeae, 
or  what  not,  any  more  than  in  his  treatment  of  the  elegy  he  is 
concerned  about  particular  titles,  which  obviously  could  be  chosen 
according  to  the  author's  fancy,  amoresy  tristia^  etc.  Suetonius 
cites  with  the  formula  in  satiris;  but  it  is  significant  that  the  facts, 
for  the  verification  of  which  he  makes  this  reference,  are  not 
contained  in  a  single  one  of  the  two  groups,  i.  e.  in  either  the 
sermones  or  in  the  epistulae  (as  Ribbeck  implies  in  citing  only 
Epp.  I  4  and  I  20),  but  in  both,  as  pointed  out  above.*  If  the 
data  to  which  he  refers  had  been  contained  alone  in  either  the 
Sermones  05  the  Epistulae,  he  might  have  referred  to  either 
group  conceivably  by  its  particular  title.  But  because  he  referred 
for  verification  of  his  statement  to  data  which  were  derived  from 
both  Sermones  and  Epistulae,  he  writes  in  satiris.  I  would  not 
seem  to  know  the  impossible  about  Suetonius'  intention;  my 
purpose  is  only  to  point  out  that  his  formula  of  citation  does  not 
mean  that  he  only  knew  the  Letters  as  Satirae,  nor  that,  on  the 
other  hand,  he  thoughtlessly  attributed  to  the  Satires  what  was 
due  to  the  Letters;  for  these  have  been  the  two  explani^tions 
offered.  What  has  already  been  said  in  explanation  of  Sueto- 
nius' allusion  to  the  Letters  as  satires,  and  of  Quintilian's  inclusion 
of  them  in  his  treatment  of  Roman  satire,  is  expressly  given  us 
by  Porphyrio.  His  words,  so  far  from  indicating  that  both 
Sermones  and  Epistulae  bore  a  common  title  Satirae,  explain 
rather  the  absence  of  that  title.      'Although   Horace   himself 

^  As  evidence  for  the  poet's  shortness  of  stature,  Serm.  II  3,  309  is  a  much 
more  emphatic  utterance  than  Epp.  I  20,  24. 


Digitized  by  VjOOQ IC 


3l8  AMERICAN  JOURNAL    OF  PHILOLOGY. 

professes  that  his  works  belong  to  the  poetical  genus  satire, 
nevertheless  he  (did  not  call  them  so  but)  gave  them  special 
titles,  Sermones  and  Epistulae' — a  paraphrase  of  the  words  of 
Porphyrio  which  I  think  will  not  be  disputed,  and  which  is 
confirmed  by  his  note  ad  Serm.  \\  \,  \\  has  duos  libros  cum 
Sermonum  inscripserii,  tamen  de  his  sic  loquitur  quasi  de  satura 
LucUium  sequens.  Finally,  in  Sidonius  we  have  the  same  point 
of  view,  per  satiras^  epistularum  sermanumque  sales,  the  generic 
name  followed  by  the  specific  titles,  as  if  he  had  said  of  Ovid, 
per  ele/^os,  tristtum  epistularumque  maesiiiiam.  I  have  stated 
the  problem  and  explained  the  evidence  hitherto  employed  with 
perhaps  unnecessary  fullness,  in  order  to  place  in  intelligible 
setting  a  bit  of  evidence  for  the  antiquity  of  the  sub-title  Epistulae 
and  the  relation  of  the  Letters  to  the  Satires,  which  has  not  to 
my  knowledge  been  employed  in  this  question. 

Statius,  in  Silv.  I  3,  describes  with  elaborate  and  obscure  detail 
the  charms  of  the  villa  of  Manilius  Vopiscus  at  Tibur.  The 
poem  is  not  without  Horatian  reminiscences,  as  would  be 
expected,  but  I  shall  not  endeavor  to  point  them  out  in  detail. 
I  will  only  say  that  in  general  the  poem  is  a  detailed  and  diffuse 
commentary  on  Horace's  me  nee  lam  peUiens  Lacedaeman  . .  • 
percussil . .  .  quxim  damus  Albuneae  resonaniis  (Carm.  I  7),  which 
it  recalls  and  alludes  to  clearly  in  vss.  83-9,  where  other  places 
famed  in  story  are  bidden  to  yield  before  its  charms:  cedanl 
TeUgoniy  cedanl  Laureniia  Tumi  iugera  etc. — this  apparently 
with  conscious  allusion  to  Horace's  laudabunl  alii^  of  the  poem 
just  quoted,  in  reference  to  the  same  place.  In  this  very  fact  we 
have  suggested  by  implication,  not  very  subtle,  a  comparison 
between  Vopiscus  and  Horace.  That  such  comparison  lay  near 
at  hand  for  the  poet  is  obvious.  Vopiscus  was  a  man  of  literary 
ambitions  who  had  apparently  dabbled  in  poetry  enough  to  give 
some  shadow  of  support  to  the  flattering  suggestion,  of  which 
Statius,  in  writing  of  Tibur,  the  favorite  country  residence  of 
Horace,  could  scarcely  fail  to  avail  himself.^  It  will  therefore,  I 
think,  be  obvious  to  every  reader   that  the  comparison  thus 

'  Concerning  the  literary  activity  of  Vopiscus,  aside  from  the  passage  about 
to  be  discussed,  we  have  only  the  following  mention,  Statius,  Silv.  I,  pref.: 
Manilius  Vopiscus  vir  eruditissimus  et  qui  praecipue  vindicat  a  situ  litteras 
iam  paene  fugientes.  And  in  this  poem,  vss.  20-33:  ipse  Anien  . . .  ponit  | 
rourmura,  ceu  placidi  veritus  turbare  Vopisci  |  Pieriosque  dies  et  alentes 
carmina  somnos. 


Digitized  by  VjOOQ IC 


ARE   THE  LETTERS  OF  HORACE  SATIRES?  319 

suggested  is  carried  out  in  the  following  passage,  where  after  a 
series  of  philosophical  reflections  in  Horatian  style,  these  lines 
occur: 

99    Hie  tua  Tiburtes  Faunos  chelys  et  iavat  ipsum 

Alciden  dicturoque  lyra  maiore  Catillam  ; 

Seu  tibi  Pindaricis  animus  contendere  plectris 

Sive  chelyn  toUas  heroa  ad  robora  sive 

Liventein  satiram  nigra  rubigine  vibres 
104    Seu  tua  non  alia  splendescat  epistola  cunu 

Digne  Midae  Croesique  bonis  et  Perside  gaza, 
106    Macte  bonis  animi !  ^ 

In  these  verses  we  have  first  (vss.  99  and  100)  a  general  state- 
ment of  poetical  activity  followed  by  the  alternatives  of  form 
which  it  might  assume.    The  general  statement  would,  it  will  be 

'  In  explanation  of  the  literary  form  of  this  passage,  it  is  perhaps  worth 
while  to  point  out  that  it  may  be  a  reminiscence  of  Horace,  Epp.  I  3>  33 ; 

Seu  linguam  causis  acuis  seu  civica  iura 
Respondere  paras  seu  condis  amabile  carmen. 
Prima  feres  hederae  victricis  praemia—- 

an  observation  which  would  suggest  that  in  the  vss.  of  Statius  a  period  is  not 
to  be  placed  after  cura^  as  is  done  by  all  editors,  but  that  a  period  (or  colon) 
should  be  placed  after  CatiUum  (vs.  100),  and  that  the  various  seu  {sive)  clauses 
should  be  carried  over  to  the  following  vss.:  digne  . . .  macte  bonis  animi. 

Statius,  with  all  his  versatility,  was  not  able  to  give  much  variety  to  the 
same  theme,  and  we  are  not  therefore  surprised  in  Silv.  II  3  (Villa  Surrentina 
Pollii  Felicis),  among  many  other  resemblances  to  this  poem,  to  find  these 
lines  (vss.  11 3-1 5): 

Hie  ubi  Pierias  exercet  Pollius  artes, 
Seu  voluit  monitus,  quos  dat  Gargettius  auctor, 
Seu  nostram  quatit  ille  chelyn  seu  dissona  nectit 
Carmina  sive  minax  ultorem  stringit  iambon. 

Finally,  our  passage  has  been  imitated  by  Ausonius,  Commemoratio  Profes- 
sornm  Burdigalensium,  I  I3  (to  Minervius): 

Teque  canam  de  te,  non  ab  honore  meo. 
Sive  panegyricis  placeat  contendere  libris 

In  Panathenaicis  tu  numerandus  eris; 
Seu  libeat  fictas  ludorum  evolvere  lites, 

Ancipitem  palmam  Quintilianus  habet. 

In  this  passage  the  idea  of  comparison  with,  or  emulation  of,  illustrious 
predecessors,  which  we  have  seen  is  contained  by  implication  in  Statius,  is 
given  direct  expression.  Both  are  modifications  of  the  motive  furnished  by 
Horace. 


Digitized  by  VjOOQ IC 


320  AMERICAN  JOURNAL    OF  PHILOLOGY. 

seen,  only  suggest  epic  poetry — a  circumstance  to  which  I  shall 
return.  Of  the  various  poetical  forms  which  succeed  and  with 
which  the  poet  imagines  Vopiscus  as  occupied,  it  will  be  observed 
that  the  reference  to  lyric  poetry  is  made  with  obvious  allusion  to 
Horace's  Pindarum  quisguis  studet  aemulari  (Carm.  IV  2). 
Concerning  the  second  form  (vs.  102),  a  word  presently.  The 
characterization  of  satire  in  vs.  103  is  thoroughly  conventional  in 
content,  with  phraseology  drawn  from  reminiscence  of  Serm.  I  4, 
100 :  hie  nigrae  sucus  lolligints^  kaec  est  aerugo  mera,  from  which 
passage  (1. 1.  92)  the  conventional  epithet  liveniem  might  also 
have  been  drawn.  Verse  104  is  a  locus  vexatus  of  long  standing, 
because  of  the  supposed  obscurity  of  the  words  nan  alia  . .  .  cura, 
Lindenbrog  and  Gevartius  conjectured  nan  alta^  which  Markland 
and  many  others  accepted.  Hand  ad  loc.  says:  Omnes  libri 
antiqui  exhibent  nan  alia^  quod  non  cum  Barthio  explicuerim, 
*  non  alia  cura  quam  quae  epistolarum  stilum  decet,'  sed  ita  ut 
sententia  in  laudem  Vopisci  dicta  sit:  si  tua  epistola,  quamvis 
levius  carminum  genus,  eadem  diligentia,  quam  in  altiori  genere 
miramur,  adhibita  splendescat.  Both  of  these  explanations  call 
for  an  ellipsis  of  thought:  that  of  Barth  revolving  in  circular 
absurdity,  that  of  Hand  unnatural  and  tortuous,  suggested  in 
well-known  philological  manner  by  the  conjecture  nan  alia. 
There  is,  of  course,  commonly  with  alius  a  slight  ellipsis  of 
variable  character,  yielded  by  the  immediate  context.  Here  it 
is  simplicity  itself,  and  nan  alia  is  nothing  more  than  a  litotes 
for  eadem,  with  immediate  reference  to  the  characterization  of 
satire  which  has  preceded,  i.  e.  cura  nan  alia  atque  in  satira 
canscribenda  adhibita.  But  one  must  have  a  better  opinion  of 
the  latinity  of  commentators  on  Statins  than  to  believe  that  they 
have  failed  to  see  so  obvious  a  thing.  The  difficulty,  I  fancy,  has 
been  that  the  sense  thus  naturally  yielded  seemed  to  involve  them 
at  once  in  a  literary  question  of  still  greater  difficulty,  viz.  the 
assumption  that  satire  and  the  poetical  epistle  were  the  same  in 
stylistic  character.  That  this  difficulty  is  a  genuine  one  we  may 
grant,  if  these  words  are  to  be  thought  of  as  a  general  character- 
ization of  the  poetical  epistle.  But,  obviously,  they  are  not 
general,  but  make  specific  allusion  to  the  relation  which  the  poet 
(in  conformity  with  our  other  witnesses  from  antiquity)  under- 
stood the  Epistles  of  Horace  to  bear  to  the  poetical  genus  satire. 
The  whole  attitude  of  mind  is  given  for  us  succinctly  enough  by 
Porphyrio  ad  Epp.  I  i,  and  his  words  are  a  sufficient  commentary 


Digitized  by  VjOOQ IC 


ARE  THE  LETTERS  OF  HORACE  SATIRES?  321 

on  the  passage  of  Statius :  Placet  episiularum  libri  iiiulo  iantum 
dissimiles  a  sermonum  sunt,  nam  ei  metrum  ei  materia  verborum 
et  communis  adsumptio  eadem  est  Thus,  in  the  correct  and 
simple  interpretation  of  the  words  of  Statius  we  have  confirmed 
the  view  expressed  above  of  an  implied  comparison  between  the 
literary  dilettante  Vdpiscus,  in  his  villa  at  Tibur,  and  Horace, 
who  by  his  residence  and  his  verse  had  given  the  place  its 
literary  associations. 

I  know  that  scholars,  so  far  as  published  utterances  are  known 
to  me,^  have  universally  used  this  passage  as  evidence  that 
Vopiscus  was  the  versatile  author  of  lyric,  epic,  satirical  and 
epistolary  poems.  But  to  interpret  such  allusions  of  the  flatter- 
ing poet  to  the  trifling  literary  labors  of  men,  whose  petty 
attempts  ranged  the  whole  gamut  of  literary  forms,*  as  cold  facts 
for  the  reconstruction  of  literary  history  displays  a  singular  lack 
of  imagination.  The  words  are  no  more  than  a  general  and 
flattering  definition  of  literary  activity  drawn  from  the  suggestion 
of  a  comparison  with  Horace  which  Vopiscus*  residence  at  Tibur 
afforded.  That  to  the  literary  activity  of  Horace  the  epos  is 
added  may  have  been  due  to  some  more  ambitious  effort  of 
Vopiscus  in  this  direction  which  called  for  special  allusion — ^a 
supposition  which  would  explain  the  fact  above  alluded  to,  that 
we  can  only  understand  the  first  general  mention  of  his  literary 
work  (vss.  99  and  loo)  to  refer  to  the  epos.  For  the  period  and 
for  the  circle  in  which  Statius  moved  nothing  would  be  more 
probable.' 

From  this  passage,  therefore,  we  gain  with  reasonable  certainty 
an  allusion  to  the  designation  Epistulae  as  applied  to  Horace's 
Letters,  accompanied  by  a  recognition  of  their  intimate  relation 
to  the  poetical  form  satire.  It  remains,  therefore,  to  consider 
whether  this  piece  of  evidence  can  be  put  in  harmony  with  the 
other  data  from  antiquity  bearing  on  this  point  and  already 
considered. 

Concerning  the  original  title  of  the  two  books  of  satires, 
scholars  are  not  yet  agreed  as  between  Satirae  and  Sermones. 

*  ForccUini-De  Vit,  Onam.  s.  voc.  Manilius  ( Vopiscus) ;  Ddlling,  Prog.  Plauen, 
1838,  p.  13;  Friedlander,  Sittengeschichte,  vol.  Ill,  453;  Teuffel-Schwabe, 
paragraph  324,  3. 

«Cf.  Martial.  Ill  20.  V  30,  XII  94 ;  Pliny,  Epp.  IV  3,  VI  21. 

'**  Von  alien  Gattungen  aber  darfte  die  epische  diejenige  gewesen  sein,  der 
sich  die  Meisten  zuwandten,*'  Friedlander,  1. 1. 


Digitized  by  VjOOQ IC 


322  AMERICAN  JOURNAL    OF  PHILOLOGY. 

The  MSS  of  Keller  and  Holder  without  exception  present 
sermonum  libri^  while  the  oldest  Blandinian  is  reported  to  have 
contained  the  title  eclogarum.  But,  on  the  other  hand,  the  latest 
editors,  Kiessling,  Orelli-Mewes  and  Hertz,  have  chosen  the  title 
Satirae,  which  is  plausibly  defended  by  Kiessling  and  put  in 
relation  to  Sermones  thus :  "  Following  Lucilius,  Horace  entitled 
these  poems  saiirae.  The  poetical  genus  to  which  they  belong 
he  designates  as  satira  by  beginning  the  second  book  with 
suni  quibus  in  scUira  videor  nimU  acer;  satira^  are  the  single 
poems:  quid prius  iUustrem  saHris  (II  6,  17).  Not  until  later, 
when  lyric  poetry  had  become  the  central  point  of  his  poetical 
productivity,  did  he  designate  them,  in  contrast  to  his  carminay 
as  sermones,  in  order  to  characterize  thereby  their  form,  verging 
on  that  of  prose"  (Hor.  Satiren,  p.  xii).  But,  clearly,  it  does  not 
at  all  follow  that  because  Horace  alluded  to  his  own  works  as 
satirae,  that  he  therefore  gave  them  this  title.  The  truth  is  that 
saiira  is  a  definite  poetical  genus  comparable  to  other  depart- 
ments of  poetry,  such  as,  for  example,  the  elegy;  but  it  was  no 
more  necessary  that  a  collection  of  satires  should  be  entitled 
saiirae  than,  for  instance,  that  a  collection  of  elegies  should  bear 
the  title  elegi.  Let  us  illustrate.  The  {Epistulae')  ex  Ponio  are 
elegies.  In  the  fifth  elegy  of  book  four  we  read  He  leves  elegi 
doctas  ad  consulis  aures,  and  I  doubt  not  if  we  had  no  other  or 
uncertain  evidence  for  the  title,  we  should  from  this  passage 
construct  elegarum  libru  Now,  the  relationship  between  soHrae 
(the  name  of  the  poems  as  representatives  of  the  poetical  genus 
satire)  and  sermones  (the  title  chosen  to  indicate  their  form)  is 
exactly  the  same  as,  in  our  example,  between  eiegi  and  epishilae 
— that  is,  between  the  generic  and  the  specific.  Horace  was  at 
liberty  to  refer  to  his  poems  by  either  name,  just  as  Ovid  does. 
Therefore,  while  we  grant  that  Horace  might  have  entitled  his 
satires  saiirae,  we  must  deny  that  this  is  a  correct  inference  from 
the  fact  that  he  makes  use  of  the  word  in  allusion  to  them.  But 
is  sermones  better  attested  ?  It  seems  to  me  clear  that  it  is.  Our 
MSS  afford  it  and  Porphyrio  certainly  found  it,  and  no  other  title, 
not  only  in  his  MSS,  but  in  the  genera]  literary  tradition  of  his 
time  (v.  supra,  p.  316). 

In  fact,  just  as  the  title  (Epistulae)  ex  Ponto  was  applied  by 
Ovid  to  his  last  four  books  of  elegies  from  Tomi  because  of  their 
form,  so  the  titles  Sermones  aM  Epistulae  may  have  been  given 
by  Horace  to  his  different  books  of  satires,  as  indicating  in  a 


Digitized  by  VjOOQ IC 


ARE  THE  LETTERS  OF  HORACE  SATIRES?  323 

general  way  the  different  forms  of  the  musa  pedesiris  which  he 
had  chosen.^  Because  in  his  later  satires  he  had  chosen  a  more 
personal  form  of  utterance  for  his  reflections  on  life  and  literature, 
he  may  have  employed  for  them  the  name  episiulae,  the  writings 
themselves  differing,  to  be  sure,  in  the  maturity  and  perfection  of 
their  thought  and  execution,  from  the  earlier  works,  but  not 
essentially  in  range  of  matter  or  method  of  treatment.'  That 
both  were  considered  in  antiquity,  so  far  as  our  scanty  record 
enables  us  to  determine,  to  be  representatives  of  the  department 
of  poetry  known  as  saiira,  we  have  seen  from  the  unanimous 
testimony  of  Quintilian,  Suetonius,  Porphyrio  and  Sidonius 
ApoUinaris,  to  whom  we  may  now  add  Statins,  who  also  affords 
us  the  separate  designation  epistula,  for  which  hitherto  Porphyrio. 
has  been  esteemed  the  earliest  witness^  Final  confirmation  of 
the  correctness  of  this  view  is  afforded,  I  believe,  by  Horace 
himself  in  a  well-known  passage  (Epp.  II  2,  58) : 

Denique  non  omnes  eadem  mirantur  amantque : 
carmine  tu  gaudes,  hie  delectatur  iambis; 
iUe  Bioneis  sermonibus  et  sale  nigro. 

Here  we  have  a  threefold  division  of  the  poet's  literary  work 
corresponding  exactly  to  that  afforded  by  Quintilian.  That  the 
satires  (by  which  I  mean  the  Sermones  and  Epistulae)  are 
designated  sermones  is  due  to  the  contrast  with  carmine  which 
the  antithetical  structure  of  the  passage  demands,  and  perhaps 
also  to  the  epithet  Bioneis^  suggesting  an  equivalent  for  diarpifial 
or  X($yoi.  Finally,  in  the  words  of  characterization,  sale  nigro,  we 
have  the  last,  and  in  the  Epistles  perhaps  the  only,  expression  of 
that  inconsistency,  to  which  I  alluded  at  the  beginning  of  this 
paper,  between  the  conventional  phraseology  of  literary  criticism 

^  Od  the  prose  serma  and  epistula  Qaintilian  has  some  very  good  observations 
which  in  many  respects  afford  an  admirable  characterization  of  the  corres- 
ponding works  of  Horace.  Inst  Or.  IX  4, 19 :  Est  igitur  ante  omnia  oratio 
alia  vincta  atque  contexta,  soluta  alia,  qualis  in  sermone  et  epistulis  . . .  quod 
non  eo  dico,  quia  non  illud  quoque  solutum  habeat  suos  quosdam  et  forsitan 
difficiliores  etiam  pedes ;  . . .  sed  non  fluunt  nee  cohaerent  nee  verba  verbis 
trahunt,  ut  potius  laxiora  in  his  vincla  quam  nulla  sint. 

*  Compare  the  words  of  Porphyrio  above  cited:  titulo  tantum  dissimiUs  sunt 
Ovid  in  the  beginning  of  the  Epistulae  ex  Ponto  (I  i,  16)  says  of  them  in 
relation  to  the  Tristia : 

Non  minus  hoc  illo  triste,  quod  ante  dedi. 
Rebus  idem,  titulo  dififert. 


Digitized  by  VjOOQ IC 


324  AMERICAN  JOURNAL   OF  PHILOLOGY. 

relative  to  satire  and  literary  practice.  Certainly,  the  Letters  are 
not  'caustic  potash'  (Greenough),  and  there  is  not  more  than  a 
bookish  suspicion  of  it  in  the  Sermones.  But  the  characterization 
belongs  to  satire,  and  since  the  Epistles  and  Sermones  were 
satires,  they  must  therefore,  though  innocent,  bear  the  opprobrium 
of  their  class. 

Our  estimate  of  the  literary  activity  of  Horace  is  scarcely 
affected  by  this  result.  For  him  it  has  mattered  little  that  the 
Letters  have  most  commonly  been  treated  as  a  separate  literary 
form.  But  for  the  history  of  Roman  satire  it  cannot  be  without 
significance  that  in  antiquity  the  Epistulae  were  reckoned  with 
the  Sermones  as  representatives  of  the  poetical  form  satira. 

G.  L.  Hendrickson. 


Digitized  by  VjOOQ IC 


v.— NOTES  ON  HORACE. 

Carm.  IV  3.  17-20: 

O  testudinis  aureae 
dukem  quae  strepitum^  Fieri,  temperas, 

O  mutis  quoque  piscibus 
donatura  cycni,  si  libeat,  sonum. 

The  words  dukem  . . .  sirepilum  deserve  more  attention  than 
they  have  thus  far  received  from  editors.  Wickham,  Schiitz, 
Kiessling,  Macleane,  Smith  and  Gow  are  silent  concerning  them. 
Dillenburger  remarks  simply  that  sirepitum  is  used  here  ''de 
grato  citharae  sonu,  ut  Ep.  I  2,  31.  14,  26."  With  this  the  com- 
ment in  Orelli-Mewes  is  practically  identical.  Page  speaks  at 
greater  length :  ^^sirepiius  being  almost  invariably  used  of  a  'din/ 
'noise/  e.  g^fori^  Romae,  valvarum,  januae  sirefiiius,  there  is  a 
tendency  to  take  dulcem  proleptically  here,  and  construe  'that 
dost  modulate  into  sweetness  the  lyre's  sound/  but,  as  Ep.  I  2, 
31  ad  strepitum  citharae  cessatum  ducere  curam,  the  word  is 
clearly  =  'music/  it  is  perhaps  simpler  to  render  here  'that  dost 
rule  the  sweet  music' "  This  utterance  seems  at  once  inadequate 
and  erroneous.  I  believe  that  Horace's  full  thought  can  be  felt 
only  by  taking  sirepitum  in  its  strictest  sense,  and  dulcem  as 
strongly  proleptic.  In  this  view  sirepitum  will  supply  the  foil  to 
mutis  in  v.  19.  The  lyre,  left  to  itself,  can  produce  nothing  but 
inharmonious  din,  even  as  the  fish  in  their  natural  state  can  make 
no  articulate  sound.  It  is  in  reducing  this  inharmonious  din  of 
the  lyre  to  harmony  that  the  power  of  the  Muse  is  shown,  even 
as  she  might  show  her  power,  should  she  feel  so  disposed,  in 
giving  voice  to  the  fish. 

Let  us  begin  by  examining  the  use  of  the  word  sirepiius  in 
Horace.  In  Odes,  I  15.  18  it  is  used  of  the  din  of  battle;  in  III 
19.  23  of  the  noise  of  a  carouse;  in  III  29.  12  of  the  din  and 
bustle  of  Rome  (cf.  Epp.  II  2.  79;  Verg.  Aen.  I  422).  Cf.  also 
Odes,  III  10.  5  Audis  quo  strepitu  ianua,  quo  nemus  . . .  remugiat 
ventis;  Sat.  I  2.  127  ff.  Nee  vereor  ne  . .  .  Ianua  frangatur,  latret 
canis,  undique  magno  Pulsa  domus  strepitu  resonet;  Sat.  II 6. 112 


Digitized  by  VjOOQ IC 


326  AMERICAN  JOURNAL    OF  PHILOLOGY.  . 

subito  ingens  Valvarum  strepitus  lectis  excussit  utrumque ;  Epp. 

I  17.  7  pulvis  strepitusque  rotarum ;  Epp.  II  i.  202  Garganum 
mugire  putes  nemus  aut  mare  Tuscum,  Tanto  cum  strepitu  ludi 
spectantur ;  A.  P.  80  ff.  pedem  . . .  populares  vincentem  strepitus. 
In  these  ten  passages  sirepiius  unmistakably  denotes  an  inharmo- 
nious, unpleasant  noise.     The  verb  sirepere  occurs  twice :  Carm. 

II  I.  17  lam  nunc  minaci  murmure  cornuum  Perstringis  auris^ 
iam  litui  strepunt;  IV  12.  3  nee  fluvii  strepunt  Hiberna  nive 
turgidi. 

Two  passages  remain,  besides  the  one  with  which  this  note  is 
especially  concerned.    One  is  Epp.  I  2.  27  ff.: 

Nos  numerus  sumus  et  fruges  consumere  nati, 
sponsi  Penelopae  nebulones  Alcinoique 
in  cute  curanda  plus  aequo  operata  iuventus, 
cui  pulchrum  fuit  in  medios  dormire  dies  et 
ad  strepitum  citharae  cessatum  ducere  somnum. 

Surely  sirepHum  is  *  din '  here.  Note  in  proof  of  this  assertion 
the  irony  that  marks  the  whole  passage.  Cf  especially  sponsi 
*  (would-be)  spouses,'  the  colloquial  in  cute  curanda,  and  pul- 
chrum. The  same  irony  appears  in  Epp.  I  14.  25,  26,  where 
Horace  says  to  his  vilicus,  "You  long  for  the  town,  simply 
because  you  can  find  no  meretrix  tibicina,  cuius  Ad  strepitum 
salias  terrae  gravis,  **to  whose  clatter  you  can  dance  like  a  lout.'^ 
In  every  passage  thus  far  considered  we  have  but  one  meaning, 
that  of  *din,  noise,  clatter.'  Why,  then,  should  we  assume  for 
this  one  passage  a  meaning  that  Horace  nowhere  else  exhibits  ? 
Further,  there  can  be  no  question  that  by  taking  strepitum  here 
as  we  must  take  it  everywhere  else  in  Horace,  we  get  the  most 
effective  interpretation,  strepitum  will  then  suggest  the  adver- 
sative notion  which  b  so  clearly  present  in  miUis,  and  the 
proleptic  dulcem  will  be  the  antithesis  to  donatura  .  . .  cycni 
sonum.  The  whole  thought  will  then  be:  "O  thou,  who  dost 
attune  to  sweetness  the  sounds  of  the  lyre,  musicless  though  they 
are  by  nattu-e,  O  thou  who  standest  ready  to  give  to  the  fish, 
even  voiceless  though  they  are,  the  swan's  song,"  etc. 

If  we  look  to  the  usage  of  strepitus  in  other  authors,  we  find 
such  phrases  as  strepitus y  fremitus ,  clamor  tonitruum;  strepitus , 
crepitus,  sonitus,  tonitrus;  inter  strepitum  tot  bellorum,  Vergil's 
usage,  as  that  of  a  contemporary  of  Horace,  is  especially  inter- 
esting. Cf.  Aen.  I  422  (cited  above'),  I  725  (of  the  din  of  conver- 
sation), VI  559  (of  the  horrid  sounds  in  Tartarus) ;  see  also  VI 


Digitized  by  VjOOQ IC 


NOTES  ON  HORACE.  32/ 

865,  IX  394,  G.  II  492.  strepere  occurs  in  Eel.  IX  36;  Aen.  VI 
709.  VIII  2,  IX  808,  X  568,  regularly  (save  perhaps  Aen.  VI  709 
strepit  omnis  murmure  campus,  said  of  bees),  of  a  decidedly 
unpleasant  noise.  How,  I  ask,  could  a  Roman  have  understood 
strepiium  in  our  Horatian  passage  save  as  has  been  maintained  in 
this  paper  ? 

In  Petronius,  §33,  there  is  a  phrase  which,  at  first  sight,  seems 
to  support  Mr.  Page's  view.  The  passage  runs  thus :  gustantibus 
adhuc  nobis  repositorium  allatum  est  cum  corbe,  in  quo  gallina 
erat  lignea  patentibus  in  orbem  alis,  quales  esse  solent  quae 
incubant  ova.  Accessere  continuo  duo  servi  et  symphania  sire- 
pente  scrutari  paleam  coeperunt  erutaque  subinde  pavonina  ova 
divisere  convivis.  But  here  again  allowance  must  be  made  for 
the  narrator's  tone,  which  is  ironical.  That  Encolpius  did  not 
enjoy  the  everlasting  music  in  Trimalchio's  house  appears  plainly 
from  §31,  especially  the  words:  Paratissimus  puer  non  minus 
me  acido  cantico  excepit  .  .  .  pantomimi  chorum,  non  patris 
familiae  triclinium  crederes.  We  must  interpret  acido  here  in  the 
light  of  §68:  Servus  qui  ad  pedes  Habinnae  sedebat,  iussus, 
credo,  a  domino  suo  proclamavit  subito  canora  voce :  .  .  .  nullus 
sonus  unquam  acidior  percussit  aures  meas  ("  Nie  verletzte  ein 
widrigerer  Ton  mein  Ohr,"  Friedlaender).  The  spirit  oi  sirepenie 
symphania^  §33,  is  well  given  by  Friedlaender's  rendering  :  **unler 
rauschender  Musikbegleitung." 

Another  passage  that  resembles  ours  is  Vergil,  Aen.  Ill  69,  70 
Inde,  ubi  prima  fides  pelago,  placataque  venti  Dant  maria  et  Ufiis 
crepitans  vocat  Auster  in  ahum,  but  the  idea  in  crepitans  is  plainly 
that  of  *  rustling,  noise,'  not  *  music,'  and  the  emphasis  is  on 
placata  and  lenis.  A  far  closer  parallel  is  to  be  found  in  Sopho- 
cles, Ajax  1 199  ff.,  where  it  is  said  of  the  man  '*who  first  taught 
the  Greeks  to  wage  confederate  war  with  hateful  arms"  that 

oifrr  paB€iav  KvktKnv 

PUfttV  fflOl  Tfp^lV  6fu\u>f, 
oCt€  yXvKvv  aiiXwk  Sropoy. 

Jebb  (edition  of  1896)  compares  Aeschylus,  P.  V.  574  Krjp6v\aaros 
oro/3ct  d6pa(.  The  references  in  Liddell  and  Scott  show  that  orofiot 
=  sirepiius.    In  language,  then,  dulcem  strepiium  (iyrae)  is  a 

*  For  convenience  I  give  Jebb*s  text  througliout. 


Digitized  by  VjOOQ IC 


328  AMERICAN  JOURNAL    OF  PHILOLOGY, 

close  equivalent  of  yXv#cw  (auXSv)  5ro/3ov.    Cf.  further  Pindar,  Pyth. 

X  39  (Gildersleeve)  vam^  dc  xop®^  napBtv^y  \  Xvpav  re  /3oai  Kapaxai  r 
ahXStv  dov€OVTat  ]  id.  Olymp.  Ill  8  fioay  avKAp;  id.  Nem.  V  38  KoKafioio 
poq, ;  Soph.  Trach.  640  ff.  6  KoXKifiSat  rdx  vfuy  I  avK6t  ovK  wapaLav  \ 
aX&v  Katfoxay  firdvttaiv,  aXkii  Btlat  \  dvriKvpov  fiov<ras,      L.  and  S.  Say 

that  Kavaxn  is  used  of  the  lyre  in  the  (Homeric)  Hymn  to  Apollo, 
vs.  185. 

It  is  quite  possible  that  Horace  had  some  one  of  these  expres- 
sions in  mind,  adapting  it,  however,  to  his  own  immediate  purpose. 
He  knew  his  Homer  well.  He  had  also  some  knowledge  of 
Pindar.  On  this  point  it  will  suffice  to  refer  to  Kiessling's  notes 
on  Carm.  I  i.  3,  I  3.  34, 1  12.  i  ff.,  and  his  introductory  remarks 
on  I  35.  Some  interesting  parallels,  in  language  and  in  thought, 
between  Horace  and  Sophocles  may  be  noted.  Compare  C.  I  3. 
9ff.: 

Illi  robur  et  aes  triplex 
circa  pectus  erat,  qui  fragilem  truci 

commisit  pelago  ratem 
primus,  nee  timuit  praecipitem  Africum,  etc., 

with  Antlg.  33^  "*  ^o^^^  ra  dtiva  Ko^dtv  dvBpinrov  dtipdrtpoy  ytAci,  k.  r.  X., 

and  Ajax  1192  ff.: 

^cXr  irp6iT9pov  aWtpa  dvpat  piyav  tj  rhv  noXvKoivov  "Aiiav 
Ktivos  Avrfp^  hs  arvytpSiv  cdri^cy  ^ttXaiv  *^EXXa(rt  KOivhv  "Apri 

(see  Jebb's  note) ;  C.  H  15.  9, 10  fervidos  .  .  .  ictus  (sc.  solis)  with 
Ajax  877  rffp  d<f}  ffXiov  /SoXtfv  k€X€vBop  (see  Jebb  ad  loc);  C.  H  16. 

21  ff.: 

Scandit  aeratas  vitiosa  naves 
Cura  nee  turmas  equitum  relinquit, 
ocior  cervis  et  agente  nimbos, 
ocior  Euro 

and  C.  Ill  I.  37-40: 

Sed  Timor  et  Minae 
scandant  eodem  quo  dominus,  neque 
decedit  aerata  triremi  et 

post  equitem  sedet  atra  Cura 


with  Antig.  951-4 : 


aXX   A  poipibla  rtr  ^vvaaii  dtivd' 

oih-*  av  i»iv  SX^os  oCt^  "AprfSf  ov  nvpyos,  ovx  dXiKrvnoi 

KtXaiyai  vacr  cm^tryoiri'. 


Digitized  by  VjOOQ IC 


NOTES  ON  HORACE,  329 

cm  24.31,  32: 

Virtatem  incolamem  odimus, 
sublatam  ex  oculis  quaerimus  invidi 

with  Ajax  961-5 : 

o\  d*  o^k  yt\oi>vT»v  KawixoipSvrtoy  /caicotr 
roiff  Tovd\     latot  rot,  ntl  fiXtiroyra  firj  n6Bovv, 
BavovT  h»  oc/x^fctav  (V  XP^h^  dopds» 
ot  yap  HQKol  yyoifiatai  rayaBhv  x*P^^^ 
€xovT€S  ovK  Xcracri,  irpiv  rtr  €KPaKjj. 

C.  Ill  30.  4,  5  innumerabilis  Annorum  series  et  fuga  temporum 
with  Ajax  646 : 

&irav6  6  fiaKp6t  KapapiBfii]TOg  XP^^°^ 
(fivti  T   adrjka  Koi  (fxuftyra  Kpxmrrrai 

(cf.  also  Hon  Epp.  I  6.  24,  25) ;  C.  IV  7.  16  pulvis  et  umbra 
sumus  (said  of  the  dead)  with  Ajax  1257  ff.: 

6ap<ray  v/Spt^rir  KafcXcv^cpoaTo/xcif 

(see  Jebb's  notes) ;  and  finally  C.  IV  13.  6-8: 

lUe  virentis  et 
doctae  psallere  Chiae 
pulchris  excubat  in  gents 

with  Antig.  781-3 : 

"Epms  aviKart  fiaxaVf  "Ept^f,  tf  tv  KrrifMai  mjmiSf 
ts  (V  fuzXaKaU  nap^ials  yedyidor  (Vyv;^cvfir* 

Satires,  I  i.  9: 

Agricolam  laudat  iuris  legumque  peritus, 
sub  galli  cantum  consultor  ubi  ostia  pulsat 

Wickham  calls  sud  galli  cantum  an  exaggeration,  comparing 
Cic.  Mur.  22  Vigilas  tu  de  nocte  ut  tuis  consultoribus  respondeas 
.  .  .  te  gallorum  . . .  cantus  . . .  exsuscitat.  Kiessling  takes  the 
same  view,  remarking,  among  other  things,  that  ''die  salutaiio 
begann  doch  erst  nach  Sonnenaufgang ;  (prima  salutantes  atque 
altera  continet  hora.  Martial.  IV  8)."  Mewes,  in  his  revision  of 
Orelli,  characterizes  sub  galli  cantum  as  a  ''ridicula  hyperbole." 


Digitized  by  VjOOQ IC 


330  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Similarly  Greenough.  Schutz  and  Palmer  evidently  regard  the 
words  as  seriously  meant,  but  their  notes  are  very  brief.  Kirkland 
is  silent  on  the  point. 

Several  considerations  may,  I  think,  be  urged  against  the  view 
that  there  is  any  great  degree  of  exaggeration  here,  (i)  There 
is  proof  that  the  regular  saluiaiio  might  begin  before  daylight. 
Cf.  e.  g.  Cic.  Cat.  I  9  dixisti  paulum  tibi  etiam  nunc  morae,  quod 
ego  viverem.  Reperti  sunt  duo  equites  Romani,  qui  te  ista  cura 
liberarent  et  sese  ilia  ipsa  node  paulo  a^iie  lucem  me  in  nieo 
lectulo  interfecturos  pollicentur.  Put  beside  this  passage  the 
words  of  Sallust,  Cat.  28  constituere  ea  node  . . .  sicuti  salulahim 
introire  ad  Ciceronem.  Had  calls  before  daylight  been  glaringly 
unusual,  any  attempt  to  visit  Cicero  at  such  an  unheard-of  hour 
would  have  awakened  his  suspicions  at  once,  and  so  have  defeated 
its  own  end.  The  conspirators  would  hardly  have  been  likely 
(stupid  though  they  proved  themselves  in  other  ways)  to  frustrate 
this  particular  scheme  themselves  by  essaying  a  call  at  an  hour 
not  sanctioned,  to  some  degree  at  least,  by  common  custom.  If» 
on  the  other  hand,  calls  even  ante  lucem  were  not  unusual,  the 
conspirators  could  rest  assured  that,  unless  Cicero's  suspicions 
were  aroused  in  some  other  way,  there  would  be  nothing  in  so 
early  a  visit  itself  to  excite  him  and  afford  him  a  chance  to 
frustrate  their  designs.  (On  the  subject  of  very  early  calls  see 
Mayor's  notes  on  Juvenal,  III  127  and  V  19-23;  Becker-GoU, 
Gallus,  II  194  ff.;  Marquardt,  Privatleben,  p.  228,  n.  2,  and  p.  259.) 

(2)  It  might  be  fairly  held  that  the  reference  in  our  passage  is 
not  to  the  regular  saluiaiio,  but  to  extraordinary  business  calls 
made  upon  the  lawyer  by  distressed  clients.  By  this  supposition 
the  case  of  the  lawyer  is  made  more  closely  parallel  to  that  of 
the  merchant,  whose  encounters  with  severe  storms  are  somewhat 
out  of  the  line  of  his  ordinary  experiences,  and  to  that  of  the 
farmer  whose  experience  with  the  law's  vexatious  processes  are 
quite  foreign  to  the  ordinary  tenor  of  his  existence.  If  this  be 
granted,  it  follows  that  no  hour  would  seem  too  early  to  a 
thoroughly  frightened  client. 

(3)  The  exaggeration  postulated  by  the  editors  named  above 
would  be  wholly  out  of  place.  Horace  is  dwelling  on  the  univer- 
sality of  human  discontent.  Four  men,  distributed  into  pairs,  are 
taken  as  types  of  all  mankind.  Each  is  brought  before  us  when 
the  hardships  or  disadvantages  of  his  lot  press  most  heavily  upon 
him.     In  speaking  of  the   first,  second   and  fourth  characters. 


Digitized  by  VjOOQ IC 


NOTES  Oy  HORACE.  33 1 

Horace  is  unquestionably  keeping  to  the  facts,  real  or  potential, 
of  their  experience.  Why  assume  that  he  is  exaggerating  in 
connection  with  the  third  ?  Such  a  proceeding  would  have  been 
inartistic,  and  would  have  weakened,  needlessly,  the  whole 
passage. 

Sat.  I  I.  93-6: 

finire  laborem 
incipias,  parto  quod  avebas,  ne  facias  quod 
Ummidius  quidam — non  longa  est  fabula—- dives 
ut  metiretur  nummos 

Bentley's  qui  tarn  for  quidam  is  indeed  tempting,  but  I  cannot 
forbear  to  enter  a  protest  against  Palmer's  method  of  supporting 
it.  He  declares  that "  Hon  never  uses  quidam  with  his  proper 
names.''  This  statement  is,  of  course,  a  peiitio  principii^  for  it 
becomes  true  only  when  Bentley's  emendation  is  accepted.  If 
we  emend  Palmer's  statement  to  read  "Horace  nowhere  else  uses 
quidam  with  proper  names,"  it  will  still  be  in  order  to  ask  "What 
of  it  ?  Is  the  fact  that  he  nowhere  else  uses  the  word  in  this  way 
sufficient  proof  that  he  cannot  have  so  used  it  here  ?  "  The  truth 
is  that  the  emendation  must  be  supported  on  other  grounds — 
those,  for  instance,  urged  by  its  author. 

The  same  sort  oi  peiitio  principii  is  perpetrated  elsewhere  in 
commentaries,  e.  g.  by  Kiessling  on  Hor.  C  I  26.  3  quis  sub 
Arcto  Rex  gelidae  metuatur  orae  . . .  unice  Securus.  He  takes 
quis  as  nominative,  giving  as  the  ground  of  this  declaration  the 
statement  that  Horace  uses  "die  veraltete  und  darum  vulgare 
Ablativform  quis^^  only  in  the  Epodes  and  the  Satires.  This 
statement  ought  not  to  be  made  at  all,  much  less  cited  here  as  a 
convincing  proof,  until  it  has  been  shown  on  good  evidence  that 
quis  does  not  stand  here  for  quibus. 

Now,  if  Vergil  could  use  quis  for  quibus  in  the  Epic  style  (e.  g. 
Aen.  I  95,  V  511),  the  form  was  surely  not  too  commonplace  or 
Wulgar'  to  be  admitted  by  Horace  into  his  Odes,  especially  in 
view  of  the  fact  that  he  is  not  always  at  the  pains  to  keep  the 
language  of  those  poems  clear  of  a  prosaic  or  even  vulgar  tone 
(see  Teuffel,  §238,  6).  It  is  the  part,  therefore,  of  the  critic  to 
show  that  it  is  preferable,  even  if  not  absolutely^^necessary,  to 
take  quis  here  as  nominative.  This  successfully  done,  it  is 
allowable  to  remark  that  Horace  nowhere  uses  quis  as  =  quibus 
in  his  Odes. 


Digitized  by  VjOOQ IC 


332  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Sat.  I  I.  io6: 

Est  modus  in  rebus,  sunt  certi  denique  fines, 
quos  ultra  citraque  neqidt  consisiere  rectum, 

Cf.  Cic.  Cato  Maior,  §41 :  Nee  enim  libidine  dominante  tempe- 
rantiae  locum  esse,  neque  omnino  in  voluptatis  regno  viriuiem 
posse  consisiere. 

Sat.  I  I.  108: 

Illuc  unde  abii  redeo,  qui  nemo  ut  avarus 
se  probet  ac  potius  laudet  diversa  sequentis 

There  can  be  no  doubt  that  qui  nemo  is  the  right  reading. 
How  shall  we  interpret  it  ?  Palmer's  view,  that  Jiat  is  to  be 
supplied  after  qui^  is,  at  the  first  blush,  the  most  tempting.  Its 
great  advantage  lies  in  the  fact  that  by  assuming  such  a  construc- 
tion we  make  Horace  go  back,  in  form  of  utterance  at  least,  to  the 
point  whence  he  started.  But  it  is  extremely  doubtful  whether 
such  an  ellipsis  is  possible.  Jio  is  not  the  kind  of  verb  to  be 
easily  omitted,  especially  when,  as  here,  the  context  in  itself  gives 
no  hint  of  such  an  ellipsis.  Again,  the  ellipsis  of  the  subjunctive 
mood  is  uncommon.  If  the  omission  oi  fiat  is  to  be  justified  at 
all,  it  must  be  done,  I  think,  by  the  argument  that  since  Horace 
is  so  manifestly  referring  back  to  the  opening  line  of  the  Satire, 
he  can  omit  here  the  essential  verb  of  that  line,  which  the  reader 
or  auditor  would  inevitably  recall.  But,  admitting  the  ellipsis  to 
be  possible,  we  meet  another  and  more  serious  difficulty.  We 
must  then  render  either  **  I  come  back  to  my  starting-point,  and 
ask  why  no  avaricious  man  praises  himself,'*  or  **why  no  man, 
because  he  is  avaricious,  praises  himself."  Neither  view  gives 
the  point  from  which  Horace  started.  Palmer,  who  adopted  the 
former,  saw  this  point,  and  so  wrote:  ''the  insertion  of  this  word 
(javarus)  is  necessary  to  fuse  Hor.'s  two  subjects  into  one.  At 
the  beginning  Hon  simply  said  nemo;  but  having  developed 
discontent  altogether  from  the  example  of  the  avarus,  he  here 
adds  that  word  as  a  sort  of  correction  to  his  exordium."  The 
other  interpretation  is  even  feebler,  since  it  makes  Horace  answer 
his  question  actually  before  he  has  fully  asked  it.  Precisely  the 
same  objection  applies  to  the  view  (held  by  Orelli-Mewes,  Miiller, 
Kiessling,  Kirkland)  which  takes  qui  nemo  directly  with  probet^ 
and  regards  ul  avarus  as  a  causal  phrase  to  be  compared  with  ui 


Digitized  by  VjOOQ IC 


NOTES  ON  HORACE,  333 

capitis  minor,  Carm.  Ill  5.  42,  ut  neque  . .  .porrectus  . . .  nee  . . . 
prodigus^  Epp.  I  7.  41,  or  tUpote  pluris  culpari  dignos,  Sat.  I 
4.  24.* 

Since  I  have  objected  to  all  the  proposed  solutions  of  the 
problem  presented  by  our  passage  (all,  at  least,  that  can  be  said 
to  have  any  degree  of  plausibility),  it  devolves  upon  me  to 
venture  to  offer  a  substitute.  For  the  past  four  years  I  have 
suggested  to  my  classes  that  qui  nemo  se  probet=  qui  omnes  se 
improbent — that  is,  that  the  clause,  though  negative  in  form,  is 
affirmative  in  sense.  If  this  be  true,  we  shall  have  to  supply 
improbai'se  with*«/  avarus^^xA  we  shall  thus  have  merely  an 
example  of  the  ordinary  comparative  {^/-clause  which  is  so 
common  in  connection  with  affirmative  clauses.  I  am  inclined  to 
think  that  this  explanation  is  tenable.  The  fact  that  ac,  not  sed, 
is  the  connective  seems  to  afford  it  support. 

Recently  I  noted  the  following  passage  (Cic.  Fin.  II,  §18)  :  Sed 
dum  dialecticam,  Torquate,  contemnit  Epicurus,  quae  una  con- 
tinet  omnem  et  perspiciendi  quid  in  quaque  re  sit  scientiam  et 
iudicandi  quale  quidque  sit,  et  ratione  ac  via  disputandi,  ruit  in 
dicendo,  ut  mihi  quidem  videtur,  nee  ea,  quae  docere  vult,  ulla 
arte  distinguit,  ut  haec  ipsa,  quae  modo  loquebamur.  Only  two 
explanations  of  ui  haec  ipsa  are  (to  me,  at  least)  conceivable. 
One  is  that  nee  .  .  .  distinguit  is  in  spirit  affirmative :  "  he  fails  to 
distinguish  ...  as,  for  instance,  in  the  case  just  mentioned";  the 
other  is  that  with  ut  haee  ipsa  we  must  supply  neque  or  non 
distinguit  If  we  take  the  latter  course,  we  get  the  very  construc- 
tion which  has  been  declared — e.  g.  by  Palmer — impossible  for 
our  Horatian  passage.  In  either  case  the  passage  from  Cicero 
seems  to  me  to  be  a  complete  parallel  to  that  in  Horace.  The 
explanation  applying  to  the  one  will  apply,  then,  to  the  other.  It 
goes  without  saying  that  none  of  the  explanations  of  Horace's 
words,  rejected  above,  could  be  forced  to  apply  to  Cicero's 
sentence. 

Another  general  parallel,  as  it  seems  to  me,  to  our  Horatian 
passage  is  to  be  found  in  Terence,  Phormio,  279-81  (Demipho  is 
the  speaker) : 

^  If  we  read  nemo  tU  avarus  seprobei^  we  are  confronted  by  two  difficulties : 
(i)  the  harsh  hiatus  in  nemo  ut^  and  (2)  the  difficulty  of  satisfactory  interpre- 
tation, for  we  must  take  ui  as  '  how,'  and  closely  join  nemo  and  avarus.  Such 
an  interpretation  would  be  practically  identical  with  Palmer*s  and  so  be  open 
to  the  objections  urged  above  against  that  view. 


Digitized  by  VjOOQ IC 


334  AMERICAN  JO URNAL    OF  PHIL  OLOGY, 

An  quisquam  iudex  est,  qui  possit  noscere 
tua  iusta,  ubi  tute  verbam  non  respondeas, 
Ita  ut  ille  fecit  ? 

Here  fecit  =  non  respandit  Had  the  sentence  concluded  with  a 
simple  iia  ut  Hie,  its  close  resemblance  to  our  passage  would  be 
more  immediately  apparent. 

With  either  of  the  proposed  explanations,  we  get  a  view  simple 
in  the  extreme.  The  framework  of  Horace's  question  here,  qui 
nemo  . .  .  se  probet,  will  be  identical  with  that  of  the  query  with 
which  the  Satire  opens,  qui  fit .  .  .  contentus  vivai,  and  the  illus- 
tration, "just  as  the  miser  fails  to  praise  himself,"  will  be  entirely 
in  point,  since  for  seventy  or  eighty  lines  he  has  been  using  the 
avarus  as  the  most  typical  example  of  discontent. 

Sat.  I  5-  43 : 

O  qui  complexus  et  gaudia  quanta  fuerunt. 

No  editor  (at  least  in  recent  times)  has  made  any  comment  on 
complexus.  It  may  be  interesting,  therefore,  to  compare  Cic. 
Acad.  Post.  I  I  In  Cumano  nuper  cum  mecum  Atticus  noster 
esset,  nuntiatum  est  nobis  a  M.  Varrone  venisse  eum  Roma  pridie 
vesperi .  . .  Itaque  confestim  ad  eum  ire  perreximus,  paulumque 
cum  ab  eius  villa  abessemus,  ipsum  ad  nos  venientem  vidimus; 
atque  ilium  complexi,  ut  mos  amicorum  est ...  ad  suam  villam 
reduximus.    See  Reid  ad  loc. 

Sat.  I  5.  7  : 

Hie  ego  propter  aquam,  quod  erat  deterrima,  ventri 
indico  bellum 

Neither  Schiitz,  Kiessling,  Kirkland,  Wickham,  nor  Greenough 
makes  any  comment  directly  on  the  words  ventri  indico  bellum. 
Mewes  defines  them  by  cena  abstineo,  comparing  Cicero,  Cato 
Maior,  §46:  ne  omnino  bellum  indixisse  videar  voluptati.  Does 
not  indico  bellum,  here  as  elsewhere,  mean  *I  make  a  formal 
declaration  of  war  against/  and  was  it  not  meant  tp  suggest  to 
Horace's  contemporaries  the  fetial  ceremonies  traditionally  con- 
nected with  such  a  declaration,  as  described  later  by  Livy,  I  32. 
5  tf.  ?  If  this  is  right,  we  get  added  humor  in  the  ridiculously 
exaggerated  tone  of  the  passage.  The  mock  heroics  will  then 
run  through  four  full  verses,  and  the  drop  to  the  '  Dutch  picture ' 
(Wickham)   in    Tum  pueri  nautis,  etc.,  will  be  all  the   more 


Digitized  by  VjOOQ IC 


N07'ES  ON  HORACE,  335 

marked.  The  mock  heroic  tone  is  resumed  in  vss.  51  ff.:  Nunc 
mihi  paucis  Sarmenti  scurrae  pugnam,  etc.,  and  in  vss.  73,  74: 
Nam  vaga . . .  tectum.  Cf.  also  Sat.  II  6.  100  lamque  tenebat 
Nox  medium  caeli  spatium,  etc.  See  also  Juvenal,  e.  g.  IV  29- 
39,  and  Friedlaender  ad  loc,  also  his  Einleitung,  p.  57.  One 
cannot  help  thinking  that  Juvenal  had  in  mind  the  passages  of 
Horace  quoted  above. 

Sat.  I  5.  51  ff.: 

Nunc  mihi  paucis 
Sarmenti  scurrae  pugnam  Messique  Cicirri, 
Musa,  velim  memores 

The  mock  heroic  character  of  these  verses  (of.  preceding  note) 
has  often  been  remarked  by  editors.  To  their  comments,  how- 
ever, something  can  be  added.  Horace  pretends  to  be  here  the 
mere  recorder  of  the  strains  sung  by  the  muse.  Cf.  then  the 
thought  of  Carm.  I  i.  32  Si  neque  tibias  Euterpe  cohibet  neque 
Polyhymnia  Lesboum  refugit  tendere  barbiton,  and  the  language 
of  Carm.  I  12.  i  ff.,  I  24.  2-4,  and  III  4.  i  ff.  This  fiction, 
whereby  the  muse  is  represented  as  the  real  singer  and  the  poet 
as  merely  her  mouthpiece,  is  the  true  Homeric  or  Epic  attitude.^ 
Cf.  the  Iliad,  1 1,  Odyssey,  1 1,  and  the  opening  of  Paradise  Lost. 
Though  Vergil  begins  self-consciously  with  Arma  virumque  cano 
(see  Conington  on  cano),  he  nevertheless  comes  back  in  vs.  8, 
Musa,  mihi  causas  memora,  to  the  Homeric  model  and  throws 
himself,  as  it  were,  wholly  on  the  muse.  We  may  compare  also 
appeals  to  the  muse  for  special  help  in  special  connections,  e.  g. 
Iliad,  II  484  ff.,  and  its  imitation  in  the  Aeneid,  VII  641  ff.;  also 
Paradise  Lost,  VII  i  ff.  Now,  Horace's  careful  adherence  to  the 
highest  poetic  models  emphasizes  the  intentional  burlesque  of  the 
^description  that  follows,  by  making  the  language  out  of  all 
proportion  to  the  subject-matter. 

Sat.  I  5.  77-80 : 

Incipit  ex  illo  montis  Apulia  notos 
ostentare  mihi,  quos  torret  Atabulus  et  quos 

*The  New  York  Herald  of  Dec.  27,  1896,  contained  an  account  of  a  mosaic 
tablet,  recently  discovered  in  Tunis.  The  tablet  represents  Vergil  as  reading 
the  Aeneid,  while  behind  him  stand  two  Muses.  Part  of  the  description 
seems  worth  quoting:  *'With  head  erect,  eyes  intent,  expression  as  of  one 
inspired  and  with  his  right  hand  placed  upon  his  breast,  the  index  finger 
being  raised,  the  poet  listens  to  Clio  and  Melpomene,  who  stand  behind  him 
and  alternately  dictate  the  melodious  lines  of  the  poem.*' 


Digitized  by  VjOOQ IC 


336  AMERICAN  JOURNAL    OF  PHILOLOGY. 

numquam  erepsemus,  nisi  nos  vicina  Trivici 
villa  recepisset 

"For  this  construction  of  vicinus  L.  S.  only  quote  Lucan, 
9.  432  ora  vicina  perusti  aetheris."  So  Palmer.  Had  he  seen,  as 
did  Wickham,  that  vicina  is  here  a  noun,  he  might  have  noted 
Cic.  Off.  Ill  104  Fidem  .  • .  in  Capitolio  vicinam  lovis,  and  Ovid, 
Fasti,  VI  399  anus  vicina  loci,  both  cited  by  L.  and  S.  Add 
Vergil,  Aen.  Ill  500  Si  quando  Thybrim  vicinaque  Thybridis 
arva  Intraro.  Cf.  finally  the  use  of  the  genitive  with  amicus 
when  construed  as  a  noun,  and  the  dative  with  the  same  word 
considered  as  an  adjective. 

Sat.  I  9.  11: 

^  "  O  te,  Bolane,  cerebri 

felicem/'  aiebam  tacitus,  cum  quidlibet  iUe 
garriret,  vicos,  urbem  laudaret. 

Most  editors  follow  Porphyrion  in  thinking  of  Bolanus  as  some 
hot-headed  individual,  who  would  long  ere  this  have  taken  the 
law  into  his  own  hands,  and  have  forcibly  rid  himself  of  so 
disagreeable  a  companion.  The  same  view  was  taken  by  Ben 
Jonson,  Poetaster,  III  i,  in  his  imitation  of  this  Satire.  Schiitz, 
after  mentioning  this  view,  continues:  "Aber  Hor.  nennt  sich 
selbst  reizbar  und  heftig  sat.  II  3,  323.  7,  35  und  44.  epist  I  20, 
25.  Vielleicht  besser :  Bol.  hat  ein  solches  Phlegma,  dass  er  sich 
nicht  leicht  erhitzt."  He  had  been  anticipated  in  this  view  by 
Gifford,  who,  in  his  edition^  of  Jonson's  works  (18 16),  writes: 
"But  no  one  could  show  more  fretfulness  and  impatience  than 
Horace  himself  does.  Surely  the  felicity  of  Bolanus  must  have 
consisted  in  an  impenetrable,  rather  than  a  ticklish  and  tender 
scull :  a  comfortable  indifference  to  all  attacks ;  a  good-humoured 
stupidity  that  dosed  over  all  impertinence;  this,  indeed,  was  to 
be  envied.*' 

Sat.  I  9.  22  ff.: 

Si  bene  me  novi,  non  Viscum  pluris  amicum, 
non  Varium  facies ;  nam  quis  me  scribere  pluris 
aut  citius  possit  versus  ?    quis  membra  movere 
mollius?    invideat  quod  et  Hermogenes  ego  canto. 

The  attempts  to  identify  the  bore  of  this  Satire — e.  g.  with 
Propertius — are  well  known.    For  myself,  I  am  inclined  to  believe 

^11.  p.  435. 


Digitized  by  VjOOQ IC 


NOTES  ON  HORACE, 


337 


that  the  four  verses  quoted  above  are  enough  of  themselves  to 
show  that  all  such  attempts  are  but  a  waste  of  time,  for  they  seem 
to  prove  conclusively  that  Horace  had  no  one  person  particularly 
in  mind.  The  unknown  bore  prides  himself  on  the  very  three 
things  which  were  Horace's  pet  aversions.  His  estimate  of  the 
credit  to  be  accorded  to  rapid  and  voluminous  composition 
appears  from  Sat.  I  i.  120,  121,  I  4.  8-21,  I  10.  50,  51,  and  I  10. 
64-71.  Such  writing  formed  the  opposite  pole  to  the  ideal  which 
he  set  before  himself,  as  suggested  in  Sat.  I  10.  72-74,  and  more 
fully  in  Carm.  IV  2.  25-32.  Cf.  Netdeship's  Vergil,  p.  17  :  "For 
the  first  time  in  the  history  of  Italian  literature  they  (Vergil  and 
Horace)  practically  laid  down  the  principle  that  no  amount  of 
labour  could  be  too  great  to  expend  upon  poetical  expression ; 
that  genius,  power,  freedom  of  utterance,  were  not  enough  to 
make  a  perfect  poet.  Like  Cicero  in  the  sphere  of  oratorical 
prose,  Vergil  and  Horace  are  never  satisfied  with  the  form  of 
their  work;  they  know  no  end  to  the  striving  for  perfection." 
See  also  Sellar's  Horace,  pp.  182  ff. 

How  Horace  valued  the  bore's  second  accomplishment,  skill  in 
dancing,  is  shown  (i)  by  the  invidious  mollius]  (2)  by  Sat.  II  i. 
24  Saltat  Milonius,  ut  semel  icto  Accessit  fervor  capiti  numerus- 
que  lacernis;  and  (3)  by  Cicero's  well-known  words  in  Mur.,  §13, 
especially  the  clause  nemo  fere  saltat  sobrius,  nisi  forte  insanit. 
Horace's  attitude  toward  Tigellius  requires  no  illustration. 

We  thus,  as  already  suggested,  find  the  bore  seeking  to  curry 
favor  with  Horace  by  the  aid  of  the  very  things  which  would 
have  been  sure  to  earn  him  Horace's  hearty  dislike.  A  man  who 
combined  the  attributes  of  a  Crispinus,  a  Milonius,  and  a  Her- 
mogenes  Tigellius  would  have  had  no  chance  whatever  to  secure 
the  poet's  favor.  My  point,  then,  is  that  the  description  is  obvi- 
ously made  to  order,  to  paint  the  bore  in  as  ludicrous  a  light  as 
possible,  and  that  it  therefore  of  itself  proves  that  Horace  had  no 
real  individual  especially  in  mind,  since  it  is  hardly  likely  that 
any  individual  would  have  possessed  such  a  curiously  composite 
character,  or  have  been  so  dull  of  perception,  so  utterly  a  stranger 
to  the  likes  and  dislikes  of  the  man  whose  favor  he  was  seeking 
to  win  as  to  endeavor  to  commend  himself  to  that  man  via  his 
pet  aversions.  Such  a  coincidence  is,  indeed,  possible;  yet  it 
hardly  seems  probable. 

I  should  rather  hold  that  Horace  has  in  mind  more  than  one 
individual,  that  he  is  dramatizing  in  the  form  of  a  single  incident 


Digiti 


338  AMERICAN  JOURNAL    OF  PHILOLOGY. 

his  experiences  with  numerous  personages  who  sought  his  aid  to 
further  their  social  or  literary  ambitions.  I  am  inclined  to  believe 
also  that,  if  we  regard  the  fifth,  sixth  and  ninth  Satires  of  this 
book  as  parts  of  one  whole,  our  interpretations  of  these  poems 
will  be  materially  assisted.  The  sixth  Satire  tells  how  naturally 
and  honesdy  the  friendship  between  Maecenas  and  Horace  arose, 
the  fifth  gives  a  glimpse  of  the  real  nature  of  that  friendship,  and 
the  ninth  tells  us  what  that  friendship  is  not.  This  view  will 
account  satisfactorily  for  the  absence  of  all  political  allusions  (save 
that  in  vss.  28  and  29)  from  the  fifth  Satire.  Horace  would  hint 
that  such  matters  form  no  part  of  the  friendship,  which  is  one 
between  men  allied,  the  one  to  the  other,  by  similarity  of  tastes 
in  matters  wholly  removed  from  the  sphere  of  politics  and 
statecraft. 

barkakd  coiLBOB.  Charles  Knapp. 


Digitized  by  VjOOQ IC 


NOTE. 
On  Lucian's  Nigrinus. 

The  difficulty  of  distinguishing  between  the  formal  and  the 
spontaneous  in  Lucian's  work  leaves  commentators  grotesquely 
at  variance  in  regard  to  the  intention  of  certain  of  his  writings, 
and  notably  of  the  Nigrinus.  From  Wieland  to  Croiset  there  is 
a  series  of  critics  who  accept  it  as  a  serious  document,  either  a 
'confession'  or  a  tract  for  the  times.  On  the  other  hand,  Ant. 
Schwarz  will  have  it  pure  satire,  a  parody  of  the  sudden  conver- 
sions aimed  at  by  certain  philosophers.  P.  M.  Boldermann,  in  a 
really  valuable  contribution  to  Lucianic  doctrine,^  by  misappli- 
cation of  a  fruitful  theory,  classes  the  dialogue  among  those  in 
which  Lucian  imitates  comedy  in  the  manner  of  Menippus: 
Nigrinus  is  a  comic  character,  but  his  talk  is  pure  Cynic  doctrine. 
E.  Schwarz,  reviewing  Boldermann's  essay  in  the  Philologische 
Wochenschrift  for  March  21,  1896,  refutes  this  notion  easily 
enough,  but  sets  up  in  place  of  it  one  of  his  own  which  will 
hardly  be  more  satisfactory  to  Lucianic  scholars.  "Die  Schilde- 
rung  des  Klientenelends  im  Nigrin,"  he  says,  "ist  nichts  als  eine 
kurze  Wiederholung  der  ausfiihrlichen  Darstellung  in  den 
Klienten,  sie  gipfelt  hier  wie  dort  in  einem  scharfen  AngrifTe  auf 
die  Philosophen,  welche  sich  freiwillig,  nur  durch  den  Glanz  des 
Reichthums  und  der  Vornehmheit  geblendet,  zu  solcher  Mis^re 
hergeben,  nur  mit  dem  Unterschiede,  dasz  die  Klienten  nur  eine 
scharfe  Invective  gegen  solche,  die  hellenische  Philosophic 
diskreditierenden  Philosophen  sind,  der  Nigrin  das  positive 
Gegenstuck  liefert  in  dem  Bilde  des  Platonikers,  der  zwar  das 
stille  und  feine  Athen  mit  dem  larmenden,  groszstadtisch  rohen 
Rom  vertauscht  hat,  aber  gerade  hier  ein  leuchtender  Exempel 
fiir  echthellenische  Weisheit  ist.  . . .  er  ist  ebenso  zu  der  Klien- 
tenschrift  das  Gegenstuck  wie  die  Entlaufenen  Sklaven  zum 
Peregrinus,  wie  der  Fischer  zu  der  B/»y  npaau,  wie  der  Ikaro- 
menipp  .  .  .  zum  Al^  lean^yopov/icyor." 

^Diss.,  Leyden,  1893. 


Digitized  by  VjOOQ IC 


340  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Now,  it  is  undeniable  that  the  discourse  of  Nigrinus  gives  a 
shorter  description  of  the  matters  set  forth  at  length  in  De 
Mercede  Conductis,  but  if  it  was  intended  as  a  pendant  and  in 
some  degree  a  corrective  thereof,  it  is  unfortunate  that  while  De 
Mercede  Conductis  was  written  in  a  realistic  spirit,  with  the 
author's  eye  on  the  object,  the  Nigrinus  was  composed  on  a 
palpably  sophistic  plan  with  argumentation  that  will  not  bear 
scrutiny.  The  comparison  of  Athens  and  Rome  is  manifestly  ad 
captandum.  Of  the  generations  of  critics  that  have  taken  it  as  a 
serious  contribution  to  Sittengeschichte,  not  one  seems  to  have 
noticed  that,  with  the  exception  of  their  undue  interest  in  horse- 
racing,  the  Romans  are  not  charged  with  a  single  vice  or  folly 
which  is  not  also  laid  at  the  door  of  the  Athenians  in  other  of  the 
Lucianic  writings.  (Cp.  Nig.  23  with  Epist.  Cron.  35  and  Gall. 
9 ;  30  with  Charon  22  and  De  Luctu  passim ;  31  with  Navig.  23 ; 
etc.)  The  dialogue  is  interesting  as  proving  that  it  was  fashion- 
able once  more  to  write  in  praise  of  Athens ;  but  if  it  be  com- 
pared with  Navigium  and  Gallus,  we  must  admit  that,  as  far  as 
Lucian's  evidence  goes,  we  have  nothing  to  prove  that  society  in 
Athens  differed  more  widely  from  society  in  Rome  than  the 
province  always  differs  from  the  capital. 

The  mystery  of  the  Nigrinus,  as  of  other  Lucianic  works,  has 
remained  unsolved  because  critics  find  it  hard  to  believe  in 
Lucian  as  a  sophist.  Few  casual  readers  take  the  trouble  to 
study  the  contemporaries  whom  he  outshone.  Even  Lucianic 
scholars,  after  formally  crediting  the  sophistic  with  their  author *s 
early  training  and  first  success,  are  wont  to  state  that  he  broke 
with  the  system  midway  in  his  career,  and  to  treat  of  him  there- 
after as  though  he  were  as  completely  a  law  unto  himself  as 
Thucydides  or  Plato.  We  find  even  recent  commentators  like 
Croiset  speaking  of  him  as  returning  to  rhetoric  in  his  old  age,  as 
though  there  were  any  evidence  to  show  that  he  ever  abandoned 
it  The  famous  passage  in  Bis  Accusatus  cannot  be  pressed  to 
mean  more  than  that  he  substituted  the  composition  and  delivery 
of  dialogues  for  that  of  meletae.  The  other  part  of  the  sophistic 
programme,  the  prolalia,  he  apparently  retained ;  the  Prometheus 
in  Verbis  is  such  a  work  and  was  designed,  by  internal  evidence, 
as  the  introduction  to  a  dialogue  recitation.  It  is  my  belief,  then, 
that  the  Nigrinus  not  only  shows  traces  of  the  sophistic  style,  as 
all  allow,  but  is  actually  a  sophistic  work.  I  conceive  that  Lucian, 
having  occasion  to  address  an  Athenian  audience,  determined  on 


Digitized  by  VjOOQ IC 


NOTE.  341 

two  points :  that  he  would  give  them  a  sort  of  Panathenaic  oration/ 
and  that  he  would  do  it  in  the  Platonic  manner,  of  which  he  had 
made  a  special  study.  In  pursuit  of  inspiration  for  the  first  point 
he  read  Pericles'  funeral  oration  in  Thucydides'  and  Plato's 
Menexenus.  Catching  from  the  Menexenus  the  tone  of  ironical 
admiration  which  aptly  secured  the  lightness  of  touch  called  for 
by  the  occasion,  he  enforced  it  by  dipping  into  the  Protagoras.' 
We  are  accustomed  to  dismiss  as  pedantic  and  futile  the  methods 
by  which  the  later  sophists  studied  the  great  classical  writers  and 
aimed  at  reproducing  their  effects,  assigning  to  Lucian's  'sophistic 
period*  none  but  works  of  hopeless  frigidity.  The  Nigrinus 
should  save  us  from  such  blunders.  We  can  hardly  doubt  its 
success  with  the  audience  for  whom  it  was  written;  still  more 
gratifying  to  the  author  would  have  been  the  knowledge  that 
posterity  would,  on  the  strength  of  it,  write  him  down  a  moralist, 
a  patriot,  and  sometime  a  Platonist. 

Emily  James  Smith. 

'  When  the  sophist  Alexander  visited  Athens,  ^  f^  Si^  dtdXe^ig  Ittoivoi  ^oav 
Tov  A<rrcof  .  .  .,  TLavadrfva'iKOv  yap  T^yav  iiriTOfiy  elKoaro,  Philostratus,  Vit.  Soph., 
d,  78, 13  ff.,  ed.  Kayser.    Cp.  Aristides'  Panathenaic  Oration. 

'  Besides  the  explicit  quotation  in  the  preface,  cp.  Nigr.  13  with  Thuc.  II 
37,  Nigr.  14  with  Thuc.  II  40, 

»Cp.  Nigr.  35  with  Prot.  328  D. 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES. 

A  Literary  History  of  the  English  People.  From  the  Origins  to  the  Renais- 
sance.  By  J.  J.  Jusserand.  G.  P.  Putnam's  Sons,  New  York  and  London^ 
1895. 

This  volume  forms  Part  I  of  a  work  to  be  completed  in  three  volumes^ 
Part  II  extending  "From  the  Renaissance  to  Pope,"  and  Part  III, "From 
Pope  to  the  Present  Day."  The  author  is  particular  to  call  his  work  not  "  a 
•  History  of  English  Literature,'  but  rather  a  *  Literary  History  of  the  English 
People.'"  It  is  thus  distinguished  from  Taine's  work,  a  brilliant  essay  on 
English  literature,  though  often  mistaken  in  its  critical  fudgments,  rather  than 
a  history  of  English  literature. 

M.  Jusserand  has  devoted  much  labor  and  study  to  the  composition  of  his 
work.  He  has  gone  to  original  sources,  and  not  the  least  valuable  portion  of 
the  volume  is  that  containing  the  numerous  bibliographical  notes  which  he 
has  appended  to  almost  every  page.  These  are  what  we  miss  in  ten  Brink, 
with  whose  valuable  history  of  English  literature  this  volume  invites  com- 
parison. It  is  true  that  ten  Brink  intended  to  add  a  very  full  bibliography,, 
but  his  most  unfortunate  death  prevented  it ;  it  is  better  to  give  the  bibliog- 
raphy in  the  course  of  the  work,  for  the  reader  desires  to  know  the  authorities 
as  he  progresses  in  his  reading. 

The  impression  produced  upon  the  mind  of  the  impartial  reader  after  a 
careful  perusal  of  the  volume,  is  that  it  is  well  done.  If  he  should  not,  in  all 
cases,  have  adopted  the  same  perspective,  and  if  he  considers  that  some 
writers  deserved  a  less  cursory  treatment,  that  is  a  matter  of  private  judgment. 
The  volume  is  divided  into  three  books  treating  respectively  the  Origins,  the 
French  Invasion,  and  England  to  the  English.  These  are  subdivided  into 
chapters,  whose  titles  deserve  mention,  both  as  giving  a  summary  of  the  work, 
and  as  illustrating  the  author's  statement,  "  To  be  easily  understood  one  must 
be  clear,"  and  throughout  the  volume  one  is  reminded  of  the  French  adage,. 
"  Ce  gut  rCestptLs  clair,  n*est pas  franfois,** 

The  first  book  contains  four  chapters:  Britannia,  the  Germanic  invasion, 
the  national  poetry  of  the  Anglo-Saxons,  and  Christian  literature  and  prose 
literature  of  the  Anglo-Saxons.  The  second  book  has  also  four  chapters: 
Battle,  literature  in  the  French  language  under  the  Norman  and  the  Angevin 
kings,  Latin,  and  literature  in  the  English  language.  The  third  book 
includes  seven  chapters :  The  new  nation,  Chaucer,  the  group  of  poets,  Wil- 
liam Langland  and  his  visions,  prose  in  the  fourteenth  century,  the  theatre, 
and  the  end  of  the  Middle  Ages.  The  several  chapters  are  subdivided  into 
sections  with  separate  headings,  so  that  at  no  moment  is  one  at  a  loss  to  know 
what  the  author  is  talking  about,  for  each  subject,  however  briefly  or  fully 
treated,  is  kept  separate  and  distinct  from  every  other  subject.    This  analytical 


Digitized  by  VjOOQ IC 


warn 


HEVIEIVS  AND  BOOK  NOTICES.  343 

subdivision  should  not  be  disregarded,  but  it  is  specially  senriceable  in  a  his- 
tory of  literature. 

In  any  work  by  a  French  writer  we  naturally  look  for  an  exaltation  of  the 
Kelts,  their  brilliancy,  their  power  of  imagination,  their  fertility  of  invention. 
Granting  all  that  may  be  said  on  this  subject  by  Keltophilists,  we  have  always 
thought  that  the  Keltic  literature  of  Britain,  and  the  Kelts  them»dlyes,  before 
the  Norman  Conquest,  had  as  little  to  do  with  English  literature  proper  as  the 
Greeks  and  Greek  literature  of  the  Byzantine  empire.  Whatever  views  we 
may  hold  as  to  the  greater  or  less  extermination  of  the  Kelts  by  the  Germanic 
invaders,  and  the  greater  or  less  incorporation  of  them  into  the  body  of  the 
English, — and  we  are  ready  to  grant,  as  the  author  holds,  that  this  incorpora- 
tion was  greater  than  was  formerly  thought, — the  literature  does  not  seem  to 
us  to  have  had  one  particle  of  influence  on  the  Old  English,  or  Anglo-Saxon 
literature,  and  we  see  no  advantage  in  beginning  with  it  a  literary  history  of 
the  English  people.  English  literature  begins  with  "  the  national  poetry  of 
the  Anglo-Saxons,'*  and  here  we  see  its  true  genius  and  spirit,  allied  to  that  of 
other  Germanic  and  Scandinavian  tribes.  The  author  is  very  right  in  saying 
that  the  Anglo-Saxons  **  did  not  allow  the  traditions  of  the  vanquished  Celts 
to  blend  with  theirs,  and  in  spite  of  their  conversion  to  Christianity,  they  pre- 
served, almost  without  change,  the  main  characteristics  of  the  race  from  which 
they  were  descended"  (p.  36),  We  must  come  down  to  the  twelfth  and  thir- 
teenth centuries  before  we  find  Keltic  traditions  beginning  to  infiltrate 
English  literature,  and  then  through  the  Normans  as  intermediaries.  What- 
ever Keltic  blood  may  survive  in  the  present  English  people,  it  did  not  affect 
the  Old  English  people,  and  whatever  Kelts  survived  the  Germanic  invasion 
became  hewers  of  wood  and  drawers  of  water,  and  the  oil  and  water  of  that 
day  did  not  mix. 

The  sketch  of  Old  English  poetry  is  all  too  brief,  and  the  account  of  the 
versification  is  quoted  from  Sweet's  "Sketch"  in  Hazlitt's  Warton,  but  that 
was  belore  the  days  of  Sievers,  so  the  student  must  revise  the  account  given  of 
it,  and  not  take  as  literal  truth  the  statement,  "  The  rules  of  this  prosody,  not 
very  difficult  in  themselves,  are  made  still  easier  by  a  number  of  licenses  and 
exceptions  "  (p.  37).  The  dispute  as  to  the  works  of  Cynewulf  is  mentioned* 
but  the  works  themselves  receive  scant  treatment,  the  doubtful  Andreas  being 
the  only  one  of  which  a  brief  outline  is  given.  The  conjecture  of  Earle  that 
he  "  lived  in  the  eleventh  century "  is  quoted  as  on  a  par  with  that  of  ten 
Brink  that  "  he  was  born  between  720  and  730"  (p.  39) ;  but  Earle's  mere  con- 
jecture is  utterly  untenable.  While  the  Brummbnrh^  or  Athelstan^  is  found  in 
the  Anglo-Saxon  Chronicle  under  the  year  937,  the  MaUon  or  Byrkinoth  (991) 
is  not  found  therein,  and  this  should  not  have  been  so  stated  (p.  47,  note  i). 
As  was  to  be  expected,  an  outline  of  the  Beowulf  is  given,  but  Beowulf  did  not 
**  return  to  his  own  country  "  after  the  fight  with  Grendel  and  before  that  with 
Grendel's  mother  (p.  52) :  he  simply  slept  elsewhere,  and  so  was  not  on  hand 
when  the  latter  came  to  Heorot  and  devoured  Aeschere.  The  author  in  his 
Preface  laments  "  misprints,"  and  it  must  be  acknowledged  that  there  is  quite 
a  number  of  them,  which  even  the  best  of  eyes  cannot  avoid,  but  ChlochilaUus 
occurs  three  times  in  note  at  foot  of  p.  50.  M.  Jusserand  thinks  that  Beowulf 
is  **  very  different  from  Roland,  the  hero  of  France,  he  too  of  Germanic  origin, 


Digitized  by  VjOOQ IC 


344  AMERICAN  JOURNAL    OF  PHILOLOGY. 

but  living  in  a  different  milieu  "  (p.  54) ;  but  surely  some  allowance  should  be 
made  for  chronology. 

The  introduction  of  Latin  letters  and  their  effect  upon  Old  English  writing, 
as  seen  in  the  works  of  Baeda  and  in  the  poems  of  Caedmon  and  Cynewulf,  is 
noted  ;  also,  the  contributions  of  Alfred  to  a  dissemination  of  a  knowledge  of 
literature  am^g  his  people  in  the  first  revival  of  learning,  and  the  later  works 
of  iElfric  and  of  Wulfstan,  receive  mention  ;  but  fault  is  found  with  this  Old 
English  literature  in  that  it  is  "  almost  stationary;  it  does  not  perceptibly  move 
and  develop ;  a  graft  is  wanted  '*  (p.  92).  This  graft  was  to  come  from  the 
Norman  invasion. 

Notwithstanding  the  comparatively  brief  sketch  of  Old  English  literature, — 
for,  in  a  work  of  this  compass,  we  cannot  expect  the  treatment  given  by  ten 
Brink,  Stopford  Brooke,  or  Morley, — the  author  appreciates  highly  what  has 
been  handed  down  to  us  from  the  Anglo-Saxon  period,  and  well  says :  "  The 
English  country  can  thus  pride  itself  upon  a  literature  which  for  antiquity  is 
unparalleled  in  Europe "  (p.  79).  This  has  not  always  been  realized  even  by 
English  writers,  and  although  a  knowledge  of  this  literature  is  easily  acces- 
sible, it  does  not  seem  to  have  penetrated  beyond  scholars. 

The  Norman  Conquest  and  its  effects  follow.  William  the  Conqueror  is 
highly  exalted.  **  The  qualities  of  which  William  gave  the  example  were  rare 
in  England,  but  common  in  France ;  they  were  those  of  his  race  and  country, 
those  of  his  lieutenants ;  they  naturally  reappear  in  many  of  his  successors  *' 
(p.  106).  The  Normans  are  duly  praised,  perhaps  over-duly.  "  Their  match- 
less strength  and  their  indomitable  will  further  one  particular  cause:  the 
infusion  of  French  and  Latin  ideas  in  the  Anglo-Saxon  people,  and  the  con- 
nection of  England  with  the  civilization  of  the  South."  The  fact  that  '*  the 
chiefs  of  the  nation  are  French,*'  and  "their  wives  are  mostly  French  too" 
(p.  108),  is  put  prominently  before  us,  so  that  one  would  infer  that  all  the 
good  in  England  came  from  France.  Now,  while  the  impartial  historian  will 
readily  acknowledge  the  good  accruing  to  England  from  the  Norman  invasion, 
he  should  never  forget  the  heroic  qualities  lying  at  the  basis  of  the  English 
people,  qualities  seen  in  other  Germanic  peoples,  which  asserted  themselves 
two  hundred  and  fifty  years  later  in  that  composite  Saxon  and  Norman,  that 
is,  English  people,  almost  as  distinct  from  the  French  of  that  day  as  were  the 
Saxon  and  Norman  at  the  time  of  the  Conquest ;  the  development  had  been 
different,  and  the  Saxon  had  predominated. 

The  diffusion  of  the  French  language  and  the  rise  of  Anglo-French  literature 
are  quite  fully  treated.  The  French  language  and  French  ideas  were  undoubt- 
edly prevalent  in  England  for  three  hundred  years.  The  author  rightly  says : 
"  It  matters  little  whether  these  ideas  went  across  the  Channel  carried  over  by 
poets  or  by  manuscripts.  What  is  important  is  to  see  and  ascertain  that  works 
of  a  new  style,  with  new  aims  in  them,  and  belonging  to  a  new  school  of  art, 
enjoyed  in  England  a  wide  popularity  after  the  Conquest,  with  the  result  that 
deep  and  lasting  transformations  affected  the  aesthetic  ideal  and  even  the 
way  of  thinking  of  the  inhabitants"  (p.  120). 

The  different  kinds  of  this  literature  are  enumerated  and  discussed.  His- 
tories, romances,  religious  works,  fabliaux^  prose  and  poetry,  all  are  found 
and  in  great  numbers.    Geoffrey  Gaimar,  Wace,  Benoit  de  Sainte-More,  the 


Digitized  by  VjOOQ IC 


HE  VIE  IV S  AND  BOOK  NOTICES.  345 

Chanson  di  Roland^ "  the  national  song  of  the  Normans  as  well  as  of  all  French- 
men** (p.  125),  the  authors  of  the  Alexander,  Charlemagne,  and  Arthur  cycles 
of  romances, — ^all  show  the  activity  of  the  Anglo-Norman  intellect,  and  the 
avidity  of  Anglo-Norman  society  for  literature  of  all  kinds.  As  a  specimen  of 
these  romances  one  may  take  the  Arthurian  cycle.  **  One  thing,  however,  was 
lacking  for  a  time  to  the  complete  success  of  the  Arthurian  epic :  the  stamp  of 
authenticity,  the  Latin  starting-point . .  .  Geoffrey  of  Monmouth  makes  up  for 
this  deficiency*'  (p.  132).  Here  comes  in  the  Keltic  influence  on  subsequent 
English  literature.  These  legends  had  developed  in  Wales,  Brittany,  and 
Cornwall.  '*  The  Briton  harpists  had,  by  the  beauty  of  their  tales,  and  the 
sweetness  of  their  music,  early  acquired  a  great  reputation,"  and  the  great 
service  of  Geoffrey  of  Monmouth  was  that  he  collected  these  tales  in  Latin 
form,  and  passed  them  on  to  Wace  and  other  Norman-French  writers,  whence 
they  were  taken  by  Layamon  first,  and  after  the  lapse  of  more  than  two 
hundred  and  fifty  years  by  Sir  Thomas  Malory,  and  so  have  been  perpetuated 
as  the  only  well-developed  cycle  of  romance  in  English  literature.  What 
difference  does  it  make  if  William  of  Newbury  says  that  Geoffrey  made 
"Arthur's  little  finger  bigger  than  Alexander's  back,"  that  he  "lies  about 
almost  everything"?  Welcome  the  Jies,  if  we  take  them  at  their  true  value ! 
They  were  *'  turned  into  Latin  verse,  into  French  alexandrines,  into  Welsh 
prose  *' ;  "  the  finest  poems  the  Middle  Ages  devoted  to  them  were  written  on 
English  ground"  (p.  134).  But  while  the  chief  one,  this  romantic  cycle  was 
not  the  only  one,  and  romance  was  not  the  only  literary  form  that  these  Anglo- 
French  poets  and  prose-writers  cultivated.  '*  They  have  also  shorter  narratives 
in  prose  and  verse,  the  subject  of  which  is  generally  love,  drawn  from  French, 
Latin,  Greek,  and  even  Hindu  legends"  (pp.  141-2).  A  very  modem  spirit 
pervades  these  love-poems.  Human  nature  was  not  so  very  different  in  the 
thirteenth  century  from  what  it  is  in  the  nineteenth,  but  it  has  progressed 
since  the  eighth.  "To  sum  up  in  a  word  which  will  show  the  difference 
between  the  first  and  second  period :  on  the  lips  of  the  conquerors  of  Hastings 
odes  have  become  chansons" 

In  addition  to  the  love-literature,  the  literature  of  wit  and  humor  was 
developed.  **  The  French  who  were  now  living  in  England  in  large  numbers 
introduced  there  the  taste  for  merry  tales  of  trickery  and  funny  adventures, 
stories  of  curious  mishaps  of  all  kinds"  (p.  155).  "All  this  literature  went 
over  the  Channel  with  the  conquerors."  Thus  new  ideas  were  introduced 
among  the  English  people,  which,  prevalent  for  a  time  in  a  distinct  grade  of 
society,  gradually  permeated  the  mass  and  aided  the  development  of  English 
thought. 

The  Latin  literature  of  the  Anglo-Norman  period  is  next  treated,  the 
period  of  the  lengthy  Chronicles  of  English  history  written  during  the  twelfth, 
thirteenth  and  fourteenth  centuries,  from  Henry  of  Huntingdon  to  Ralph  Hig- 
don.  These  have  been  made  accessible  of  late  years  in  the  Rolls  series,  and 
constitute  a  unique  possession,  furnishing  very  full  materials  for  the  history  of 
the  times.  M.  Jusserand  gives  an  interesting  account  of  the  Latin  education  of 
the  time,  the  establishment  of  monasteries,  and  of  schools  and  libraries  under 
their  walls.  Paris  was  the  literary  capital,  and  its  University  in  the  twelfth 
century  was  a  great  resort.    After  its  model  the  universities  of  Oxford  and 


346  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Cambridge  were  formed,  "  but  their  celebrity  was  chiefly  local,  and  they  never 
reached  the  international  reputatipn  of  the  one  at  Paris"  (p.  173). 

English  writers  were  great  Latin  scholars.  "  They  handle  the  language 
with  such  facility  in  the  twelfth  century,  one  might  believe  it  to  be  their 
mother-tongue;  the  chief  monuments  of  English  thought  at  this  time  are 
Latin  writings.  Latin  tales,  chronicles,  satires,  sermons,  scienti6c  and 
medical  works,  treatises  on  style,  prose  romances,  and  epics  in  verse,  all  kinds 
of  composition  are  produced  by  Englishmen  in  considerable  numbers"  (p.  176). 

The  poem  of  Joseph  of  Exeter  on  the  Trojan  war,  long  attributed  to  Cor* 
nelius  Nepos,  is  aa  example  of  the  facility  with  which  Englishmen  wrote 
Latin  verse.  Mr.  Wright  has  made  the  works  of  these  Anglo-Latin  poets 
accessible  in  the  Rolls  series,  and  has  given  an  account  of  them  in  his  Bi^ 
grapkia  BHtannica  LiUraria^VoX,  II,  the  Anglo-Norman  period.  A  most 
interesting  satire  is  the  S^eulum  SiuUortim  of  Nigel  Wireker,  in  which  the 
stupid  monk  is  taken  off  in  the  person  of  Burnellus  the  ass,  who  visits  many 
nniversities,  finally  that  of  Paris,  where  he  matriculates  among  the  English 
nation,  and  after  seven  years  study  has  learnt  nothing  but  "  ya" : 

*'  Cum  nihil  ex  toto  quodcunque  docente  magistro 
Aut  socio  potuit  discere  praeter  ya." 

Finally  his  master,  who  has  been  searching  for  him  far  and  wide,  carries  him 
back  to  his  usual  duties. 

Geoffrey  de  Vinsauf  writes  his  Nova  Poetria^  in  which  he  lays  down  new 
rules  for  the  art  of  poetry.  Latin  prose  writings  are  more  abundant  than 
poetry !  witness  the  PoHcrtUicut^  sive  de  Nugis  Curialtum  of  John  of  Salisbury, 
whose  alternative  title  is  taken  by  Walter  Map  for  one  of  his  witty  works. 
Much  has  been  fathered  upon  Map,  which  he  most  probably  never  wrote, 
particularly  the  so-called  Apocalypse  and  Confession  of  Golias. 

The  last  chapter  of  the  second  book  treats  the  literature  in  the  English 
language,  and  here  we  have  a  succinct  account  of  the  revival  of  the  vernacular 
literature  after  a  long  repose.  It  had  never  been  extinct,  but  for  a  hundred 
and  fifty  years  after  the  Norman  Conquest  there  had  scarcely  been  more  than 
enough  to  preserve  the  continuity.  **  The  twelfth  century,  so  fertile  in  Latin 
and  French  works,  only  counts,  as  far  as  English  works  are  concerned,  devo- 
tional books  in  prose  and  verse  "  (p.  205).  We  may  add  the  later  entries  in 
the  Anglo-Saxon  Chronicle,  but  these,  with  the  modernisations  of  the  Gospel 
version,  the  series  of  Homilies,  the  so-called  Moral  Ode,  and  a  few  other  poems, 
constitute  all  that  we  have  in  English  before  the  year  1200.  The  native 
English  mind,  as  distinguished  from  the  Norman  mind,  was  repressed,  felt  no 
impulse  to  produce,  and  time  was  needed  to  amalgamate  the  two.  The 
thirteenth  century  saw  the  beginning  of  this  amalgamation,  and  before  its 
close  we  find  the  first  English  king  on  the  throne  in  the  person  of  Edward  I, 
and  the  beginnings  of  an  English  literature  consequent  upon  the  fusion  of  the 
two  races.  This  is  the  century  of  Layamon's  Brut,  the  Ormu/um,  the  Anavn 
RiwU,  the  Genesis  and  Exodus,  the  Owl  and  NighHngaU,  the  most  original 
poem  of  the  century,  the  romances  of  ffavelok  the  Dane,  and  King  Horn,  and 
the  continuation  of  chronicle  history  in  English  in  the  work  of  Robert  of 
Gloucester.    The  Pialier  is  paraphrased  in  English  verse,  and  metrical  lives 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES.  347 

of  saints  in  both  Northern  and  Southern  English  abound.  While  not  much 
originality  is  manifested,  English  authors  can  write  in  their  own  language  as 
well  as  in  Lathi  and  French,  and  the  demand  for  such  works  shows  the 
increasing  prevalence  of  the  language,  but  we  must  wait  until  the  fourteenth 
century  before  the  preponderance  is  on  that  side.  *'  Most  of  the  religious 
treatises  in  English  that  have  come  down  to  us  . . .  belong  to  the  first  half  of 
the  fourteenth  century.  In  the  thirteenth  . .  .  many  Englishmen  considered 
French  to  be»  together  with  Latin,  the  literary  language  of  the  country ;  they 
endeavored  to  handle  it,  but  not  always  with  great  success.  .  .  .  These 
attempts  become  rare  as  we  approach  the  fourteenth  century,  and  English 
translations  and  imitations,  on  the  contrary,  multiply'*  (pp.  2I3~4).  This 
represents  the  true  state  of  affairs.  In  the  first  half  of  the  fourteenth  century 
we  have  Robert  of  Brunne's  Handfyng  Synne  and  his  Chronicle,  both  transla- 
tions of  Anglo-French  works,  the  English  version  of  Robert  Grosseteste*s 
CasUl  of  Love^  the  voluminous  Cursor  Afundi,  the  great  repertory  of  legends, 
Dan  Michel's  Xemarse  of  Conscience,  and  Richard  RoHe  of  Hampole's  Prick  of 
Conscience  and  other  works,  issuing  from  Kent  and  Yorkshire  respectively, 
and  both  about  the  same  year,  1340,  and  many  others.  Thus  at  the  time 
of  Chaucer's  birth  there  was  an  English  people,  and  an  English  literature 
in  embryo,  and  the  time  was  ripe  for  a  great  English  poet.  *'  In  the  course 
of  the  fourteenth  century,  under  Edward  III  and  Richard  II,  a  double 
fusion,  which  had  been  slowly  preparing  during  the  preceding  reigns,  is  com- 
pleted and  sealed  forever;  the  races  established  on  English  ground  are 
fused  into  one,  and  the  languages  they  spoke  become  one  also.  The  French 
are  no  longer  superposed  on  the  natives,  henceforth  there  are  only  English  in 
the  English  island  '*  (p.  236).  It  had  taken  a  long  time,  about  three  hundred 
years,  but  it  had  been  finally  accomplished.  M.  Jusserand  sets  this  well  before 
us,  and  not  the  least  interesting  fact  stated  in  this  connection,  on  the  authority 
of  Bracton,  is  that,  in  the  thirteenth  century,  if  a  murder  had  been  committed, 
an  inquest  was  necessary  to  determine  whether  the  murdered  man  was  an 
Englishman  or  of  French  birth ;  in  the  former  case  no  fine  was  imposed.  The 
statute  of  1340  abolishes  the  "  presentement  d'Englescherie,"  showing  that  by 
this  time  no  distinction  was  made  as  to  the  genealogy  of  the  slain,  French- 
man and  Englishman  being  on  a  par. 

The  chapter  on  **  the  new  nation  "  is  one  of  the  most  interesting  in  the  book 
and  brings  before  us  the  life  of  the  fourteenth  century.  The  disappearance  of 
French  and  the  rise  of  English  is  traced :  the  race  as  well  as  the  language 
is  transformed.  A  real  English  Parliament  is  constituted,  and  *'  from  the  end 
of  the  fourteenth  century  an  Englishman  could  already  say  as  he  does  to-day : 
'  My  business  is  not  the  business  of  the  State,  but  the  business  of  the  State  is 
my  business.'  The  whole  of  the  English  constitution,  from  the  vote  on  the 
taxes  to  the  Aa6eas  corpus,  is  comprised  in  this  formula."  This  is  a  compendium 
of  the  principles  of  liberty,  and  is  very  different  from  the  French  king's  dic- 
tum :  "Z'//a/,  c^est  moi**  A  nation  in  which  political  liberty  had  been  thus 
developed,  in  which  trade,  commerce,  architecture,  art,  social  life,  had  pro- 
gressed, must  needs  have  a  progressive  literature.  The  sense  of  beauty  that 
had  been  manifested  in  other  directions  must  manifest  itself  in  literature,  and 
the  touch  of  a  great  poet  was  wanting  for  this  purpose.    The  hero  is  the  pro- 


348  AMERICAN  JOURNAL    OF  PHILOLOGY. 

duct  of  his  time»  but  the  hero  too  carries  forward  his  time  to  a  greater  perfec* 
tion.  The  first  half  of  the  fourteenth  century  was  preparatory  to  the  efflor- 
escence of  literature  in  the  second  half. 

The  leader  in  this  movement  was,  of  course,  Geoffrey  Chaucer,  the  poet  of  the 
new  nation  and  the  representative  of  the  new  age  ;  "  he  paints  it  from  nature 
and  is  a  part  of  it "  (p.  267).  We  have  a  full  chapter  on  Chaucer  and  his  works 
treated  chronologically,  based  on  the  publications  of  the  Chaucer  Society. 
His  early  reception  of  French  influence  is  noted,  his  acquaintance  with  the 
works  of  Deguileville,  Machault,  Des  Champs,  and  Granson,  but  we  do  not 
find  him  regarded,  with  Sandras,  as  a  mere  imitator  of  the  French  trouvhrs. 
The  critical  question  as  to  his  authorship  of  the  existing  version  of  the  Romauni 
of  the  Rose  is  barely  touched  upon.  '*  The  first  fragment  alone  might,  on  account 
of  its  style  and  versification,  be  the  work  of  Chaucer,  but  this  is  only  a  surmise, 
and  we  have  no  direct  proof  of  it ''  (p.  278,  note  2).  That  M.  Jusserand  is  very 
doubtful  about  it  is  shown  by  his  statement  in  the  text  that  *'this  trans- 
lation by  Chaucer  is  lost."  The  spurious  works  are  duly  enumerated  (p.  279, 
note  i).  Very  brief  outlines  of  the  Boke  of  the  Duchesse  and  of  the  House  of  Fame 
are  given,  and  a  much  fuller  one  of  Troilus  and  Criseyde^  in  which  poem  **  he 
surpasses  now  even  the  Italians  whom  he  had  taken  for  his  models,  and  writes 
the  first  great  poem  of  renewed  English  literature."  The  Canterbury  Tales 
are  duly  described  and  commented  on  with  sympathetic  appreciation.  **  There 
appear  in  perfect  light  his  masterly  gifts  of  observation,  of  comprehension,  and 
of  sympathy ;  we  well  see  with  what  art  he  can  make  his  characters  stand 
forth,  and  how  skilfully  they  are  chosen  to  represent  all  contemporaneous 
England  "  (p.  334).  With  respect  to  Chaucer's  skill  as  a  poet  and  its  effect 
upon  the  undeveloped  language,  M.  Jusserand  remarks:  "The  brilliancy  with 
which  Chaucer  used  this  new  tongue,  the  instant  fame  of  his  works,  the  clear 
proof  afforded  by  his  writings  that  English  could  fit  the  highest  and  the  lowest 
themes,  assured  to  that  idiom  its  definitive  place  among  the  great  literary 
languages  *'  (p.  338).  "  Chaucer's  efforts  were  not  exercised  in  vain ;  they 
assisted  the  work  of  concentration.  After  him  the  dialects  lost  their  import- 
ance ;  the  one  he  used,  the  East  Midland  dialect,  has  since  become  the  language 
of  the  nation"  (p.  309).  It  is  impossible  to  over-estimate  the  position  of 
Chaucer  in  a  history  of  English  literature.  His  influence  is  seen  throughout 
the  fifteenth  and  sixteenth  centuries  down  to  the  new  impulse  given  by  the 
study  of  Italian  and  classical  literature  in  the  reign  of  Elizabeth.  The  poets 
immediately  succeeding  Chaucer  are  simply  his  imitators,  and  serve  to  per- 
petuate his  influence  for  a  hundred  and  fifty  years.  In  "  the  group  of  poets  " 
authors  are  included  who  preceded,  as  well  as  those  who  were  contemporary 
with,  and  those  who  succeeded,  Chaucer,  as  witness  the  author  of  Sir  Gawayne 
and  the  Green  Knight  and  The  Pearl,  Lawrence  Minot's  poems  on  the  French 
and  Scottish  wars  of  Edward  III  are  touched  upon,  and  the  later  romance  of 
The  Bruce  by  John  Barbour,  circa  1375.  Gower  comes  in  for  some  considera- 
tion, but  "  he  is  aristocratic  and  conservative  by  nature,  so  that  he  belongs  to 
old  England  as  much  as  to  the  new  nation,  and  is  the  last  in  date  of  recog- 
nizable representatives  of  Angevin  Britain  "  (p.  364). 

A  much  greater  poet  than  Gower,  William  Langland,  has  a  chapter  devoted 
to  him  and  his  visions.    Professor  Skeat  has  made  us  well  acquainted  with  him. 


Digitized  by 


Google 


REVIEIVS  AND  BOOK  NOTICES,  349 

and  M.  Jusserand  has  devoted  one  of  his  works  to  a  study  of  Piers  Plowman^ 
some  extracts  from  which  are  included  in  this  chapter.  They  may  be  distin- 
guished by  the  more  rhetorical  style,  and  serve  to  acquaint  the  reader  still 
further  with  the  contents  and  value  of  Langland's  great  work,  for  he  stands 
next  to  Chaucer,  and  gives  us  a  view  of  another  side  of  life  in  the  fourteenth 
century.  Chaucer  takes  things  easy  and  simply  makes  merry  over  abases: 
Langland  is  the  genuine  reformer,  who  scores  abuses  in  Church  and  State  with 
the  spirit  of  Elijah  or  John  the  Baptist,  and  is  the  defender  of  the  rights  of  the 
people.  M.  Jusserand  places  the  revision  of  the  third  text  "  in  1398  or  shortly 
after,"  correcting  Professor  Skeat,  who  places  it  in  1393  (p.  375,  note  2).  A 
slip  of  ** Shrewsbury "  for  '* Shropshire "  on  p.  375  maybe  noted  in  passing. 
Langland's  work  is  Indispensable  for  a  knowledge  of  the  times,  and  serves  as 
a  complement  to  the  works  of  Chaucer.  "  Chaucer  and  Langland,  the  two 
great  poets  of  the  period,  represent  excellently  the  English  genius,  and  the  two 
races  that  have  formed  the  nation"  (p.  402),  the  former  representing  *'the 
latinized  Celts  "  ;  the  latter,  *•  the  race  of  the  Anglo-Saxons." 

A  chapter  on  the  prose  of  the  period  follows,  and  here  we  have  Sir  John 
Mandeville  relegated  to  the  region  of  myth,  a  conclusion  that  is  a  step  beyond 
the  results  reached  by  the  investigations  of  Col.  Yule  and  Mr.  Nicholson  in 
their  Encyclopedia  Britannica  article,  and  is  founded  on  the  work  of  Warner 
(1889),  and  later  discoveries  of  Mr.  Nicholson.  Maetzner  long  since  showed 
that  the  English  version  conld  not  have  been  written  by  the  author  of  the 
French  one,  and  its  language  plainly  shows  that  it  is  much  nearer  1400  than 
1350.  The  French  version  is  now  assigned  to  Jean  de  Bourgogne,  or  **  Joannes 
ad  Barbam,"  a  physician,  who  died  in  1372  at  Liige,  "where  his  tomb  was 
still  to  be  seen  at  the  time  of  the  French  Revolution"  (p.  407).  This  is  the 
tomb  that  was  formerly  considered  to  be  that  of  Sir  John  Mandeville.  We 
give  up  Sir  John  with  regret,  but  we  keep  his  Travels  as  "  one  of  the  best  and 
oldest  specimens  of  simple  and  flowing  English  prose,**  even  if  we  do  not  know 
who  wrote  the  English  version,  which  *'  was  made  after  1377,  and  twice  revised 
in  the  beginning  of  the  fifteenth  century."  This  investigation  leaves  Wyclif 
as  the  true  "father  of  English  prose,"  and  his  life  and  works,  in  both  Latin 
and  English,  which  are  very  numerous,  are  next  considered.  His  unfortunate 
death  postponed  the  Reformation  in  England  for  a  hundred  and  fifty  years, 
but  perhaps  the  time  was  not  yet  ripe  for  it. 

An  interesting  chapter  on  the  stage,  with  accounts  of  the  ancient  Mysteries 
and  a  brief  notice  of  the  Moralities^  succeeds,  and  the  volume  closes  with  too 
brief  an  account  of  fifteenth  and  early  sixteenth  century  literature.  It  pro- 
duces the  impression  of  having  been  written  in  a  hurry.  While  Lydgate  and 
Hoccleve,  and  even  James  I  of  Scotland,  are  re-echoes  of  Chaucer,  and  Hawes 
and  Henryson  but  continue  the  literary  tradition,  Douglas  and  Dunbar, 
especially  the  latter,  deserved  fuller  consideration,  and  to'  these  Sir  David 
Lyndsay  might  have  been  added,  but  perhaps  he  is  only  postponed.  Bishop 
Pecock  too  was  a  man  of  mark,  even  if  he  was  not  the  stuff  of  which  martyrs 
are  made.  Sir  Thomas  Malory  is  barely  named,  and  we  miss  any  treatment 
of  the  important  ballad  literature  of  this  century.  We  shall,  however,  await 
with  interest  the  succeeding  volumes,  for  M.  Jusserand  has  given  us  a  very 
readable  book. 

James  M.  Gaknett. 


Digitized  by  VjOOQ IC 


350  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Alfred  Hillebrandt.  Ritual-Literator.  Vedische  Opfer  und  Zauber.  Being 
volume  III,  part  2  of  Grundriss  der  Indo-Arischen  Philologie  and 
Altertttmskunde  (Encyclopedia  of  Indo- Aryan  Research),  edited  by 
Georg  BOhler.  'Strassburg,  Karl  J.  Trflbner,  1897. 

The  philological  status  of  no  less  than  six  important  sections  of  the 
Indo-European  community  of  peoples  has  been  summarized,  or  is  being 
summarized,  by  groups  of  competent  scholars.  Greek  and  Latin  philology 
took  the  lead,  and  Iwan  MttUer's  Handbuch  der  klassischen  Altertums- 
Wissenschaft  stimulated  the  production  of  no  less  than  four  other  *Grund- 
risse,*  all  of  which  were  undertaken  by  the  enterprising  firm  of  Triibner  in 
Strassburg  :  one  of  Germanic  philology,  edited  by  Hermann  Paul ;  another 
of  Romance  philology,  edited  by  Gustav  GrOber ;  next  that  of  Iranian 
philology,  edited  by  Wilhelm  Geiger  and  Ernst  Kuhn ;  and  finally  one  of 
Hindu  philology,  in  so  far  as  it  concerns  the  *  Aryan'  peoples  of  India, 
edited  by  the  eminent  Vienna  Indologist,  Professor  Georg  Bflhler,  with  the 
aid  of  about  thirty  scholars,  German,  Austrian,  English,  Dutch,  American, 
Indian  (both  native  and  Anglo-Indian).  Thus  far  there  have  appeared,  in 
addition  to  the  work  at  the  head  of  this  notice,  the  parts  containing 
Speyer's  Vedic  and  Sanskrit  Syntax,  Buhler's  Indian  Palaeography,  Jolly's 
Laws  and  Customs  (Recht  und  Sitte),  Macdonell's  Vedic  Mythology, 
Garbe's  Samkbya  and  Yoga  Philosophy,  and  Kern's  Manual  of  Indian 
Buddhism.  These  treatises  exhibit  fairly  the  scope  of  the  work  which 
proposes  to  deal  with  the  languages,  literatures,  history,  religion,  laws  and 
customs,  science  and  art  of  the  Aryan  Hindus.  Of  the  more  important 
Indo-European  philologies,  Indian  philology  is  the  most  recent,  and  stands 
in  need  of  concinnate  treatment.  Indeed,  rather  more  than  half  of  the 
subjects  outlined  in  the  prospectus  have  never  before  precipitated  them- 
selves from  out  of  the  amorphous  state  of  article  and  dissertation  into  a 
connected  form  of  treatment.  The  freshness  of  the  subjects  had  invited  in 
the  past  rather  the  edition  and  elucidation  of  the  difficult  texts,  the  state- 
ment of  strong,  salient,  interesting  points,  and  the  striking  of  the  paths  that 
were  to  be  the  familiar  exercise-ground  of  the  enquirer.  Of  this  there  is 
still  a  vast  deal  to  do.  We  need  but  mention  the  gaps  in  the  list  of  even 
the  first  editions  of  important  texts ;  but  there  is  plenty  of  good  timber  for 
the  rearing  of  a  provisional  house.  The  present  work  is  timely  and  being 
executed  by  strong  and  deft  hands.  The  Sanskritist  by  profession,  as  he 
glances  over  the  compact  pages  of  these  encyclopedic  treatises,  realizes 
that  his  knowledge  has  been  enlarged  and  the  basis  of  his  researches 
broadened ;  were  there  nothing  but  the  sifted  bibliographies  in  orderly 
array,  which  are  one  of  the  regular  requirements  of  each  contribution,  these 
treatises  would  not  have  been  written  in  vain. 

But  if  we  mistake  not,  this  series  is  destined  to  exercise  an  unusual 
amount  of  influence  in  broadening  and  solidifying  historical  and  institu- 
tional sciences  in  general.  India,  on  account  of  the  singular  nature  of  her 
literary  tradition,  is  destined  to  remain  a  very  permanent  source  of  knowl- 
edge, as  indeed  she  has  in  the  past  proved  herself  to  be  the  originator  of 
important  branches  of  historical  and  institutional  science.     The  compara- 


Digitized  by  VjOOQ IC 


RE  VIE  WS  AND  BOOK  NO  TICES.  3  5 1 

tive  absence  of  disturbance  from  the  oatside  has  ensured  her  continuous 
development  with  little  foreign  admixture  ;  her  own  unrivalled  systematic 
presentations  in  formal  treatises  of  her  religions,  laws  and  customs  have 
preserved  a  relatively  perfect  and  unbroken  record  of  that  development. 
India  is  largely  responsible  for  the  new  so-called  Science  of  Religions 
which  is  at  this  moment  profoundly  and  wholesomely  modifying  men's 
minds  in  their  views  of  religion  and  philosophy.  Students  of  comparative 
and  historical  jurisprudence  have  also  been  long  accustomed  to  turn  in  the 
same  direction  for  materials  and  for  organic  correlation  of  the  tissue  and 
bones  of  law.  Professor  Jolly's  work,  *  Kecht  und  Sitte,'  catalogued  above, 
offers  an  invaluable  digest  of  Hindu  law,  and  points  the  way  in  the  intricate 
maze  of  native  literature,  and  now  Professor  Hillebrandt's  contribution 
distinctly,  for  the  first  time,  assembles  and  summarizes  the  exceedingly 
systematic  and  painfully  painstaking  Vedic  treatises  on  home-life  and 
house-customs,  on  the  ritualistic  practices  of  the  Brahmans,  and  on  witch- 
craft, incantation,  exorcism  and  superstitions  in  general. 

No  student  of  India  will  say  that  a  more  ideally  competent  scholar  than 
Professor  Hillebrandt  could  have  been  called  to  this  particular  task.  He 
is  to  begin  with  an  all-round  Vedic  scholar  of  the  first  rank.  But  his 
special  qualification  is  found  in  his  prolonged,  patient  studies  of  the 
so-called  ^rauta-literature,  the  literature  of  the  great  Vedic  sacrifices, 
having  himself  edited  one  of  the  most  important  texts  of  that  ciass,  the, 
^ankhayana-^rautasutra,  and  having  elaborated  a  number  of  connected 
treatises  on  special  phases  of  this  literature — witness,  e.  g.,  his  essays  on 
the  New-moon  and  Full-moon  sacrifice,  on  the  Solstitial  Festivals  (Sonn- 
wendfeste),  and  others.  His  sketch,  as  he  modestly  calls  it,  of  the  contents 
of  the  ^rautasutras  (pp.  97-166),  though  based  to  a  considerable  extent  on 
Professor  Weber's  pioneer  labor  in  the  same  field  (Zur  Kenntniss  des 
Vedischen  Opferrituals,  Indische  Studien,  X  and  XIII),  is  the  piece  de 
resistance  of  the  entire  work.  To  this  Vedic  scholars  will  turn  most 
frequently  for  information  on  the  literature  of  the  subject,  for  guidance 
through  the  intricate  performances  of  the  numerous  priests,  for  explanation 
of  the  well-nigh  countless  technical  terms,  and  for  correlation,  where 
possible,  of  these  rigid  technical  performances  with  the  living  world;  in 
other  words,  for  an  account  of  their  development  out  of  popular  (ethno- 
logical) needs  and  beliefs. 

One  wish  connected  with  this  very  part  of  the  work  is  not  easily 
suppressed.  The  general  plan  of  the  series  follows  the  native  division  of 
the  Vedic  literature  into  revealed  texts  (^ruti\  and  traditional  texts  {smpt). 
This  division  is  both  mythical  and  unpractical,  and  it  is  to  be  regretted 
that  it  has  been  introduced  at  all.  Mantra  and  Brahma^a  (fruti)^  as  far  as 
their  subject-matter  is  concerned,  are  not  nearly  so  closely  allied  as  Brah- 
mana  and  Qrautasutra  (xmi*/!).  The  separation  of  the  last  two  is  in  reality 
impossible.  Professor  Geldner  has  in  charge  the  fruti  of  the  three  Vedas, 
and,  if  we  mistake  not  the  temper  and  the  trend  of  his  previous  investi- 
gations, he  will  deal  with  the  Brahma^as  of  the  three  Vedas  from  the  point 
of  view  of  literary  history  rather  than  from  the  point  of  view  of  ritualistic 


Digitized  by  VjOOQ IC 


352  AMERICAN  JOURNAL    OF  PHILOLOGY. 

detail.  The  character  of  the  Brahmanas  as  a  mixture  of  sacrificial  prescript 
with  legendary  illustration — hallacha  and  haggada^  as  they  are  called  in  the 
Talmudic  systematic  view — is  likely  to  engage  his  attention  and  to  preempt 
the  space  allotted  to  him  so  much  that  he  will  lack  the  opportunity  to 
present  a  complete  sketch  of  the  various  sacrifices  as  treated  in  the  Brih- 
manas  with  any  detail  whatsoever.  Indeed,  were  he,  after  all,  to  do  this, 
he  would  be  doing  over  again  what  Professor  Hillebrandt  has  done  so 
excellently  upon  the  basis  of  the  closely  allied  ^rautasutras.  It  is  not 
unlikely  that  this  encyclopedia  will  pass  into  subsequent  editions :  would 
it  be  too  much,  then,  to  ask  Professor  Hillebrandt  to  take  courage  and  break 
down  the  artificial  barrier,  and  to  treat  each  grauta-sacrifice  both  in  its 
Brahmana  and  Sutra  form  ?  At  least  he  might  add  to  his  citations,  without 
great  trouble  to  himself  and  without  the  need  of  great  additional  space,  the 
places  in  the  voluminous  Brahmanas  in  which  each  sacrifice  is  treated,, 
even  though  he  restricted  himself,  in  the  main,  to  his  original  expositions 
in  the  first  edition. 

The  entire  work  is  divided  into  four  parts.  The  first  deals  with  the 
beginnings,  the  literary  sources,  and  the  significance  of  the  ritual  practices  ; 
the  second  with  the  practices  of  home-life  (Grhyasutras);  the  third  with 
the  Vedic  sacrifice  (^rautasutras) ;  and  the  fourth  with  Vedic  witchcraft. 
All  four  are  products  of  scholarship  so  profound  and  judgment  so  nice  as 
to  leave  one  well  satisfied  that  this  compact  treatise  presents  a  picture 
whose  general  outline  will  never  be  altered  materially.  A  little  more 
breadth  might  have  been  desirable  for  the  last  chapter.  The  literature  of 
Hindu  superstition  is  so  extensive,  it  is  so  largely  dominated  by  transparent 
symbolism  and  by  concomitant  explanatory  circumstances  as  to  ensure  for 
it  that  same  basic  importance  in  general  ethnology  and  folklore  which 
confessedly  belongs  to  Hindu  religion,  law  and  house-customs.  The 
subject  of  omens  and  portents  alone  ^  would  justify  an  independent  treatise, 
as  would  also  the  subject  of  Vedic  physiology,  anatomy  and  medicine. 
But  even  in  these  matters  a  gratifying  beginning  has  been  made. 

A  few  details  may  be  added  to  this  notice.  P.  36, 1.  4  :  the  fdunakayaina 
is  after  all  not  original  with  the  Vaitana-sutra,  since  it  is  found  also  KB.  iv. 
6 ;  ^^.  3.  10.  7 ;  AQ.  9.  7.  i.  But  its  correlation  with  the  abhicdrakdma 
remains  interesting  as  illustrating  the  probably  apocryphal  name  of  the 
redactor  of  the  vulgate  version  of  the  AV. — On  the  same  page  near  the 
end  the  author  Upavar^a  must  be  identical  with  the  one  mentioned  JAOS. 
XI  376;  Kaugika,  Introduction,  p.  xvii. — On  the  same  page,  note  i,  the 
Yajfiaprayagcittasutra  is  doubtless  identical  with  the  six  prayagcittadhyiyas 
of  the  Vait.;  see  Garbe,  Introduction  to  his  edition  of  that  text,  p.  v; 
Weber,  Verzeichniss  der  Sanskrit-Handschriften,  vol.  II,  p.  83. — P.  41  (cf. 
also  p.  71)  :  for  the  division  of  the  sacrifice  into  pdkayajna^  etc.,  see  Gop. 
Br.  I.  5.  7,  and  23. — P.  64,  middle  :  for  an  attempt  to  explain  the  so-called 
Indrapl-rite  as  a  practice  to  prevent  the  death  of  a  husband  and  consequent 
widowhood,  see  the  present  writer,  ZDMG.  XLVIII  553,  note  2. — P.  76, 

1  Add  to  the  litenture  on  the  subject  (p.  184)  Hatfield's  treatment  of  the  Aufanasadbhutani 
(JAOS.  XV,  pp.  907  ff.),  and  the  adbhuu-texts  at  the  end  of  Athanra-Pari9Utas. 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES.  353 

bottom  :  Qveta  of  course  is  =  Paidya,  the  white  horse  of  Pedu  that  kills 
serpents  from  the  time  of  the  RV.  on. — P.  80,  I.  10:  the  practice  of 
slaughtering  a  cow  in  honor  of  a  guest  (obviously  obsolescent  in  the 
Grhyasutras)  is  embalmed  in  the  Vedlc  proper  name  Atithigva ;  see  AJP. 
XVII,  pp.  424  ft. — On  the  same  page,  middle :  To  the  practices  connected 
with  the  building  of  a  house  add  the  so-called  fyenaydga  or  fyenejyd, 
unearthed  by  the  present  writer,  JAOS.  XVI,  pp.  12  £f. — P.  90,  middle  : 
for  a  somewhat  more  precise  explanation  of  the  word  frdddha  see  AJP. 
XVII  411.— P.  169,  middle :  see  SBE.  XLII,  pp.  20  ff. 

Maurice  Bloomfield. 


Sophokles  Elektra.     ErkUrt  yon  Georg  Kaibel.     Leipzig,  B.  G.  Teubner, 
1896. 

The  new  Teubner  Sammlung  wissenschafilicher  Commentare  zu  grie^ 
chisehen  u,  rSmischen  Schriftstellern  challenges  attention  by  its  title  and 
still  more  by  its  programme.  No  concession  is  to  be  made  to  practical 
needs.  The  commentaries  are  to  address  themselves  to  mature  scholars, 
and  consequently  invite  the  most  rigorous  scrutiny.  To  teach  teachers  is 
a  perilous  task,  and  the  publishers  have  made  a  wise  selection  in  the 
editor  of  the  first  commentary,  and  the  editor  a  wise  selection  in  the  choice 
of  his  text  Apart  from  his  long  and  close  association  with  Wilamowitz, 
Kaibsl's  independent  work  would  lead  us  to  expect  a  penetrating  treat- 
ment of  his  author,  and  the  Elektra  of  Sophokles  is  just  the  play  to  bring 
out  the  value  of  the  principles  that  Kaibel  advocates.  By  a  rare  kindness 
of  fortune  we  are  able  to  compare  the  dramatic  methods  of  the  three  great 
coryphaei  of  Attic  tragedy  in  handling  the  same  theme,  and  interpretation 
necessarily  plays  a  conspicuous  part  in  the  Elektra.  True,  textual  criticism 
will  never  cease  from  troubling,  but  exegesis  must  come  to  the  front  when 
so  many  problems  of  tragic  psychology  are  involved  as  one  finds  in  this 
play  of  Sophokles.  **  Exhaust  interpretation  before  you  attack  the  text '^ 
is  a  wise  rule  of  a  great  teacher,  but,  unfortunately,  the  interpreter  too 
often  becomes  exhausted  before  the  interpretation  and  conjectural  criticism 
is  summoned  to  the  relief.  To  be  sure,  what  is  sometimes  called  conjec- 
ture is  not,  properly  speaking,  conjecture.  It  is  a  manner  of  proof-reading 
for  which  modern  slaves  of  the  vernacular  press  take  no  credit  to  them- 
selves, as  every  man  that  has  served  in  the  humble  capacity  of  reader 
makes  daily  *  emendations '  that  would  be  the  fortune  of  some  scholars,  if 
the  operations  were  performed  on  the  body  of  the  classic  texts.  It  is 
purely  a  matter  of  familiarity  with  the  range  of  thought  and  expressioni 
and  is  less  a  wonder,  the  more  one  is  at  home  in  a  given  language. 
Indeed,  it  is  very  questionable  whether  Hellenists  of  the  old  time  plumed 
themselves  to  much  on  their  corrections  as  do  men  of  our  day,  and  the 
praises  that  have  been  showered  on  some  of  Keiske's  work  in  that  line 
would  doubtless  have  astonished  that  large-limbed  scholar  himself.  But 
a  homily  on  the  abuses  of  conjectural  criticism  would  be  sadly  out  of  place 
in  a  review  of  Kaibkl's  Elektra,  for  in  the  very  first  lines  of  his  prelimi* 


Digitized  by  VjOOQ IC 


354  AMERICAN  JOURNAL   OF  PHILOLOGY. 

nary  observations  he  gives  us  to  understand  that  he  does  not  favor  a  ready 
resort  to  conjecture.  *  Emendation,'  he  says,  *  is  a  rare  flower,  which  grows, 
if  anywhere,  on  the  rock  of  interpretation,  and  the  longer  one  pursues  this 
flower,  the  better  does  he  know  how  hard  it  is  to  find  or  to  pluck.'  The 
application  of  this  remark  to  the  text  of  Sophokles  is  near  at  hand,  and 
Kaibel's  protest  against  *  the  flood  of  conjectures  by  which  the  text  of 
Sophokles  has  been  marred'  will  be  more  readily  echoed  to-day  than  it 
would  have  been  twenty  or  thirty  years  ago.  Especially  to  be  taken  to 
heart  are  Kaibel's  words  about  the  patching  of  texts  by  parallel  passages, 
a  kind  of  skin-grafting  to  which  critics  are  prone  ;  and  he  is  very  emphatic, 
as  emphatic  as  was  the  late  Mr.  Matthew  Arnold,  on  the  rights  of  poetic 
individuality.  'Every  great  poet,'  he  says,  *  himself  creates  his  language 
and  bis  art.  Developed  with  him,  they  grow  with  him»  and  die  with  him  as 
his  daifiuv.  He  has  not  inherited  them  and  cannot  transmit  them.'  And 
yet  these  words  must  be  taken  with  some  reserves.  In  Greek  the  type  of 
each  form  of  art,  of  each  sphere  of  language,  is  more  potent  than  is 
commonly  supposed,  more  potent  against  other  forms  of  arts,  other  spheres 
of  language,  than  any  amount  of  individual  sympathy.  The  lyric  poet  is 
nearer  to  the  lyric  poet,  no  matter  how  diverse  bis  temperament,  than  he 
is  to  the  tragic  poet  of  kindred  genius.  Self-ravelling  is  the  best  material 
for  darning  the  poet's  text,  just  as  a  poet  is  always  bis  own  best  interpreter, 
but  in  default  of  that  one  may  well  resort  to  rival  looms. 

After  Kaibel's  frank  statement  of  his  attitude  towards  conjectural 
criticism,  the  student  of  this  edition  of  Sophokles'  Elektra  will  not  expect 
to  encounter  a  host  of  irritating  and  inconclusive  conjectures,  and  at  least 
one  old  fellow-student  of  Vahlen's  has  read  with  a  certain  satisfaction  the 
tribute  that  Kaibbl  has  paid  to  the  sound  methods  of  that  eminent  scholar, 
by  whose  example  conservative  souls  have  been  strengthened  in  their, 
adherence  to  the  precept  of  the  great  master  already  cited. 

I  have  given  at  some  length  this  confession  of  faith  because  Kaibel's 
Elektra  is  the  initial  volume  of  a  series  that  seems  destined  to  have  a 
decided  influence  on  the  editorial  work  of  classical  scholars.  At  first,  as  I 
have  intimated,  the  veteran  student  may  not  be  willing  to  grant  that  so 
much  remains  to  be  done  for  the  interpretation  of  Sophokles,  may  resent 
the  assumption  of  superior  insight  into  the  meaning  of  a  poet  who  has 
claimed  the  study  of  so  many  gifted  scholars;  but  there  is,  after  all,  no 
arrogance  in  Kaibel's  claims,  as  there  is  no  arrogance  in  any  form  of 
devotion.  The  secret  that  is  revealed  to  kindred  genius  at  a  glance  may 
be  won  by  lesser  spirits  through  steadfast  and  loving  contemplation.  That 
a  new  Elektra  has  emerged  from  this  study  of  Kaibel's  would  be  saying 
too  much,  but  passage  after  passage,  scene  after  scene,  has  received 
welcome  light  and  fresh  color. 

It  is  to  be  regretted  that  Kaibel  has  not  followed  Wilamowitz  in  accom- 
panying his  edition  with  a  translation.  A  translation  is  often  the  best 
commentary,  and  this  Kaibel  recognizes.  If  Christ  had  given  us  a  Latin 
rendering  of  Pindar,  we  should  not  be  in  doubt  as  to  his  judgment  on 
moot-points  without  number.     But  Kaibel  evidently  subscribes  to  Wilam* 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES.  355 

owitz's  doctrine  that  a  poet  must  be  translated  into  poetry,  and  for  this  he 
professes  not  to  have  the  necessary 'gift,  and  Jebb's  illustrious  example 
has  not  induced  him  to  attempt  a  prose  version. 

The  editor's  judgment  as  to  the  MS  basis  of  the  text  may  be  given  in  a 
few  words.  A  theoretical  A  and  a  theoretical  11  are  the  ultimate  sources 
of  L  and  P.  L'  presents  a  very  faulty  text,  corrected  by  L'  after  a  member 
of  the  P  family.  Whoso  admits  the  indispensableness  of  the  corrections 
of  L'  admits  the  value  of  11  and  the  consequent  value  of  P.  No  new 
collation  has  been  found  necessary,  and  Kaibel  does  not  attach  any 
countervailing  importance  to  the  facsimile. 

The  Introduction  is  largely  taken  up  with  an  analysis  of  the  drama  and  a 
suggestive  comparison  of  the  Elektra  of  Sophokles  and  the  Elektra  of 
Euripides,  in  which  the  idealism  of  the  elder  poet  is  not  extolled  at  the 
expense  of  the  naturalism  of  the  younger.  *  Euripides,'  says  Kaibel  in 
substance,  *has  shown  wonderful  power  of  invention  in  creating  a  heroine 
of  which  no  representative  of  the  modern  school,  of  the  "experimental 
romance,"  need  be  ashamed.'  He  had  no  *  documents';  he  was  hampered 
by  a  mass  of  traditions,  which  had  to  be  respected  and  yet  so  reinterpreted 
and  so  readapted  as  to  explain  out  of  the  environment  the  character  of  the 
heroine,  which  he  conceived  and  created  in  his  own  way.  The  thought  of 
evolving  Elektra  as  a  necessary  product  of  her  milieu  was  not  his.  It  was 
due  to  Sophokles.  But  Euripides  felt  that  Sophokles  had  not  made  the 
most  of  it,  that  the  fruitful  idea  had  not  been  made  to  yield  all  its  dramatic 
possibilities,  and  the  Euripidean  Elektra  was  the  result — not  separated  far 
in  time  from  that  of  Sophokles. 

This,  it  will  be  observed,  is  a  very  different  tone  from  that  of  the 
traditional  criticism  of  the  Euripidean  Elektra.  The  *  dramatic  possi- 
bilities'  of  the  life  of  an  old  maid  in  Greek  antiquity  cannot  be  measured 
by  modern  standards.  Nay,  unless  the  process  of  transformation  is 
arrested,  fifty  years  hence  Americans  will  need  a  learned  apparatus  in 
order  to  understand  the  old  maid  of  the  nineteenth  century,  and  perhaps 
even  Frenchmen  will  be  at  a  loss  to  comprehend  the  vieilhfiUe  of  Balzac. 

As  has  been  intimated,  the  commentary  shows  on  almost  every  page  the 
value  of  fresh,  independent  study,  but  the  character  of  the  work  precludes 
the  production  of  specimens.  More  open  to  comment  are  the  grammatical 
notes.  Indeed,  the  admissibility  of  grammatical  notes  in  an  edition  of  so 
high  a  reach  as  this  will  be  questioned  by  some.  Assuredly,  trivial 
matters  ought  not  to  be  treated,  but  what  is  trivial,  what  not,  is  always 
doubtful.  However,  the  points  that  are  taken  up  are  usually  despatched 
in  a  few  words  with  Wilamowitzian  resoluteness.  And  yet  the  conclusion 
reached  is  not  always  self-evident.  For  instance,  some  note  on  ^poifietf 
(v.  58)  seems  to  be  necessary.  Yet  the  explanation  given,  which  is  credited 
to  Vahlen,  is  not  satisfactory.  To  say  that  the  optative  makes  the  action 
designed  depend  on  circumstances  is  too  vague.  The  irregular  sequence 
might  be  explained  by  the  intrusion  of  the  wish,  but  a  more  simple  expla- 
nation is  at  hand.  KeKpv/i/ihov  involves  an  action  prior  to  that  of  the  leading 
verb,  so  it  is  at  once  a  perfect  and  a  pluperfect,  and  these  intercalated 


Digitized  by  VjOOQ IC 


356  AMERICAN  JOURNAL    OF  PHILOLOGY, 

clauses  are  responsible  for  many  such  shifts,  "kafiuv  re  koI  rrrraadat  (v.  274) 
is  not  happily  interpreted  by  hipelv  re  koI  /i^  hipeiv.  Like  all  verbs  of  priva- 
tion, TTfTdaOcu  is  much  more  than  fi^  ^Belv  (see  on  Find.  Pyth.  6,  22),  and 
there  is  bitter  emphasis  in  the  article  hipelv  9  dfwioc  koX  rd  TfrraaOatf  the 
same  bitter  emphasis  that  Kaibel  himself  recognizes  in  the  article  flse- 
where,  v.  166  ;  rdv  avffWTw  \  olrov  i;fOMTa  KwcStv.  On  v.  318 :  tov  Koaiyv^ov  ri 
^^r*  V^ovTo^  ij  fjiTJkovTdq^  K.  explains  the  genitive  as  a  partitive.  '  Nicht  das 
ganze  Wesen  oder  Handeln  des  Bruders  kommt  in  Betracht,  sondem  nur 
cin  Theil.'  rot)  KaaiyviiTov  is  equivalent  to  rb  tov  k.,  and  if  any  one  chooses 
to  call  that  a  partitive,  he  is  welcome.  But  a  parallel  with  irulv  tov  olvov 
does  not  commend  itself,  and  rd  TCnfd'  ebvow  ndpa  (v.  1203)  is  possessive 
rather  than  partitive,  as  is  shown  by  parallels  with  the  possessive  pronoun. 
On  v.  590 :  EKpakowf  ix^ic,  Kaibbl  decides  the  moot-point  as  to  the  transi- 
tiveness  or  intransitiveness  of  ix^  with  a  positiveness  hardly  justified  by 
recent  surveys  of  the  history  of  this  construction.  *Ueberall  kann  und 
muss  ix^  intransitiv  "sich  verhalten"  sein.'  The  quarrel  is  one  between 
historical  growth,  which  favors  the  transitive  view,  and  logical  consistency, 
which  favors  the  intransitive  view,  and  is  not  to  be  settled  by  a  ukase. 
These  are  a  few  of  the  various  grammatical  points  in  which  agreement 
with  Kaibel  is  not  inevitable;  nor  am  I  inclined  to  subscribe  to  the 
sweeping  sentence  (p.  90)  that  the  ethical  character  of  Greek  metres 
cannot  be  determined  by  universal  rules,  that  it  depends  on  poetical 
handling,  and  especially  on  the  environment.  This  is  what  Wilamowitz 
maintains  for  Glyconic  verses,  and  the  flutter  of  the  Glyconic  metres  may, 
indeed,  serve  to  express  a  variety  of  emotions,  but  there  are  certain 
logaoedic  measures — notably  those  in  which  syncope  abounds — which  have 
a  more  uniform  tone,  and  Kaibel  is  somewhat  inconsistent  with  himself 
when  he  remarks  on  v.  171  =  192  that  the  bitter  blame  at  the  close  is  well 
marked  by  the  strongly  syncopated  iambic  verses.  There  is  no  reason  to 
me  discernible  why  similar  checks  to  movement  should  not  produce  the 
same  results  in  logaoedic  measures. 

B.  L.  GiLDBRSLEEVE. 


Digitized  by  VjOOQ IC 


REPORTS. 

Romania,  Vol.  XXIV  (1895).    First  half. 

Janvier.. 

Victor  Friedel.  Deux  fragments  du  Fierabras.  Etude  critique  sur  la  tra- 
dition de  ce  roman.  55  pages,  with  facsimile  of  one  of  the  fragments.  Two 
.scraps,  consisting  of  a  single  leaf  from  each  of  two  separate  MSS,  are  made 
the  basis  of  an  elaborate  investigation,  valuable  as  a  showing  of  method  rather 
than  of  results. 

C.  Boser.  Le  remaniement  proven^al  de  la  Somme  U  Rot  et  ses  derives. 
30  pages.  The  Somme  le  Roi  (i.  e.  '  Summa'  Regis),  otherwise  entitled  in  the 
French  texts  Mirdr  du  Monde  and  Livre  des  vices  et  des  vertus^  exists  in 
Provencal  in  three  different  forms,  the  relations  of  which  to  each  other  and  to 
the  French  are  here  elucidated,  with  copious  comparisons. 

J.  Bedier.  Fragment  d'un  ancien  mystire.  '*  On  ne  saurait  se  representer 
ce  mystdre  si  modeste  que  compost,  monte,  joue  dans  quelque  petite  et  pauvre 
ville.  Cela  m£me  est  significatif  et  prouve  que  le  genre  ^tait  ixhs  repandu, 
Xxhs  aime.  Que  les  manuscrits  de  ces  myst^res,  si  nombreux  qu'on  les 
suppose,  se  soient  tous  perdus,  c'est  encore  ce  que  notre  fragment  nous  fait 
comprendre:  c'etaient  des  po^mes  d'occasion,  rimes  sans  nuUe  pretention 
litteraire :  la  f6te  pass^e,  nul  ne  s*en  souciait  plus." 

R.-J.  Cuervo.  Los  casos  enclfticos  y  proclfticos  del  pronombre  de  tercera 
persona  en  castellano.  19  pages  {it  stdvre).  An  admirable  historical  study. — 
Illam^  ilium,  illas,  ilhs  gave  la^  lo,  las,  los ;  fV/t ,  illis  became  K,  lis^  and  later  le, 
les.  Accordingly,  lo  (masc.  and  neut.),  la,  hs,  las  are,  etymologically,  accusa- 
tives and  le,  les  datives ;  but  the  case-forms  began  early  to  be  confused,  le 
being  first  substituted  for  lo,  then  les  for  los,  and  finally  la^  las  and  lo,  los  for  le, 
les.  From  a  long  statistical  list  chronologically  arranged,  giving  the  usage  of 
numerous  authors  in  regard  to  this  point,  it  appears  that  el  leismo  (the  substi- 
tution oile  for  lo  in  the  accusative)  culminates  in  the  i6th  and  17th  centuries, 
among  the  writers  of  Madrid  and  the  surrounding  provinces.  The  usage 
found  in  a  given  work  does  not  always  correspond  to  that  of  the  native  region 
of  its  author.  If  we  were  to  judge  from  Saavedra  of  the  usage  of  Murcia  in 
the  17th  century,  or  from  Valera  [lately  Spanish  Minister  at  Washington]  of 
that  of  Cordova  in  the  19th,  we  should  be  completely  deceived;  modem 
writings  of  a  local  character  reveal  to  us  that  the  peasants  of  the  author  of 
Pepita  Jim/nez  are  to-day  as  devoted  to  lo  as  were,  in  earlier  centuries,  Feman 
Perez  de  Oliva  and  Juan  de  Mena. 

Melanges.  J.  Cornu.  Combre  et  derives.  Refers  the  group  of  words  to 
which  belong  Fr.  encombrer,  decombrer  and  their  congeners,  usually  connected 


Digitized  by  ^ 


358  AMERICAN  JOURNAL    OF  PHILOLOGY. 

heretofore  with  Lat.  cumulare,  Fr.  combUr  (but  cf.  A.  J.  P.  XVII  500),  to  Lat. 
cumera^  cumerus,  'grand  panier  d*osier  pour  conserver  le  ble/  "mot  qui  a  fort 
bien  pu  avoir  aussi  le  sens  de  nasse  ou  de  tel  autre  engin  semblable  servant  k 
prendre  du  poisson.  A  combre  od  s'encombrent  les  poissons  peuvent  dtre 
aisement  ramenes  les  difFerents  sens  qu'ont  pris  ses  derives.'* — A.  Thomas. 
Fr.  cormoran  [=  Eng.  carmorani],  4  pages.  '*  Je  considire  cormarant  {cartn»ran 
est  une  alteration  inexpliquee  de  cormaratty  qui  est  pour  cormarant)  comme 
representant  un  plus  ancien  corp  marenc,  et  je  n'hesite  pas  k  admettre  en  ancien 
frangais  I'existence  d'un  adjectif  marenc^  tire  du  latin  mare  k  Taide  du  sufHxe 
germanique  -ing,^* — A.  Thomas.  Fr.  girouette,  '^  Gyrovagum^  passant  par 
gyrovaoy  gyrovo  a  d(i  aboutir  4  *giroUy  d'oii  les  diminutifs  girouet^  girouette** — 
A.Thomas.  Yx,hampe\  prov,  mod,  gamo^gamoun.  Littr^  gives  to  the  word 
hampe  five  different  meanings,  the  first  three  of  which  are  closely  related  to 
each  other  (shaft  of  halberd,  etc.).  "  Aux  sens  1-3,  hampe  est  une  alteration 
recente  de  karute^  qui  s*emploie  encore,  sous  la  forme  haiite  ou  ante,  dans  les 
deux  premieres  acceptions.  Je  considire  les  sens  4-5  [*  breast  of  stag,'  in 
venery,  etc.]  comme  appartenant  4  un  mot  different.  Le  dictionnaire  frangais- 
allemand  de  Mozin  traduit  *  la  hampe  du  cerf '  par  *  die  Wamme  oder  Brust  des 
Hirsches.'  Or  Wamme  presente  en  allemand  la  forme  parall^le  wampe .  . . 
M.  Godefroy  a  recueilli  deux  anciens  exemples  de  wampe^  vampe  au  sens  de 
'  empeigne  de  Soulier':  c*est  une  extension  toute  naturelle  du  sens  de  'ventre, 
peau  de  ventre.*  *'  M.  Thomas  is  apparently  unaware  that  this  O.Fr.  word 
vampe  survives  with  its  ancient  meaning  in  the  Eng.  vamp,  and  that  he  has 
here  brought  incidentally  to  light,  for  the  first  time,  the  true  etymology  of  a 
word  which  the  English  lexicographers  refer,  awkwardly  enough,  to  the  Fr. 
avant'pied  (which,  however,  itself  means  'vamp').  Avampi^  as  a  gloss  to 
antipodium  occurs  in  a  MS  of  the  second  half  of  the  13th  century  (cf.  Rom. 
24,  171). — J.  J.  Jusserand.  Les  contes  4  rire  et  la  vie  des  recluses  au  XII^ 
siicle  d'apr^s  Aelred,  abbe  de  Rievaulx.  6  pages.  " '  Vix  aliquam  inclusarum 
hujus  temporis  solam  invenies,  ante  cujus  fenestram  non  anus  garrula  vel 
nugigerula  mulier  sedeat,  quae  eam  fabulis  occupet,  rumoribus  aut  detractio- 
nibus  pascat,  illius  vel  illius  monaci  vel  clerici  vel  alterius  cujuslibet  ordinis 
viri  formam,  vultum  moresque  describat.  Illecebrosa  quoque  interserat,  puel- 
larum  lasciviam,  viduarum,  quibus  licet  quidquid  libet,  libertatem,  conjugum 
in  viris  fallendis  explendisque  voluptatibus  astutiam  depingat.'  Et  il  s'agit 
bien  de  contes  4  rire,  de  vrais  fabliaux  4  I'etat  embryonnaire,  car  Aelred 
ajoute:  Os  interea  in  risus  cachinnosque  dissolvitur,  et  venenum  6um  suavitate 
bibitum  per  viscera  membraque  diffunditur."— P.  Meyer.  Guillem  d'Autpol 
et  Daspol.  Identification. — G.  Paris.  La  Dance  Mctcabri  de  Jean  le  F^vre. 
**  Ce  qui  est  certain,  c'est  que  le  po^me  de  Jean  Le  Fivre  portait  le  nom  de 
Dante  Macabre',  J'ai  dej4  indique  que  c'est  cette  forme,  et  non  macabre,  qui 
est  la  seule  authentique  .  .  .  Mais  qu'est-ce  que  ce  Macabre,  dont  le  nom 
etait  accole  4  celui  de  dance^  pour  signifier  la  danse  de  la  mort,  dds  le  XlVe 
si^cle?  On  ne  pent  14  dessus  que  faire  des  conjectures.  Le  nom  de  MacabrJ 
est  une  prononciation  populaire  de  MacabS-=.  Macchabaeum  ...  Je  serais  plus 
port^  4  voir  dans  Macabre  le  peintre  qui  avait,  le  premier  peut-fitre,  represenie 
sous  la  forme  d'une  danse  menee  par  la  mort  I'appel  fatal  qu'elle  adresse  4 
tous  les  humains." 


Digitized  by  VjOOQ IC 


REPORTS.  359 

Comptes  rendus.  Vicenzo  Crescini.  Manualetto  provenzale  (P.  Meyer). 
**Cel  utile  manuel  nous  apporte  une  nouvelle  preuvc  du  developpement  que 
les  Etudes  proven^ales  prennent  en  Italie  .  . .  Tel  qu'il  est,  le  Manualetto  est 
un  ouvrage  fort  recommendable." — ^Joseph  Bedier.  Les  Fabliaux,  a*  Edition 
(Ch.-M.  dcs  Granges).  8  pages.  A  penetrating  review,  devoted  chiefly  to  a 
scrutiny  of  Brier's  somewhat  startling  attempt  at  a  refutation  of  the  Orien- 
talist or  'Indian'  theory  of  the  dissemination  of  folk-tales.  "Toute  une 
partie  de  la  discussion  contenue  dans  les  chapitres  suivants  est  done  excel- 
lente.  Abondance  de  documents,  sOret^  d'interpretation,  rien  n'y  manque ; 
une  seule  chose  y  est  de  trop :  la  perpetuelle  demangeaison  de  vexer  les  orien- 
talistes,  qui  am^ne  M.  B.  i  mesurer  ses  conclusions  beaucoup  moins  d'apris 
leur  valeur  rdelle  qu'en  proportion  du  depit  qu'elles  peuvent  causer  aux  autres 
.  .  .  Sur  la  forme  m^me  du  mot  fabliau  adopte  par  M.  B.,  on  pourrait  faire 
quelques  reserves.  M.  B.  s'autorise  d'une  raison  ^^usage:  on  a  dit,  on  dit,  on 
ecrit  partout  fabliau.  Soit.  Mais  cet  ouvrage  est  le  plus  approfondi  et  le 
plus  complet  qui  ait  ete  consacre  au  genre ;  il  va  faire  desormais  autorite :  son 
titre  se  fftt  impose,  avec  la  forme  correcte  fableau^  sMl  avait  plu  4  I'auteur  de 
heurter  de  front  un  prejuge."  Even  the  philologists  (it  is  Gaston  Paris  that 
stands  sponsor  for  the  new  form  fabUau) — who  know  so  much  better  than  any 
one  else  that  human  speech,  even  at  its  highest  and  best  estate,  is  fairly 
honeycombed  with  so-called  errors  and  with  real  inconsistencies — are  only 
too  much  like  the  every-day  language-tinkers  and  authors  of  Dites — ne  dites 
pas  and  DonU^  in  their  proneness  to  forget,  in  dealing  with  questions  of 
linguistic  correctness,  that,  in  face  of  the  gentle  yet  imperious  authority  of 
established  usage,  the  *  etymological  argument '  ought  to  have  about  as  little 
weight  in  orthology  (the  word  is  needed)  as  in  logic. — D.  Merlini.  Saggio  di 
ricerche  suUa  satira  contro  ii  villano  (G.  Paris).  **  La  partie  la  meilleure  de 
son  livre  est  consacree  k  cette  litt^rature  (de  I'antagonisme,  k  peu  pr^s  restreint 
k  I'ltalie,  entre  les  citadins  et  les  campagnards) ;  il  a  montre  notamment  avec 
succ^s,  k  ce  qn'il  me  semble,  le  rapport  o{i  elle  est  avec  le  developpement  de 
la  Gftnmedia  delV  arte  et  la  creation  du  type  du  Zanni  (=  Gianni  =  Giovanni), 
divis^  plus  tard  en  Arlequin  et  Brighella:  Zanni  est  primitivement  lefaccAino 
venu  des  montagnes  bergamesques  qui  faisait  k  Venise,  aux  XV«  et  XVI« 
siicles,  tous  les  gros  ouvrages  du  port  et  s'attirait  les  railleries  des  gens  de  la 
ville  (quelque  chose  comme  TAuvergnat  k  Paris  [and  the  Gallego  at  Madrid])." 

Chronique.  Thor  Sundby,  professor  of  the  French  language  and  literature 
at  the  University  of  Copenhagen,  died  Nov.  19, 1894,  at  the  age  of  64  years. 
His  chief  works  were  a  study  of  the  life  and  writings  of  Brunetto  Latino 
(1869;  translated  into  Italian,  1884),  ^  study  of  Pascal  (1877),  ^"d  a  Diction- 
naire  danois-fran^ais  et  fran^ais-danois^  2  vols.,  1883-84).  He  is  succeeded  in 
his  chair  by  Kr.  Nyrop,  the  eminent  Romance  scholar,  who  was  adjunct 
professor  in  the  same  university. — W.  Borsdorf  has  been  appointed  professor 
of  Romance  philology  at  the  recently  founded  Welsh  *  University  College  *  of 
Aberystwyth. — For  the  marriage  (October,  1893)  of  Vittorio  Cian,  the  young 
Italian  literary  historian,  his  friends  inaugurated  a  newly  devised  sort  of /rr 
ncBU.  Twenty-five  of  them  combined  in  the  publication  of  a  magnificent 
quarto  volume  of  more  than  450  pages,  in  which  their  contributions,  relating 
chiefly  to  Italian  literature,  were  printed  in  the  chronological  order  of  the 


Digitized  by  VjOOQ IC 


360  AMERICAN  JOURNAL    OF  PHILOLOGY. 

subjects  treated. — G.  Weigand,  author  of  Nouvelles  recherches  sur  h  roumain  de 
ristrie  and  of  Die  Aromunen^  has  organized  at  Leipzig,  with  the  support  of  the 
Rumanian  Government,  an  Institut  fUr  romdnische  Sprache. — On  the  occasion 
of  the  fourth  centenary  of  the  death  of  Matteo  Maria  Boiardo  (December  19, 
1894),  the  commune  and  province  of  Reggio  d*£milia  published,  under  the 
editorship  of  N.  Campanini,  and  with  the  collaboration  of  a  number  of  the 
leading  Italian  litterateurs,  a  handsome  volume  devoted  to  the  life  and  work 
of  the  Count  of  Scandiano  (Bologna,  pp.  479). — The  Societe  de  Thistoire  de 
France  has  published  the  first  volume  of  a  new  edition  of  x)\^  Journal  de  Jean 
de  Roye^  commonly  known  as  *la  chronique  scandaleuse'  of  King  Louis  XL 

Livres  annonc^s  sommairement.     16  titles. 

Avril. 

P.  Meyer.  Anciennes  gloses  frangaises.  Two  sets  of  glosses  of  Latin 
terms,  numbering  respectively  117  and  55  items,  indexed  and  instructively 
annotated. 

H.  Morf.  Notes  pour  servir  i  Thistoire  de  la  l^gende  de  Troye  en  Italic 
(suite  et  fin).  IV.  La  version  venitienne.  23  pages.  "C'est  un  texte  tr^s 
interessant  au  point  de  vue  linguistique  et  tris  curieux  au  point  de  vue  de 
I'histoire  litteraire,  et  qui,  4  ce  double  titre,  meriterait  bien  que  quelque 
editeur  intelligent  lui  donnat  ses  soins." 

P.  Meyer  et  N.  Valois.  Po^me  en  quatrains  sur  le  Grand  Schisme  (1381)- 
**  Ayant  eu  la  curiosite  de  le  lire,  je  le  jugeai  assez  interessant  pour  meriter 
d'etre  public.  Je  le  copiai,  et  M.  Noel  Valois,  mieux  prepare  que  personne, 
par  ses  travaux  sur  le  Grand  Schisme,  i  apprecier  un  po^me  qui  reflate  les 
sentiments  de  TUniversite  de  Paris  sur  ce  grand  evinement,  voulut  bien,  4 
ma  demande,  rediger  la  preface  qu^on  va  lire.''  The  poem  consists  of  73  four- 
line  strophes,  in  monorime. 

Ou  nom  de  Yhesucrist  qui  fut  vray  Dieu  et  homme 

Je  weil  ycy  dieter  et  compter  une  somme 

Ou  sera  recite  le  fait  et  la  voie  comme 

Est  tires  et  sachez  le  saint  siege  de  Romme. 

Berthelemieu  du  Bar  si  se  dist  estre  pape, 
Et  Robert  de  Geneve  lui  veult  oster  la  chappe. 
Chascun  comme  saint  pere  lez  biens  de  Tautel  happe, 
Et  pour  lez  soustenir  Tun  fiert  et  I'autre  frappe. 

R.-J.  Cuervo.  Los  casos  encliticos  y  procliticos  t/J«).  45  pages.  '*  Resu- 
miendo  esta  disertacion  ya  demasiado  larga,  dire  que  en  gran  parte  de  los 
dominios  del  castellano  se  ha  conservado  y  se  conserva  con  precision  el  uso 
etimol6gico  de  los  casos  de  //;  que  habiendo  nacido  la  confusi6n  entre  el 
acusativo  lo  y  el  dativo  le  por  causas  morfologicas,  se  ha  extendido  por  causas 
sintaticas,  y  al  fin  por  extension  abusiva  hasta  predominar  notablemente  el  le 
en  el  lenguaje  comun  de  Castilla  ...  A  esto  [la  distinci6n  de  oficios]  han 
encaminado  sus  esfuerzos  Salva  y  posteriormente  la  Academia  proponiendo  6 
preceptuando  que  le  se  aplique  a  personas  y  i^  a  cosas :  al  tiempo  toca  declarar 


Digitized  by  VjOOQ IC 


REPORTS,  361 

si  por  ahf  va  en  Castilla  la  corriente  del  uso/*  In  an  appendix  the  author 
establishes  once  for  all  the  correct  doctrine  of  the  pronoun,  reflexive  and 
personal,  in  impersonal  phrases.  *'Las  f rases  se  les  eastiga^  se  Us  admira^ 
nacidas  de  la  analogia  con  se  les  dice  6  se  les  ruega  esto  6  lo  otro^  se  les  aplUa  el 
casHgOt '  •  •  conservando  el  dativo,  aparecen  sin  sujeto.  Para  hacerlas  entrar 
en  la  sintaxis  normal,  es  preciso  descubrir  el  sujeto,  y  aquf  entra  la  diver- 
gencia,  variando  las  opiniones  segiin  la  manera  de  estimar  el  complemento. 
Los  que,  habituados  al  uso  etimol6gico,  distinguen  sin  vacilaci6n  alguna  los 
casos,  sientan  en  le  les  un  dativo,  y  naturalmente  buscan  el  sujeto  del  verbo 
pasivo :  de  ahi  las  explicaciones  de  Salva  y  de  Bello.  Los  que  estin  hechos  d 
decir  y  ofr  le  les  en  lugar  de  lo,  los,  toman  aquellos  primeros  como  acusativos,  y 
no  pueden  menos  de  buscar  el  sujeto  en  el  se^  y  de  darle  en  consecuencia  el 
cali6cativo  de  pronombre  indeterminado,  como  se  hace  con  nuestro  uno,  con  el 
on  de  los  franceses  y  el  putn  de  los  alemanes  (Acad.  Gram,,  p.  349,  ed.  de 
1880V' 

Melanges.  A.  Thomas.  Etymologies  frangaises:  Aockier,  This  word  is 
used  to  translate  the  Lat.  suffocare  in  the  parable  of  the  Sower,  and  is 
probably  an  agricultural  term  corresponding  to  a  type  ^adoccare,  composed  of 
ad  and  the  well-known  word  oecare  *to  harrow \a  field  and  'to  dress  with 
earth'  the  foot  of  a  tree.  From  this  to  •  stifle,  choke*  "  il  y  a  moins  que  rien." 
— Artiller,  artilleur,  artillerie  [=  Eng.  artillery].  The  only  question  is  as  to 
how  artem  has  produced  arfiller,  O.Fr.  artillier  is  simply  an  alteration,  by 
folk-etymology  (under  the  influence  of  the  word  art),  of  the  verb  atilKer,  the 
precise  etymology  of  which  is  not  yet  established. — Goupilhn  [holy  water 
sprinkler].  Menage  was  the  first  to  connect  goupillon  with  goupil  [vulpeculum] 
*  renard.'  The  etymology  is  rather  to  be  sought  in  a  stem  vi^  or  wipp,  not 
furnished  by  the  Latin;  cf.  Dutch  wip  'rocking  motion,*  Eng.  whip,  wisp, 
wipe, — Hausse-col,  Not,  as  Littre  would  have  it,  from  hausser  and  col^  but  an 
alteration,  by  folk-etymology,  of  hauscot,  halscot  *  neck-coat'  (cf.  haubert, 
halsbere), — Penture  [part  of  hinge].  Derived  from  pendre  (cf.  Eng.  hinge  and 
hang\ — Rature  =.*raditura,  from  radere^  through  a  participle  *raditus, — G. 
Paris.  Fr.  dSme,  It  is  here  for  the  first  time  pointed  out  and  conclusively 
proven  that  the  two  meanings  *dome'  and  'cathedral'  are  of  entirely  distinct 
origin  and  history.  In  the  former  sense  the  word  is  the  Lat.  doma,  borrowed 
from  the  Greek  dcifia,  which  had  acquired  the  general  meaning  of  '  roof* ;  in 
the  latter  it  is  the  simple  reproduction  of  the  Ital.  duomo,  German  Dom, 
answering  to  Latin  domum. 

Comptes  rendus.  L.  Hervieux.  Les  fabulistes  latins  depuis  le  si^cle 
d'Auguste  jusqu'4  la  fin  du  moyen  ftge.  2^  Edition,  entiirement  refondue 
(L.  Sudre).  8  pages.  "  Si  M.  H.  n'a  pas  toujours  la  rigueur  scientifique  qui 
est  k  desirer  en  des  ouvrages  de  ce  genre,  11  a  la  patience  d'un  benedictin  et 
an  flair  incomparable  de  chercheur.  Quelque  imparfait  que  soit  encore  le 
monument  qu'il  a  eleve  k  Phidre  et  k  ses  successeurs,  il  imposera  toujours 
I'admiration  et  la  reconnaissance." — E.  Etienne.  Essai  de  grammaire  de 
I'ancien  fran^ais  (IX«-XIV«  siicles)  (G.  Paris).  "Marque  un  tris  grand 
progr^s." — G.  Schlaeger.  Stud i en  Uber  das  Tagelied  (dissertation  de  docteur) 
(A.  Jeanroy).  '*Cette  etude  est  certainement  la  plus  complete  et  la  plus 
richement  documentee  qui  ait  etc  ecrite  sur  le  sujet ;  mais  le  resultat  est  loin 


Digitized  by  ^ 


!ogIe 


362  AMERICAN  JOURNAL    OF  PHILOLOGY. 

de  repondre  k  reffort." — Thomas  A.  Jenkins.  L^Espurgatoire  seint  Patriz  of 
Marie  de  France.  Published  with  an  introduction  and  a  study  of  the  language 
of  the  author.  Johns  Hopkins  University  dissertation  (G.  Paris).  "Avec 
r impulsion  fdconde  qui  a  ete,  dans  ces  derniers  temps,  donn^e  aux  etudes 
romanes  en  Amerique,  surtout  gr^ce  a  M.  A.  Marshall  Elliott,  11  faut  nous 
attendre  a  voir  arriver  prochainement  des  Etats-Unis  des  flottes  de* disser- 
tations' dans  le  genre  de  celles  que  nous  envoie  si  abondamment, — un  pea 
moins  abondamment  depuis  quelque  temps, — rAllemagne.  Nous  ne  pourrons 
que  nous  en  fdliciter,  si  beaucoup  ont,  comme  celle-ci.  un  reel  merite  et  sont 
le  fruit  d*un  travail  intelligent,  consciencieux  et  bien  dirige." — H.  Oskar 
Sommer.  The  Recuyell  of  the  Historyes  of  Troie  (G.  Paris).  A  reproduction 
of  the  first  work  printed  in  English,  though  not  in  England,  Caxton  having 
printed  his  translation  of  Raoul  Lef&vre's  Recueil  des  histaires  de  Troye  at 
Bruges  or  at  Cologne  about  1474,  while  it  was  not  until  1477  that  he  set  up  at 
Westminster  the  first  English  printing  press.  Dr.  Sommer  has  provided  the 
work  with  a  valuable  introduction,  glossary  and  index. — F.  Araujo.  Est'udios 
de  Fon^tica  castellana  (J.  Sarol'handy).  5  pages.  "  J'eprouve  quelque  embar- 
ras  4  conclure.  M.  Araujo  est  I'un  des  premiers,  en  Espagne,  k  s*etre  tenu  au 
courant  de  la  philologie  moderne;  il  y  a  deploy^  une  grande  activite,  et  il 
m*est  penible  de  constater  que,  dans  le  travail  dont  j'ai  passe  une  partie  en 
revue,  ses  efforts  n'ont  pas  ete  suivis  d'un  succds  complet." — L.  Sainepu. 
Basmele  romslne  [the  Rumanian  folk-tales]  in  comparatiune  cu  legendele 
antice  clasice,  pp.  xiv,  1114  (G.  Paris).  "  Le  titre  de  cet  ouvrage,  la  dimension 
du  volume  et  le  nom  de  Tauteur  en  indiquent  suffisamment  I'importance." 

Periodiques. 

Chronique.  M.  Kawczynski,  author  of  a  remarkable  work  on  the  origin  and 
history  of  rhythms  (cf.  A.  J.  P.  XI  358-71),  has  been  appointed  professor  of 
Romance  philology  at  the  University  of  Cracow. — On  the  17th  of  February, 
1894,  was  appropriately  celebrated,  in  the  Aula  of  the  University  of  Vienna, 
the  sixtieth  birthday  anniversary  of  the  distinguished  professor  of  Romance 
philology,  Adolph  Mussafia. — Twenty-one  former  students  of  Professor  Adolph 
Tobler  have  signalized  his  twenty-five  years'  incumbency  of  the  chair  of 
Romance  philology  at  Berlin,  by  presenting  him  with  a  memorial  volume  of 
510  pages,  devoted  exclusively  to  Romance  studies. 

H.  A.  Todd. 


Hermes,  XXXI  (1896). 

J.  Kromayer,  Die  MilitSlrcolonien  Octavians  und  Caesars  in  Gallia  Narbo- 
nensis.  Plancus'  legio  veterana  (Cic.  ad  Fam.  XII  2)  was  the  Legio  X  dismissed 
by  Caesar  and  enrolled  again  by  Lepidus.  Caesar  in  the  autumn  of  45  had 
settled  the  tenth  legion  in  Narbo  and  the  sixth  in  Arelate,  and  these  cities 
were  called  Julia  Patema  to  distinguish  them  from  the  colonies  founded  by 
Octavian  (Dio  IL  34,  4),  Baeterrae  in  36,  Arausio  between  35  and  33,  and 
Forum  Julii  in  30.  This  meets  the  objections  raised  by  Mommsen  (IIP  553, 
N.  I),  and  shows  that  here  as  elsewhere  Octavian  completed  the  work  begun 
by  his  father. 


Digitized  by  VjOOQ IC 


REPORTS,  363 

L.  Ziegler,  Zur  Textkritik  des  Scholiasta  Bobiensis,  gives  a  collation  and 
restoration  of  the  Milan  palimpsest  for  Pro  Flacco,  Cum  senatui  gratias  egit, 
Pro  Plancio,  Pro  Milone,  Pro  Sestio. 

J.  Kromayer,  Zur  Geschichte  des  II.  Triumvirats.  IV.  Der  Partherzug  des 
Antonins.  His  roate  was  from  Zeugma  to  Melitene,  NNE  to  Satala,  £  through 
Caranitis  to  Artaxata,  through  the  Araxes  vall^v  to  Atronatene.      This  is  safer 

than  Mesopotamia  (where  the  Parti 
able,  is  the  right  distance  (8000  st 
Pompey  and  by  Caesar.  (2)  The  p 
mended  by  Artavasdes  and  had  be 
approach  and  the  retreat.  (3)  Ante 
for  he  must  have  been  two  months 
that  he  would  leave  Zeugma  in  i 
Zeugma  was  intended  to  draw  atter 
neither  energy  nor  strategic  skill, ' 
was  not  supported  by  Artavasdes. 

J.  Toeppfer,  Die  Liste  der  Athen 
only  historic  kings,  Codrus  and  Mi 
office  was  as  old  as  Akastus,  it  was 
that  time,  whereas  the  archons  me 
(777-754)  was  the  last  hereditary 
Cbarops  begin  the  ten-year  kings, 
family.  After  Hippomenes  the  ofl 
but  we  do  not  know  whether  the 
kings.  The  archon*s  power  was 
account  (FHG.  Ill  536)  is  alone  co 
though  Chalkedon  was  the  more 
Apollo,  a  lobster  and  an  anchor  poi 
Ankore.    The  city  was  destroyed  a 

U.  Koehler,  Attische  Inschriften 
favor  of  Potamodorus  and  Euryti< 
Thuc.  IV  76).  The  second  decree 
ment  of  a  tribute  list  from  439/I 
recently  ceded  lands  to  Athens,  foi 
little  fragment  which  fits  in  betwei 
between  CIA.  I  190  and  191.  (4)  a 
15,  was  inscribed  under  a  statue  oi 
by  his  son  (cf.  CIA.  II  151a).  (8) 
1299,  is  older  than  the  Peloponne 
astronomical  use. 

W.  Soltau,  P.  Scipio  Nasica  als  i, 
Aem.  Paul.  1 5)  was  written  to  Mas 
Liv.  44.  35,  14),  who  was  the  friend 
only  from  Juba's  history,  which  c< 
by  antiquarian  notes. 

C.  G.  Brandis,  Studien  zur  R5m 
consuls  governed  Bithynia  under 


Digitized  by  VjOOQ IC 


364  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Marcus  Aurelius,  we  may  conclude  that  during  the  latter's  reign  this  province 
became  imperial  territory,  and  Lycia-Pamphylia  was  ceded  to  the  senate. 

(2)  Pontus  had  a  separate  legislative  assembly  established  by  Pompey  and 
reorganized  by  Augustas.  Its  early  existence  is  attested  by  the  presence  of  a 
special  Pontic  magistrate  throughout  the  Mithridatic  territory,  while  only  in 
the  time  of  Pompey  were  the  cities  E  and  W  of  the  Halys  so  well  united  that 
they  could  form  such  an  assembly. 

E.  Assmann,  Nautica.  (i)  A  mediaeval  parallel  in  H.  Yule's  Marco  Polo 
(Introd.,  Ixix)  confirms  the  statement  of  Livy  (30.  34)  that  conquered  ships 
were  towed  stem-foremost,  (a)  Ships  with  several  beaks  are  attested  by 
Athen.  (V  204),  by  coins  and  vases,  so  that  SeicifipoXoc  must  mean  '  ten-beaked.' 

(3)  In  Hdt.  II  96  iT7uvdrj66v  refers  to  the  use  of  small  boards  edge  to  edge. 
The  hurdle  before  the  bow  warned  the  sailor  of  danger,  the  stone  at  the  stem 
aided  him  in  quickly  checking  his  course. 

R.  Reitzenstein,  Properz-Studien,  sets  forth,  with  detailed  interpretation  of 
the  context,  the  poet's  use  of  parenthesis  and  digression  in  II  i.  47-56  (not  a 
separate  poem),  10.  7-8,  34.  47-50,  3.  2^31 1  35-8. 7. 15-18.  30.  31-32.  I  16.  ^ 
12, 15.  39-33, 14.  5-6,  8. 15  (keep  Ei),  The  parenthesis  is  not  formal,  but  a 
free  and  conversational  use  of  brief  explanatory  clauses  which  are  outside  the 
logical  sequence  of  thought,  a  usage  found  more  often  in  Propertius  than  in 
any  other  poet.  Recognition  of  this  principle  will  often  prevent  transposition 
or  the  assumption  of  a  lacuna,  yet  we  find  a  gap  after  v.  34  in  II  6.  This 
poem  is  intended  to  rival  Hor.  Od.  Ill  6.  In  II  34.  31  he  defends  memorem^ 
nam  in  33,  and,  rejecting  Maass*  ascription  of  34-40  to  Callimachus  and 
Philetas,  refers  the  whole  passage  to  Antimachus'  Thebais.  II  15.  25  goes 
back  through  some  unknown  Alexandrian  to  PI.  Sym.  192  D. 

M.  Wellmann,  Aegyptisches.  Plutarch's  version  of  the  Osiris  myth  goes 
back  to  Manetho,  who,  consistent  with  the  desire  of  the  Ptolemies  to  unite 
the  Greek  and  Egyptian  religions,  identified  the  Orphic  worship  of  Dionysus 
and  Demeter  with  the  cult  of  Osiris  and  Isis,  divinities  similar  in  many 
details  of  life-history,  attribute  and  ritual,  which  are  here  enumerated.  A 
study  of  the  symbolic  interpretation  of  Egyptian  totemism  found  in  Ael., 
Plut.,  Porph.  and  Macrob.  shows  their  dependence  on  a  common  source.  By 
a  fondness  for  marvels,  by  a  use  of  Stoic  doctrines  and  of  Homer,  and  by  a 
wide  range  of  reading,  this  is  proven  to  be  Apion ;  through  him  the  myth 
reached  Plutarch.  The  combination  of  Pythagorean  theories  of  number  with 
Egyptian  legends  is  taken  from  Eudoxus. 

J.  E.  Kirchner,  BeitrSge  zur  Attischen  Prosopographie.  An  examination  of 
epigraphic  testimony  regarding  the  personal  history  of  a  dozen  Athenians. 
We  learn  the  family-tree  of  Kleon,  of  Dikaiogenes  (Isae.  V)  and  of  Kallippus, 
pupil  of  Plato.    Ankyle  was  the  deme  of  Kallias  and  Hipponicus. 

G.  Kaibel,  Zu  den  Epigrammen  des  Kallimachos.  V  interpreted.  XIII. 
With  HeTikaldv  sc.  )3o6r,  meaning  the  ox  on  a  coin  of  Pella.  XLII  alludes  to 
Archinus*  Stoic  studies.  Propertius  and  Ovid  imitate  this  naiyvtov,  XLVIII 
interpreted  by  comparison  with  Anthol.  Pal.  VI  308. 

W.  Dittenberger,  Antiphons  Tetralogien  und  das  Attische  Criminal recht. 
The  law  cited  by  Antiphon,  III  (B)  p  9,  et  al.,  which  forbids  and  punishes 


Digitized  by  VjOOQ IC 


REPORTS.  365 

justifiable  homicide,  is  a  mere  fiction  of  the  author,  and  the  tetralogies  cannot 
be  used  as  an  independent  source  of  Athenian  law. 

L.  Ziegler,  Zur  Textkritik  des  Scholiasta  Bobiensis,  concludes  his  detailed 
study  of  this  subject. 

F.  MQnzer,  Die  Zeitrechnung  des  Annalisten  Piso.  A  study  of  Dionys.  I 
74,  Plin.  VIII 16,  and  frag.  26,  27  of  Piso  shows  that  the  annalist  adopted  the 
era  of  Cato,  which  placed  the  founding  of  the  city  751  B.  C. 

J.  Ziehen,  Epencitaten  bei  Statins.  Silv.  I  a.  213,  III  i.  71  ff.,  3. 179,  V  2. 
113  ff.  show  that  the  poet  borrows  his  similes  and  allusions  because  of  their 
literary  excellence,  although  not  especially  adapted  to  his  context. 

Miscellen.— F.  Bechtel  in  the  Stratos  inscription  (BCH.  XVII  445)  reads 
(1. 9, 10)  pMjopx^  ik"  (i.  e.  ^i*)  27r<vftipof  (?).  The  datives  t«<,  ^LaKkLirKm  instead 
of  locatives  may  be  due  to  Athenian  influence. — W.  Dittenberger  reads 
Aj^r^dupoc  in  Diodor.  XVIII  7,  5. 

M.  Pohlenz,  Ueber  Plutarchs  Schrift  i^tpi  aopyifaiai.  The  source  of  this 
treatise  is  Hieronymus  of  Rhodes,  who  as  a  Peripatetic  opposes  Plato  (ch.  9 
rd  vevpa  r^  in/xv^f  cf.  Rep.  Ill  41 1  6)  rather  than  Aristotle.  In  ch.  4  and  in 
the  introductions  to  ch.  6-10  Plutarch  presents  his  own  views  and  uses  his 
own  phraseology,  hence  some  obscurity.  The  same  source  appears  in  irepi  1% 
tfOuc^  dperiK  and  in  the  Vita  Coriolani. 

H.  Diels,  Alkmans  Partheneion.  This  consisted  of  ten  strophes,  written  in 
four  columns  of  c.  34  lines  each,  but  these  were  not  sung  by  single  maidens. 
Diodorus  (IV  33,  6),  like  Alcman,  mentions  bnt  ten  Hippokoontidae ;  v.  29-35 
refer  to  the  Giants  and  conclude  the  first  half  of  the  ode.  The  simile  in  43  ff. 
is  justified  by  the  contrast  between  the  coarse  Spartan  humor  and  the  lofty 
tone  of  the  poem.  Moreover,  aytTji  was  the  Laconian  word  for  chorus. 
Agesichora  is  the  pupil  and  kpufUvff  of  Agido,  as  Kleis  of  Sappho.  She  leads 
one  of  the  hemichonises  consisting  of  ten  maidens,  while  Agido  guides  the 
other,  which  numbers  eleven  with  the  coryphaea.  There  is  also  an  antichorus, 
called  Pleiades  (1.  59)  by  the  poet,  which  was  trained  and  led  by  Aenesimbrota. 
The  ode  was  dedicated  to  Artemis  (Orthia  in  60,  Aotis  in  87)  and  to  Helen. 
In  many  places  new  readings  are  proposed,  e.  g.  in  39,  bptjp*  uif  (=  &aTe). 

E.  Maass,  Untersuchungen  zu  Properz,  interprets  the  design  on  a  Berlin 
vase  (Furt.  2642)  as  Dionysus  Musagetes  with  Muses  who  bear  the  lyre  and 
thyrsus.  I.  This  function  of  Bacchus  as  god  of  poetry  (cf.  Hor.  Od.  I  i,  29, 
etc.)  appears  in  many  allusions  of  Propertius,  as  in  IV  3.  28,  where  the  tympana 
are  called  orgia  Musarum^  and  in  III  30.  25-39  (cf*  Orph.  Hymn  76).  In  IV 
3.  25  ff.  the  two  fountains  belong  to  Ennius  and  to  Philetas,  while  in  III  10 
the  source  of  Permessus  is  Aganippe,  one  of  the  springs  struck  out  by  the 
hoof  of  Pegasus  (a  Hippocrene,  v.  Ov.  Fast.  V  7),  and  here  again  sacred  to 
Philetas  and  to  the  Muses,  nymphs  of  the  mountain  spring.  Indeed,  inscrip- 
tions show  that  the  Muses  were  honored  at  Aganippe.  In  Verg.  Eel.  VI  64 
ad  flumina  must  follow  duxerit,  for  Permessus  and  Aonia  are  one  locality, 
while  Linos  gives  the  pipe  to  Gallus  as  an  imitator  of  Euphorion.  There 
were  two  traditions  regarding  the  springs  of  Helicon  ;  the  older,  followed  by 


Digitized  by  VjOOQ IC 


366  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Hesiod  and  Vergil,  considers  Aganippe  alone  as  the  fountain  of  the  Mases. 
the  later  distinguishes  Aganippe  as  sacred  to  the  elegy  and  Hippocrene  as  the 
source  of  the  epos.  This  version  appears  in  Philetas  and  Ennius.  II.  Brief 
interpretation  of  III  32.  33  ff.,  IV  i.  27  ff,,  5.  39  ff. 

P.  Wendland,  Philo  und  Clemens  Alexandrinus.  The  de  fort.^  de  car.,  de 
paen,  and  de  nobil,  belong  to  Philo's  treatment  of  the  Mosaic  law.  A  detailed 
comparison  of  Clement's  citations  from  Philb  with  Philo's  text  supplies 
lacunae  in  the  latter  and  shows  the  true  order  of  the  writings  as  given  above. 
The  constant  agreement  of  Clement  with  S  makes  this  MS  the  most  important 
authority  for  Philo's  text. 

E.  Thomas,  Das  Janiculum  bei  Ovid,  defends  in  Fast.  I  245  the  reading  ara 
mea  est  colli^  quern  volgtu.  Janus  may  well  have  had  an  altar  on  the  Janiculum, 
for  his  son  Fons  was  thus  honored  (Cic.  de  leg.  2. 22.  56  and  Dion.  Hal.  2.  76. 6). 

Miscellen. — E.  Lattes  shows  that  the  inscriptions  of  Novilara  are  Etruscan 
in  alphabet  and  vocabulary.  Only  6  and  U  are  new,  and  they  are,  at  least, 
not  Greek. — Th.  Gomperz  reads  in  Emped.  21  (Stein)  /zJ^e  rv\f  hf^u  nianVf  131 
ra  vvv  kaop€»fji£v  anain-a^  183  ^upa  re  -^a^  irpiv  KiKpr/ro. — B.  Keil.  The  'OAari^df 
irdXefiog  of  the  inscription  published  in  Arch.-Epigr.  Mitth.  aus  Oest.  XI  66  ff. 
shows  that'OAof  in  CIA.  I  274, 7  is  the  name  of  a  slave  taken  from  a  Thracian 
tribe  living  near  Apollonia  on  the  Pontus. — P.  Stengel  cites  other  passages  to 
support  his  definition  of  niXavoc  (Herm.  XXIX  281  ff.),  and  also  shows  that  in 
divination  from  the  entrails  of  animals  the  seers  observed  the  bursting  of  the 
gall  and  the  intensity  of  the  flame,  not  the  movements  of  the  victims. 

B.  Niese,  Zur  Geschichte  des  Pyrrhischen  Krieges.  Trogus,  preserved  in 
Justin,  is  the  only  reliable  authority,  since  other  writers  distort  the  facts  in 
their  desire  to  praise  Rome.  This  falsification  was  begun  not  by  contemporary 
Romans,  but  by  later  Greek  historians,  and  was  continued  by  the  Augustan 
writers.  It  appears  from  Justin,  with  whom  Diodorus  and  Cicero  agree,  that 
the  only  visit  made  by  Cineas  was  after  the  battle  of  Asculum,  when  the 
Romans  had  more  power  in  Italy.  The  speech  of  Ap.  Claudius  was  an  early 
invention,  for  it  was  really  the  embassy  of  Mago  that  brought  the  Romans  to 
a  decision.  The  battle  of  Beneventum  was  no  overwhelming  defeat,  nor  did 
Pyrrhus  entirely  give  up  the  war.  Tarentum  was  not  taken  by  storm,  for  it 
remained  independent  till  Hannibal's  time,  nor  was  it  betrayed  by  Milon,  for 
he  had  returned  to  Pyrrhus. 

B.  Keil,  Zur  Delphischen  Labyadeninschrift,  reads  ev  veura  in  12,  in  D  31 
ff.  interprets  ^pav  as  *day*  and  GVfiirpriiaKtif  as  *burn'  (sacrifice),  in  D  13  takes 
"kcKxol  as  dative  and  irap^  from  napitjfu,  in  D  26  writes  dp;f6w  and  supplies 
avTuv  with  ttolSvtuv^  in  C  33  explains  arpw^^  as  *  turning-places,'  and  in  C  38 
writes  h^vre  Ka  ha  BifKa  knl  yav  anoTdsQfji^  where  rd  is  due  to  progressive 
dissimilation. 

K.  J.  Neumann,  Polybiana.  The  historian  was  led  to  place  the  first  treaty 
between  Rome  and  Carthage  in  the  first  year  of  the  republic  by  the  general 
feeling  that  Punic  treachery  went  back  to  the  earliest  times,  a  feeling  reflected 
in  Naevius  and  Ennius  and  promoted  by  Cato.     He  does  not  mention  the 


Digitized  by  VjOOQ IC 


REPORTS,  367 

treaty  of  306,  simply  because  it  was  lost  and  forgotten.  Polybius  wrote  books 
1-15  before  151  B.  C,  and,  on  resuming  work  soon  after  144,  he  revised  i  and 
2,  inserted  the  discussion  oi  Punica  fides  and  the  digression  in  3.  21,  9-32,  lo, 
and  then  published  1-15.  So  this  portion  of  his  work  could  not  be  altered 
(v.  16.  20).  Books  4-7  of  Cato*s  Origines  were  written  between  154  and  149, 
whereas  1-3  were  published  in  168. 

H.  F.  Kastner,  Pseudodioscoridis  De  herbis  feminin is,  publishes  this  treatise, 
prefixing  a  description  of  the  MSS  (two  Laur.,  one  Par.)  and  a  discussion  of 
the  sources,  Dioscorides  and  Ps.-Apuleius.  The  meaning  of  the  title  is  a 
mystery. 

C.  Robert,  Die  Scenerie  des  Aias.  der  Eirene,  und  des  Prometheus.  In  the 
Ajax  the  protagonist  took  the  parts  of  Ajax  and  Teucer.  The  change  of 
scenery  was  effected  by  projecting  up  from  'Charon's  stairs'  a  platform 
covered  with  earth  and  stones,  and  up  these  steps  came  Tecmessa  (891),  like 
the  ghosts  in  Bum.,  Hec.  The  covering  of  Ajax  with  a  mantle  (915)  allows 
the  actor  to  slip  down  the  stairs,  while  a  lay  figure  is  put  in  his  place.  Here 
again  in  the  Pax  is  the  cave  where  Eirene  is  buried,  and  into  this  descends 
Trygaeus  (726).  This  underground  passage  was  originally  larger,  and  was  the 
oldest  scenic  contrivance  of  the  Greeks.  The  wooden  skene  (the  hill  in 
Aesch.  Suppl.,  the  temenos  in  Sept.)  covered  the  rear  of  the  orchestra,  yet  left 
abundant  space  for  the  chorus  to  dance.  The  skene  did  not  develop  from  the 
dressing-room.  The  wooden  houses  of  Zeus  and  Trygaeus  were  on  different 
levels,  and  were  placed  one  in  front,  the  other  at  the  side  (cf.  Ach.  Nub.  Ran. 
Thes.  657).  A  large  kicKijKXiffM  in  Thes.  277  bears  the  chorus,  but  not  the 
actors  nor  the  altar.  Over  the  stairway  was  built  the  hill  on  which  Prome- 
theus lay,  and  into  this  it  sank  at  the  end.  The  actor  entered  the  image 
through  the  passage-way,  and  through  this  the  chorus  came  to  ascend  the 
winged  chariot  on  which  they  floated  in  the  air.  They  did  not  dance  at  all, 
hence  no  dance-measure  in  the  poem.  That  the  flying-machine  is  as  old  as 
Aeschylus  appears  from  Eum.  403-5  and  the  irvxoaTcuria  (Poll.  IV  130),  and 
the  6eoh)yeiop  is  equally  ancient  (Plut.  de  aud.  poet.  17  A).  The  Prometheus 
is  played  mainly  in  the  air,  and  no  other  tragedy  makes  such  large  use  of 
machinery. 

Miscellen. — P.  Stengel  defines  Oveiv  as  'sacrifice  for  the  sake  of  the  gods' 
(to  honor  them)  and  dveaffcu  *for  man's  sake'  (to  get  help).  The  middle  is 
common  in  the  historians,  but  very  rare  elsewhere.  Upcxrwa  are  the  perqui- 
sites of  the  priests  in  general,  yipo  only  their  share  of  the  flesh.  Oeo/iopia  is 
the  part  consecrated  to  the  gods,  though  sometimes  taken  by  the  priests. — 
W.  Dittenberger  proves  by  the  use  of  arpifceia  and  of  ireidapxeiv  c.  gen.  that 
the  letter  of  Darius  I  in  BCH.  XIII  529  was  composed  in  the  Ionic  dialect 
and  later  changed  to  Attic. — F.  Skutsch  shows  that  Firmicus  was  a  Syracusan. 
— -C.  Turk  cites  the  Delphic  inscription  in  BCH.  XIX  i  to  support  the 
meaning  of  eviavrd^  as  •  anniversary.' 

Barker  Nbwhall. 


Digitized  by  VjOOQ IC 


BRIEF  MENTION, 

Professor  Michel  6r£al's  Essai  de  Shnantique^  which  bears  the  subtitle 
Science  de  significations  (Hachette),  seems  destined  to  make  both  the  name 
and  the  subject  popular.  *  Semantic  *  or  *  Semantics '  is  better  than  *  Semas- 
iology' on  the  one  hand  or  *  Meaning-lore '  on  the  other,  and  the  wider 
public  will  welcome  a  book  which  unfolds  some  of  the  most  interesting 
lessons  that  the  study  of  language  has  to  teach.  There  is  something,  says 
M.  Br£al,  in  linguistics  besides  phonetics,  *by  which  the  study  is  reduced 
to  a  secondary  branch  of  acoustics  and  physiology,'  something  besides  *an 
enumeration  of  the  losses  undergone  by  the  grammatical  mechanism,'  an 
enumeration  which  produces  'the  illusion  of  a  crumbling  structure.'  Spec- 
ulations on  the  origin  of  language  only  add,  without  great  profit,  chapter 
after  chapter  to  the  history  of  systems.  The  object  of  *  semantic'  is  to 
extract  from  linguistic  study  food  for  reflexion  and  rules  for  the  vernacular. 
Each  one  of  us  is  a  collaborator  in  the  evolution  of  human  speech,  and 
*  semantic'  appeals  to  practice  as  well  as  to  theory.  Such  is  the  programme 
of  the  Essai  de  S^mantique,  which  cannot  fail  to  attract  attention  in  profes- 
sional circles  as  well  as  in  the  larger  world  for  which  it  seems  to  be 
primarily  destined. 

In  the  introductory  chapter,  which  gives  the  scope  of  the  work.  Professor 
Br£al  clears  the  way  by  protesting  against  the  abuse  of  metaphors,  which 
leads  people  not  only  to  say  but  to  think  that  language  goes  its  own  road 
and  that  words — form  and  sense — lead  an  existence  of  their  own,  inde- 
pendent of  man.  Words  are  spoken  of  as  being  born,  as. coming  into 
conflict  with  one  another,  as  propagating  their  species,  and  as  dying  out. 
There  is  no  great  harm  in  these  phrases,  if  there  were  not  those  who  take 
them  literally,  and  hence  this  prefatory  protest  against  a  terminology  which 
is  apt  to  efface  the  real  causes  of  things.  It  is  not  the  words  that  live,  but 
the  men  that  make  the  words.  Languages  are  not  mothers,  \ior  do  they 
have  daughters.  Verbs  do  not  borrow  certain  tenses  from  other  verbs. 
They  are  neither  borrowers  nor  lenders.  Nouns  do  not  take  on  such  and 
such  a  termination  in  the  plural.  Let  us  get  rid  of  these  fanciful  expres- 
sions. Let  us  *  study  the  intellectual  causes  that  have  presided  over  the 
transformation  of  our  languages.'  And  yet  a  critic  might  say  that  causes 
do  not  preside  and  that  there  is  danger  of  conjuring  up  some  such  image 
as  Akj7  ^iveSpog  Z^vdc  apxcUoic  vdfioiQ,  In  point  of  fact,  personification  is  too 
much  for  us.  The  inheritance  from  our  imaginative  forefathers  determines 
our  thought.  We  woo  an  abstraction,  and  the  pure  creature  is  no  sooner 
won  than  it  reveals  itself  as  a  shrewish  goddess.  But  such  a  book  as  this 
incites  on  every  page  to  reflexion  and  comment,  and  further  notice  is 
reserved.     The  exigencies  of  the  Journal  compel  the  postponement  of  an 


Digitized  by  VjOOQ IC 


SXi 


BRIEF  MENTION.  369 

elaborate  review,  and  this  preliminary  mention  is  intended  only  as  an 
announcement  of  a  work  in  which  the  eminent  author  has  gathered  up  in 
an  attractive  form  the  results  of  long  and  deep  study. 


When  in  the  account  which  he  gives  of  the  first  Persian  invasion  (Legg. 
3»  698  C)  Plato  says  of  Dareios  :  Qavarov  cnjrif  (sc.  T<f)  Adridi)  npoemcjv  fi^ 
npd^avri  ravra  there  can  be  little  doubt  that  ft^  npd^avri  is  a  conscious 
abridgment  of  a  conditional  clause,  whether  edv  fi^  irpd^y  or  el  fifj  izpd^eie. 
If  Plato  had  not  been  in  such  a  hurry  he  might  have  employed  the  finite 
verb  and  have  used  el  uti  np&^ei  (Kpd^ot)^  on  account  of  the  threat.  In  the 
more  leisurely  story  of  the  Menexenos  (240  A)  we  read :  elirev  tjkciv  dyovra 
*EpeTpUac  Koi  *ABijvaiovg  ei  pobh)iTO  rrjv  kavrcv  Ket^a^^  i^eiv.  But  in  the  Laws 
Plato  is  rushing  on  with  his  participles,  and  in  this  very  passage  indulges 
in  a  curious  participial  anacoluthon.  Now,  this  use  of  the  participle  with 
fi^  as  the  conscious  shorthand  of  a  conditional  is  post-Homeric  (A.  J.  P. 
XVrti  244),  and  Dr.  Callaway's  laborious  dissertation  On  the  use  of  fiij 
with  the  participle  in  classical  Greek  exhibits  in  detail  the  evolution  of  the 
usage.  Of  course,  it  was  well  known  that  Homer  does  not  combine  fifj 
with  the  participle  freely,  and  Monro  (H.  G.',  S360)  cites  but  one  passage,  Od. 
4,  684,  where  it  is  clearly  part  of  a  wish.  However,  he  has  overlooked  II. 
i3»  48,  where  Faesi-Franke  explains  fiij  with  partic.  as  conditional,  and  the 
same  explanation  is  given  by  Paech,  Ueber  den  Gebraueh  des  Ind,  Futuri 
alt  modus  iussivus  bei  Horner^  p.  14,  to  which  Ameis-Hentze  refers.  The 
passage  runs  : 

AZavrc,  er^w  ^v  re  aa^aere  Aadv  *Axai»v 

dh(^C  fivtfaafiivu  fiTfSi  Kpvepoio  i^poio. 

True,  Dr.  Gallaway  is  not  satisfied  with  the  explanation  of  the  fin^e  as 
conditional,  but  he  does  not  mend  matters  by  connecting  it  with  Kpvepoio, 
for  fiv^k  assuredly  belongs  to  fivrfaafitvu,  and  as  a  conditional  clause  firi6h 
fivr/aafihu  would  make  sense  in  Attic.  Of  course,  Eustathios  has  no  diffi- 
culty with  the  finSi ;  it  is  the  natural  negative  of  the  participle  with  him 
(A.  J.  P.  1 55), and  he  calls  the  passage  dy6pe(i)onf  irratvoc,znd  makes  it  declar- 
ative, and  not  hortatory.  So  does  the  paraphrast,  whose  version  is  espe- 
cially interesting,  o  Alavreg,  he  writes,  vfjieic  dv  Siaa^oijre  rdv  Xabv  tuv 
*^'kfyvuv  Tfj^  laxhoc  dnofiVTfadevTec  xai  ov  ryg  ^pt^rm  ^vy^Ct  the  negative  ov 
showing  the  declarative  conception  and  dv  diaaCxrfrre  that  the  writer  had 
before  him  the  reading  of  Apollonios,  ice  diaaoaeref  and  as  dv  with  fut.  ind. 
would  have  no  terror  for  a  Greek! ing  (see  my  note  on  £p.  ad  Diogn.  4, 
17),  we  may  read  here  for  6uia6a7fTef  6iaff6aeTe,  and  not  dtae^oaire.  In  his 
commentary  (1888)  Monro  gives  the  future  in  II.  13,  47  'a  hortatory  force,' 
but  in  his  H.  G.*  (1891)  he  does  not  accept  the  'gentle  imperative'  (§326) 
theory.  Moreover,  the  hortatory  force  of  the  future  does  not  explain  the 
negative  fjiv^i,  as  Paech  has  pointed  out  (1.  c.) ;  and  compare  further 
A.  J.  P.  XV  117,  where  I  have  shown  that  an  imperative  future  with  fi^/ 
has  very  doubtful  warrant     The  trouble  is  that  aa^aere  has  not  been  recog- 


Digitized  by  VjOOQ IC 


370  AMERICAN  JOURNAL    OF  PHILOLOGY, 

nized  by  the  authorities  as  a  thematic  form  of  the  aor.  imperatiye.  Gehring 
still  classes  it  as  a  futare,  and  it  does  not  keep  ocrrere  company  in  the 
lists  usually  given.  If  Dr.  Gallaway  had  consulted  Leaf,  he  would  haye 
found  what  seems  to  be  the  true  solution  of  the  problem :  ^aaLatrt  is  the 
imperative  of  the  sigmatic  aorist,  as  is  shewn  by  the  following  /ii^,'  for 
fivTfGafiivii  clearly  represents  the  imperative.  Compare  the  phrase  II.  6, 
112:  fiv^aaOe  6k  dohptdoq  ahajg.  The  point  is  a  small  point  even  for  Brief 
Mention,  which  is  a  manner  of  pin-cushion  for  small  points,  but  it  is  not 
altogether  uninteresting  as  a  specimen  of  the  progress  of  doctrine. 


Mr.  £.  F.  Benecke  *  met  with  his  death  in  Switzerland,  on  July  i6th, 
1895,  in  his  twenty-sixth  year,'  and  his  friends  have  published  for  the  use 
of  scholars  a  fragment  which  he  left  behind,  Antimachus  of  Colophon  and 
the  Position  of  Women  in  Greeh  Poetry  (Swan  Sonnenschein],  in  the  hope 
that  the  material  may.be  of  service  to  those  engaged  in  similar  studies, 
as  it  undoubtedly  will.  And,  furthermore,  there  will  be  no  lack  of  fhose 
engaged  in  similar  studies,  for  the  subject  has  a  fatal  fascination,  especially 
for  young  men.  So  there  is  an  essay  in  the  Harvard  Studies,  vol.  I,  by 
another  young  scholar,  entitled  The  Position  of  Women  in  Aristophanes^ 
which  covers  part  of  the  ground  traversed  in  Mr.  Bsnecke's  posthumous 
book.  Under  the  sad  circumstances,  criticism  would  be  out  of  place,  and 
yet  one  who  has  studied  the  subject  enough  to  despair  of  attaining  may  be 
allowed  to  remark  resignedly  that  Greek  and  women,  apart  or  together, 
are  lessons  never  learned  to  an  end.  Dies  diei  eructat  verbum  may  answer 
for  the  one,  nox  nocti  indicat  scientiam  may  answer  for  the  other.  No  better 
average  woman,  Greek  or  other,  than  the  seawoman  of  the  old  iambist  of 
Amorgos.  Find  out  the  attitude  of  the  Greek  to  the  sea,  and  you  have 
the  answer  to  the  other  problem. 


Apart  from  the  contents  of  his  writings,  Josephus  may  be  made  to  render 
good  service  as  a  specimen  of  the  kind  of  Greek  that  ought  not  to  be 
written,  and  as  we  want  the  best  texts  attainable  for  all  samples,  whether 
good  or  bad,  we  welcome  the  completion  of  Niese's  monumental  edition 
of  Josephus  in  the  closing  volume,  Flavii  losephi  An{t)iguitatum  ludaiearum 
Epitoma  (Weidmann). 


Professor  Gudeman*s  handy  little  manual.  Outlines  of  the  History  of 
Classical  Philology  (Boston,  Ginn  &  Co.),  has  vindicated  its  usefulness  by 
reaching  a  third  edition,  and  it  is  to  be  regretted  that  the  *  thorough 
revision'  claimed  for  it  has  not  extended  to  a  number  of  eccentric  Greek 
accents. 


Digitized  by  VjOOQ IC 


NECROLOGY. 

George  Martin  Lane. 

On  the  last  day  of  June,  George  Martin  Lane  ended  his  useful  and  noble 
life  in  Cambridge.  He  was  born  in  Charlestown  on  Christmas  "Eve,  1823, 
but  his  parents  removed  to  Cambridge  during  his  infancy,  and  he  never 
knew  any  other  home.  His  early  education  was  in  the  common  schools  of 
the  town,  but  in  his  seventeenth  year  a  new  bent  was  suddenly  given  to  his 
life.  A  circular  left  at  his  father's  house  described  the  Hopkins  bequest 
for  a  classical  school  to  fit  Cambridge  boys  for  college,  and  on  reading  it, 
young  Lane  said  at  once,  **I  must  go."  Two  years  later  he  entered 
Harvard,  and  upon  his  graduation.  Dr.  Beck,  who  then  went  to  Europe  for 
a  year,  selected  him  as  his  substitute,  saying  that  he  had  never  had  a  pupil 
who  could  write  Latin  so  well.  As  tutor  he  did  excellent  work,  and  then 
went  to  GQttingen,  where  he  received  his  degree  of  Ph.  D.,  in  1851,  pre- 
senting a  thesis  which  is  still  an  authority  upon  the  history  and  antiquities 
of  Smyrna.  He  was  at  once  appointed  Professor  of  Latin  in  Harvard 
College,  and  continued  the  active  work  of  his  chair  until  1894,  when  he 
was  made  professor  emeritus  with  the  degree  of  LL.  D.;  but  until  his 
strength  failed,  in  1896,  he  still  gave  instruction  to  advanced  classes  in 
the  graduate  department. 

Dr.  Lane  was  one  of  the  greatest  teachers  of  his  time.  All  who  knew 
him  felt  in  his  society  the  presence  of  an  intelligence  at  once  broad  and 
acute,  profound  and  lively,  richly  famished  with  acquired  knowledge 
which  had  been  assimilated  and  organized  by  original  thought.  But  his 
pupils  found  in  him  more  than  this :  a  mind  with  the  peculiar  power  of 
stimulating  other  minds  to  do  their  best,  a  perpetual  source  of  impulse  and 
zeal  in  the  search  for  truth.  The  least  dogmatic  of  men,  the  most  modest 
in  asserting  even  cherished  convictions,  he  was  always  found  to  have  deep 
and  strong  foundations  for  his  slightest  suggestions  of  belief  or  doubt  A 
controversy  in  which  he  engaged  always  took  the  form  of  unprejudiced 
inquiry.  No  tradition,  no  doctrine,  no  belief,  in  scholarship  or  in  life,  had 
any  value  for  him  save  as  it  could  endure  the  most  rigid  examination  of 
proofs.  His  catholic  welcome  of  every  new  idea,  even  if  it  challenged  his 
life-long  habits  of  thought,  might  suggest  to  strangers  a  feeble  grasp  of 
principles ;  but  not  to  his  friends,  who  knew  that  none  was  more  tenacious 
of  his  reasoned  beliefs,  though  none  was  more  free  from  prejudice,  or  more 
abhorrent  of  the  influences  of  feeling  upon  mental  processes.  Hardly 
surpassed  in  minute  accuracy  of  learning  within  his  special  province,  he 
was  yet  more  remarkable  for  the  instinctive  and  indomitable  habit  of 
linking  the  whole  with  every  detail ;  of  finding  analogies  between  the  dust 
and  the  stars  of  thought;  of  illuminating  and  ennobling  what  seemed 
trifling,  by  side-lights  from  high  places.  This  peculiar  perception  of 
likeness  and  of  contrast,  ever  playing  upon  the  stores  of  a  wonderful 


Digitized  by^ 


372  AMERICAN  JOURNAL    OF  PHILOLOGY. 

memory  drawn  from  a  vast  range  of  reading  and  observation,  enlivened 
his  class-room  and  his  conversation,  and  made  his  wit  famous  far  beyond 
the  college  circle.  Indeed,  if  any  passion  ever  threatened  his  equanimity 
and  disturbed  his  judgment,  it  was  the  love  of  fun,  and  the  only  cruelty 
with  which  he  could  ever  reproach  himself  was  when  his  perfect  courtesy, 
so  kindly  sensitive  for  all  around  him,  was  for  a  moment  qualified  by  his 
vivid  sense  of  the  ludicrous  in  some  expression  of  dullness  or  ignorance. 
Yet  even  then  his  sympathy  was  keen,  and  he  suffered  more  than  the 
victim  of  his  epigram.  Throughout  a  life  which,  though  not  exempt  from 
burdens  and  deep  sorrow,  was  yet  one  of  the  happiest,  his  most  unfailing 
pleasure  was  to  indulge  and  foster  the  sense  of  the  ludicrous,  and  had  he 
not  been  the  first  Latinist,  he  might  have  been  the  first  humorist  of 
America.  No  other  shade  could  give  him  so  fraternal  a  welcome  to  the 
spirit  land  as  Desiderius  Erasmus. 

The  works  he  has  left,  the  fruits  of  his  life,  are  to  be  found  almost  wholly 
in  the  minds  he  helped  to  form.  He  made  important  contributions  to 
Latin  lexicography,  but  only  as  gifts  to  a  friend,  and  to  the  cause  of  sound 
scholarship,  and  with  no  personal  stamp  upon  them.  The  one  book  to 
which  he  gave  many  years  of  labor,  the  Latin  grammar  which  he  left 
unpublished,  will  mark  an  epoch  in  the  study  of  the  laws  of  the  language, 
by  its  clearness,  completeness  and  accuracy,  while  excelling  its  prede- 
cessors above  all  in  felicity  of  expression.  But  Professor  Lane's  published 
and  posthumous  writings  together  were  but  a  meagre  product  for  such  a 
mind.  On  the  other  hand,  no  one  ever  studied  under  him  but  found  in 
after-life  the  pathway  of  truth  smoothed  and  the  best  use  of  his  own 
faculties  made  easier  by  that  companionship  and  guidance.  A  large 
proportion  of  the  graduates  of  Harvard  for  the  last  half- century,  gratefully 
recognize  his  contributions  to  whatever  of  culture  and  of  power  they  have 
acquired.  And  in  them,  and  most  of  all  in  those  among  them  who  have 
carried  on  his  methods  and  spirit  as  a  teacher  for  the  benefit  of  yet  another 
generation,  is  his  true  monument. 

It  has  seemed  needful  here  to  speak  only  of  his  professional  career  and 
of  his  work  for  mankind  at  large.  But  any  tribute  to  these,  were  it  even 
far  less  inadequate,  would  seem  to  the  large  but  rapidly  diminishing 
number  of  those  in  whose  life  he  was  for  many  years  a  valued  part,  to  be 
less  than  the  shadow  of  the  man.  It  is  as  the  brightest  of  companions, 
the  most  generous  of  hosts,  the  wittiest  and  cheeriest  of  talkers,  the  most 
sympathizing  of  counsellors,  the  most  affectionate  of  friends,  that  they 
remember  him  and  will  ever  cherish  his  memory.  His  teachings  and 
writings  have  well  merited  his  fame,  but  we  have  known  something  far 
greater  than  they  are — himself. 

Charlton  T.  Lewis. 


Frederic  De  Forest  Allen. 

Frederic  De  Forest  Allen,  whose  sudden  death  on  August  4,  1897,  was 
the  heaviest  affliction  which  classical  philology  in  our  country  has  suffered 
for  nearly  a  quarter  of  a  century, — since  the  death  of  James  Hadley  in 


Digitized  by  VjOOQ IC 


wwmmmm9^9^T^ 


NECROLOGY.  373 

1872,  at  about  the  same  age, — was  born  in  Oberlin,  Ohio,  on  May  25, 1844, 
of  true  New  England  stock.  His  father,  George  N.  Allen,  set  out  for  Ohio 
as  a  young  man  in  1832,  under  the  influence  of  his  pastor,  Lyman  Beecher, 
who  went  to  live  in  Cincinnati  in  that  year.  Being  taken  ill  on  the 
journey,  and  kindly  cared  tor  at  Hudson,  Ohio,  where  the  Western  Reserve 
College  had  been  recently  established,  he  remained  there  and  spent  five 
years  in  study  in  the  preparatory  school  and  college,  but  at  the  close  of 
his  junior  year  went  to  the  new  college  at  Oberlin,  where  be  graduated  in 
1838.  Three  years  later  he  married  Miss  Rudd  of  Stratford,  Conn.,  who 
had  just  received  the  degree  of  bachelor  of  arts  at  Oberlin, — being  with 
two  classmates  doubtless  the  first  women  of  the  world  to  receive  this 
degree  in  course, — and  was  appointed  instructor  in  the  college,  where  he 
taught  music  and  (for  many  yearsj  natural  science,  until  187 1.  Mrs.  Allen 
had  great  strength  of  mind  and  of  character,  with  marked  scholarly  tastes, 
and  maintained  a  deep  and  close  interest  in  her  son's  philological  studies. 
Frederic  Allen  graduated  from  college  in  1863,  a  few  weeks  after  he  was 
nineteen  years  of  age.  In  his  undergraduate  days  his  philological  tastes 
were  not  awakened ;  he  was  still  immature,  and  had  not  come  to  his  own ; 
but  he  read  widely  in  the  best  French  literature  and  gave  indications  of 
unusual  tastes  and  powers.  Soon  after  graduation  he  seems  to  have 
turned  with  eagerness  to  classical  studies,  and  was  appointed  professor  of 
Greek  atid  Latin  in  the  University  of  East  Tennessee  before  he  was 
twenty-two  years  old.  He  soon  felt  the  need  of  better  training  for  his 
work,  and  in  1868  took  a  leave  of  absence  in  order  to  study  in  Leipzig  with 
Georg  Curtius,  who  was  then  the  leader  in  the  application  of  the  results  of 
the  study  of  comparative  philology  to  the  older  science  of  classical  phil- 
ology. In  Leipzig  Allen  remained  two  years,  taking  an  active  part  in  the 
work  of  Curtius's  Grammatische  Gesellschaft,  and  winning  the  hearty 
respect  of  the  scholars  of  the  university.  His  dissertation  for  the  degree 
of  doctor  of  philosophy,  de  DiaUcto  Locrensium^  was  highly  commended  by' 
Curtius,  in  public  and  in  private.  Returning  to  Knozville,  he  resumed  his 
former  place,  but  in  1873  he  was  called  to  Harvard  as  tutor  in  Greek. 
There  he  had  his  first  opportunity  of  giying  instruction  to  advanced 
students,  since  Professor  Goodwin,  who  spent  that  year  abroad,  left  part 
of  his  work  in  his  care.  The  philological  companionship  which  he  found 
there,  and  the  treasures  of  the  Harvard  library,  were  a  peculiar  delight  to 
him  after  his  comparative  isolation  in  Tennessee;  but  his  sojourn  in 
Cambridge  at  this  time  was  only  for  a  single  year,  for  he  was  called  to  the 
chair  of  ancient  languages  in  the  newly-founded  University  of  Cincinnati, 
and  felt  obliged  for  pecuniary  reasons  to  decline  overtures  to  remain  at 
Harvard.  His  new  work  in  Cincinnati  interested  him  greatly.  He  was 
pleasantly  associated  with  other  well-trained  young  scholars,  and  all 
worked  together  in  harmony.  There  he  made  his  excellent  edition  of 
Enripides's  Medea  for  the  use  of  college  classes,  and  prepared  his 
Remnants  of  Early  Latin,  which  not  only  has  been  used  in  this  country 
and  England,  but  also  has  been  the  basis  of  university  lectures  in  Germany, 
and  wrote  his  important  tract  on  Homeric  Verse,  which  was  published  in 
Kuhn's  Zeitschrift.    In  1879  ^^  received  with  pleasure  a  call  to  the  chair 


Digitized  by  VjOOQ IC 


374  AMERICAN  JOURNAL    OF  PHILOLOGY. 

at  Yale  College  which  had  stood  vacant  since  the  death  of  Professor 
Hadley  in  1872,  bat  he  was  hardly  yet  habilitated  at  Yale  when  he  was 
invited  in  the  spring  of  1880  to  do  more  congenial  work  at  Harvard. 
According  to  the  custom  of  the  time  and  the  constitution  of  the  depart- 
ment, he  had  at  Yale  fifteen  hours  of  instruction  each  week  with  sopho- 
mores and  very  limited  opportunities  for  giving  instruction  to  advanced 
students.  At  Harvard  he  was  to  have  fewer  hours  in  the  classroom,  and 
most  of  these  were  to  be  with  graduate  students.  Yale  has  the  honor  of 
being  the  first  institution  in  this  country  to  offer  systematic  instruction  to 
graduate  students  in  philology,  and  in  the  early  seventies,  with  Whitney^ 
Hadley,  Thacher,  and  Packard,  stimulating  lectures  were  given  and  good 
work  was  done,  as  a  partial  list  of  the  students  would  testify, — I.  T.  Beck- 
with,  M.  W.  Easton,  C.  R.  Lanman,  Jules  Luquiens,  I.  J.  Manatt,  B.  Perrin» 
W.  R.  Harper,  H.  P.  Wright,  J.  P.  Peters,  R.  B.  Richardson.  But  the 
death  of  Hadley  and  the  ill-health  of  Thaeher  and  Packard  had  broken  up 
the  well-laid  plans  and  the  possibilities  of  such  work  at  Yale  for  the  present* 
while  President  Eliot  was  then  giving  special  prominence  to  graduate 
work  in  philology  at  Harvard.  Allen  was  attracted  also  by  the  call  to  be 
professor  of  Classical  Philology ;  he  had  recently  published  his  Remnant» 
of  Early  Latin,  and  always  enjoyed  maintaining  his  Latin  and  his  Greek 
studies  side  by  side.  Some  of  his  more  important  courses  of  lectures  at 
Harvard  were  :  Religion  and  Worship  of  the  Greeks ;  Roman  Religion 
and  Worship;  Greek  Grammar  with  study  of  dialectic  inscriptions;  Latin 
Grammar :  sounds  and  inflexions ;  Elements  of  Oscan  and  Umbrian ; 
History  of  Greek  Literature;  Roman  Comedy;  Homer's  Iliad,  with  an 
elaborate  introduction.  For  some  of  these  courses  the  MS  is  so  carefully 
prepared  and  complete  that  we  may  hope  that  no  wrong  would  be  done  to 
the  author's  memory  by  publication.  In  addition  to  his  published  works 
to  which  reference  has  already  been  made,  the  following  deserve  mention  : 
a  revision  of  Hadley's  Greek  Grammar,  1884 ;  a  translation  and  edition  of 
Wecklein's  Prometheus  of  Aeschylus,  1891 ;  Greek  Versification  in  Inscrip- 
tions (100  pp.)  in  the  fourth  volume  of  the  Papers  of  the  School  at  Athens* 
1888.  He  contributed  Etymological  and  Grammatical  Notes  to  the  first 
volume  of  this  Journal,  and  in  later  volumes  published  some  Greek  and 
Latin  inscriptions  and  an  article  on  *  Prometheus  and  the  Caucasus* 
(Journal,  vol.  XIII).  He  read  at  least  five  papers  before  the  American 
Philological  Association.  His  papers  in  the  Harvard  Studies  on  *  Gains 
or  Ga![us?,'  ^Manus  Consertio,'  and  *Os  columnatum'  are  characteristic 
He  wrote  a  few  excellent  articles  for  encyclopaedias,  and  reviews  and 
notes  for  the  Nation  and  the  Classical  Review.  He  was  never  in  haste  ta 
publish  what  he  had  written,  or  to  put  into  writing  his  observations  and 
discoveries,  and  doubtless  the  later  years  of  his  life,  if  he  had  lived  to 
"three  score  years  and  ten,"  would  have  seemed  more  productive.  Per- 
haps the  most  important  work  which  he  left  incomplete  is  an  edition  of  the 
Scholia  to  Plato.  He  gave  the  better  part  of  his  last  *  Sabbatical  year,' 
1891-92,  to  the  careful  collation  of  the  Clarkianus  and  Parisinus  A  MSS, 
and  found  to  his  surprise  that  this  work  had  never  been  done  properly 
before.    The  inspection  and  possibly  the  collation  of  the  Venetian  MSS 


Digitized  by  VjOOQ IC 


^m 


NECROLOGY,  375 

remained,  to  be  undertaken  before  the  publication  of  the  results  of  his 
labor,  and  for  this  he  was  already  planning  as  the  chief  occupation  of  his 
next  Sabbatical  year,  1898-99.  Another  important  work  in  which  he  was 
greatly  interested,  and  for  which  he  had  made  extensive  researches,  was 
on  the  history  of  religions. 

The  academic  year  1885-86  Professor  Allen  spent  in  Greece  as  the 
Director  of  the  American  School  of  Classical  Studies  at  Athens,  but  the 
year  proved  one  of  disappointment  and  deepest  sorrow :  his  oldest  and 
only  then-living  child  died  in  Athens,  and  his  health  was  so  wretched  that 
he  was  obliged  to  abandon  his  plans  for  archaeological  explorations. 

Professor  Allen  was  never  robust.  He  suffered  sadly  from  sciatica  and 
asthma  in  Ohio,  from  asthma  in  Greece,  from  hay-fever  for  at  least  the 
latter  half  of  his  life,  and  particularly  from  disabling  sick  headaches, — but 
he  performed  much  intellectual  labor.  When  in  this  country  in  the  summer 
he  sought  refuge  in  the  White  Mountains  from  the  distress  of  hay-fever. 
For  many  years  he  spent  weeks  each  summer  in  climbing  and  tramping  in 
northern  New  Hampshire,  visiting  repeatedly  the  summits  of  the  Presi- 
dential range,  and  being  specially  fond  of  the  summit  of  Mt.  Moosilauke, 
where  he  often  sojourned  for  a  week  or  two  among  the  clouds,  with  one  or 
two  friends  and  some  philological  books. 

The  chief  recreation  of  Professor  Allen,  from  his  boyhood,  was  found  in 
music.  His  knowledge  of  the  art  of  music  was  thoroughly  scientific,  and 
probably  no  other  American  scholar  understood  ancient  Greek  music  so 
well  as  he.  He  found  relief  from  his  severer  studies  not  simply  in  hearing 
but  also  in  composing  music.  In  addition  to  setting  occasional  verses  to 
music,  he  composed  the  music  for  a  pantomime  and  an  operetta  of  his 
friend  Professor  Greenough ;  and  in  his  desk  after  his  death  was  found  a 
MS  of  which  no  one  knew  anything  except  that  he  had  said  it  was  for  this 
same  friend, — an  operetta,  with  words  and  music  complete. 

Professor  Allen  was  married  on  Dec  26, 1878,  to  Miss  Emmeline  Laighton 
of  Portsmouth.  Their  eldest  child,  a  daughter,  died  in  Athens,  as  has  been 
stated.     A  son  and  a  daughter,  with  their  mother,  survive  him. 

The  scholarship  of  Professor  Allen  was  admirable.  His  command  of 
philological  methods  was  unusual,  and  his  presentation  of  arguments  in 
behalf  of  any  thesis  was  most  attractive, — never  seeming  merely  plausible, 
but  commanding  acceptance  as  necessary  truth.  Never  was  mind  more 
open  than  his  to  the  receipt  of  light  from  any  quarter, — like  Socrates,  ever 
glad  to  be  refuted,  and  abandoning  old  views  without  a  shadow  of  regret 
when  these  were  shown  to  be  false.  His  kindly  patience,  his  accuracy,  his 
absolute  sanity,  and  his  clearness  of  exposition  made  him  a  remarkable 
teacher  as  well  as  a  great  scholar.  His  criticisms  were  absolutely  frank, 
but  assumed  so  fully  that  he  whom  he  criticised  was  seeking  the  truth  like 
himself,  that  they  left  no  sting.  Those  who  knew  him  well  admired  in  him 
the  man  even  more  than  the  scholar.  He  was  ever  simple,  straightforward, 
kindly,  affectionate.  His  friends  depended  more  than  they  knew  on  him 
and  his  judgment.     With  him  a  great  and  pure  light  has  gone  oat  of  their 

lives. 

Thomas  Day  Seymour. 


Digitized  by  VjOOQ IC 


RECENT  PUBLICATIONS. 

Thanks  are  due  to  Messrs.  Lemcke  &  Buechner,  8x2  Broadway,  New 
York  City,  for  material  furnished. 

AMERICAN. 

Caesar  (C.  J.)  The  Second  Book  of  Caesar's  Gallic  War;  ed.  for  the 
use  of  schools  by  W.  C.  Collar.  Boston,  Ginn  &*  Co,.,  x^97>  9  +  9^  PP- 
Map,  i6mo,  cl.,  40  cts. 

Cicero  (Marcus  TuUius).  Pro  Flancio ;  ed.  with  introd.  and  notes  by 
H.  W.  Auden.  New  York,  The  Macmillan  Co,,  1897.  84+150  pp.  i6mo, 
cl.,  net  80  cts. 

Hempl  (G.)  German  Orthography  and  Phonology.  Boston,  Ginn  6r*  Co., 
1897.     c.     27 +  303  pp.     i2mo,  cl.,  $2.10. 

Homer.  Eight  Books  of  Homer's  Odyssey;  with  introd.,  commentary 
and  vocabulary,  by  B.  Perrin  and  T.  D.  Seymour.  Boston,  Ginn  fir*  Co., 
1897.     c.     68+ 107  pp.     i2mo,  $1.50. 

Hutson  (C.  W.)  The  Story  of  Language.  Chicago,  A.  C.  McClurg  &» 
Ctf.,  1897.     c.     2  +  392  pp.     i2mo,  cl.,  I1.50. 

Pearson  (H.  C.)  Greek  Prose  Composition.  New  York,  American  Book 
Co,,  1897.     c.     187  pp.     i2mo,  cl.,90  cts. 

Snider  (D.  J.)  Homer's  Odyssey  :  a  commentary.  St.  Louis,  Mo.,  T^e 
Sigma  Publishing  Co,,  1897.     c.  '95.     534  pp.     8vo,  cl,,  $z. 

Sophocles.  Tragoediae  ;  ed.  by  Rob.  Y.  Tyrrell.  (Parnassus  Library  of 
Greek  and  Latin  Classics.)  New  York,  The  Macmillan  Co.,  1897.  25  + 
272  pp.     i6mo,  bds.,  net  $1.60. 

Xenophon.  Works ;  tr.  into  English  by  H,  G.  Dakyns.  In  4  v.  V.  3, 
pt.  I.  The  Memorabilia,  and  Apology,  The  Economist,  The  Symposium, 
and  Hiero.  Pt.  2.  Three  essays  :  on  the  duties  of  a  cavalry  general,  on 
horsemanship,  and  on  hunting.  New  York,  The  Macmillan  Co,,  1897.  70 
+  130  pp.     i2mo,  cl.,  net  $1.25. 

ENGLISH. 

Cicero  Correspondence.  Revised  text,  notes,  etc.,  by  Rb.  Y.  Tyrrell 
and  L.  C.  Purser.     Vol.  V.     London,  1897.     8vo,  500  pp.     14s.     Vol.  I-IV, 

£2  IIS. 

Elliott  (W.  A.)  Dictionary  of  the  Tebele  and  Shuna  Languages.  Lon- 
don, 1897.    8vo.     I  OS. 

Hill  (G.  F.)  Sources  for  Greek  History  between  the  Persian  and  Pelo- 
ponnesian  Wars.     London,  1897.    8vo,  434  pp.     los.  6d. 

Jaschke  (H.  A.)  Tibetan- English  Dictionary.  English-Tibetan  Dic- 
tionary.   London,  1897.     8vo,  694  pp.     £1  ids. 


Digitized  by  VjOOQ IC 


RECENT  PUBLICATIONS,  m 

Lang  (A.)    Modern  Mythology.    London,  1897.    8vo,  238  pp.    9s. 

de  Larajasse  (F.)    Somali-English,  English-Somali  Dictionary.    London, 

1897.      8V0.       I2S. 

and  de  Sampont  (F.  C.)  Practical  Grammar  of  the  Somali  Lan- 
guage.    London,  1897.    8vo.     12s. 

Larsen.      Dano-Norwegian   English   Dictionary.      London,   1897.      8vo, 

687  pp.      1 28.  6d. 

Robinson  (C.  H.)  Hausa  Grammar,  Exercises,  Readings,  Vocabulary. 
London,  1897.    8to.     5s. 

Wright  (J.)  The  English  Dialect  Dictionary.  Part  3.  London,  1897. 
4to.     S.  Nr.  108 1.     15s. 

FRENCH. 

de  Julleville  (P.  L.)  Histoire  de  la  langue  et  de  la  litterature  frangaise. 
Vol.  Ill,  fasc.  4-1 1.     Paris,  1897.     8vo.     Vol.  Ill  cplt.,  16  £r. 

Masqueray  (E.)  Observations  grammaticales  sur  la  grammaire  Touareg 
(<  Publications  de  TEcole  des  lettres  d'Alger,'  tome  XVIII,  fasc.  3).  Paris, 
1897.     8vo,  193  pp.     5  £r. 

Vandaele  (H.)  L'optatif  grec.  Essai  de  syntaxe  historique.  Paris, 
1897.    8vo.    8  £r. 

GERMAN. 

'Arlb,  Tabarf  continuatus,  quem  edidit,  indicibus  et  glossario  instruxit 
M.  J.  de  Goeje.  gr.  8.  xxvii,  213  S.  Leiden,  Buchhandlg,  u.  Druckerei 
vorm.  E.  J,  Brill,     m.  6. 

'AfHGTo^vovc  'Etlpffvri,  Cum  scholiorum  antiquorum  excerptis  passim  emen- 
datis  recognovit  et  adnotavit  Henr.  van  Herwerden.  2  partes,  gr.  8. 
"Ltiden,  A,  IV,  Sij'lAoff.     m.  12.50.    . 

Beneze  (Emil).  Sagen-  u.  litterarhistorische  Untersuchungen.  Nr.  i  u. 
2.  gr.  8.  Halle,  Jif.  Niemeyer, — i.  Traummotiv,  das,  in  der  mittelhoch- 
deutschen  Dichtung  bis  1250  u.  in  alten  deutschen  Volksliedern.  82  S. 
m.  2.40. — 2.  Orendel,  Wilhelm  v.  Orense  u.  Robert  der  Teufel.  Eine 
Studie  zur  deutschen  u.  franzds.  Sagengeschichte.     v,  112  S.     m.  2.80. 

Bibliothek  indogermanischer  Grammatiken.  Bearb.  v.  B.  DelbrUck,  K. 
Foy,  G.  N.  Hatzidakis,  etc.  6.  Bd.  i.  Thl.  2.  Abth.  gr.  8.  L.,  Breitkopf 
^  Hdrtel, — 6.  HUbschmann  (H.)  Armenische  Grammatik.  i.  Thl.;  Ar- 
menische  Etymologie.  2.  Abth.:  Die  syr.  u.  griech.  LehnwOrter  im  Alt- 
armenischen  u.  die  echtarmen.  WOrter.     xxiii  u.  S.  281-575.     m.  8. 

—  phonetiscber  WOrtcrbttcher.  Hrsg.  v.  H.  Michaelis.  i.  Bd.  gr.  8. 
Hannover,  C,  Meyer, — x.  Michaelis  (H.)  u.  Passy  (P.)  Dictionnaire  pho- 
netique  de  la  langue  fran^aise.  Avec  preface  de  Gaston  Paris,  xvi,  319  S. 
m.  4 ;  geb.,  m.  4.80. 

Blanckenburg  (Curt.)  Studien  Ub.  die  Sprache  Abrahams  a  S.  Clara, 
gr.  8.     iv,  87  S.     Halle,  M,  Niemeyer,     m.  2.40. 

Braune  (Wilh.)  Althochdeutsches  Lesebuch.  4.  Aufl.  gr.  8.  viii,  249  S. 
Halle,  M,  Niemeyer,     m.  4.50. 

Bursy  (Bernh.)  De  Aristotelis  IIoA^rcazc  ^Adrfvaiuv  partis  alterius  fonte  et 
auctoritate.    gr.  8.    viii,  148  S.    Jurjewi.    "L,,^  K,  F,  Koehler'* s  Sort,    m.2.50. 


Digitized  by  VjOOQ IC 


378  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Corpus  scriptorum  historiae  Byzantinae.  Editio  emendatior  et  copiosior, 
consilio  B.  G.  Niebuhrii  C.  F.  instituta,  auctoritate  academiae  litterarum 
regiae  borussicae  continaata.  Vol.  I  (Schluss).  loannis  Zonarae  epitomae 
historiarum  libri  XVIII.  £z  recensione  Maur.  Pinderi.  Tom.  III.  Libri 
XIII-XVIII.  Ed.  Thdr.  Battner-Wobst.  gr.  8.  xxi,  933  S.  Bonn,  E, 
Weber,     m.  24. 

Dienel  (Rich.)  Untersuchungen  ttb.  den  Taciteischen  Rednerdialog. 
Progr.    gr.  8.     20  S.     St.  Polten,  Selbstverlag,    m.  — 40. 

Dissertationes  philologicae  Halenses.  Vol.  XIII.  Pars  3.  gr.  8.  Halle» 
M,  Niemeyer. — 3.  Haym  (Conr.)  De  puerorum  in  re  scaenica  Graecorum 
partibus.     iii  a.  S.  217-295.     m.  2. 

Dittmar  (Armin).  Studien  zur  lateinischen  Moduslehre.  gr.  8.  xi, 
346  S.     L.,  B,  G.  Teubner,     m.  8. 

Enneccerus  (M.)  Die  altesten  deutschen  Sprach-Denkmaler.  In  Licht- 
drucken  hrsg.  Fol.  44  Taf.  auf  18  Bl.  m.  3  S.  Text.  Frankfurt  a.  M., 
F.  Enneccerus.    In  Mappe,  m.  27. 

Eudociae  Augustae,  Procli  Lycii,  Claudiani  carminum  graecorum  reli- 
quiae.    Recensuit  Arth.  Ludwich.    8.    vi|  241  S.    L.,,  B.  G.  Teubner.   m.4. 

Grammatiken,  altera  deutsche,  in  Neudrucken,  hrsg.  v.  John  Meier.  IV. 
8.  Halle,  Af,  Niemeyer, — IV.  Olinger  (Alb.)  Deutsche  Grammatik,  hrsg. 
V.  Willy  Scheel.     ix,  Ixii,  129  S.  m.  2  Fcsms.     m.  5. 

Grundriss  der  germanischen  Fhilologie,  hrsg.  v.  Herm.  Paul.  2.  Aufl. 
3.  Bd.    I.  Lfg.     gr.  8.     S.  1-256.     Strassburg,  K.J,  TrUbner  Verl.     m.  4. 

der  indo-arischen  Philologie  u.  Altertumskunde.     Hrsg.  v.  Geo. 

Btthler.     i.  Bd.   3.  Hft.   B.     gr.  8.     Strassburg,  K,  J,  TrUbner  Verl,--!  3, 
B.  Zachariae  (Thdr.)    Die  indischen  WOrterbUcber  (Kosa).    42  S.    m.  2.50. 

der  iranischen  Philologie.     Hrsg.  v.  Wilh.  Geiger  u.  Ernst  Kuhn. 

2.  Bd.   3.  Lfg.     gr.  8.     S.  321-480.     Ebd.     m.  8. 

Jacob  (A.)  Die  Glossen  des  Cod.  S.  Pauli  D.  82.  Diss.  gr.  8.  vi,  52  S. 
Halle,  M,  Niemeyer,     m.  1.20. 

JahrbUcher,  neue,  f.  das  klassische  Altertum,  Geschichte  u.  deutsche 
Litteratur  u.  f.  Padagogik,  hrsg.  v.  Johs.  Ilberg  u.  Rich.  Richter.  i.  Jahrg. 
1898.  I.  u.  2.  Bd.  10  Hfte.  Lex.^.  i.  Bd.  i.  Hft.  80  u.  64  S.  L.,  B.  G, 
Teubner.     m.  28. 

Jahresbericht  des  Instituts  f.  rumanische  Sprache  (rumanisches  Seminar) 
zn  Leipzig.  Hrsg.  v.  Gust.  Weigand.  gr.  8.  ix,  336  S.  L.,  J,  A,  Barth. 
m.  6. 

Jahresberichte  f.  neuere  deutsche  Litteraturgeschichte,  hrsg.  v.  Jul.  Elias 
u.  Max  Osborn.  6.  Bd.  (J.  1895).  i.  Abtlg.  Lex.-8.  141  S.  L.,  G,  /. 
Gdschen.     m.  6. 

lungius  (C.  L.)  De  vocabulis  antiquae  cumoediae  atticae  quae  apud 
solos  comicos  aut  omnino  inveniuntur  aut  peculiari  notione  praedita  occur- 
runt.  gr.  8.  xxiy,  358  u.  6  S.  Traiecti  ad  Rhenum.  Amsterdam,  J. 
Mailer,    ro.  10. 

Kluge  (Frdr.)  Angelsfichsisches  Lesebuch,  zusammengestellt  u.  m. 
Glossar  versehen.     2.  Aufl.     gr.  8.     iv,  214  S.     Halle,  M,  Niemtyer,    m.  5. 

Ktlnstlinger  (Dav.)  Die  Theorie  der  ZahlwOrter  in  den  semitischeo 
Spiachen.    gr.  8.     32  S.     B.,  S,  Calvary  &*  Co,     m.  i. 


Digitized  by  VjOOQ IC 


■^i 


RECENT  PUBLICATIONS,  3/9 

Laurin  u.  Der  kleine  Rosengarten.  Hrsg.  v.  Geo.  Holz.  gr.  8.  v,  xWi, 
213  S.     Halle,  M,  Niemeytr.    m.  7. 

LehrbUcher  des  Seminars  f.  oriental iscbe  Sprachen  zu  Berlin.  Hrsg.  v. 
dem  Director  des  Seminars.  XVI.  Bd.  gr.  8.  B.,  IV.  Spemann,  Geb.  in 
Leinw.— XVI.  Viehe  (G.)  Grammatik  des  Otjiherero,  nebst  WQrterbach. 
zii,  140  S.     m.  12. 

Lemcke  (Ernst).  Teztkritische  Untersuchangen  zu  den  Liedern  Hein- 
richs  V.  Morungen.     gr.  8.     no  S.     Jena,  O,  Rassmann,     m.  1.60. 

Mitteilungen  der  vorderasiatischen  Gesellschaft.  1897.  2  u.  3.  gr.  8. 
B.,  IV,  Fetter  Verl.  in  Kotnm, — 2.  Rost  (Paul).  Untersuchungen  zur  alt- 
orientalischen  Geschichte.  iii,  141  u.  8  autogr.  S.  m.  8. — 3.  Sammelheft. 
Glaser  (Ed.)  Das  Alter  der  minaischen  Inschriften  u.  der  Ursprung  des 
Namens  der  Ebraer. — Hommel  (Fritz).  Das  graphische  T\  im  Minaischen 
a.  das  Alter  der  minaischen  Inschriften. — MuUer  (W.  M.)  Geographische 
Einzelheiten.— Miiller  (W.  M.)  u.  Winckler  (Hugo).  Pap^hu  Hierapolis. 
— Winckler  (Hugo).  OTU  nriK'.— Winckler  (Hugo).  Tel-Amarna  125.— 
Winckler  (Hugo).  Die  Istar  v.  Ninive  in  Aegypten. — Niebuhr  (C.)  Die 
erste  Dynastie  v.  Babel.    48  S.     m.  2. 

1897.     4-6  (Schluss).     gr.  8.      B.,  W,  Peiser  Verl,  in  JKomm,-'^, 

Peiser  (F.  E.)  Studien  zur  orientalischen  Altertumskunde.  32  S.  m.  i. 
— 5.  Winckler  (Hugo).  Die  sabsiischen  Inschriften  der  Zeit  Alhan  Nah- 
fan's.  32  S.  m.  1.50. — 6.  Glaser  (Ed.)  Zwei  Inschriften  Ub.  den  Darom- 
bruch  V.  Mirib.  Ein  Beitrag  zur  Geschichte  Arabiens  im  5.  u.  6.  Jahrh.  n. 
Chr.     iii,  128  S.    m.  6. 

Miinzer  (F.)  Beitr&ge  zur  Quellenkritik  der  Naturgeschichte  des  Plinius. 
gr.  8.     zi,  432  S.     B.,  IVeidmann,     m.  12. 

Neoglottia.  Kritische  Rundschau  auf  dem  Gebiete  der  neueren  Sprachen 
o.  Litteraturen.  Hrsg.  ▼.  Adf.  Kressner.  i.  Jahrg.  Oktbr.  1897-Septbr. 
1898.     24  Nrn.     hoch  4.     Nr.  i  u.  2, 16  S.     L.,  Renger,     Halbjahrlich  m.  5. 

Saga-Bibliotheky  altnordische.  Hrsg.  v.  Gust.  CederschiOld,  Hugo 
Gering  u.  Eug.  Mogk.  6.  Hft.  gr.  8.  Halle,  M,  Niemeyer, — 6.  Eyrbyggja 
saga.     Hrsg.  v.  Hugo  Gering.     zxxi,  264  S.     m.  8. 

Sammlung  kurzer  Grammatiken  germanischer  Dialekte.  Hrsg.  y.  Wilh. 
Braune.  VIII.  gr.  8.  Halle,  M.  Niemeyer.^Y 111,  Noreen  (Adf.)  Alt- 
nordische Grammatik.  II.  Altschwedische  Grammatik  m.  Einschluss  des 
Altgutnischen.     i.  Lfg.  (Einleitung,  Sonanten).     173  S.     m.  3.60. 

Schmid  (Wilh.)  Der  Atticismus  in  seinen  Hauptvertretern  von  Dionysius 
V.  Halikarnass  bis  auf  den  zweiten  Philostratus.  Registerbd.  gr.  8.  iii, 
234  S.     St,  IV,  Kohlhammer,     m.  6. 

Stucken  (Ed.)  Astralmythen  der  Hebraer,  Babylonier  u.  Aegypter. 
2.  Tl.  LoL     gr.  8.     S.  81-125.     L.,  E.  Pfeiffer,     m.  5. 

Studien  zur  englischen  Philologie,  hrsg.  v.  Lor.  Morsbach.  I.  u.  II.  Hft. 
gr.  8.  Halle,  M,  Niemeyer. — I.  Spies  (Heinr.)  Studien  zur  Geschichte 
des  englischen  Pronomens  im  XV.  u.  XVI.  Jahrh.  (Flezionslehre  u. 
Syntax.)  xix,  311  S.  m.  8.— II.  Herzfeld  (Geo.)  William  Taylor  v. 
Norwich,    viii,  71S.     m.  2. 

Tibullas  (Albius).  Untersuchung  u.  Text  y.  H.  Belling.  2  Tie.  8.  B., 
R,  Gaertner,     m.  9. 


Digitized  by  VjOOQ IC 


380  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Toepffer  (Jobs.)  Beitr2i);e  zur  griechischen  AUertumswissenschaft. 
gr.  8.     xvi,  384  S.  m.  i  Abbildg.  u.  Bildnis.     B.,  IVeidmann.    m.  10. 

U8ener(H.)  Der  Stoff  des  griechischen  Epos.  [Aus  *Sitzungsber.  d. 
k.  Akad.  d.  Wiss.']  gr.  8.  63  S.  Wien,  C.  GerolePs  Sohn  in  JComm. 
m.  1.50. 

Vandaele  (Hilarius).  Qua  mente  Phaeder  fabellas  scripserit.  Thesis, 
gr.  8.     iv,  105  S.    Paris,  E,  Bouillon,    m.  3.20. 

Winckler  (Hugo).  Altorientaliscbe  Forschungen.  i.  Bd.  VI.  gr.  8. 
V  u.  S.  469-573.     L.,  E.  Pfeiffer,     m.  6;  geb.,  m.  6.50. 

Witkowski  (Stanisl.)  Prodromus  grammaticae  papyrorum  graecarum 
aetatis  Lagidarum.  gr.  8.  65  S.  Krakau,  Buchn.  der  poln.  Verlags- 
Gesellschaft  in  JComm,     m.  3. 

ITALIAN. 

Cicero  (M.  Tullius).  Actionis  in  C.  Verrem  secundae  liber  quartps  [de 
signis].  Commentato  da  Vittorio  Brugnola.  Torino.  8vo,  pagine  151. 
L.  1.80. 

Fregni  (Giuseppe).  Studi  storici,  filologici  e  letterari.  Modena.  8to, 
fig'»  pagine  164.     L.  5. 

Pizzi  (Italo).  Grammatica  elementare  dell'  antico  iranico  (zendo  e 
persiano  -  antico),  con  antologia  e  relativo  vocabolario.  Torino.  8vo,  p. 
92.     L.  2.50. 


Digitized  by  VjOOQ IC 


BOOKS  RECEIVED. 


Digitized  by 


Google 


382  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Hermes,  herausg.  v.  G.  Kaibel  u.  Carl  Robert.  XXXII.  Band.  2.  u.  3. 
Heft,     Berlin,  Weidmannsche  Buchhandlung^  1897. 

Journal  (The)  of  Germanic  Philology.     Vol.  I,  No.  2.     1897. 

Journal  of  Philology  (The).  Ed.  by  W.  A.  Wright,  Ingram  Bywater  and 
Henry  Jackson.  Vol.  XXV,  No.  50.  London  and  New  York,  Macmillan 
&>  Co.,  1897.     4s.  6d. 

Knauth(H.)  UebungsstUcke  zum  Uebersetzen  in  das  Lateinische.  Deut- 
scher  Text  u.  Lateinische  Uebersetzung.     2te  Aufl.     Leipzig,  G.  Freytag. 

Leo  (Fr.)  Die  plautinischen  Cantica  u.  die  hellenistische  Lyrik.  Ab- 
handlungen  der  KOnigl.  Gesellsch.  der  Wissenschaften  za  Gdttingen. 
Philolog.-histor.  Klasse.  Neue  Folge.  Bd.  I,  No.  7.  Berlin,  Weidmann- 
sche Buchhandlung^  1897.     7  ra.  50  pf. 

Michaelis  (H.)  et  Passy  (P.)  Dictionnaire  phonetique  de  la  langue 
f  rangaise.     Hannover,  Carl  Meyer^  1897.     4  m. 

Mnemosyne.  Bibliotheca  Philologica  Batava.  Nova  series.  Vol.  XXV. 
Partes  I,  II,  III.     L.-B.,  E,  J.  BrilL    Lipsiae,  O,  Harrassowitz,  1897. 

Modern  Language  Association  of  America,  Publications  of  the.  Ed.  by 
James  W.  Bright,  Sec'y.  Vol.  XII,  No.  3.  N.  S.,  vol.  V,  No.  3.  Contents  ; 
Literature  and  Personality  (Calvin  Thomas);  Learned  and  LearnM  (Geo. 
Hempl) ;  A  Study  of  the  Metrical  Structure  of  The  Pearl  (C.  S.  Northup) ; 
Gaston  Paris  (H.  A.  Todd);  Pastoral  Influence  in  the  English  Drama  (H. 
Smith).     Baltimore,  y<?A«  Murphy  &*  Co.,  1897. 

Moore  (Clifford  H.)  Julius  Firmicus  Maternus,  der  Heide  o.  der  Christ. 
(MUnchener  Diss.)     (1897.) 

MUnzer(F.)  Beitrage  zur  Quellenkritik  der  Naturgeschichte  des  Plinius. 
Berlin,  Weidmannsche  Buchhandlung,  1897.     12  m. 

New  York,  University  of  the  State  of.  Examination  Bulletin.  No.  13, 
June,  1897.  College-Entrance  English.  Albany,  University  of  the  State  of 
N.  K.,  1897. 

JIpoKTiKa  TTjq  kv  'ABffvatg  apxaioh)yiKf}c  iTaipiac  tov  Itovq  1896.    'Aft^Tfftv,  1897. 

Keditus  Augusti,  carmen  praemio  aureo  ornatum  in  certamine  poetico 
Hoeufftiano.  Accedunt  quattuor  poematalaudata.  Amsterdam,  Muller,  1897. 

Sollima  (Francesco).  Le  Fonti  di  Strabone  nella  geografia  della  Sicilia. 
Messina,  1897. 

Sophokles'  Oidipus  auf  Kolonos.  Fttr  d.  Schulgebr.  herausg.  v.  Fr. 
Schuber.     zte  verb.  Aufl.     1897.     60  pf. 

Syms  (L.  C.)     Third  Year  in  French.     N.  Y.,  Cine,  Chic,  1897.     I1.20. 

Tropea  (Giacomo).  II  mito  di  Crono  in  Sicilia  e  la  ragione  del  nome 
Zancle.     Messina,  1S97. 

Weidner  (Andreas).  SchUler-Kommentar  zu  Tacitus*  historischen  Schrif- 
ten  in  Auswahl.     Leipzig,  G,  Freytag,  1897.     i  m.  70  pf. 

Xenophon's  Cyropaedia.  Ed.  by  C.  W.  Gleason.  New  York,  Cine, 
Chic,  American  Book  Co,,  1897.     $1.25. 

Zeitschrift  far  franzbsische  Sprache  u.  Litteratur,  unter  besonderer  Mit- 
wirkung  ihrer  BegrUnder,  Kdrting  a.  Koschwitz,  herausg.  v.  Behrens.  Bd. 
XIX,  Heft  5  u.  7.  Der  Abhandlungen  3.  u.  4.  Heft.  Berlin,  W,  Gronau, 
18.  Sept.  1897. 


Digitized  by  VjOOQ IC 


AMERICAN 
JOURNAL    OF    PHILOLOGY 

Vol.  XVIII,  4.  Whole  No.  72. 


I.— THE  SUBJUNCTIVE  IN  INDEPENDENT  SEN- 
TENCES  IN  PLAUTUS. 

III.— Optative  and  Potential. 

The  immediate  result  of  the  working  of  the  forces  discussed  in 
the  previous  paper  has  been  the  formation  of  usages  and  idioms 
in  which  the  meaning  of  the  subjunctive  form  is  restricted  more 
or  less  closely,  as  the  usage  is  narrow  or  broad,  to  a  particular 
function.  Therefore  the  explanation  of  any  particular  case  of  the 
subjunctive  on  historical  principles  must  begin  with  the  determi- 
nation of  the  extent  to  which  one  or  more  of  these  forces  has  acted 
upon  it.  The  explanation  of  caperes  fustem  is  that  the  mode  is 
limited  by  the  past  tense  to  the  expression  of  obligation ;  nan 
meream  by  person  and  by  verb-meaning  to  negative  determina- 
tion in  regard  to  a  supposed  case ;  eamus  by  person  and  number 
to  an  exhortation,  and  eamus  iUy  Una,  in  ius  by  the  emphasis  upon 
the  2d  pers.  becomes  a  command,  egon  . . .  patiar  is  determined 
by  the  form  of  sentence  and  by  the  meaning  of  paiior.  The 
added  words  and  the  meaning  of  the  verb  limit  sit  per  me  quidem 
to  an  expression  of  indifference.  In  saluos  sis  the  verb-meaning 
and  the  person  are  enough  to  restrict  the  phrase  to  use  in  greet- 
ings, as  the  same  influences  restrict  salue.  Three  forces — the 
verb-meaning,  the  passive  voice  and  the  3d  pers.  (impers.) — limit 
Jiat  to  an  expression  of  assent.  The  extension  of  the  modal  force 
by  suggestion  is  illustrated  in  quo  mode  ego  uiuam  sine  tef  with 
its  apparent  can  force,  in  quor  with  the  accompanying  should 
sense,  and  in  qui  ego  isiuc  credamf,  where^a  special  form  of 


Digitized  by  VjOOQ IC 


384  AMERICAN  JOURNAL    OF  PHILOLOGY. 

interrogation  unites  with  a  particular  verb-meaning  to  give  the 
mode  a  can  sense  in  a  stereotyped  phrase. 

Between  the  usages  thus  fixed  it  is  at  times  possible  to  trace,  or 
at  least  to  suspect,  a  real  historical  connection.  The  various 
forms  of  guid /aciam  f  which  in  the  movement  toward  precision 
expand  along  one  line  into  gntd  ego  nunc  faciam?  and  along 
another  line  into  quid  uis  faciam  f  illustrate  such  a  connection. 
The  differentiation  of  will  from  wish  by  uolo  and  uelim  with  a 
paratactic  subjunctive  and  the  subsequent  use  of  uelim  with 
infinitives  and  even  with  an  object  or  absolutely,  the  optative 
sense  becoming  less  marked  in  the  later  uses,  is  another  illus- 
tration. Some  other  indications  of  relationship  have  been  noted 
above,  but  for  the  most  part  it  is  not  possible  to  prove  an  actual 
connection  between  different  usages.  Schemes  of  relationship, 
intended  to  exhibit  the  many  varieties  of  subjunctive  usage  in  a 
kind  of  modal  genealogical  tree,  are  speculations.  But  two 
groups  of  usage,  the  potential*  and  the  optative,  are  so  evident 
and  so  generally  accepted  that  they  must  be  recognized  in  any 
discussion  of  the  subjunctive. 

The  subjunctives  which  do  not  express  any  of  the  varying 
shades  of  will  and  desire  have  been  noted  in  the  lists  above. 
Excluding  the  questions  and  the  few  paratactic  cases,  there  are 
about  140  such  cases. 

About  40  of  them  have  a  protasis  in  the  context,  though  not  in 
the  same  sentence.  They  are  found  in  all  forms,  mainly  in  3d 
sing,  (about  12  cases)  and  in  3d  plur.  (about  20,  of  which  10  are 
in  Merc.  407  ff.).  They  are  true  potentials,  or  at  least  are  hypo- 
thetical, but  most,  if  not  all,  are  so  influenced  by  the  protasis  that 
they  cannot  be  made  the  basis  of  any  reasoning  about  the  sub- 
junctive in  independent  sentences.  They  are  material  for  study 
of  the  conditional  sentence. 

With  these  must  go  a  smaller  group  of  cases  where  a  subjunc- 
tive clause  (with  dum^  ubiy  quamuis^  qui)  accompanies  the  leading 
clause.  Thus,  qui  culpet .  . .,  shilius  sit.  It  is  impossible  to 
exclude  the  influence  of  these  clauses,  which  usually  precede  the 
main  verb  and  may  affect  it  as  a  protasis  would. 

In  discussions  of  potential  use  examples  are  largely  drawn  from 
the  interrogative  sentence,  especially  from  quis  questions  with  the 
verb  in  ist  sing,  of  the  present.    These  cases  have  been  suffi- 

*I  have  used  the  term  potential,  not  because  it  is  correct,  but  because  it  is 
convenient  and  in  common  use. 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  385 

ciently  commented  upon  above.  They  are  not  very  frequent, 
possibly  30  or  40  cases  out  of  more  than  200 ;  they  are  not 
hypothetical,  except  when  a  definite  protasis  is  expressed ;  and 
they  are  connected  most  closely  and  distinctly  with  the  subjunc- 
tive of  will.  The  meaning  of  the  mode  is,  however,  turned  from 
direct  will  to  obligation,^  first  by  the  interrogative  form  in  general 
and  also  by  the  special  influences  which  have  been  noted  above. 
In  quid  ego  nunc  faciamf  it  is  the  self-address;  in  other  cases  it 
is  the  special  form  of  qtiis,  as  quid  *why,'  quor^  qui.  In  the 
phrase  qui  ego  istuc  credamf  the  verb-meaning  helps  to  give  the 
can  sense,  but  if  it  had  been  quor  ego  istuc  credamf  the  apparent 
sense  would  have  been  should,  A  sense  which  varies  with  the 
varying  form  of  sentence,  instead  of  remaining  fixed  with  the 
fixed  mode,  is  not  a  meaning  of  the  mode,  but  of  the  whole 
sentence,  including  the  mode.  It  is  not  of  great  importance  to 
decide  whether  these  uses  should  be  called  by  the  name  potential ; 
that  depends  upon  the  definition  given  to  the  term;  but  it  is 
important  that  we  should  look  upon  them  rather  as  a  group  of 
usages  which  have  grown  out  of  the  subjunctive  of  will  or  desire 
in  sentences  of  peculiar  form,  than  as  the  result  of  a  supposed 
potential  force,  inherited  from  the  I.E.  stage. 

A  fourth  group  consists  of  some  16  or  18  cases,  aU  but  one  in 
pres.  ist  sing.,  in  which  there  is  an  element  of  will,  though  they 
are  negatived  by  non^  and  which  relate,  for  the  most  part,  to  a 
hypothetical  act  or  situation.  Of  hand  (non)  dicam  dolo  and 
deum,  (maioruni)  uiriute  dicam  it  has  been  said  above  that  they 
are  almost  futures,  used  as  eloquar  is  frequently  used,  to  introduce 
a  statement  of  determination  in  regard  to  a  future  act  (cf.  ibo\ 
But  an  unmodified  expression  of  determination  is,  strictly,  incom- 
patible with  haud  dolo  and  still  more  plainly  with  deum  uiriuie ; 
the  addition  of  the  idea  'without  concealment'  or  'thanks  to  the 
gods,'  'by  the  goodness  of  the  gods,'  brings  in  the  suggested 
thought '  I  am  able  to  say,' '  I  may  say.'  That  is,  deum  uiriute 
dicam^  expanded  to  its  full  meaning,  is  'by  the  kindness  of  the 
gods  I  may  say  and  will  now  say,  that . . .,'  and  so  dicam  gets  in 
these  phrases  a  slight  potential  coloring  and  lies,  as  I  have  said, 
between  the  future  and  the  subjunctive. 

The  other  cases  of  this  group  express  a  determination,  usually 
negative,  in  regard  to  the  speaker's  action  in  a  supposed  case. 
There  is  in  non .  . .  mereamy  non  duim,  an  element  of  will,  not 

» Cf.  Elmer,  Prohibitive,  p.  36  (=  A.  J.  P.  XV  3,  p.  313). 


Digitized  by  VjOOQ IC 


386  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Strong,  but  sufficient  to  distinguish  them  from  cases  which  express 
an  opinion  or  even  a  conviction,  like  nimis  nili  tibicen  siem.  But 
a  determination  is  separated  from  a  conviction  by  so  small  an 
interval  that  Trin.  758,  ab  amico  . .  .  mutuom  argentum  rogem,  or 
True.  495,  sine  uirtute  argutum  ciuem  mihi  habeam  pro  praefica, 
might  be  either.  But  the  potential  force  of  all  these  cases  is 
shown  by  the  fact  that  they  deal  with  a  supposed  situation. 

A  fifth  group  is  made  up  of  the  sigmatic  aorists,  ist  sing.,  18 
cases.  The  MSS  give  only  attsim  and  faxiniy  but  negassim  and 
empsim  are  not  improbable  conjectures.  In  use  they  are  of  two 
kinds,  ausim,  negassiniy  empsim  wth  a  negative  expressing  deter- 
mination, like  nan  meream^  and  faxim  in  hypothetical  sentences, 
usually  with  a  protasis.  Functionally,  therefore,  they  belong 
either  in  the  fourth  group  or  in  the  first,  but  they  are  put 
together  here  because  the  form,  which  was  already  becoming 
obsolete  with  most  verbs,  doubtless  aided  in  producing  a  sharp- 
ness of  definition,  a  limitation  of  use,  which  they  would  not  have 
shown  if  the  form  had  been  in  free  use.  Archaic  forms  survive 
in  idioms.  The  same  fact  would  limit  their  influence  upon  later 
constructions. 

The  cases  of  the  indefinite  second  person  have  been  given 
above,  about  20  in  number.  This  usage  is  by  no  means  easy  to 
define,  and  some  of  the  cases  might  doubdess  be  understood  to  refer 
to  a  definite  person.  The  doctrine  of  Madvig,  that  the  subjunc- 
tive is  always  used  with  the  indefinite  2d  pers.,  is  now  known  to 
be  too  sweeping,^  since  cases  exist  with  the  indie,  e.  g.  Asin.  242. 
But  a  general  tendency  to  associate  the  indef.  2d  sing,  with  the 
subjunctive  evidently  exists.  In  Plautus  the  verbs  are  uideas  (5), 
censeas  (3),  audias  (2),  scias  (2),  nescias  (2),  conspicias,  cupias^ 
desideres^  uelis,  inuenias^  perdas^  noceas ;  but  the  last  two  are 
used  after  quom  ferias  and  exorUur  opiio^  a  situation  in  which  a 
verb  with  definite  subject  would  be  subjunctive.  The  rest  are  all 
verbs  of  mental  action. 

This  rather  specialized  idiom,  then,  is  marked  by  three  charac- 
teristics— the  meaning  of  the  verb,  the  potential  tone  of  the  mode, 
and  the  indefinite  subject — and  the  explanation  of  it  must  be 
sought  along  these  lines.  In  the  first  place,  only  one  of  these 
verbs  occurs  in  2d  sing,  in  a  direct  expression  of  will,  and  in  that 
passage  (M.  G.  282  tute  scias  soli  tibi)  it  is  distinctly  less  jussive 

^KUhner,  II,  p.  480;  Hoffmann,  Das  Modusgesetz,  28  fT.;  Blase,  WOlfflin's 
Archiv,  IX  i,  p.  19  f. 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  387 

than  a  verb  of  action  like  aieas,  accipias.  The  meaning  of  the 
verbs  almost  excludes  them  from  the  jussive  uses  and  confines 
uideas,  nescias^  if  the  forms  are  employed  at  all,  to  potential 
functions.  Only  the  impv.  form  could  give  a  direct  will-force  to 
these  verbs,  and  even  in  this  form  uide  is  rather  *see!'  than 
'perceive,'  audi  is  'listen !'  and  sciio  has  a  peculiar  sense.  In  the 
second  place,  a  potential  or  conditioned  use  in  the  definite  2d 
sing,  passes  easily  into  the  indefinite,  because  what  is  true  of  the 
definite  'you'  is  true  generally,  especially  when  the  verb  is  of 
mental  action.  The  statement '  If  this  should  be  so,  yoy  would 
think  thus  and  so,'  is  easily  extended  from  the  definite  'you'  to 
the  general  'you,'  when  it  means  'Under  certain  conditions  you 
would  necessarily  think  in  a  certain  way.'  In  such  reasoning  it  is 
really  a  matter  of  indifference  whether  the  2d  person  is  definite 
or  general,  and  only  the  context  would  tell  whether  una  opera 
postules  is  definite  (as  it  happens  to  be  in  all  three  cases)  or 
indefinite.^  But  the  use  of  these  verbs  of  mental  action  with  a 
general  subject  still  further  separated  them  from  the  direct  will- 
force,  since  the  will  cannot  well  be  generalized,  as  it  is  when  it  is 
directed  upon  an  indefinite  'you,'  and  still  retain  its  simple  force. 
It  becomes  rather  an  expression  of  propriety,  as  in  cases  in  3d 
plur.  where  the  subject  is  a  class  of  persons.  It  is  to  the  combi- 
nation of  these  two  forces,  weakening  the  meaning  of  the  mode, 
with  the  influence  of  a  preceding  thought  setting  a  hypothetical 
tone  (see  below),  that  the  highly  specialized  idiom  of  the  indef. 
2d  pers.  is  due.  Or,  in  other  words,  the  indef.  2d  pers.  is  a 
variety  of  the  hypothetical  use,  made  idiomatic  by  the  meaning 
of  the  verb  and  by  the  indefinite  direction  of  the  will.  The 
withdrawal  of  these  verbs  from  ordinary  uses  leaves  four  usages 
in  2d  sing,  which  supplement  each  other  with  little  over-lapping : 
verbs  of  physical  activity  in  the  various  expressions  of  desire,  the 
same  verbs  in  potential  uses  with  preceding  clause  (very  rarely 
without  such  introduction),  verbs  of  mental  action  in  the  impv., 
and  the  same  verbs  in  the  indef.  2d  pers.,  potential.  The  last  use 
is  evidently  the  most  specialized  and  presumably  the  latest  of 
them  all. 

Beside  the  groups  of  usage  already  mentioned,  there  are  about 
25  cases:  esse  is  used  9  times  {siem^  sif,  stent,  esses,  esset,fuisses), 
possiet  once,  dicat  once,  duaniimce,  auiumeni  twice,  postules  (or 

^  There  is  a  like  tendency  to  pass  from  the  particular  to  the  general  in 
phrases  of  the  m  tufrustra  sis,  ne  erres  class. 


Digitized  by^ 


388  AMERICAN  JOURNAL   OF  PHILOLOGY. 

iubeas)  with  una  opera  5  times,  habeam  '  I  should  consider/  uide-- 
aiur,  ecficiatur,  once  each,  neque  mereai  (which  is  like  non 
meream)  and  seruareiur  (but  this  may  express  obligation). 
These  are  chiefly  verbs  of  saying  and  esse^  and  they  are  mainly 
in  the  3d  person.  As  to  sU^  it  is  used  in  direct  expressions  of  will 
only  in  wishes  (never  in  commands)  and  in  expressions  of  pro- 
priety like  bonus  sit  bonis ^  malus  sit  malis,  describing  an  ideal. 
With  its  passive  meaning  it  is  little  likely  to  be  used  in  any 
jussive  sense,  though  it  is  freely  used  in  questions.  But  the  chief 
influence  in  all  these  cases  is  not  the  verb-meaning,  but  other 
words  in  the  sentence  which  influence  the  mode.  Every  word  or 
phrase  which  is  added  to  the  nucleus  of  the  sentence  both  brings 
into  clearer  light  the  germ  of  the  thought  and  modifies  that 
thought  by  its  own  associations  and  color.  I  have  tried  above  to 
show  how  this  is  true  of  added  verbs  and  of  some  adverbs  and 
particles,  which  strengthen  or  weaken  or  color  the  idea  of  will. 
In  a  like  manner  a  sentence  may  begin  with  a  phrase  which  is 
incompatible  with  the  direct  forms  of  will  and  therefore  excludes 
them,  but  which  heightens  and  fosters  a  potential  force.  Thus 
sine  uirttde  ciuem  introduces  the  thought  in  such  a  way  that  the 
verb,  haieam  pro  praejica,  already  by  person  and  verb-meaning^ 
inclined  toward  the  potential,  can  mean  nothing  but  'I  should 
consider.'  In  Ba.  312  the  father  is  told  that  his  money  is  in  public 
guard  at  Ephesus;  he  begins  his  reply,  occidisiis  me:  nimio  hie 
priuatim,  and  into  this  setting  the  verb  seruareiur  must  fit  with 
a  potential  force.  So  una  opera  introduces  a  comparison  between 
an  ideal  and  the  proposed  act ;  it  is  in  itself  hypothetical,  and 
when  it  is  followed  by  posiules  (with  the  Plautine  meaning,  'to 
expect')  the  potential  tone  is  doubled.  Cf.  Cas.  309  ff.  una 
edepol  opera  in  furnum  calidum  condito  atque  ibi  torreto  me  pro 
pane  rubido,  ere,  qua  istuc  opera  a  me  impetres  quod  postulas, 
where  una  opera  is  strong  enough  to  give  a  potential  force  even 
to  the  imp  v.,  condiio^  torreto.  So  non  par  uidetur  expresses  an 
opinion  and  prepares  for  the  further  expression  of  opinion  in 
neque  sit  consentaneum.  It  is  the  modifying  effect  of  a  compar- 
ative which  gives  to  malim  a  potential  force  that  uelitn  does  not 
have,  and  potius  gives  potential  meaning  to  the  future  i^dic.^  and 
even  to  an  impv.,  Rud.  1048  uos  confugite  in  aram  potius  quam 

^Neumann,  de  fut.  in  prise.  Lat. .  . .  ui  et  usu,  Breslau,  1888,  p.  28.  This 
dissertation  contains  some  very  acute  observations  upon  the  over-lapping  of 
temporal  and  modal  meanings. 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  389 

ego.  Cf.  also  the  use  of  the  subj.  after  potius  quant.  So  in  Rud. 
780  f.  utrum  tu  .  .  .  cum  malo  lubentius  quiescis  an  sic  sine  malo 
• .  .  ?  the  comparative  lubentius  in  the  interrogative  sentence 
makes  quiescis  almost  potential.  It  is  perhaps  not  fanciful  to  feel 
a  modal  difference  between  faciam  sedulo  and  faciam  lubens^ 
though  the  English  phrases  which  we  instinctively  recall  may 
easily  mislead  us.  These  illustrations,  taken  almost  at  random, 
may  serve  to  show  how  great  the  influence  of  the  preceding 
thought  upon  the  modal  sense  may  be.  A  force  which  can 
suggest  potential  meaning  in  an  impv.  or  a  fut.  or  pres.  indie, 
might  easily  be  sufficient  to  confine  the  subj.  to  potential  func- 
tions. This  influence  of  the  preceding  thought  is,  in  fact,  recog- 
nized by  the  phrases  *  suppressed  condition,'  'disguised  condition,' 
which  are  sometimes  employed ;  but  these  phrases  imply  that  the 
conditioning  expressions  modify  the  modal  meaning  only  because 
they  are  substitutes  for  a  protasis.  This  is  to  explain  the  simpler 
structure  by  the  more  complex,  fac  ualeas  by  fac  ut  ualeas^  an 
in  simple  questions  by  an  in  double  questions.  Such  simplifica- 
tion of  language  may  no  doubt  take  place,  but  it  caii  be  accepted 
only  on  sufficient  evidence.  The  general  trend  of  thought  and 
language  is  the  other  way,  and  when  a  simple  structure  and  a 
complex  one  are  surely  connected,  the  probability  is  that  the 
simpler  form  is  the  earlier.  A  phrase  like  una  opera  is  rather  a 
protasis  in  embryo  than  a  suppressed  protasis.  As  it  becomes 
more  distinct  in  thought  it  takes  on  more  distinct  form,  as  a 
clause  or  a  formal  protasis,  and  gives  a  more  fixed  and  definite 
potential  sense  to  the  main  verb. 

The  potential  in  Plautus  is  not  a  single  and  unified  usage ;  it  is 
a  group  of  usages,  similar  but  by  no  means  identical  in  meaning, 
and,  in  this  early  stage  at  least,  scarcely  enough  alike  to  influence 
each  other  through  analogy.  They  have  followed  distinct  lines 
of  analogical  connection,  have  been  influenced  in  their  changes  of 
meaning  by  distinct  sets  of  forces,  and  have  assumed  meanings 
which,  on  any  careful  analysis,  are  also  distinct,  quid  ego  nunc 
faciam?  starts  from  a  subjunctive  in  which  speaker  and  wilier 
are  different  persons,  is  turned  from  its  original  meaning  by  being 
addressed  by  the  speaker  to  himself,  and  results  in  a  meaning 
like  the  English  'What  am  I  to  do?'  nan  meream  starts  from  a 
subjunctive  in  which  speaker  and  wilier  are  one,  is  influenced  by 
person  and  number  and  by  verb-meaning,  and  results  in  an 
expression  of  negative  determination.  So  faxim^  uideas^  qui  ego 
istuc  credamf    una  opera  postules^  sit^  has  each  its  separate 


Digitized  by  VjOOQ IC 


390  AMERICAN  JOURNAL    OF  PHILOLOGY, 

semasiological  history  and  its  distinct  result.  Of  all  these  forces 
that  which  is  exerted  by  a  preceding  thought,  giving  a  hypo- 
thetical or  ideal  tone  to  the  sentence,  is  the  strongest  and 
produces  the  widest  departure  from  the  ordinary  meanings  of  the 
subjunctive. 

It  is  extremely  difficult  to  say  at  what  point  the  similarities 
between  different  usages  begin  to  be  recognized  or  felt,  so  that 
the  usages  exert  an  influence  upon  each  other  through  analogy. 
It  is  clear  enough  that  all  these  developments  are  part  of  one 
general  process,  an  extension  of  subjunctive  meaning,  but  it  is 
difficult  to  think  that  in  the  time  of  Plautus  the  subjunctive  in 
quid  fcLciamf  can  have  influenced  uideas  or  that  una  opera 
postules  can  have  suggested  the  use  of  nimis  7iilt  tibicen  stents 
And  if  they  were  not  united  into  a  single  group,  they  can  have 
exerted  no  common  influence  upon  later  usage,  e.  g.  upon  the 
subordinate  clause.  The  only  kind  of  potential  use  which  can 
have  affected  the  subordinate  clause  is  that  in  which  the  mode  is 
ideal  or  hypothetical,  a  rare  usage,  not  more  than  one-twentieth 
of  the  true  independent  subjunctives.  This  is  connected  by  clear 
lines  with  the  conditional  sentence,  but  the  connection  with  the 
relative  clause,  which  is  often  assumed,  is  less  certain. 

It  is  generally  held  that  the  potential  use  of  the  subjunctive^ 
which  is  regarded  as  a  single  use,  is  descended  from  an  I.E. 
future  or  contingent  future.  This  would  involve  the  belief  that 
the  potential  is  an  early  development.  On  the  contrary,  most  of 
these  usages  appear  to  be  comparatively  late.  This  is  certainly  true 
of  the  deliberative  question,  of  the  expressions  of  negative  determi- 
nation and  of  the  indefinite  2d  person.  The  archaic  form  oifaxim 
might  seem  to  indicate  an  early  use,  but  the  complete  separation 
olfaxim.faxiSyfaxit  into  three  widely  different  functions  would 
imply  a  long  period  of  slow  change  from  the  original  single 
function  or  closely  related  functions.  As  to  the  purely  hypo- 
thetical use,  it  is  probably  earlier  than  the  others,  since  the  condi- 
tional sentence  in  the  time  of  Plautus  had  already  advanced  so 
far,  but  it  is  the  result  of  the  influence  of  thought  upon  thought 
and  cannot  have  come  into  use  until  language  and  thought  were 
somewhat  complex.  If  E.  Hermann^  is  right  in  thinking  that 
there  was  no  hypotaxis  in  language  in  the  pro-ethnic  period,  then 
the  I.E.  potential,  if  any  such  usage  existed,  must  have  been 
merely  rudimentary.     This  would  of  course  involve  the  hypoth- 

*  K.  Z.  33  (1895),  pp.  481  flf.    Gab  es  im  Indogcrmanischen  nebensatzc  ? 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS,  39 1 

esis  that  the  developments  in  other  languages,  e.  g.  in  Greek, 
were  distinct  from  the  Italic. 

I  have  thus  far  avoided  using  the  negative  as  a  test  of  the 
potential,  not  because  it  is  not,  in  the  main,  a  correct  test,  but 
because  it  should  not  be  used  without  some  consideration,  of  its 
meaning.    The  composition  of  ne  with  all  kinds  of  words,  and 
especially  with   indicative  verbs,  nescio,  nequeo,  nolo,  can  be 
explained  only  on  the  hypothesis  that  ne  was  once  the  general 
Italic  negative  and  that  non  is  a  later  strengthened  or  compounded 
form  of  ne.    Whatever  may  have  been  its  original  composition  or 
form,  something  in  the  meaning  of  non  made  it  a  stronger  word 
than  ne  for  statement,  and  it  therefore  displaced  ne  from  its  use 
with  the  indie,  leaving  to  it  only  the  function  of  negativing 
expressions  of  will.    And  by  a  well-known  law  of  language,  ne 
then  lost  the  power  of  serving  as  the  negative  of  a  statement. 
Meanwhile,  with  the  general    movement  of  language    toward 
complexity  and  precision,  the  functions  of  the  subjunctive  were 
extended    and    new    functions    were    added.      Some    of  these 
approached  the  indicative  so  closely  that  there  is  a  common 
ground  where  either  mode  may  be  used.    Thus  oporiet  esse  is 
used  as  a  parallel  to  sit  and  aequom  fuit  disperiisse  to  dareni ; 
the  subjunctive  is  the  expression  of  an  obligation,  the  indicative 
is  a  statement  of  it.    As  ne  had  so  narrowed  its  function  that  it 
could  be  used  only  in  prohibitions,  it  was  unfitted  for  use  in  these 
statements  of  obligation,  or  in  hypothetical  or  ideal  statements. 
The  use  of  fion  therefore  indicates  only  that  to  the  Roman  these 
uses  were  more  akin  to  the  statement  than  to  the  prohibition. 
But  a  division  of  all  sentences  into  expressions  of  will  or  state- 
ments of  fact  is  an  extremely  rough  classification ;  between  these 
extremes  lie  many  shades  of  meaning,  and  it  should  not  be 
thought  that  in  the  choice  of  non  or  ne  language  has  been  always 
precise.    There  are  cases  where  9ton  is  used  with  a  subjunctive  of 
will  or  wish  (Cist.  555  utinam  audire  non  queas,  though  gueo  is 
regularly  compounded  with  ne)  occurring  all  through  Lai  * 
ature  (Schmalz*,  §31),  which,  though   they  usually  nej 
single  word,  are  yet  evidence  that  the  fields  of  non  an 
separated  not  by  a  sharp  line,  but  by  a  strip  of  neutral 
Still  less  is  the  use  of  non  evidence  as  to  the  origin  of  a  p 
modal  meaning.    For  non  is  applied  to  the  result  of  the 
it  shows  that  the  shift  of  meaning  has  produced  a  functic 
more  nearly  resembles  the  indie  than  the  subjunctive,  bi 
little  of  the  process  and  nothing  of  the  starting-point. 


Digitized  by 


Google 


392  AMERICAN  JOURNAL    OF  PHILOLOGY. 

can  be  no  doubt  that  in  abi.  ||  abeamf  the  subjunctive  is  one  of 
will,  but  such  a  subjunctive  is  negatived  by  rum  (Capt.  139  ne 
fie.  II  egone  ilium  non  fleam  ?).  It  is  likely  enough  that  the  use 
of  -ne  in  such  questions,  with  a  negative  force  still  somewhat  felt 
in  the  time  of  Plautus,  may  have  prevented  the  use  of  the  nega- 
tive ne  in  the  same  sentence.  The  use  oini  as  the  negative  after 
quid,  quippe,  offers  a  curious  illustration  of  the  persistence  of  ne 
with  a  subjunctive  which  is  as  nearly  potential  as  that  of  any 
quis  question.  All  subjunctives  after  quid  'why'  have  this 
sense,  yet  they  are  negatived  by  ni  (which  is  here  only  another 
form  of  ne^  and  not  conditional)  because  the  use  olni  is  preserved 
by  the  association,  almost  composition,  with  quid,  quippe.  These 
cases  show  that  ne  was  once  the  negative  of  such  questions,  just 
as  ne-scio  shows  that  it  was  the  negative  of  the  indicative. 

While,  therefore,  the  use  of  non  is  evidence,  in  a  general  way, 
of  a  potential  sense  and  indicates  a  likeness  to  the  indie,  it  is 
neither  a  perfect  test,  since  some  potentials  have  ni  and  some 
expressions  of  will  have  nan,  nor  is  it  of  any  value  for  determining 
the  history  of  a  construction.  It  is  a  test  which  cannot  be  trusted 
implicitly  nor  used  mechanically,  as  it  is  not  infrequently  used  in 
syntactical  work.  Least  of  all  can  the  use  of  non,  a  purely  Italic 
particle  and  of  comparatively  late  origin,  indicate  anjrthing  as  to 
the  supposed  connection  between  the  potential  and  an  I.E.  future. 

The  accepted  theory  of  the  Latin  subjunctive,  that  it  is  the 
result  of  the  amalgamation  of  subjunctive  and  optative,  rests 
upon  two  lines  of  argument.  With  the  argument  from  the 
comparative  stand-point  I  have  nothing  to  do,  except  to  say  that 
the  existence  of  the  two  modes  in  Sanskrit  and  Greek  may  be 
perfectly  well  explained  as  a  separate  and  later  development. 
But  the  argument  from  the  Latin,  which,  next  to  these  two 
languages,  has  employed  the  subjunctive  (and  optative)  most 
widely,  falls  within  the  scope  of  this  paper,  and  I  shall  try  to 
show  that  it  has  less  weight  than  is  usually  ascribed  to  it. 

The  optative  forms  in  Plautus  are  these : 

uelim  and  compounds  (74  in  ist  sing.),  78 

sim  and  compounds  (in  all  persons),  97 

duim  and  compounds  (18  in  3d  pers.),  21 

Sigmatic  aorists,  79 

Perfect  tense,  70 

345 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  393 

Counting  all  cases  of  uelim  and  compounds  as  wishes,  there  are 
about  125  wishes  in  this  number,  roughly  one  in  three.  Not 
including  uelim  there  are  about  50  wishes,  one  in  seven.  If  the 
forms  of  the  ist  conj.,^  amtni,  infelicet,  etc.,  were  included,  the 
proportions  would  not  be  greatly  changed.  Of  the  1250  subjunc- 
tive forms  about  1 25  are  wishes,  one  in  ten.  This  preponderance 
of  optative  forms  with  optative  functions  is,  however,  misleading. 
With  a  very  few  exceptions,  the  50  wishes  fall  into  three  classes : 
saluos  sis  in  greetings,  bene  (jnale^foriunaiuni)  sii,  and  curses  or 
blessings  with  di  or  the  name  of  a  god  and  the  verbs  perduity 
perduini^faxini  and  duint.  The  range  of  usage  is  very  narrow. 
If  all  significant  distinctions  were  made,  there  would  not  be  more 
than  six  or  eight  phrases.  Mere  number  of  cases  may  mean 
nothing ;  the  17  cases  of  saluos  sis  indicate  only  the  frequent 
recurreace  of  a  certain  dramatic  situation,  and  are  no  more  signi- 
ficant of  the  extent  of  modal  use  than  is  the  fact  that  of  the  94 
cases  in  ist  plur.,  42  are  of  eamus  and  compounds.  The  wishes 
with  subjunctive  forms  show  something  of  the  same  tendency  to 
run  in  ruts  (jii  perdant  30  times,  di  anient  14  times),  but  the 
variety  of  phrase  is,  on  the  whole,  greater  than  with  the  optative 
forms.  These  considerations  are,  I  think,  sufficient  to  remove 
the  impression  which  the  statistics  at  first  make  and  to  justify  the 
statement  that  there  is  no  real  preponderance  of  optative  forms 
in  wishes,  but  only  a  frequent  repetition  of  a  few  specialized 
phrases. 

From  the  functional  side  the  wish  is  an  expression  of  simple 
desire,  unmixed  or  but  slightly  mixed  with  intention  or  determin- 
ation or  expectation.     But  such  a  definition  is  general  and  does 
not  take  into  account  the  minor  varieties,  in  which  the  optati 
force  is  heightened  or  lowered  according  to  the  nature  of  tl 
wished-for  act  and  the  personality  of  the  actor.    These  elemer 
combine  in  many  and  somewhat  complex  ways,  but  two  or  thr 
main  groups  may  be  noted: — a)  A  simple  expression  of  desi 
in  regard  to  the  action  of  another  person,  such  as  is  expressed  1 
the  English  '  I  wish  that  he  would  come.'    ^)  Desire  mixed  wi 
hope  in  regard  to  the  circumstances  or  health  or  prosperity 
another  person,    c)  Desire  taking  the  form  of  an  indirect  appc 
to  the  gods  to  act.     Further  subdivisions  might  be  made,  if  tl 
nature  of  the  act  were  to  be  more  fully  analyzed.     E.  g.,  tl 

»Cf.  Stolz  in   Mailer's  Handbuch,  II,  §115,  2d  ed.,  p.  378;    Bnigmar 
Gnindr.  II  3,  §946,  p.  1309. 


Digitized  by  VjOOQ IC 


394  AMERICAN  JOURNAL    OF  PHILOLOGY. 

optative  force  is  milder  if  the  desire  is  that  a  certain  person  shall 
come,  stronger  and  more  easily  recognizable  as  a  wish  if  the 
desire  is  that  he  shall  perish  or  suffer  harm.  But  these  minor 
variations  of  thought,  though  they  affect  the  language,  are  rarely 
distinct  enough  to  produce  special  forms  of  wish,  and  may  there- 
fore be  neglected. 

There  are  a  few  cases  in  which  there  is  nothing  in  the  form  to 
distinguish  the  wish  from  ordinary  expressions  of  will.  Thus 
Cas.  6ii  ducas  easque  in  maxumam  malam  crucem,  M.  G.  936 
bene  ambula,  bene  rem  geras,  Cas.  822  uir  te  uestiat,  tu  uirum 
despolies,  Trin.  351  quod  habes  ne  habeas;  these  and  a  few  other 
cases  like  them  lack  the  element  of  determination  which  distin- 
guishes the  will  from  the  wish,  but  the  optative  sense  is  not 
strong,  and  they  illustrate  the  faintness  of  the  line  which  divides 
the  two  fields.  But  for  the  most  part  the  wish  is  marked  by  a 
distinct  form.  The  milder  expressions  of  desire  in  regard  to  the 
action  of  another  person  have  uHnam  or  ueliniy  which  are  also 
employed  (the  latter  more  often  than  the  former)  in  curses, 
usually  comic  and  elaborate  curses.  Wishes  for  the  health  or 
prosperity  (or  the  reverse)  of  another  person  are  expressed  by 
the  impersonal  forms,  bene  (male)  sit,  marked  by  the  verb  and 
adverb.  The  phrases  of  greeting,  saluos  sis,  ualeas,  were  origin- 
ally of  this  sort,  but  became  formulaic  and  lost  something  of 
their  meaning.  Asseverations,  iia  me  di  ament,  are  still  more 
distinctly  differentiated  by  iia^  and  when  the  gods  are  mentioned, 
as  in  these  forms  and  in  di  te  perduint  (J>erdant^,  the  fullest 
optative  meaning  is  brought  out. 

Now,  in  all  these  phrases  the  forces  which  give  the  optative 
sense  are  perfectly  clear.  Except  in  some  of  the  cases  with  uelim> 
and  utinam,  the  action  contemplated  is  one  which  it  is  out  of  the 
power  of  a  human  actor  to  perform  and  out  of  the  power  of  the 
speaker  to  influence.  The  element  of  determination  can  have 
nothing  to  do  with  the  action  or  state  of  ualere,  saluos  esse,  bene 
esse.  In  other  words,  these  verbs,  if  they  are  employed  at  all 
in  expressions  of  desire,  must  be  optative;  the  verb-meaning 
excludes  any  other  sense.  So  also  the  actor  is,  in  all  the  most 
distinct  expressions  of  wish,  either  left  out  of  view  entirely,  as  in 
the  impersonals,  or  is  superhuman  and  so  beyond  the  reach  of 
human  determination.  These  two  forces,  the  verb-meaning  and 
the  actor  (person  and'  number)  absolutely  fix  the  meaning  of  di 
te  perdant,  so  that  it  very  rarely  takes  utinam.     But  in  the  more 


Digitized  by  VjOOQ IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  395 

elaborate  or  less  formulaic  optations,  where  the  force  of  the  verb- 
meaning  is  less  clear,  and  especially  in  the  2d  and  3d  sing.,  where 
a  definite  person  is  the  subject  and,  in  a  sense,  the  actor,  some 
specific  sign  of  the  wish  is  needed,  and  here  vtinam  and  uelim 
are  employed. 

The  optative  function,  therefore — ^that  is,  the  capacity  to  express 
a  wish — is  not  so  distinct  as  to  require  us  to  explain  it  by  refer- 
ring it  to  an  I.E.  optative  function.  Any  mode  which  expresses 
desire  would  of  necessity  express  optation  also,  if  the  desire  was 
that  the  gods  should  bring  a  certain  man  to  ruin.  To  explain  di 
ieperdant  and  ualeas  by  saying  that  they  got  the  optative  func- 
tion from  di  ieperduini  and  saluos  sis,  and  that  these  acquired  it 
at  some  remote  time  in  some  unexplained  way,  is  to  turn  away 
from  simple  forces,  working  under  our  eyes,  to  a  vague  hypothesis 
which,  after  all,  explains  nothing.  The  forces  which  can  give  an 
optative  sense  to  saiu€  or  di  te  amabuni  will  explain  all  optative 
subjunctives.^ 

The  state  of  things  which  we  find  existing  in  Plautus  as  to 
optative  forms  and  functions  is  best  explained  not  by  the  hypo- 
thesis that  the  Italic  had  an  optative  mode,  a  system  of  special 
terminations  applicable  to  any  verb-stem  and  having  as  their 
most  distinct  function  the  expression  of  a  wish,  but  by  the 
hypothesis  that  the  two  modal  formations  were  along  the  Italic 
line  of  descent  never  clearly  differentiated.  It  is  unlikely  that 
both  forms  were  in  general  use  with  the  same  verb-stem,  and 
there  is  no  evidence  to  show  that  they  had  that  universality  and 
system  which  would  justify  the  use  of  the  term  mode.*  Out  of 
this  undifferentiated  or  but  partially  differentiated  modal  material, 
the  growing  consciousness  of  the  wish,  as  distinct  from  will, 
working  through  the  person  and  number  and  the  verb-meaning, 
produced  the  optative  forms  of  sentence.  In  general,  these  were 
not  sufficiently  specific  without  the  addition  of  specializing  words, 
utinam,  uelim,  but  certain  forms,  still  further  separated  from  the 
ordinary  uses  of  the  subjunctive  by  being  employed  as  greetings, 

^  Compare  also  the  future  with  the  subjunctive  in  Pers.  160  Sagaristio,  di 
ament  te.  |  o  Toxile,  dabunt  di  quae  exoptes ;  Capt.  877  ff.  ita  me  amabit 
sancta  Saturitas,  Hegio,  itaque  suo  me  semper  condecoret  cognomine,  ut  ego 
uidi. 

'Cf.  Streitberg  in  Paul  and  Braune's  Beitr.  15,  p.  116:  "damals  [in  idg. 
urzeit]  existierten  (iberhaupt  keine  *  tempora,*  d.  h.  keine  formalen  kategorien, 
deren  ursprtlngliche  function  es  war,  zur  bezeichnung  der  relativen  zeitstufen 
zu  dienen." 


Digitized  by^ 


396  AMERICAN  JOURNAL    OF  PHILOLOGY. 

asseverations,  curses  or  blessings,  became  idiomatic  and  needed 
no  distinguishing  particle. 

The  potential  and  the  optative  uses  are  not  parallel,  though 
they  are  in  some  respects  similar.  Both  are  in  part  due  to  the 
influence  of  person  and  number  (the  subject)  and  of  verb- 
meaning,  but  these  influences  are  stronger  in  the  optative  than  in 
the  potential ;  in«4he  latter  they  produce  conditions  favorable  to 
the  potential,  rather  than  true  potential  uses.  These  appear  only 
when  other  forces  are  brought  into  operation.  The  optative  is 
a  specialized  use,  the  result  of  convergence  and  of  increasing^ 
isolation,  whether  it  be  in  phrases  like  saluos  sis,  di  te  perdant,  or 
in  uses  which  require  a  special  optative  particle,  iiiinam.  The 
close  resemblance  to  ordinary  expressions  of  desire  is  most 
apparent  in  the  wishes  of  a  general  character  and  content,  such 
as  take  uiinam,  but  it  is  also  plain  in  the  more  specialized  and 
formulaic  wishes,  and  there  is  no  form  of  optation  in  which  this 
connection  cannot  be  clearly  seen ;  the  element  of  determination 
is  lessened  or  dropped  out,  but  the  element  of  desire  is  intensified. 
In  the  potential  uses  the  tendency  toward  isolation,  which  appears 
in  non  meream,  quid  ego  nunc  faciam?  or  deum  uirtute  dicam^ 
is  one  which  increases  the  element  of  determination  and  the 
meanings  allied  to  the  future  and  which  lessens  the  element  of 
desire.  But  desire  in  some  form  is  the  most  common  meaning 
of  the  subjunctive,  the  meaning  which  a  subjunctive  form  at  once 
suggests,  and  when  it  is  weakened  the  form  is  left  to  a  consid- 
erable extent  meaningless,  emptied  of  its  normal  meaning.  This 
is  one  of  the  processes  by  which  a  word  or  a  syntactical  group  is 
prepared  to  assume  new  meanings,  new  functions.  The  form 
meream,  existing  through  the  analogy  of  other  verbs  of  the  2d 
conjugation,  and  in  part  excluded  by  its  meaning  from  the 
expression  of  desire,  is  speech-material  ready  to  take  on  any  new 
function,  not  too  far  removed  from  the  old.  The  suggestion  of 
new  meaning  comes  from  the  context,  the  preceding  thought. 
For  the  subjunctive  is  never  potential — differing  in  this  altogether 
from  the  optative  uses — when  it  is  alone.  A  form  like  ualeas, 
saluos  sis,  pereat  (the  last  not  in  Plautus)  can  convey  a  wish 
without  context,  except  as  the  attendant  circumstances  always 
supply  a  context,  or  a  brief  phrase  like  di  te perdant  with  almost 
no  context;  but  this  is  rarely,  if  ever,  true  of  the  potential. 
faciam  alone,  without  the  help  of  the  interrogative,  is  not  poten- 
tial;   meream  alone  might  have  somewhat  more  of  potential 


-  Digitized-by  VjOOQIC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  397 

suggestion,  because  the  suggestion  comes  largely  from  the  verb- 
meaning,  but  sii^  dicai,  postules,  duceret  do  not  convey  potential 
meaning  except  in  a  certain  context.  It  is  only  when  these  forms, 
unfitted  or  only  partially  fitted  for  the  expression  of  desire,  are 
used  with  phrases  which  set  a  hypothetical  tone — the  interroga- 
tives,  deum  uirtute,  una  opera^  a  subjunctive  clause,  a  formal 
protasis — that  they  take  on  new  meaning  from  their  surroundings, 
and  become  most  widely  separated  from  the  usual  subjunctive 
functions.  To  a  slight  degree  an  optative  phrase,  ualeas,  saluos 
sts,  may  also  acquire  new  meaning  from  its  use  in  greetings,  but 
for  the  most  part  it  is  correct  to  say  that  the  optative  function  is 
only  an  intensification  and  isolation  of  a  meaning  inherent  in  the 
subjunctive,  while  the  potential  force  is  an  acquired  function,  not 
inherent  in  the  mode,  but  rather,  in  its  extreme  development, 
showing  but  slight  trace  of  its  connection  with  the  mode  of 
determination  and  desire. 

^The  potential  and  the  optative  uses  are  only  the  most  striking 
illustrations  of  the  process  which  went  on  over  the  whole  range 
of  modal  expression,  and  which  in  the  end  produced  the  many 
varieties  of  subjunctive  usage  to  be  found  in  any  author  or  at  any 
period.  Looked  at  from  the  functional  side,  this  process,  if  its 
steps  could  be  traced,  would  be  one  of  constant  progress  from 
vague  thinking  toward  precise  thinking,  from  undiscriminating 
and  vague  desire  with  reference  to  all  kinds  of  action  under  all 
varieties  of  circumstance  toward  differentiated  will,  entreaty, 
exhortation,  command,  permission,  direction,  advice,  and  so  on. 
In  other  words,  we  must  suppose  that,  though  man  in  the  primi- 
tive stage  entreated,  exhorted,  commanded,  and  though  these 
emotions  were  really  different,  the  difference .  was  not  strong 
enough  to  find  expression  in  language.  But  it  is  not  to  be 
supposed  that  this  movement  toward  differentiation,  though 
constant,  was  regular.  The  need  of  expression  varied,  and  was 
felt  in  some  directions  (e.  g.  perhaps  in  curses  or  in  direct  com- 
mands in  2d  sing.)  before  it  was  felt  in  others.  The  means  of 
expression,  also,  would  lie  near  at  hand  for  such  a  use  as  ist  plur., 

^  The  paragraphs  which  follow  I  present  as  a  hypothesis  merely ;  to  others 
they  may  seem  no  more  than  speculation.  And  the  same  thing  may  be  said 
of  some  of  the  preceding  remarks  upon  the  potential  and  the  optative,  though 
I  venture  to  think  that  they  rest  more  firmly  upon  direct  inference  from  the 
facts.  I  have  also  examined  somewhat  carefully  all  the  corresponding  uses  in 
a  dozen  or  more  of  other  writers,  Cato.  Terence,  Varro,  Lucretius,  Catullus, 
Vergil,  Iforace,  Caesar,  Pliny,  and  some  others,  but  not  Cicero. 


Digitized  by  VjOOQ IC 


398  AMERICAN  JOURNAL    OF  PHILOLOGY. 

but  for  other  uses  (deliberative  questions,  una  opera  posiuUs) 
would  have  to  be  shaped  by  long  use.    Looked  at  from  the 
formal  side,  the  process  was  equally  irregular.    The  forces  of 
analogy  and  assimilation,  by  which  the  modal  forms  were  slowly 
worked  into  a  systematic  mode,  afforded  at  first  and  for  a  long^ 
time  an  irregular  and  unsystematized  speech-material.     Some 
verbs  had  both  subjunctive  and  optative  forms,  some  had  only  one, 
some  had  neither,  and  we  can  hardly  doubt  that  still  other  modal 
formations  existed  which  have  not  been  preserved.*    Nor  is  it  to 
be  supposed  that  all  verbs  began  to  be  used  in  subjunctive  forms 
at  the  same  instant.     I  have  said  above  that  some  verbs  were 
peculiarly  fitted  by  their  meaning  for  use  in  expressions  of  will, 
and  that  other  verbs  were  excluded  by  their  meaning  or  by  tense 
or  voice  or  person  and  number  from  the  expression  of  the  simpler 
forms  of  will.    That,  of  course,  does  not  mean  that  such  forms  as 
dicaptt  meream^  caperes^  ecficiatur  existed  but  were  not  used,  but 
only  that  they  may  be  said  to  have  had  a  potential  existence  from 
the  time  when  similar  forms  came  into  use ;  an  actual  existence, 
as  a  part  of  the  language,  they  did  not  have  until  the  modified 
and  specialized  kinds  of  will,  which  are  all  that  they  are  capable 
of  expressing,  came  to  be  felt,    mereo  had  no  pres.  subj.  ist  sing, 
until  the  idea  expressed  in  nan  meream  (or  some  similar  idea) 
called  for  expression;  the  form  dicam  was  very  rare  until  the 
idea  of  futurity  or  something  like  it  was  felt,     uelim  is  a  striking 
illustration  of  this.     Unless  we  suppose  that  it  meant  'I  should 
wish' — a  hypothesis  which  really  rests  upon  classical  and  later 
usage  and  is  absolutely  contradicted  by  the  usage  of  Plautus  and, 
indeed,  by  much  of  the  later,  especially  the  colloquial,  usage — 
uelim  had  no  proper  meaning,  and  uelle  had  therefore  no  modal 
form  in  the  ist  sing,  until  the  need  arose  of  distinguishing,  e.  g. 
ueniai  the  will  from  ueniat  the  wish.     Then  uolo  ueniat  expressed 
the  one  and  ueniat  uelim  the  other,  the  form  uelim  coming  into 
existence  through  the  analogy  of  similar  forms  under  the  influence 
of  the  optative  force  of  ueniat    Such  independent  meaning  as 
uelim  has,  it  acquired  from  this  association  with  an  optation  and 
carried  over  into  the  uses  with  infin.  and  ptc.  and  adjective ;  in 
fact,  the  other  persons,  uelis,  uelit^  are  rare  in  ^ny  really  inde- 
pendent use. 

*Quintil.  I  7,  23  quid?  non  Cato  Censorius  *  dicam*  et  'faciam*  *diccin'  et 
*  faciem'  scripsit  eundemque  in  ceteris,  quae  similiter  cadunt,  modum  tenuit? 


Di( 


igilized  by  Vj  O  O^  IC 


THE  SUBJUNCTIVE  IN  PLAUTUS.  399 

Thus  upon  the  formal  side,  as  well  as  upon  the  functional,  the 
spread  of  the  mode  was  irregular,  and  the  formation  of  the 
specialized  usages,  which,  taken  together,  are  the  basis  for  our 
generalizations  as  to  the  subjunctive,  was  for  both  reasons,  the 
formal  and  the  functional,  irregular  and  unsystematic.  There 
was  no  formation  of  broad  types,  like  the  potential,  but  only  of 
restricted  usages,  and  the  analogies  which  determined  the  direc- 
tion of  the  process,  the  lines  of  cause  and  effect,  the  retarding 
influences,  must  all  be  sought  in  the  specialized  usages.  It  is 
true  that  such  usages  may  sometimes  be  grouped  together  on  the 
basis  of  a  functional  resemblance,  but  the  separate  members  of 
such  a  group  have  no  organic  connection  until  they  begin  to 
influence  each  other;  that  is,  until  the  resemblance  in  function 
becomes  clearer  and  stronger  than  the  similarity  in  verb-form,  in 
sentence-form  or  in  verb-meaning.  Such  a  stage  was  most  cer- 
tainly not  reached  by  the  various  members  of  the  potential  group 
— the  type  which  is  most  frequently  treated  as  having  an  actual 
existence — in  the  I.E.  period ;  if  it  was  beginning  at  all  in  the 
time  of  Plautus,  its  range  was  still  very  narrow.  The  use  of  the 
term  Grundbegriff  to  describe  the  sphere  of  application  of  a 
group  of  syntactical  forms  ^  is  unfortunate,  if  it  encourages  the 
belief  that  the  members  of  such  a  group  exerted  a  common 
influence  in  addition  to  the  influence  of  each  individual  usage. 
Common  influence  can  be  exerted  only  when  the  usages  have 
become  bound  together  by  mutual  analogies. 

The  formation  of  specialized  usages  in  which  the  tense  or  the 
verb-meaning  or  some  other  force  influenced  the  meaning  of  the 
mode  either  intensified  the  subjunctive  sense  or  weakened  it; 
where  will  and  desire  were  weakened,  the  subjunctive  form  took 
on,  in  part,  new  meaning.  The  result  was  an  extension  of  the 
field  of  the  mode,  on  the  one  side  toward  the  most  explicit 
expression  of  will,  the  impv.,  on  the  other  side  toward  the  mode 
of  statement,  the  indicative.  In  the  one  direction  the  subjunctive 
was  extended  to  uses  (e.  g.  some  kinds  of  prohibition)  in  which  it 
is  impossible  to  detect  any  difference  between  its  force  and  that 
of  the  impv.;  on  the  other  side  it  was  extended  until  it  reached 
ideas  of  expectation,  determination,  propriety,  necessity,  obliga- 
tion, which  could  be  more  precisely  expressed  by  the  future,  by 
modal  verbs  {pportet^  debet,  uelle.posse),  by  phrases  like  aequom 

^  It  is  nsed  in  somewhat  this  way  by  DelbrUck  in  Brugmann's  Grundr.  Ill  i, 
p.  81,  and  by  Bragmann,  Indog.  Forsch.  V,  p.  93,  n.  2. 


Digitized  by  VjOOQ IC 


400  AMERICAN  JOURNAL    OF  PHILOLOGY. 

esty  by  the  periphrastic  forms  or  by  verbal  nouns.  For  the  precise 
determination  of  the  meaning  of  the  mode — that  is,  of  the  range 
of  its  application — a  study  of  the  neutral  territory  in  which  these 
phrases  are  employed  as  parallels  to  subjunctive  forms  is  to  be 
desired.^  Within  the  historical  period  the  competition  between 
the  mode  and  the  other  expressions  of  modality  went  on  until  the 
modal  forms  were  to  a  considerable  extent  driven  out  by  the 
analytic  forms,  and  passed  over  into  the  subordinate  clause  or 
survived  in  the  Romance  languages  with  changed  functions.  In 
Plautus  the  competition  is  just  beginning. 

An  actual  historical  connection  between  different  usages  can  be 
established  by  clear  evidence  in  only  a  few  instances.  Thus  the 
expansion  oiquidfaciamf  in  one  direction  into  quid  uis  facianif 
and  in  the  other  into  quid  ego  nunc  faciamf  is  clear,  and  I  have 
attempted  to  show  above  how  the  use  of  the  indef.  2d  sing,  with 
verbs  of  mental  action  is  an  off-shoot  of  the  ordinary  hypothetical 
use.  In  such  cases  the  connection  shows  which  usage  is  the 
older.  With  somewhat  less  of  probability,  it  is  possible  to  con- 
jecture that  the  simple  and  direct  expressions  of  desire  preceded 
the  more  complex.  The  hortatory  ist  plur.  must  have  begun, 
very  much  as  it  appears  in  Plautus,  as  soon  as  the  analogies  of 
other  subjunctive  forms  led  to  the  use  of  the  ist  plur.  ending.  It 
is  still  in  Plautus  used  only  with  verbs  of  physical  action  and  with 
a  simple  meaning,  except  where  the  addition  of  tu  or  a  vocative 
suggests  a  jussive  force ;  it  shows  no  signs  of  the  rhetorical  uses, 
uideamus,  transeamus^  which  are  found  in  Cicero.  It  can  hardly 
be  doubted  that  such  a  use  is  earlier  than  a  phrase  like  qui  ego 
isiuc  credamf  or  than  the  entirely  distinct  uses  of  the  three 
persons,  faxim,  faxis^  faxit,  which  can  have  come  about  only 
by  a  long  process  of  shift  of  meaning.  A  still  slighter  degree  of 
probability  attaches  to  attempts  to  prove  relationship  and  com- 
parative age  by  reasoning  based  upon  English  auxiliary  verbs, 
would,  should,  can,  will,  shall]  such  reasoning  suggests  specula- 
tion and  may  thus  become  fruitful,  but  it  does  not  of  itself  estab- 
lish facts.  The  difficulty  of  proving  relationship  between  the 
different  uses  of  the  subjunctive  is  not  due,  however,  to  insuffi- 
cient data  for  the  early  periods  or  to  inaccurate  observation.     It 

^  Some  suggestions  bearing  npon  this  kind  of  definition  of  the  mode  may  be 
found  in  J.  Lattmann,  Die  deutschen  Modalit&tsverba,  Progr.,  Clausthal,  1879; 
"iohviXi^oTiytUbere, posse  sim.,  Upsala,  1868  ;  Neumann,  De  futuri . . .  usu,  Breslao, 

1888. 


Digitized  by 


Google 


^ 


THE  SUBJUNCTIVE  IN  PLAUTUS.  4OI 

is  rather  evidence  that  but  few  such  relationships  existed,  except 
as  all  uses  of  like  forms  are  related,  and  that  the  specialized 
usages  grew  up  separately  through  the  complex  working  of  a 
number  of  different  forces. 

It  is  not  less  difficult  to  determine  with  any  considerable  degree 
of  probability  which  usages  of  the  Latin  mode  go  back  to  the 
I.E.  stage.  That  modal  forms  then  existed  is  clear  enough,  and 
it  is  plain  that  the  forms  were  used  to  express  some  kind  of  desire 
or  will ;  just  how  precise  this  statement  may  be  made  is  far  less 
clear,  and  the  value  of  it  to  the  student  of  a  single  language  is 
often  over-estimated.  For  to  one  who  is  endeavoring  to  under- 
stand the  phenomena  of  a  single  language,  Greek  or  German  or 
Latin,  the  course  of  that  language  from  the  beginning  of  speech 
must  be  regarded  as  continuous ;  as  to  the  student  of  the  Romance 
languages,  French  or  Portuguese  presents  an  unbroken  line  of 
development  from  Plautus  to  the  present  time.  The  certaifity,  if 
it  could  be  reached,  that  a  particular  special  usage  dated  from  the 
I.E.  stage  would  make  it  possible  to  reason  with  somewhat  greater 
exactness  from  the  forms  which  that  usage  took  in  other  languages, 
but  the  gain  would  not  be  great.  The  phrase  "  of  I.E.  origin," 
which  one  not  infrequently  meets  in  syntactical  work,  is  not  in 
fact  an  explanation,  but  a  very  vague  date.  But  it  is  not  the 
when  that  is  of  primary  importance  in  syntax :  it  is  the  haw  and, 
if  possible,  the  why.  It  is  the  process  of  change,  the  laws,  the 
forces,  the  causes,  that  historical  syntax  must  follow  out ;  the  date 
is  only  a  means  to  the  accomplishment  of  this  task. 

I  have  not  cared  to  make  in  this  paper  what  is  called  a  ''  critical 
examination  "  of  the  theories  of  the  mode  current  in  America  and 
to  some  extent  in  England.  The  differences  between  them  and 
the  view  which  I  have  been  trying  to  present  will,  I  hope,  be 
evident,  and  there  is  no  work  in  which  philologists  engage  more 
useless,  in  my  opinion,  than  negative  criticism.  If  a  theory  is 
wrong,  it  can  be  disproved  only  by  supplying  a  better  theory. 
This  paper  is  an  attempt  to  suggest  a  new  and,  possibly,  more 
fruitful  method  of  studying  the  subjunctive. 

E.  P.  Morris. 


Digitized  by^ 


II.— THE  USE  OF  ENIM  IN  PLAUTUS  AND  TERENCE. 

In  the  winter  of  1885-86,  Goetz,  who  was  busy  with  his  new 
edition  of  the  Bacchides,  suggested  to  me,  then  a  student  at  Jena, 
the  investigation  of  the  use  ofenim  in  the  earlier  language.  The 
following  pages  present  the  results  of  that  study,  delayed  and 
postponed  for  various  causes  these  eleven  years.  While  the 
conclusions  may  not  be  all  that  were  hoped  at  the  time  the 
investigation  was  begun,  from  over  twenty  readings  and  compari- 
sons of  the  entire  text  of  Plautus  and  Terence,  it  is  felt  that  the 
classification  is  more  thorough  and  systematic  than  has  been 
previously  attempted.  While  my  views  on  minor  points  have 
sometimes  changed,  my  opinions  and  convictions  on  the  most 
important  usages  have  been  strengthened  by  successive  compari- 
sons of  the  text  Of  the  numerous  conjectures  that  would  intro- 
duce enim  into  the  text,  only  the  more  plausible  have  been 
noticed.  It  has  not  been  deemed  advisable  to  cumber  the  page 
with  improbable  emendations.  Only  disputed  or  typical  passages 
have  been  quoted  in  full. 

Nearly  thirty  years  ago,  Ramsay,  in  his  edition  of  the  Mostel- 
laria,^  stated :  *'we  maintain  that  in  the  earlier  writers  entmvero 
always  signifies  ' for  in  truth'  as  emm  always  signifies  ' for/ and 
that  both  are  uniformly  employed  to  introduce  an  explanation.*' 
Eleven  years  later  Langen,'  who  devoted  considerable  space  to 
the  discussion  and  gave  the  most  complete  classification  hitherto 
attempted,  asserted  with  equal  positiveness  (p.  262) :  "  Ich  glaube 
behaupten  zu  diirfen  enim  ist  bei  Plautus  ausschliesslich  Betheuer- 
ungspartikel,  es  wird  von  ihm  iiberhaupt  nicht  zur  Begriindung 
eines  vorhergehenden  Gedankens  gebraucht."  This  latter  view 
became  at  once  the  prevalent  one  among  Plautine  students, 
although  a  number  of  prominent  editors  and  critics  have  taken 
more  or  less  exception  to  its  sweeping  conclusions.  There  is,  it 
would  seem,  a  position  between  these  two  extremes,  which  we  are 
warranted  in  taking. 

^  London,  1869,  p.  206. 

^  Beitrage  zur  Kritik  und  Erklarung  des  Plautus.     Leipzig,  1880,  pp.  261-71. 


Digitized  by  VjOOQ IC 


USE  OF  ENIM  IN  PLAUTUS  AND  TERENCE,  403 

I. —  The  Position  of  Enim, 

The  position  of  enim  may  best  be  considered  under  two  heads : 
(i)  the  position  of  enim  in  the  sentence;  (2)  the  position  of  enim 
in  relation  to  other  words. 

(i)  TAe  position  of  enim  in  the  sentence. — In  Plautus  enim 
stands  as  the  first  word  in  the  sentence  in  the  following  passages : 
Aul.  500/  Capt.  592,  Cas.  890,  Cist.  777,  Mil.  429,  1018,  Most. 
1 144,  Pers.  236,  319,  612,  Trin.  1134.  In  Epid.  701  I  should  read 
enim  istaec  captiost,  as  I  see  no  good  reason  for  discarding  the 
manuscript  reading  for  Brix's  conjecture,  em  istaec  captiost^ 
adopted  by  Gotz.  In  Men.  846  I  read  enim  haereo^  with  Brix. 
This,  suggested  by  Ussing  in  his  note  to  Aul.  492,  is  much  to  be 
preferred  to  enim  periculum  est,  which  he  adopts  in  his  later 
edition.  SchoU  transfers  the  words  from  Menaechmus  to  Matrona, 
and  so  is  compelled  to  change  haereo  to  censeo.  In  Trin.  806 
enim-  is  to  be  preferred  to  the  manuscript  at  enim. 

The  two  most  probable  conjectures  that  would  give  enim  first 
place  in  the  sentence  are  Lachmann*s  enim  verbis  prohus  for  in 
verbis  probus  in  Amph.  838,  and  Ribbeck's  ingenious  emendation 
of  Mil.  1319,  which  will  be  discussed  more  fully  below. 

Enim  vero  stands  at  the  beginning  of  the  sentence  in  the 
following  passages:  Amph.  723,  771,  Asin.  688,  Capt.  628,  Cas. 
475,  728,  Cist.  519,  Cure.  175,  608,  Men.  860,  1075,  Merc.  739, 
Pers.  349,  Poen.  296,  435  (where  its  parenthetical  position  really 
gives  it  first  place),  Rud.  1003,  Stich.  398,  616,  Trin.  958,  989 ; 
probably  in  Capt.  22  and  Poen.  280.' 

Enim  is  first*  in  four  passages  in  Terence:  Ad.  168,  H.  T.  72, 
Hec.  238,  Phorm.  983,  and  enim  vero  in  eight :  And.  91,  206,  H. 
T.  320,  1045,  Hec.  673,  Phorm.  465,  937,  1036. 

Enim  is  found  in  the  second  place  (when  not  joined  with  other 
particles)  21  times  in  Plautus  and  7  in  Terence.     In  only  one 

^  The  citations  for  Plautus  (both  plays  and  fragments)  are  made  according 
to  the  edition  of  GOtz,  LOwe  and  SchoU ;  for  Terence,  that  of  Dziatzko. 

'  Compare  Capt.  532,  Merc.  739. 

'Langen's  statement  (p.  263):  **Weit  haufiger  (am  Anfang  des  Satzes)  ist 
aber  die  Verst&rkung  durch  vero^  mindestens  dreissigmal,"  is  incorrect 

♦This  position  oi enim  in  the  earlier  language  is  not  noted  by  the  majority 
of  grammars  in  general  use.  Roby,  II,  p.  22 ;  Harkness,  §569,  III ;  Allen 
and  Greenough,  §§156,  R,  345,  b,  and  Bennett,  §345,  are  all  guilty  of  the  same 
omission.  Gildersleeve-Lodge,  §498,  n.  i,  recognizes  the  usage.  A  number 
of  equally  dogmatic  and  incorrect  additional  statements  could  easily  be  secured. 


Digitized  by  VjOOQ IC 


404  AMERICAN  JOURNAL    OF  PHILOLOGY. 

passage  does  it  occupy  third  place,  Cas.  525  em,  nunc  enim  etc, 
where  its  position  can  be  accounted  for  by  its  close  connection 
with  nunc. 

(ii)  The  position  of  enim  with  reference  to  certain  words. — An 
examination  of  the  passages  shows  that  enim^  both  alone  and  in 
compounds,  is  often  found  associated  with  certain  words.  The 
following  collocations  are  worthy  of  note: 

(i)  Enim  with  pronouns. — cu  With  personal  pronouns:  enitn^ 
ego,  Cas.  280,  Merc.  251,  Mil.  809,  Most.  888,  926,  Poen.  604; 
mihi,  Aul.  500,  Amph.  733,  Cas.  366;  me,  Trin.  1134;  /w,  Capt. 
568 ;  enim  vero  ego,  Capt.  534,  Pseud.  979,  Trin.  958 ;  certo  enim, 
ego,  Aul.  811;  mihi,  Stich.  88;  at  enim  mihi,  Stich.  738;  nos^ 
Stich.  129;  tu^  Epid.  94;  quia  enim  me,  Merc.  248,  True.  266; 
te,  Amph.  606,  Pers.  592 ;  certe  enim  tu,  Asin.  614 ;  nempe  enim. 
tu,  Trin.  60 ;  verum  enim  tu.  Mil.  293 ;  nan  enim  tu,  Rud.  989. 

b.  With  demonstrative  pronouns:  enim  id,  Men.  163,  Ad.  730; 
enimvero  id.  And.  848 ;  verum  enim  vero  id.  Ad.  255 ;  at  enim  id^ 
Bacch.  793,  1080 ;  quia  enim  id.  Most.  1098 ;  enim  ilia,  Phorm. 
113;  illoc.  Men.  249;  etenim  ille,  Amph.  266;  enim  vero  ille, 
Amph.  771 ;  illud^  Men.  860;  at  enim  ille,  Cist.  739,  Men.  790; 
illi,  Pers.  569;  quia  enim  ille.  Cure.  667;  ne  enim  illi.  Most. 
1095;  non  enim  ilium,  Rud.  922;  neque  enim  illi,  Trin.  585; 
enim  ipsi,  Cas.  323 ;  etenim  ipsus.  And.  442 ;  enim  istaec,  Epid. 
701,  Most.  1 144;  ai  enim  istaec,  Eun.  381;  isioc,  H.  T.  699; 
enim  hie,  Bacch.  457;  at  enim.  hoc,  Poen.  1197;  ^^^  enim  haec^ 
Most.  827. 

c.  With  relative  or  interrogative  pronouns:  cerio  enim  quod^ 
Poen.  1 182;  at  enim  quod,  Pers.  832;  quia  enim  qui,  Hec.  311 ; 
verum  enim  vero  qui,  Poen.  874;  quid  enim,  Amph.  694. 

(2)  Enim  with  adverbs, — It  is  also  joined  with  many  adverbs, 
especially  those  of  time.  Thus,  nunc  enim,  Asin.  598,  Cas.  525, 
Epid.  162,  648  (non  enim  nunc),  Capt.  534  (nunc  enim  vero). 
And.  823  (immo  enim  nunc) ;  enim  iam,  Cas.  890 ;  verum  enim 
quando,  Ad.  201. 

Four  examples  are  found  of  the  combination  ita  enim  vero^ 
Amph.  410,  Asin.  339,  Cist.  519  (enim  vero  ita).  Most.  920. 

(3)  Enim  with  negative  a?id  final  particles. — It  is  joined  with 
negative  and  final  particles :  enim  non.  Cist.  777,  Pers,  236,  Ad. 

^  These  lists  are  arranged  for  the  sake  of  brevity  with  enim  first,  even  if  it 
be  postpositive. 


Digitized  by  VjOOQ IC 


USE  OF  ENIM  IN  PLAUTUS  AND  TERENCE,  4O5 

168,  Capt  628  (enim  vero  non),  Merc.  395,  Mil.  1139,  Pseud.  325 
(quia  enim  non);  haud eniniy  Capt.  592;  enim  Tie,  Mil.  429,  Cist. 
235,  Most.  922  (at  enim  ne). 

(4)  Enim  with  the  first  person  of  verbs. — In  mimerous  cases  it 
is  joined  with  the  first  person  of  verbs.  Examples  are :  aio  enim 
vero,  Amph.  344,  Pers.  185 ;  ego  enim  dicam,  Cas.  372. 

II. —  The  Force  of  Simple  Enim. 

(i)  Enim  with  corroborative  force. — In  both  Plautus  and  Ter- 
ence enim  has  in  the  majority  of  cases  an  affirmative  or  corrobo- 
rative force,  corresponding  to  our  'indeed,  certainly,  to  be  sure,' 
and  the  German  'fiirwahr,  wahrhaftig.' 

1.  With  this  corroborative  force  enim  occupies  the  first  place 
in  the  sentence  in  Aul.  500,  Capt.  592,  Cas.  890,  Cist.  777,  Epid. 
701,  Men.  846,  Mil.  1018,  Most.  1144,  Pers.  236,  319,  612,  Trin. 
806,  1 134,  H.  T.  72,  Hec.  238,  Phorm.  983.  I  do  not  find  any 
passage,  resting  on  manuscript  authority,  uhere  enim  in  the  first 
place  has  any  other  force.  Lachmann's  conjecture,  Amph.  838 
enim  (MSS  /«,  Uss.  Id  tu)  verbis  probas,  has  the  same  meaning, 
with  a  tinge  of  irony. 

2.  It  is  similarly  employed  in  the  second  place  in  the  sentence 
with  no  unusual  emphasis:  Amph.  333,  Asin.  598,  Bacch.  457, 
Cas.  525,  Epid.  648,  Men.  251,  Merc.  251,  Phorm.  113. 

3.  In  answers  it  is  frequently  employed  with  the  same  signifi- 
cation :  Cas.  279-80  Ch.  Te  uxor  dicebat  tua  Me  vocare.  Lys. 
Ego  enim  vocari  iussi,  323,  366,*  372,  Men.  162,  Mil.  429  {enim 
first),  810,  Most.  888,  Pers.  670,  Poen.  387,  Ad.  168  (enim  first), 

730- 

Nil  is  sometimes  joined  with  enim  in  the  reply :  Bacch.  701-2 
Pist.  Nunc  quid  nos  vis  facere?  Chrys.  Nil  enim  (Nihil  Uss., 
enim  nihil  R.,  Lang.)  nisi  ut  ametis  impero;  Most.  551,  Ad.  656, 
921,  Hec.  850. 

An  isolated  example  that  may  be  quoted  here  is  H.  T.  317  CI. 
Quid  ilia  facias?     Sy.  At  enim.     CI.  Quid  enim? 

4.  The  corroborative  force  sometimes  takes  an  ironical  turn 
like  vero  or  the  German  freilich :  Capt.  568  Tu  enim  repertu's, 
Philocratem  qui  superes  veriverbio ;  Amph.  836-8  Ale.  Quae  non 

'  SchSll's  arrangement  and  panctuation  of  the  line  removed  the  objections 
to  the  rare  and  doubtful  use  of  enim  in  questions.  One  other  case  will  be 
considered  below. 


Digitized  by  VjOOQ IC 


406  AMERICAN  JOURNAL    OF  PHILOLOGY. 

deliquit,  decet  Audacem  esse,  confidenter  pro  se  et  proterve  loquL 
Amph.  Satis  audacter.  Ale.  Ut  pudicam  decet.  Amph.  Enim 
verbis  probas. 

5.  In  Amph.  •694  is  found  the  only  example  of  a  usage  sa 
familiar  in  Ciceronian  Latin,  quid  enim  in  Quid  enim  censes  ?  te 
ut  deludam  coptra  lusorem  meum  ?  Langen  (p.  267)  denies  its 
genuineness,  and  declares :  "  Plautus  hat  gewiss  quidnam  censes^ 
geschrieben."  While  there  is  no  other  example  in  the  writers  of 
the  period  based  on  as  good  MS  authority  (^Quid  enim,  Cure. 
273,  being  a  conjecture ;  quis  enim,  Enn.  1 14  (M.),  depending  oa 
the  reading  of  the  scholiast,  and  quis  enimf  ex  inc.  inc.  fab.  i 
(R.  I),  having  so  uncertain  a  date),  there  seems  no  reason  for 
making  the  change.  There  are  other  readings  of  equal  authority 
and  rarity  in  Plautus. 

(ii)  Enim  with  causal  force. — Most.  925-6  reads:  TV.  Quid? 
tibin  umquam  quicquam,  postquam  tuos  sum,  verborum  dedi? 
Th.  Ego  enim  recte  cavi.  Lorenz,  in  his  note  to  the  passage, 
recognizing  its  causal  force,  and  the  implied  ellipsis,  translates : 
^'Ego  enim,  *nein,  denn  ich' — eine  bei  nam  und  enim  wie  bei  yaf^ 
haufige  und  bekannte  Ellipse." 

In  Poen.  604,  Milphio  exclaims:  En,  edepol  mortales  malos! 
whereat  Agorastocles  proudly  replies:  Ego  enim  docui.  The 
passage  is  similar  to  the  preceding,  and  the  simplest  and  most 
natural  way  to  interpret  it  is  by  supplying  the  evident  ellipsis : 
'Certainly  they  are,  for  I  taught  them.'  To  explain  enim  as 
equivalent  to  profecio  is  to  decidedly  weaken  the  force  of  the 
reply. 

I  have  always  been  sorely  tempted  to  regard  a  similar  ellipsis 
as  existing  in  Cas.  279-80  Lys.  Te  uxor  aiebat  tua  Me  vocare. 
Ch.  Ego  enim  vocari  iussi,  though  the  causal  force  does  not  seem 
as  strong  as  in  the  two  preceding  passages. 

Pseud.  133  seqq.  Ballio  comes  out  heaping  abuse  on  the  heads 
of  his  slaves :  Exite,  agite  exite,  ignavi,  male  habiti  et  male  con- 
ciliati  Quorum  numquam  quicquam  quoiquam  venit  in  mentem 
ut  recte  faciant  Quibus  nisi  ad  hoc  exemplum  experior,  non 
potest  usura  usurpari,  Neque  homines  magis  asinos  umquam  vidi» 
ita  plagis  costae  callent,  Quos  quom  ferias,  tibi  plus  noceas,  eo 
enim  ingenio  hi  sunt  etc.  Lorenz,  properly  regarding  enim  as 
causal,  explains  the  passage:  '^noceas,  theils  weil  sie  dann  an 
Diebstahl,  Raub  und  Flucht  denken;  denn  eo  ingenio  sunt  etc.** 


-Digife< 


odbyGooqt^ 


3* 


USE  OF  ENIM  IN  PLAUTUS  AND  TERENCE.         407 

There  is  no  necessity  of  thus  straining  the  meaning  of  noceas, 
Ussing  gives,  to  my  mind,  the  true  explanation  by  regarding  the 
clause  beginning  neque  homines  as  parenthetical  and  referring 
etiim  to  the  lines  preceding. 

Terence  furnishes  one  example,  And.  808-9  nam  pol  si  id 
scissem,  numquam  hue  tetulissem  pedem;  semper  enim  dictast 
esse  atque  habitast  soror. 

Of  the  half  dozen  or  more  examples  of  enim  that  have  found 
their  way  into  the  text  of  Plautus  by  conjecture,  I  shall  mention 
only  one,  Ribbeck's  emendation  of  Mil.  1319  Enim  pietas  sic 
hortat.  Two  objections  have  been  urged  against  the  conjecture, 
the  use  of  enim  as  causal  and  the  active  form  horiai.  The  first 
has  been  already  disposed  of.  The  second  is  stronger,  though 
examples  of  the  active  forms  of  horior  are  cited  by  Ribbeck  in 
his  critical  notes  and  the  lexicons.  It  must  be  admitted  from 
Langen's  statistics  (p.  63)  as  to  the  forms  of  horior  in  Plautus 
based  on  manuscript  authority  that  the  active  form  is  improbable, 
though  possible.  Still  the  strongest  argument  against  the  reading 
is  the  position  of  enim.  It  has  been  shown  that  in  all  passages,  in 
both  Plautus  and  Terence,  where  enim  holds  the  first  place,  its 
force  is  corroborative.  Indeed,  we  are  justified  in  regarding  this 
as  a  rule.  Enim,  in  Ribbeck's  text  is  nothing  if  not  causal,  and  in 
its  position  lies  the  gravest  objection  to  its  adoption. 

It  is  in  place  to  state  Langen's  argument  as  to  the  non-existence 
of  causal  enim  in  Plautus.  Briefly  put,  it  is  as  follows :  In  a  large 
majority  of  passages  in  Plautus  enim  has  the  corroborative  force 
and  no  other  meaning  is  possible.  In  the  remaining  examples  a 
causal  force  is  possible,  though  a  corroborative  force  can  be  given. 
Therefore  there  is  no  passage  in  which  the  corroborative  force  is 
impossible.  Let  us  test  this  argument  with  reference  to  Terence. 
It  is  agreed  that  the  investigation  must  start  with  simple  enim> 
and  proceed  to  its  compounds.  In  Terence  there  are  10  instances 
of  simple  enim  with  corroborative  force  to  one  with  causal. 
Applying  Langen's  reasoning,  as  the  overwhelming  majority  of 
passages  are  corroborative,  all  may  well  be,  and  the  one  causal 
instance  vanishes.  But  it  does  not.  Langen  admits  that  it  is 
causal  and  cannot  be  otherwise.  If  one  example  in  1 1  can  be 
causal  in  Terence,  is  the  proportion  so  great  as  to  make  it  impos- 
sible for  3  out  of  37  or  4  out  of  38  in  Plautus,  as  shown  above  ? 
It  may  be  answered  that  the  causal  meaning  is  the  only  permis- 
sible one  in  the  Terentian  passage,  but  only  one  of  two  and 


Digitized  by  VjOOQ IC 


408  AMERICAN  JOURNAL    OF  PHILOLOGY, 

perhaps  not  the  better  in  the  four  Plautine  examples.  It  is  no 
greater  feat  of  mental  gymnastics  to  read  a  corroborative  force 
into  the  passage  from  the  Andria  than  into  the  passages  dted 
from  Plautus.  The  causal  force  of  enim  in  a  number  of  passages 
in  Plautus  yet  to  be  discussed  is  as  plain  to  me  as  the  majority 
admittedly  so  in  Terence.  Each  reading  only  emphasizes  this 
view.  Tests  made  with  others,  who  could  not  be  accused  of  bias 
toward  either  view,  favor  the  causal  interpretation  as  the  only 
reasonable  one,  and  as  the  clearest  and  most  emphatic.  I  can 
see  no  special  difference  in  usage  between  Plautus  and  Terence 
in  this  regard.  Any  preconceived  idea,  carried  out  to  its  logical 
result,  will  be  as  sweeping  in  its  conclusions  as  Langen's  on  this 
subject.  That  the  conclusions  are  always  correct,  and  the  process 
a  laudable  one,  is  deserving  of  serious  question. 

III. — Enim  with  Affirmative  Particles. 

(i)  Enim  vero, — From  enim  we  pass  to  the  strengthened  form 
enim  vero^  which  simply  increases  the  force  of  the  affirmation. 
The  view  of  some  early  grammarians,  that  it  may  have  an  adver- 
sative force  like  sed^  is  not  sustained  by  the  examples.  Drager^ 
shows  that  its  occurrence  with  this  meaning  is  only  in  later  prose. 

1.  It  is  found  in  simple  assertions:  Amph.  266,  723,  771,  Capt 
22,  Cas.  475,  Cist.  519,  Men.  860,  Stich.  398,  Trin.  958,  And.  91, 
206. 

2.  It  is  often  used  to  denote  a  state  or  condition,  and  then  is 
frequently  accompanied  by  a  temporal  particle :  Capt.  534  Nunc 
enim  vero  occidi;  Cure.  175,  608,  Merc.  739,  Hec.  673. 

3.  It  is  used  in  statements  expressing  indignation  or  irony :  H. 
T.  1045,  Phor.  465. 

4.  It  is  found  in  answers.  These  are  of  two  kinds:  (i)  where 
the  answer  is  suggested  by  the  statements  of  the  preceding 
speaker:  Capt.  628  Heg.  Fuistin  liber?  Tyn,  Fui.  Ar,  Enim 
vero  non  fuit,  nugas  agit ;  Most.  920,  Pers.  349,  Poen.  280,  296; 
435,  Rud.  1003,  Stich.  616,  Trin.  989,  And.  848,  H.  T.  320, 
Phorm.  937,  985 ;  (2)  where  the  answer  is  a  direct  reply  to  the 
preceding  question:  Amph.  344  Merc.  Ain  vero?  So.  Aioenim 
vero;  410,  759,  Asin.  339,  688,  Cas.  728,  Men.  1075,  Pers.  185, 
Pseud.  979,  Trin.  987,  Phorm.  I036. 

'Historische  Syntax,  II,  p.  131. 


Digitized  by.VjO.Q2l£_ 


J 


USE  OF  ENIM  IN  PLAUTUS  AND  TERENCE.         4O9 

(ii)  Certe  enim  and  certo  enim, — Langen*  in  an  exhaustive 
study  and  citation  of  the  Plautine  and  Terentian  passages  in 
which  the  words  occur,  reaches  the  conclusion  that  in  Plautus 
eerie  expresses  'subjective  certainty'  and  certo  'objective  cer- 
tainty.' In  Terence  we  find  certe  in  its  later  classical  usage 
taking  the  place  of  certo  in  expressions  of  'objective  certainty.' 
The  same  results  apply  in  the  use  of  the  words  when  strengthened 
by  entm.  The  examples  are  not  numerous — six  in  Plautus  and 
one  in  Terence. 

1.  Certe  enim  is  found  Amph.  331,  658,  Asin.  614,  And.  503. 
Aul.  811  the  manuscripts  read:  Certo  enim  ego  vocem  hie 
loquentis  modo  mi  audire  visus  sum.  This  should  be  changed 
to  Langen's  reading  certe^  in  conformity  to  his  rule. 

2.  Certo  enim  occurs  in  two  passages:  Poen.  1182  Certo  enim, 
quod  ad  nos  attinuit,  Pulchrae  praepollentesque,  soror,  fuimus; 
Stich.  88  Certo  enim  mihi  paternae  vocis  sonitus  auris  accidit. 
Terence  has  no  example  of  the  combination. 

(iii)  Nempe  enim. — Trin.  61  Ritschl  and  SchoU  read :  Nempe 
enim  tu,  credo,  me  inprudentem  obrepseris.  The  manuscript 
reading  Tiamque  enim  is  adopted  and  defended  by  Brix,  Hand,' 
Langen'  and  others.  It  must  be  admitted  that  Ritschl's  conjec- 
ture is  unusual,  it  being  the  first  instance  of  the  usage  outside  of 
the  writers  of  the  Silver  Age.  On  the  other  hand,  namque  enim 
occurs  nowhere  else.  Hand  would  explain  it  as  a  colloquialism, 
comparing  it  with  neque  hand.  This  explanation  is  far  from 
acceptable.  Ritschi's  conjecture^  has  two  reasons  to  commend 
it:  (i)  it  is  probable  from  the  frequent  interchange  of  nempe, 
namque,  neque  in  the  manuscripts,  and  (2)  more  important  still, 
it  is  in  perfect  harmony  with  Megaronides'  remark.* 

IV. — Enim  with  Adversative  Particles, 

(i)  At  enim. — Enim  is  frequently  joined  with  the  adversative 
particle  a/,  having  in  most  cases  the  affirmative  or  corroborative 
force  already  noticed.  It  may  then  be  translated  'but  indeed,  but 
surely.' 

I.  Examples  of  such  usage  are:  Bacch.  993,  1080  {et  MSS,  sed 
Acidalius),  Cist  235,  739,  Epid.  94,  Men.  790,  Merc,  159,  Most. 

^  BeitrSge,  pp.  22-31.  'Vol.  IV,  p.  12.  ' BeitrSge,  p.  261. 

*  Prolegomena,  p.  Ixxv  (reprinted  in  his  Opuscnla,  vol.  V,  p.  332). 
^For  other  conjectures  and  discussions  of  this  much-disputed  passage,  see 
Schttll,  App.  Crit.,  p.  127. 


Digitized  by^ 


4IO  AMERICAN  JOURNAL   OF  PHILOLOGY. 

808,  Pers.  569,  832,  Poen.  914,  1197  (twice),  Pseud.  436,  Stich. 
129,  738,  Trin.  919,  Eun.  381,  751,  H.  T.  317,  699,  713. 

2.  In  Most  922  At  enim  ne  caption!  mihi  sit,  si  dederim  tibi  is 
an  example  of  the  common  ellipsis  of  metuo}  In  Ad.  830  seq. 
we  have  At  enim  metuas,  ne  ab  re  sint  tamen  Omissiores  paulo. 

3.  It  is  used  in  the  reply  expressing  indignation  or  some  other 
emotion:  Phorm.  487  Ph.  Audi  quod  dicam.  Do.  At  enim 
taedet  iam  audire  eadem  milia. 

4.  It  is  found  twice  in  questions  in  connection  with  scin :  Pseud. 
538,  641. 

5.  One  example  is  found  of  a  comical  play  on  the  particle : 
Epid.  95  At  enim, — bat  enim.  With  this  can  be  compared  Pseud. 
236  Cal.  At.    Ps.  Bat;  and  Pers.  213  P<ug.  Heia.    Soph.  Beia. 

(ii)  Verum  enim. — Langen'  shows  that  verum  has  only  adver- 
sative force.  Any  interpretation  (like  that  of  Ussing  in  his  note 
to  Asin.  790,  who  translates  it  by  sane)  which  would  regard  it  as 
synonymous  with  vero  is  incorrect. 

1.  Six  examples  of  verum  enim  are  found  in  Plautus  and 
Terence:  Cist.  80,  Mil.  293,  Poen.  874,  Ad.  201,  Eun.  742, 
Phorm.  555.  In  five  of  these  verum  has  plainly  the  force  oised. 
The  sixth  presents  unusual  difficulties.  It  is  Poen.  873-4,  where 
Goetz  reads:  Syn.  I  in  malam  rem.  Mil.  I  tu  atque  herus. 
Syn.  Verum  enim  qui  homo  eum  norit,  cito  homo  pervorti. 
Geppert  changed  the  second  verse  so  as  to  read :  Verum  enim,  si 
modo  eum  noris  etc.,  where  verum  enim  caij  only  have  the  force 
of  enim  vero^  and  the  answer  is  not  in  harmony  with  what 
precedes. 

Two  ways  out  of  the  difficulty  suggest  themselves.  Enim  vero 
can  be  read,  in  harmony  with  the  numerous  passages  where  its 
Plautine  force  has  been  shown,  or  we  can  suppose  that  some 
passage  or  lines  containing  Synecratus'  reply  has  been  lost,  and 
that  the  statement  of  the  text  is  its  continuation. 

2.  A  strengthened  form  of  verum  enim  is  verum  enim  vero. 
An  example  of  this  is  found  in  each  author:  Capt.  599,  Ad.  255. 

(iii)  Sed  enim.^ — No  example  oised  enim  is  found  in  the  manu- 
scripts of  Plautus  or  Terence.    Three  conjectures  have  introduced 

^  Lorenz  in  his  note  on  the  passage  cites  other  examples  of  the  same  ellipsis. 
•  BeitrSge,  pp.  1 13-21.  'See  Brix's  note  on  Mil.  983. 


Digitized  by  VjOOQ IC 


USE  OF  ENIM  IN  PLAUTUS  AND  TERENCE,  4II 

it  into  the  text.  So  Acidalius  in  Bacch.  1080  in  place  of  etenim 
(at  enim  Pareus),  Ritschl  in  Bacch.  1083,  while  Goetz  prefers  to 
follow  the  manuscripts,  and  again  in  Mil.  983,  with  Fleckeisen 
and  Lorenz.  Ribbeck,  Brix  and  Goetz,  however,  read  sed  ne  et 
isiam  instead  of  sed  enim  ne  istam^  which  removed  the  faulty 
hiatus  sed  ne  isiam.  The  first  example  based  on  manuscript 
authority  is  in  Cato,  Or.  pro  Rhod.  (Jordan,  23,  9). 

(iv)  Immo  enim. — Immo  enim  is  used  whenever  an  opinion 
opposed  to  what  has  just  been  expressed  is  to  be  emphatically 
stated :  Pseud.  31  Call.  Lege  vel  tabellas  redde.  Ps.  Immo  enim 
pellegam ;  Stich.  699,  And.  823,  Phorm.  337.  Enim  has  in  these 
examples  its  corroborative  force. 

The  stronger  form  immo  enim  vera  occurs  with  substantially 
the  same  force:  Capt.  608,  Eun.  329,  Phorm.  528. 

V. — Enim  with  Causal  and  Final  Particles. 

(i)  Quia  enim. — Enim  is  often  joined  with  quia^  strengthening 
or  intensifying  its  causal  force.  It  is  thus  found  in  answers  to 
questions  introduced  by 

1.  Qui,  Amph.  266,  1034,  Pers.  228,  True.  733. 

2.  Quiistuc,  Phorm.  331. 

3.  Qui  dum,  Epid.  299,  Rud.  11 16. 

4.  Qui  vero,  Merc.  395  (Ritschl). 

5.  Quid,  Capt.  884,  Cas.  385,  Cure.  449,  Mil.  1139,  Poen.  1344, 
True.  266  Quia  enim  me  truculentum  nominas. 

6.  Quid  iia,  Pers.  592. 

7.  Quid  iamy  Bacch.  50,  Mil.  834,  Pseud.  325.^ 

8.  Quo  argtimento.  Mil.  looi. 

9.  Qua  istuc  ratione.  Pseud.  804. 

10.  Quamobrem,  Cure.  443,  667,  H.  T.  800. 

11.  CuTy  Merc.  648,  Most.  1097. 

12.  Qua  propter y  H.  T.  188,  Hec.  311. 

(ii)  Ut  enimy  ne  enim. — In  a  similar  way  enim  with  its  affirma- 
tive force  is  joined  with  the  final  particles  ut  and  ne.    Thus  with 

1.  Uty  Cas.  268,  Epid.  277,  Poen.  855. 

2.  Ney  Most.  1095. 

*  Lorenz,  by  comparing  this  passage  with  318  (qtdapol)  and  345  {quia  edepol\ 
shows  the  connection  between  enimy  polssi^  edepol. 


Digitized  bT^^^"""" 


412  AMERICAN  JOURNAL    OF  PHILOLOGY. 

VI. — Enim  with  Negative  Particles. 

(i)  Non  enim, — There  are  two  distinct  usages  of  nan  enim,  as 
has  been  found  to  be  the  case  with  simple  enim, 

1.  It  has  already  been  shown  that  in  a  large  majority  of  the 
passages  in  Plautus  and  Terence  in  which  it  occurs,  the  force  of 
enim  is  merely  corroborative.  A  comparison  of  the  passages 
containing  non  enim  in  the  light  of  these  results  will  give  the 
same  conclusions.  Accustomed  as  most  scholars  of  Plautus  have 
been  to  Ciceronian  usage,  they  have  often  been  led  astray  by  the 
discovery  that  enim  with  causal  force,  in  negative  sense,  is  not  in 
place  in  several  passages.  To  remove  this  difficulty  the  archaic 
negative  noenum  or  noenu  is  substituted,  as  by  Ritschl  in  Trin. 
705  and  Bucheler  in  Asin.  808.  It  is  questionable  if  this  is  either 
necessary  or  based  on  good  reasons.  The  examples  of  noenum 
(u)  based  on  manuscript  authority  are  so  rare  that  conjectures 
increasing  their  number  must  be  regarded  as  venturesome.* 
With  the  corroborative  force  of  enim,  so  generally  admitted,  no 
change  is  necessary. 

This  corroborative  force  is  shown  by  the  following  examples: 
Aul.  594,  Cist.  562,  Epid.  162,  Most.  1133,  Pseud.  1266,  Rud.  989, 
Stich.  600,  True.  309. 

Three  passages  similar  in  construction  are :  Mil.  283  Non  enim 
faciam  quin  scias ;  Stich.  302  Non  enim  possum  quin  revortar ; 
Trin.  705  Non  enim  possum  quin  exclamem. 

Non  enim  is  used  once  in  Terence  to  express  a  strong,  confident 
denial :  Phorm.  694  An.  Quid  fiet?    Ge,  Non  enim  ducet. 

2.  In  the  following  passages  the  causal  force  of  enim  is  far  more 
in  place  than  the  corroborative ;  indeed,  in  several  it  is  the  only 
possible  one. 

Capt.  860  /feg.  Non  sentio.  Ergas,  Non  enim  es  in  senticeto, 
eo  non  sentis.  Brix,  striving  to  reproduce  the  pun  and  at  the 
same  time  preserve  the  corroborative  force  of  enim^  translates : 
**Ja,  du  bist  auch  kein  Marker."  A  correct  translation  must 
bring  out  the  causal  force  of  enim:  *You  don't  feel,  because  you 
are  not  in  the  briars.' 

Most.  827-8,  Tranio  says  of  the  door-posts:  Atque  etiam  nunc 
satis  boni  sunt,  si  sunt  inducti  pice.    Non  enim  haec  pultifagus 

^The  only  cases  I  have  discovered  are  Aal.  67  and  Lucr.  3, 199;  4,  712. 
The  three  passages  in  Ennius,  A.  287,  479  (M)  and  F.  doi  (R)  are  all  dae  to 
conjecture.  See  L.  MQUer,  Lucilius,  30,  23  (p.  267) ;  Ritschl,  Opus,  vol.  II,  p^ 
242. 


Digitized  by  VjOOQ IC 


USE  OF  ENIM  IN  PLAUTUS.  AND  TERENCE.  4I3 

opufex  opera  fecit  barbarus.  Sonnenschein,  in  his  note  on  the 
passage,  says:  ''This  is  one  of  the  few  passages  in  Plaut.,  in 
which  enim  seems  to  approach  very  nearly  to  the  meaning  of 
'for/  but  it  may  be  translated  'look  you.'"  I  cannot  see  how 
any  translation  but  the  causal  can  be  defended  here.  Tranio 
plainly  assigns  his  reason  for  the  good  condition  of  the  posts. 
He  does  not  stop  and  turn  to  Theopropides  with  the  exclamation 
'Look  you,  no  pottage-eating  artisan  from  foreign  parts  made 
them.' 

So  Poen.  285-6  Nam  pro  erilei  et  nostro  quaestu  satis  bene 
ornatae  sumus.  Non  enim  pote  quaestus  fieri,  nisi  sumptus 
sequitur  scio,  and  True.  907-8  Numquam  uno  hoc  die  ecficiatur 
opus  quin  opus  semper  siet.  Non  enim  possunt  militares  pueri 
ut  alii  (Bugge,  avis^  Schbll)  educier. 

Rud.  921-2  Gripus  in  his  monologue  says:  Vigilare  decet 
hominem  qui  volt  sua  temperi  conficere  officia,  adding  as  his 
reason,  not  as  a  parallel  statement,  non  enim  ilium  exspectare 
oportet  dum  erus  se  ad  suom  suscitet. 

Bucheler  read  noenum  Asin.  808,  where  the  text  has  Haec  sunt 
non  nugae :  non  enim  mortualia.  Ussing  would  explain  it  thus : 
"  Haec  seria  sunt,  non  nugae ;  neque  enim  mortuis  haec  cantantur, 
sed  vivis."  This  explanation  is  designed  to  meet  Langen's  objec- 
tion as  to  its  causal  force,  since  anything  can  be  nugae,  without 
being  necessarily  mortualia.  Ussing*s  interpretation  has  much 
to  commend  it,  though  I  doubt  the  genuineness  of  the  verse. 

Eun.  453  Th.  Bene  dixti  ac  mi  istuc  non  in  mentem  venerat. 
Gn.  Ridiculum !  non  enim  cogitaras.  We  may  translate :  '  Ab- 
surd !  why,  you  had  not  thought  of  it'  it  is  not  difficult  to  see 
something  of  the  causal  force  in  the  passage. 

(li)  Neque  enim, — The  same  peculiarities  of  usage  are  exhibited 
in  neque  enim. 

1.  Its  corroborative  force,  in  a  negative  sense,  is  shown  in  Cas. 
888  Reppulit  mihi  manum ;  neque  enim  dare  sibi  savium  me  sinit. 

2.  Its  causal  force  is  evident  in  Pers.  63  seq.  Neque  quadru- 
plari  me  volo;  neque  enim  decet  Sine  meo  periculo  ire  aliena 
ereptum  bona,  Neque  illi  qui  faciunt,  mihi  placent;  Trin.  584 
Les.  Nam  certumst  sine  dote  baud  dare.  Sias,  Quin  tu  i  modo. 
Les,  Neque  enim  illi  damno  umquam  esse  patiar. 

This  force  is  still  more  marked  in  two  passages  from  Terence : 
Ad.  647  Habitant  hie  quaedam  mulieres  pauperculae ;  Ut  opinor 


Digitized  by  VjOOQ IC 


414  AMERICAN  JOURNAL   OF  PHILOLOGY. 

eas  non  nosse  se  et  certo  scio ;  Neque  e;pim  diu  hue  migrarunt ; 
Hec.  833-5  Haec  tot  propter  me  gaudia  illi  contigisse  laetor: 
Etsi  hoc  meretrices  aliae  nolunt ;  neque  enim  est  in  rem  nostram 
Ut  quisquam  amator  in  nuptiis  laetetur.^ 

(iii)  Numquam  enim. — Numquam  occurs  with  enim  in  corrob- 
orative force  in  Pers.  489,  Stich.  96,  751. 

Vll.—Etenim. 

Etenim  (a  word  as  peculiar  and  mysterious  in  its  formation  as 
namque)  is  foreign  to  Plautus,  the  only  passage  in  which  it  is 
retained  in  the  text  being  in  the  late  prologue  to  the  Amphitruo, 
where  (v.  26)  we  read:  Etenim  ille,  quoius  hue  iussu  venio, 
Juppiter,  Non  minus  quam  nostrum  quivis  formidat  malum.  The 
two  passages  Cist.  777  and  Baeeh.  1080,  in  which  the  manuscript 
reading  has  been  changed  in  our  texts,  have  already  been 
discussed.  The  causal  force,  shown  in  the  Amphitruo  passage, 
occurs  in  three  passages  from  Terence :  And.  442  Deinde  desinet. 
Etenim  ipsus  secum  eam  rem  reputavit  via ;  Eun.  1074  Ut  luben- 
ter  vivis  (etenim  bene  lubenter  victitas);  H.  T.  546-7  Facile 
equidem  facere  possum  si  iubes.  Etenim  quo  pacto  id  fieri  soleat 
calleo.  I  see  no  reason  to  read  with  Langen,  in  the  Eunuchus 
passage,  et  enim  ("und  waHrlich").  The  causal  force  is  not  as 
strong,  it  is  true,  as  in  the  other  two  examples.' 

Omitting  all  conjectures  that  would  introduce  causal  enim  into 
the  text  and  all  examples  of  etenim^  there  are,  at  a  rough  estimate, 
14  examples  oi  enim  corroborative  to  i  oi  enim  causal  in  Plautus, 
while  in  Terence  the  prpportion  is  13  to  i.  The  causal  force  in 
the  examples  from  Plautus  is  clear,  in  most  cases  as  much  so  as 
those  of  Terence.  The  percentage  of  causal  examples,  though 
not  large,  is  respectable  enough  not  to  be  rejected  through  mere 
devotion  to  a  theory.  The  proportion,  too,  it  will  be  noted,  is 
nearly  as  large  as  in  Terence.  With  the  exception  of  etenim^ 
which  leads  a  peculiar  existence  in  most  of  the  poets  before  the 
Silver  Age,  the  use  of  enim  and  its  compounds  in  the  two  poets 

^  Drager's  statement  (Syntax,  vol.  II,  p.  68) :  **Nec  enim  statt  non  enim  findet 
sich  zuerst  Ter.  Hec.  V  3,  36/'  is  shown  by  the  above  examples  from  Plautus 
to  be  incorrect. 

2  Drager's  statement  (Syntax,  vol.  II,  p.  171)  that  etenim  occurs  twice  in 
Plautus  and  four  times  in  Terence  must  be  corrected. 


Digitized  by  VjOOQ IC 


USE  OF  ENIM  IN  PLAUTUS  AND  TERENCE,  415 

seems  to  harmonize.  Under  such  conditions,  it  is  impossible  to 
give  any  dogmatic  assertion  regarding  the  origin  and  growth  of 
the  causal  usage.  Had  more  of  the  earlier  language,  outside  of 
the  two  dramatists,  been  preserved,  such  a  statement  might  be 
risked.  With  the  scant  remains  at  our  disposal,  and  these  largely 
conjectural,  no  satisfactory  results  can  be  gained.  It  is  for  this 
reason  that  other  writers  of  the  period,  though  examined,  have 
not  been  drawn  into  the  discussion.^ 

UmvBxsxTT  OF  Idaho»  WiLLARD  K,  ClEMENT. 

Aug.  6, 1897. 

^  Remoteness  from  large  libraries  and  philological  centres  makes  it  impos> 
sible  for  me  to  familiarize  myself  with  more  than  the  names  of  many  of  the 
German  dissertations  or  programs  bearing  more  or  less  directly  on  the  theme. 
Experiences  while  a  student  abroad  convinced  me  that  it  is  often  impossible 
to  secure  certain  much-quoted  pamphlets  or  articles.  Omissions  or  failures  to 
make  proper  reference  or  give  due  credit  are  not  always  the  result  of  care- 
lessness or  ignorance. 


Digitized  by  VjOOQ IC 


IIL—ON  THE  CHARACTER  OF  INFERRED   PARENT 
LANGUAGES. 

I. — The  idea  of  inferentially  constructing  a  parent  language  on 
the  basis  of  actually  existing  cognate  languages  or  dialects  seems 
to  have  originated  with  Schleicher.  In  his  Linguistische  Unter- 
suchungen,  vol.  II  (Die  Sprachen  Europa*s  in  systematischer 
Uebersicht),  published  in  1850  at  Bonn,  he  speaks  of  'primary 
languages'  (^Primdrsprachtn^  e.  g.  pp.  29-30)  such  as  the  Latin 
and  the  Sanskrit,  contrasting  them  with  'secondary  languages* 
such  as  the  Romance  languages  and  the  modem  Hindu  vernacu- 
lars. In  some  cases,  he  adds,  such  primary  languages  are  not 
extant,  but  must  be  constructed  from  their  descendants  (secondary 
languages).  These  primary  languages,  in  turn,  he  regards  as 
daughters  of  one  mother,  the  parent  language  (  Urspracke).  Two 
years  later,  in  his  Formenlehre  der  Kirchenslavischen  Sprache 
(1852),  he  expresses  himself  similarly.  A  Parent  Slavonic  is 
posited  there  as  the  common  source  from  which  the  different 
Slavonic  idioms  must  be  derived  and  which  may  be  inferred  by  a 
comparison  of  these  idioms  (p.  27).  And  by  way  of  illustration 
he  constructs  (p.  28)  the  Parent  Slavonic  active  present  participle 
on  the  basis  of  the  Church  Slavonic,  Serbo-IUyrian,  Russian, 
Polish,  and  Bohemian  forms. 

What  is  done  here  for  the  Slavonic  dialects  he  considers 
possible  for  the  Indo-European  languages:  'From  a  comparison 
of  the  oldest  extant  languages  of  the  different  Indo-European 
families,  with  due  regard  to  the  laws  of  historical  grammar,  we 
may  form  a  comparatively  clear  conception  of  the  Indo-European 
parent  language  from  which  the  mothers  of  the  different  families 
[=  Schleicher's  primary  languages]  developed  in  a  manner  anal- 
ogous to  that  in  which  the  Romance  languages  were  evolved  from 
the  Latin'  (p.  4, 1.  c).  All  the  derived  languages,  he  maintains, 
must  form  the  basis  on  which  the  Indo-European  parent  language 
is  to  be  constructed,  since  all  of  them  have  originally  flowed  alike 
from  this  common  source.  But  the  varying  degree  of  faithfulness 
with  which  the  different  languages  have  preserved  old  sounds 
and  forms  makes,  according  to  Schleicher,  those  languages  of 


Digitized  by  VjOOQ IC 


INFERRED  PARENT  LANGUAGES,  417 

especial  importance  which  have  remained  nearest  to  the  original 
home  of  the  Indo-European  parent  people. 

It  was  nine  years  later,  viz.  in  1861,  when  this  plan  of  recon- 
'struction  was  actually  carried  out  As  the  subtitle  of  the  com- 
pendium 'Kurzer  Abriss  einer  Lautlehre  der  Indogermanischen 
Ursprache,  des  Altindischen,  Alteranischen,  Altgriechischeni  Alt- 
italischen,  Altkeltischen,  Altslawischen,  Litauischen  und  Altdeut- 
schen'  shows,  'the  attempt  has  here  been  made  to  place  the 
inferred  Indo-European  parent  language  alongside  of  its  really 
existing  descendants'  (',  p.  8,  note).  Summing  up  the  results  of 
comparative  grammar  of  the  preceding  half-century,  the  compen- 
dium closes  the  first  period.  It  opens  the  second  period  in  that 
it  endeavors  to  trace  the  &cts  of  the  various  Indo-European 
idioms  back  into  prehistoric  times,  in  order  to  reconstruct  from 
the  data  of  the  individual  languages  the  parent  language  from 
which  all  of  them  are  descended.^  For  a  great  deal  of  the  work 
of  the  last  forty  years  has  been  done  along  these  very  lines,  and 
in  the  eyes  of  many  the  ultimate  reconstruction  of  the  Indo- 
European  parent  language  has  been  the  ideal  of  all  special 
comparative  investigation,  the  more  so  as  it  seemed  the  key  to 
open  to  us  the  mysteries  of  a  prehistoric  civilization.  *  I  had 
originally  intended,'  says  Fick  in  the  preface  to  the  fourth  edition 
of  his  Comparative  Dictionary  (1890),  *a  work  on  a  much  larger 
scale.  I  had  in  view  to  add  to  the  lexicon  of  the  Indo-European 
parent  language  also  its  grammar,  and,  furthermore,  a  sketch  of 
the  civilization  of  the  parent  people.  But  the  time  for  doing  this 
has  not  yet  come.  There  is  need  of  more  works  like  J.  Schmidt's 
Pluralbildungen,  before  we  may  dare  approach  the  reconstruction 
of  the  grammar  oi  the  parent  language  . . .' 

To  be  sure,  the  parent  language  as  now  reconstructed  looks 
very  different  from  that  inferred  by  Schleicher. 

2. — ^We  have  ceased  to  look  with  Schleicher  for  absolute 
simplicity  in  the  parent  language.  To  him  the  morphological 
elements  of  a  word  were  then  still  intact,  for  successive  vowels 
and  consonants  had  not  yet  begun  to  react  on  each  other.  The 
diversity  and  manifoldness  in  sounds  and  inflection  of  the  various 
Indo-European  idioms  as  they  appear  in  historical  times  are  to 
Schleicher  the  results  of  decay  and  degeneration.  This  theory 
was  gradually  abandoned  for  two  reasons. 

^  Cf.  Bechtel,  Hauptprobleme,  Einleit.,  p.  i. 


Digitized  by  VjOOQ IC 


4l8  AMERICAN  JOURNAL    OF  PHILOLOGY. 

First,  because  such  an  a  priori  postulate  of  simplicity  could 
only  reasonably  be  made  for  the  very  first  period  of  language- 
production.  But  this  period  is  absolutely  beyond  our  reach  and 
separated  by  a  vast  gulf  from  the  periods  amenable  to  recon- 
struction. 

Second,  because  this  principle  conflicts  with  Schleicher's  second 
methodological  principle,  that  the  parent  form  must  be  of  such  a 
character  that  all  really  existing  forms  of  the  Indo-European 
languages  may  be  derived  from  it  by  regular  laws.  The  more 
consistendy  this  principle  has  been  applied,  the  more  has  sim- 
plicity given  way  to  complexity,  and  in  consequence  of  it  the 
parent  language  as  now  reconstructed  is,  in  some  respects,  richer 
than  any  of  its  descendants. 

3. — An  inquiry  into  the  nature  and  character  of  the  parent 
language,  thus  reconstructed,  will  naturally  fall  into  two  parts. 

First,  we  must  examine  the  various  limitations  to  which  this 
method  of  reconstruction  is  necessarily  subject. 

Second,  we  must  determine  how  these  limitations  affect  the 
object  reconstructed  according  to  this  comparative  method; 
whether,  namely,  they  imply  quantitative  imperfections  only,  or 
whether  their  influence  is  so  vital  as  to  touch  upon  the  very 
essence  and  quality  of  the  reconstructed  object. 

4. — Philology,  like  all  historical  sciences,  requires  an  object 
clearly  defined  in  time  and  in  space.  It  is  here  that  we  find  our 
comparative  method  most  seriously  defective. 

The  chief  characteristic  of  all  modem  grammatical  investiga- 
tions is  that  they  are  historical,  i.  e.  that  they  do  not  treat  a 
language  as  if  it  were  fixed  and  immovable,  but  as  a  growth 
whose  changing  phases  should  be  outlined  in  a  connected  series 
of  successive  periods.  The  very  attempt  to  reconstruct  a  parent 
language  is  due  to  this  historical  treatment,  for  its  aim  is  simply 
to  extend  the  continuity  of  development  beyond  historical  times. 

5. — But  the  question,  To  which  period  of  the  prehistoric  Indo- 
European  does  a  given  reconstructed  form  belong  ?  is,  unfortun- 
ately, unanswerable.  'When  we  speak  of  Indo-European  forms/ 
says  Brugmann  (Compendium,  Engl,  tr.,  I,  p.  13,  §12),  *we  gen- 
erally mean  those  forms  which  were  in  use  toward  the  close  of 
the  primitive  period.'     But  we  also  often  mean  such  forms  as 

^  The  vagueness  of  this  limit  is  pointed  out  below,  §10. 


Digitized  by  VjOOQ IC 


^mm^mmmmmmmmmmmmms^^mmBmmmmmmmmim 


INFERRED  PARENT  LANGUAGES,  419 

belonged  to  an  earlier  period  of  this  stage  and  which  had  already 
undergone  a  change  toward  its  termination.  Forms  put  down  by 
us  as  primitive  Indo-European  . . .  are  therefore  not  to  be  indis- 
criminately regarded  as  belonging  to  the  same  period.^  The 
result  of  this  uncertainty  becomes  glaringly  apparent  if  we 
imagine  an  English  grammar  or  dictionary  constructed  according 
to  a  method  by  which  Anglo-Saxon,  Chaucerian,  and  nineteenth- 
century  forms  could  not  be  separated  but  would  all  stand  on  the 
same  plane. 

6. — It  is  only  another  aspect  of  the  same  fundamental  difficulty 
that  we  are  unable  to  fix  accurately  the  time  and  extent  of  opera- 
tion of  inferred  phonetic  laws.  Ignorant  of  the  exact  time  during 
which  they  were  operative  and  of  the  relative  chronology  of 
different  laws,  it  is  inevitable  that  we  must  be  constantly  commit- 
ting the  gravest  anachronisms  in  our  reconstruction  of  Indo- 
European  forms,  combining  in  the  same  form  laws  which  oper- 
ated at  entirely  different  periods.  As  early  as  1869  Johannes 
Schmidt  called  attention  to  this  danger.  In  the  preface  (p.  ix)  to 
the  second  edition  of  Schleicher's  Die  deutsche  Sprache  (revised 
by  him  after  Schleicher's  death)  he  says :  *  The  forms  of  the 
German  parent  speech  I  have  left  as  Schleicher  wrote  them  .  .  . 
It  was  of  no  importance  to  reconstruct  here  the  words  in  all  their 
parts  just  as  they  actually  existed  at  some  one  definite  prehistoric 
point  of  time,  but  simply  to  restore  the  old  endings  for  the  better 
understanding  of  their  later  forms.  Whether,  for  instance,  the 
gen.  plur.  dag&m  ever  existed  in  this  form,  or  whether,  at  the 
time  when  the  gen.  plur.  terminated  in  -dm,  the  shifting  of  mutes 
had  not  yet  taken  place  and  the  form  was,  consequently,  daghdm^ 
while  after  the  shifting  of  mutes  the  real  form  was  dagd^  is  imma- 
terial for  the  purposes  of  this  book.  In  this  respect  all  forms  of 
the  German  parent  speech  are  merely  hypothetical.' 

If  we  were  to  adopt  this  method  in  the  reconstruction  of  an 
English  word  we  should  run  the  risk  of  joining  to  a  Chaucerian 
stem  an  Anglo-Saxon  prefix  and  a  nineteenth-century  suffix, 
begetting  a  monster  not  unlike  the  Chimaera,  np6a6€  XcW,  ^tti^cv  dc 
dpaxtip,  fXfiroTi  d«  xlfiaipa. 

7. — In  a  very  suggestive  article  on  *  Relative  Sprachchronologie ' 
(Indogermanische  Forschungen,  IV  (1894)),  Otto  Bremer,  after 
aUuding  (p.  8)  to  the  chronological  difficulties  just  treated,  which 


Digitized  by  VjOOQ IC 


420  AMERICAN  JOURNAL    OF  PHILOLOGY. 

lead  us  to  assign  two  words,  of  which  one  is  perhaps  much  older 
than  the  other,  to  the  same  preliterary  period,  or  cause  a  similar 
error  by  uniting  in  the  same  word  phonetic  changes  belonging  to 
entirely  different  periods,  offers  a  most  excellent  illustration  of  a 
third  obstacle  in  the  way  of  reconstruction  of  parental  forms,  viz. 
that  the  antiquity  of  a  sound-change  cannot  be  measured  by  the 
frequency  of  its  occurrence  in  the  individual  languages.^  This 
theory  stands  and  falls  with  the  linear  theoiy  of  the  disintegration 
of  the  Indo-European  primitive  stock,  which  pictures  it  in  the 
form  of  a  genealogical  tree.  It  becomes  theoretically  untenable 
as  soon  as  the  latter  hypothesis  is  discarded.  And  Bremer  has 
shown  \\.  practically  ^NXOXi%  in  the  concrete  example  of  a  change 
confined  to  the  Anglo-Frisian  for  which,  on  direct  evidence, 
greater  antiquity  may  be  claimed  than  for  certain  changes  which 
are  common  to  all  Germanic  dialects. 

8. — And  finally  we  must  base  our  reconstructions  on  individual 
forms  which  have  behind  them  a  most  unequal  stretch  of  inde- 
pendent development.  A  Vedic  form  is  separated  from  an  Alba- 
nian form  by  at  least  3000  years.  And  the  problem,  as  was  early 
enough  recognized  and  admitted,  is  in  reality  not  the  reconstruc- 
tion of  the  parent  on  the  basis  of  descendants  of  the  first  degree, 
but  on  the  basis  of  an  aggregate  of  descendants  of  very  different 
degrees,  descendants  which  have  undergone  an  independent 
development  of  very  unequal  duration,  during  which  unknown 
external  forces  have  had  an  opportunity  of  variously  affecting 
them. 

9. — We  have  seen  above  how  intimately  linked  questions  of 
chronology  are  with  the  geographical  notions  held  with  reference 
to  the  spread  of  the  Indo-Europeans  over  the  territory  which 
they  hold  in  historical  times. 

To  Schleicher  the  Indo-European  parent  people  was  a  nation 
limited  in  numbers,  inhabiting  a  comparatively  small  area  some- 
where in  Asia,  whence  issued  forth,  from  time  to  time,  migratory 
expeditions  which  settled  down  in  new  homes  more  or  less 
removed  from  the  old  parent  stock,  and,  breaking  intercourse 
with  it,  started  on  a  line  of  independent  development.  Though 
remnants  of  this  theory  yet  linger  on,  as  is  shown  by  expressions 
like  'the  parting  of  the  peoples'  or  'the  end  of  the  primitive 

^  Cf.  Paul,  Principien',  p.  41  {§46  of  the  Engl.  tr.). 


Digitized  by  VjOOQ IC 


INFERRED  PARENT  LANGUAGES,  421 

period'  and  the  like,  it  is  safe  to  say  that  it  is  at  present  generally 
discarded  and  has  been  superseded  by  the  theory  of  gradual 
expansion,  which  may  be  regarded  as  a  transfer  of  the  leading 
idea  of  Schuchardt-Schmidt's  linguistic  wave-theory  to  the  ethno- 
logical problem  of  the  spread  of  the  Indo-Europeans.  Now,  if 
migration,  as  conceived  by  Schleicher,  played  at  best  but  a  very 
small  part  in  locating  the  Indo-Europeans  in  their  present  quar- 
ters, and  we  substitute  for  it  a  gradual  expansion  progressing 
in  (roughly  speaking)  concentric  circles,  belt  after  belt  being 
added  as  time  passed  and  numbers  increased,  the  whole  aspect  is 
considerably  changed.  Instead  of  assuming  a  series  of  sudden  . 
interruptions  of  intercourse  between  members  of  the  outer  belts 
and  the  central  stock,  we  shall  rather  have  to  admit  a  constant 
communication  of  members  of  inner  and  outer  belts,  varying,  of 
course,  in  the  degree  of  intensity,  which  would  depend  on  a 
variety  of  causes  which  need  not  be  treated  here.  The  effect  of 
this  continual  interdependence,  which  thus  takes  the  place  of 
Schleicher's  independence,  is  a  slow  dissemination  over  an  ever- 
increasing  area  of  whatever  developments  in  language  or  institu- 
tions or  art  or  manufacture  may  arise  in  any  one  locality. 

lo. — But  as  soon  as  we  substitute  this  theory  of  gradual  expan- 
sion for  that  of  disintegration,  we  can  no  longer  contrast  the 
parent  stock  with  the  individual  members,  as  Schleicher  does. 
Instead  of  a  parent  stock  which  is  broken  up  into  a  number  of 
smaller  units  which,  stopping  their  intercourse  with  the  main 
body,  start  on  a  new  and  independent  course  of  development,  we 
must  posit  a  nucleus  which  develops  proportionately  to  the  outer 
belts,  which  continually  reacts  on  them,  as  they  react  on  it,  which, 
in  a  word,  is  ever  changing. 

Leaving  out  of  consideration,  for  the  moment,  all  possible 
contact  with  foreign  tribes  (though  the  influence  from  that  source 
was  certainly  very  great),  the  territory  occupied  by  the  Indo- 
Europeans,  even  in  the  course  of  a  wholly  normal  expansion, 
must  have  represented  thousands  of  years  before  our  oldest 
historical  records  begin,  an  aspect  so  diversified  that  it  would  be 
impossible  to  regard  it  as  one  national  unit,  much  less  to  assign 
to  it  one  dialect.  But  this  is  the  very  period  for  which  alone  a 
reconstruction  can  be  attempted.  The  time  for  which  uniformity 
might  theoretically  be  postulated  lies  in  so  distant  a  past  that  it  is 
altogether  beyond  our  reach. 


Digitized  by  VjOOQ IC 


422  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Brugmann  very  clearly  states  that  he  is  not  *  operating  with  the 
idea  of  one  primitive  nation  and  a  primitive,  homogeneous  parent 
speech/  a  fault  which  E.  Meyer  had  urged  against  comparative 
philologists  (Geschichte  des  Alterthums  (1884),  I,  pp.  7-8,  note) 
when  he  says  (Compend.,  Engl,  tr.,  I,  p.  2,  §3)  that  *it  is  impos- 
sible to  suppose  that  a  language  [like  the  Indo-European]  should 
have  gone  through  a  long  course  of  development  and  be  spoken 
by  a  people  of  any  considerable  number  without  a  certain  amount 
of  dialectic  variations,  and  hence  we  cannot  look  upon  the  speech 
of  the  Indo-Europeans,  even  while  they  still  occupied  a  compar- 
.  atively  small  territory  and  maintained  a  fairly  close  degree  of 
intercourse  with  one  another,  as  bearing,  in  any  strict  sense,  a 
uniform  character.  Local  differences  had,  no  doubt,  already 
arisen  .  . .  We  may  take  it  for  granted  that  the  differentiation  of 
dialects  about  the  year  2000  B.  C.  had  gone  so  far  that  a  number 
of  communities  existed  side  by  side  which  could  no  longer  or 
only  with  difficulty  understand  each  other.*  And  Bremer  almost 
verbally  coincides  with  E.  Meyer's  remark  (1.  c.) :  '  Nowhere  does 
a  homogeneous  parent  speech  exist,  but  everywhere  we  have 
dialects  influencing  one  another,*  when  he  says  (Indog.  Forsch. 
IV  (1894),  p.  10):  'Within  every  parent  language  there  existed 
at  all  times  dialectic  variations.  I  do  not  believe  that  we  shall 
ever  succeed  in  reconstructing  the  posited  Indo-European  parent 
language  in  its  main  features.  We  shall  have  to  content  ourselves 
with  the  reconstruction  of  the  dialectically  differentiated  compo- 
nents.* 

II. — After  thus  summing  up  the  various  limitations  which  are 
inherent  in  the  method  followed  in  the  work  of  reconstructing  the 
parent  language,  we  are  in  a  position  to  approach  the  second 
question  propounded  above:  Is  the  result  of  these  limitations 
merely  a  greater  or  smaller  number  of  quaniitaiive  imperfections 
in  the  reconstructed  object,  or  are  they  of  such  a  character  as  to 
affect  it  qualitatvvely  ? 

12. — It  is  plain  that  the  term  Indo-European  parent  language 
is  parallel  to  terms  like  Greek  language,  German  language  and 
the  like. 

These  latter  terms  we  use,  however,  in  two  entirely  different 
senses.  For,  as  sometimes  used,  the  term  German  language 
refers  to  the  literary  language  of  Germany,  and  the  term  Greek 


Digitized  by  VjOOQ IC 


INFERRED  PARENT  LANGUAGES,  423 

language  is  not  infrequently  used  to  designate  the  literary  Attic 
dialect.  If  thus  employed,  the  term  language  simply  denotes  a 
certain  dialect  which,  for  one  reason  or  another,  and  often  with 
admixtures  from  other  dialects,  has  gained  a  supremacy  over  its 
competitors  and  is  accepted  as  the  general  means  of  (chiefly 
literary)  communication.  The  elevation  of  a  dialect  to  a  literary 
language,  however,  is  always  a  very  late  development.  It  pre- 
supposes a  literature  and  a  strong  national  feeling  tending  toward 
centralization.  But  the  farther  back  we  go  the  weaker  becomes 
this  feeling.  So  little  realized,  in  fact,  is  the  homogeneous  char- 
acter of  large  aggregates  that  during  the  earlier  periods  very 
frequently  a  name  for  these  larger  aggregates  is  wanting.  With 
Meyer  (Gesch.  des  Alterth.  I,  p.  7,  §7,  note),  we  must  regard  the 
creation  of  a  literary  language — like  the  creation  of  a  nation — as 
the  goal  of  historical  evolution.  It  is  apparent  that  the  term 
Indo-European  parent  language  cannot  be  used  in  this  sense. 

13. — There  remains,  then,  the  second  sense  in  which  terms  like 
English  language,  German  language,  etc.,  are  used,  viz.  when  we 
employ  these  abstractions  as  classificatory  devices,  in  order  to 
arrange  a  large  mass  of  more  or  less  similar  units.  Is  the  nature 
of  these  generic  abstractions  such  as  to  make  reconstruction 
possible?  This  question  can  only  be  answered  on  the  basis  of  a 
minute  examination  of  the  method  by  which  they  are  formed. 
The  result  of  our  reconstruction  is  forms,  and  forms  are  per- 
ceptual objects.  If  a  language-form  is  a  perceptual  object  it 
permits,  theoretically,  of  reconstruction ;  if  not,  reconstruction  of 
a  language-form  is  an  impossibility.  In  the  following  paragraphs 
we  shall  therefore  attempt  an  analysis  of  the  generic  terms 
'dialect'  and  language'  with  a  view  of  ultimately  determining 
whether  dialect-forms  and  language-forms  are  perceptual  objects 
or  not. 

14. — Like  all  historical  objects,  the  language  of  a  people 
presents  static  and  dynamic  problems.  In  the  first  case  it  is 
necessary  to  regard  the  object  as  stationary,  and  our  task  is  to 
examine  the  qualities  exhibited  by  the  object  at  a  given  point  of 
time.  Extending  such  an  examination  over  a  number  of  succes- 
sive stages,  the  result  of  the  examination  of  each  stage  marks  a 
point  through  which  the  object  in  its  development  passed.  In 
the  second  case  the  object  is  considered  as  being  in  continual 


Digitized  by  VjOOQ IC 


424  AMERICAN  JOURNAL    OF  PHILOLOGY. 

motion,  and  our  task  now  is  to  determine  the  forces  which  govern 
this  motion.^ 

For  the  full  understanding  of  a  given  object  it  is  essential  that 
it  should  be  investigated  by  both  methods.  Such  knowledge  is 
made  up  of  the  results  of  both  descriptive-historical  and  explan- 
atory-analytical treatment. 

15. — But  popular  concepts  are  not  the  result  of  such  a  purely 
scientific  investigation.  A  large  part  of  the  elements  of  which 
popular  concepts  consist  is,  no  doubt,  of  a  static  or  dynamic 
character.  But  a  naive  observation  couples  with  them  elements 
which  cannot  be  assigned  to  either  class,  elements  which  are  in 
no  way  inherent  in  the  object  itself,  but  connected  with  it  by 
external  ties,  viz.  temporal  or  local  contiguity.  I  propose  to  call 
these  elements  *  associative  elements.'  So  elements  of  the  percept 
*lamp'  may  associatively  enter  into  the  concept  'light,'  or  those 
of  *a  court  of  law'  into  that  of  'justice."  For  practical  purposes 
this  associative  admixture  causes  little  or  no  inconvenience  because 
the  total  picture  is  of  sufficient  clearness.  When,  however,  these 
same  terms  are  used  for  scientific  purposes  the  heterogeneous 
character  of  their  composition  gives  rise  to  much  ambiguity  and, 
consequently,  of  controversy.  In  this  case  it  becomes  a  matter  of 
importance  to  distinguish  between  the  various  elements  which 
make  up  the  concept,  especially  with  a  view  to  remove  the 
dangerous  associative  elements. 

16. — How,  then,  does  the  concept  of  a  dialect*  originate,  and  of 
what  character  are  the  elements  composing  it? 

The  concept  originally  is  not  the  result  of  scientific  investigation, 
but  of  naive  observation.  The  naive  person  expects  every  one  to 
talk  like  himself.    Hence  the  fact  that  his  neighbor  talks  like 

^  Quite  similarly  we  may,  in  mathematics,  regard  a  curve  eithfr  as  a  system 
of  discrete  points  or  as  the  track  of  a  point  moving  under  the  influence  of 
certain  forces.  And  no  bridge  leads  from  the  system  of  discrete  points  to  the 
continuum. 

'  It  is  especially  where  the  other  elements  are  weak,  indistinct,  and  insuffi- 
cient to  produce  a  clear  concept,  that  the  latter  is  supported,  as  it  were,  by  a 
frame-work  of  associative  elements. 

'  The  most  important  points  affecting  the  scientific  study  of  dialects  were 
brought  out  in  the  controversy  regarding  the  boundaries  of  Romance  dialects, 
which  is  admirably  summarized  by  A.  Homing  in  Zt.  f.  roman.  Philol.  (1893), 
XVIII  160  c  ff,  Cf.  also  Paul's  second  chapter,  where  im  wesentlichm  einkeit- 
lick  (pp.  35,  37)  equals  my  'subjectively  uniform.* 


Digitized  by  VjOOQIC 


INFERRED  PARENT  LANGUAGES,  42$ 

himself  fails  to  arouse  his  attention  or  interest.  This  fact,  indeed, 
is  not  noted  by  him  until  he  is  confronted  by  a  group  of  indi- 
viduals differing  from  him  in  their  speech.  The  canlrasi  for  the 
first  time  makes  him  realize  the  identity  of  speech  of  himself  and 
the  members  of  his  group.  This  speech-identity  of  his  group  he 
conceives  of  as  the  dialect  of  his  group. 

Dynamic  elements,  therefore,  originally  never  enter  into  the 
make-up  of  this  concept  They  are  without  value  for  the  imme- 
diate purpose  for  which  the  concept  was  created ;  for  the  forces  to 
which  identity  and  diversity  of  speech  are  due  have  no  direct 
bearing  on  the  contrast  between  'like  speech'  and  'unlike  speech.' 

Static  elements,  on  the  other  hand,  are  largely  present.  In 
calling  the  speech-identity  of  a  group  its  dialect  we  have  com- 
bined in  this  concept  a  large  number  of  judgments  passed  on  the 
quality  of  the  speech  of  a  certain  number  of  individuals,  singling 
out  their  speech  from  that  of  the  rest  and  claiming  likeness  for  it. 
The  term  dialect  thus  expresses  a  certain  relation  of  the  speech 
of  some  individuals  to  that  of  other  individuals.  It  must  vary  as 
this  relation  varies,  and  a  dialect,  Z7,  may  be  regarded  as  a 
function  of  this  relation,  R  :  D^ F {R). 

But  the  concept  of  a  dialect  is  not  wholly  made  up  of  static 
elements.  Speech  is  indissolubly  linked  to  the  speaking  indi- 
viduaL  And,  consequently,  wholly  heterogeneous  elements  asso- 
ciatively  enter  into  our  concept  which,  for  want  of  a  better  term, 
might  be  called  '  ethnological.' 

After  the  contrast  of  speech  of  two  groups  A  and  B  had  been 
noted  and  found  expression  in  the  formation  of  the  concepts 
'/^-dialect'  and  '^-dialect,'  it  became  evident  that  these  dicUecial 
groups  corresponded  to  certain  political  groups.  And  the  more 
normal  and  primitive  the  conditions,  the  closer  must  have  been 
the  similarity  between  these  groups,  the  stronger,  therefore,  also 
the  associative  tie  by  which  they  were  held  together.  The  inevi- 
table result  was  2l  fusion  in  which  elements  of  one  concept  passed 
over  into  the  other.  Thus  the  concept  of  a  dialect,  which  arose 
from  the  necessity  of  marking  the  relation  of  a  certain  kind  of 
speech  to  another  kind  of  speech,  by  this  admixture  of  ethno- 
logical elements  departs  somewhat  from  its  original  connotation 
and  comes  to  be  used  not  only  with  reference  to  a  certain  relation 
existing  between  two  kinds  of  speech,  but  also  denoting  a  given 
speech  as  characteristic  of  a  given  political  group ;  and  thus  part 
of  its  purely  abstract  character  is  lost. 


Digitized  by  VjOOQ IC 


426  AMERICAN  JOURNAL   OF  PHILOLOGY. 

17. — ^After  we  have  thus  determined  the  character  of  the 
elements  of  which  the  popular  concept  of  a  dialect  is  composed, 
we  turn  to  examine  somewhat  minutely  the  exact  manner  of 
procedure  in  the  formation  of  this  concept. 

The  knowledge  which  we  obtain  concerning  speech  is  either 
subjective  or  objective. 

The  knowledge  which  is  based  upon  the  direct  acoustic  sense- 
impressions  conveyed  to  our  brain  by  the  speech-sounds  I  term 
subjecHve, 

Objective  knowledge  of  speech,  on  the  other  hand,  is  based  on 
a  direct  examination  of  the  stimuli  producing  our  sensations. 

Neither  one  of  these  two  methods  can  rightly  claim  a  superiority 
over  the  other.  Both  alike  are  experimental.  They  differ  only 
in  that  the  objects  of  investigation  differ.  In  the  former  case  we 
examine  sensations^  in  the  latter  case  stimuli.  Their  results, 
therefore,  can  never  be  said  to  conflict.  For,  if  the  results 
obtained  by  one  method  are  not  like  those  obtained  by  means  of 
the  other,  the  diversity  merely  shows  that  sensation  and  stimulus 
are  two  different  things. 

18. — In  the  naive  observation  which  formed  the  concept  of  a 
dialect  the  objective  method  played  no  part.  It  was  formed 
wholly  subjectively y  i.  e.  it  is  based  on  sensations  only,  not  on  a 
knowledge  of  the  stimuli  which  gave  rise  to  these  sensations. 

Such  subjective  knowledge  is  characterized  by  these  qualities : 

I.  Our  sensations  are  imperfect.     For — 

(a)  They  are  of  moderate  sensitiveness.  Certain  stimuli  are 
not  perceived  at  all.  There  is  an  upper  and  lower  limit  for 
audible  tones ;  variations  of  a  stimulus  within  certain  bounds  are 
not  discovered ;  etc. 

{V)  They  are  subject  to  deception,  as  in  the  case  of  visual 
illusions. 

II.  They  lack  uniformity.  For  their  degree  of  accuracy 
depends — 

{a)  On  practice,  as  in  judging  distances,  weights,  etc.,  and 
(^)  On  attention.  This  is  of  especial  importance  where,  as  in 
speaking,  a  complex  object  (the  spoken  word)  can  be  observed 
for  a  short  time  only.  As  it  is  possible  to  attend  to  only  one 
thing  at  a  time,  a  short  observation-time  will  necessarily  prevent 
all  qualities  of  the  object  from  being  equally  attended  to. 


Digitized  by  VjOOQ IC 


INFERRED  PARENT  LANGUAGES,  AV 

From  this  it  appears  that  in  subjectively  forming  the  concept  of 
a  dialect  we  may  d.  priori  assume — 

{a)  That  certain  stimuli,  though  present,  were  disregarded 
because  they  were  not  perceived. 

{p)  That  the  ratio  of  two  or  more  sensations  permits  no  direct 
inference  as  to  the  ratio  of  the  corresponding  stimuli. 

{c)  That  the  results  must  vary  in  direct  proportion  to  both 
practice  and  attention  of  the  observer. 

19. — By  this  method  the  naive  observer  classifies  the  speech  of 
the  individuals  surrounding  him,  and,  as  we  saw  above,  by  a 
ficrd/3a<riff  into  the  ethnological  ycVor,  these  individuals  themselves. 
The  speech  which  is  like  his  he  groups  into  one  class ;  the  speech 
which  is  different  from  his  into  a  second  dass.  As  in  all  classifi- 
cation, he  thus  simplifies  the  comprehension  of  a  large  number  of 
individual  objects.  Like  all  generic  names,  the  name  of  a  dialect 
does  not  stand  for  any  perceptual  object,  but  expresses  a  peculiar 
relation  of  a  series  of  perceptual  objects.  It  stands,  not  for  a 
sense-percept,  but  for  the  particular  manner  in  which  we  have 
viewed  and  grouped  a  number  of  sense-percepts. 

20. — Two  ways  are  open  for  such  classification : 

I.  We  may  begin  by  tracing  a  certain  system  of  boundary  lines 
within  which  we  include  kindred  objects.  But  very  frequently 
such  boundary  lines  cannot  be  drawn,  and  in  their  stead  we  have 
boundary  zones.  J.  Simon  and  others  experimentally  showed 
this  to  be  the  case  in  dialect-boundaries,  and  many  similar 
instances  might  be  added. 

II.  But  instead  of  starting  from  the  periphery,  we  may  also 
select  a  center  around  which  a  number  of  kindred  objects  are 
grouped  in  concentric  circles,  the  radius  of  these  circles  being 
inversely  proportional  to  the  degree  of  similarity  with  the  center. 

If  we  choose  for  such  a  center  one  of  the  many  concrete  objects 
which  are  to  be  classified,  I  propose  to  call  this  a  concrete  center. 
If,  on  the  other  hand,  we  construct  the  center  on  the  basis  of  the 
concrete  objects,  none  of  them  being  absolutely  identical  with  it, 
I  will  call  this  an  ideal  center. 

21. — Let  us  first  examine  the  manner  in  which  such  an  ideal 
center  may  be  constructed.  We  must  distinguish  here  between 
two  possibilities : 


Digitized  by  VjOOQ IC 


428  AMERICAN  JOURNAL    OF  PHILOLOGY. 

I.  If  we  classify  single  qualities  expressible  in  numbers  (e.  g. 
weight,  distance,  etc.),  the  ideal  center  is  equal  to  the  mean  of 
these  quality-numbers.^ 

Around  the  mean  thus  obtained  the  variations  may  be  grouped. 

And  such  a  classification  is  of  especial  interest  because  certain 
mathematical  theories  may  be  directly  brought  to  bear  on  it. 
For  Quetelet  showed  in  1846  (Lettres  sur  la  th6orie  des  probab. 
appliq.  aux  sciences  mor.  et  pol.  Lett.  XVIII  119)  that  the 
different  variations  grouped  around  such  a  mean  may  be  regarded 
as  so  many  fallible  measurements  of  this  same  mean,  and  that 
therefore  the  Law  of  the  Frequency  of  Error  may  be  applied  to 
them.'    We  shall  return  to  this  point  below  (§26). 

II.  But  if  we  classify  not  single  qualities^  but  whole  objects,  we 
construct  our  ideal  center  in  a  somewhat  different  way. 

We  begin  by  comparing  all  objects  {0^0^ . . .  ^«)  as  to  their 
qualities.     It  will  then  appear — 

(a)  That  certain  qualities  are  present  in  the  same  degree  or 
manner  in  all  objects  (constant  qualities). 

ip)  Certain  qualities  are  present  in  all  objects,  but  not  in  the 
same  degree  or  manner  (variable  qualities). 

(^)  Certain  qualities  are  present  in  some  objects  and  absent  in 
others  (variable  qualities). 

Our  ideal  center,  (?,  must  then  be  constructed  in  such  a  manner 
that  it  will  contain  all  qualities  enumerated  above  under  (a)  and 
those  qualities  enumerated  under  (^)  and  (c)  in  the  most  charac- 
teristic manner  or  degree,  by  which  is  meant  that  manner  or 
degree  which  will  permit  the  variations  as  they  appear  in  the 

^  Whether  this  mean  should  be  the  arithmetical  mean, 


nf     ^i  +  ga +g» 


or  the  geometric  mean. 


M^z^/a^a^  ....  a.. 


must  depend  on  the  nature  of  the  case.  If,  e.  g.,  we  have  a  series  of  weights, 
the  weight-center  will  be  the  arithmetical  mean.  If,  on  the  other  hand,  we 
deal  with  such  sensations  as  come  under  Weber-Fechner's  law  (viz.  sensation 
=  log.  stimulus),  the  geometric  mean  must  be  substituted  (cf.  Galton,  Proc. 
Roy.  Soc.  Lond.  (1879),  XXIX  365). 

'Cf.  on  this  also  Stieda,  Archiv  f.  Anthropol.  XIV  167;  Galton,  Proc.  Roy. 
Soc.  (1879),  XXIX  365 ;  McAlister,  ibid.,  p.  367 ;  Galton,  ibid.  (1889).  XLV 
135,  and  the  applications  by  Galton  (above),  Davenport  and  BuUard,  Proc.  Am. 
Ac.  Arts  and  Sci.  (1897),  XXXII,  No.  4,  and  Brewster,  ibid..  No.  15. 


Digitized  by  VjOOQ IC 


INFERRED  PARENT  LANGUAGES.  429 

concrete  objects  to  be  most  easily  deduced  from  it.  This  deduc- 
tion in  all  historical  sciences  is  genetic.  The  qualities  falling 
under  {U)  and  {c)  will  therefore  be  given  to  O  in  such  degree  and 
manner  as  will  make  it  possible  to  regard  them  as  the  sources 
from  which  the  individual  variations  may  have  developed.  I  say 
advisedly  ^may  have  developed/  because  by  such  a  method  of 
comparison  we  can  never  obtain  results  for  which  more  than  a 
possibility  can  be  claimed.  Certainty  could  only  be  gained  by 
experimentally  watching  the  actual  progress  of  development 
Strictly  speaking,  we  do  not  reconstruct  'parent '-forms,  but  we 
construct  them.  Only  when  we  have  been  able  to  observe  an 
object  during  its  period  of  evolution  are  we  able  to  reconstruct  by 
retracing  the  course  of  development.  For  in  this  case  the  course 
of  development  is  given.  In  the  construction  of  'parent '-forms 
by  the  comparative  method,  however,  the  course  of  development 
from  the  unknown  'parent '-form, _;/,  to  the  present  form,  a,  is  not 
experimentally  determined;  it  is  not  given,  but  inferred.  We 
deal  here  with  two  unknown  quantities.  And  this  makes  it  very 
problematic  that  the  result  of  such  a  construction  will  be  exactly 
identical  with  the  real  prehistoric  'parent '-form.  It  will  be  more 
or  less  similar  to  it ;  but  real  identity  would  be  a  mere  matter  of 
chance,  obtained  rather  despite  of  our  method  than  by  means  of 
it.    And  cogent  proof  of  sUch  identity  must  always  be  lacking.^ 

A  comparison  of  any  one  concrete  object  o,  with  the  ideal 
center  O  will  then  show  that  o^  varies  from  O  either  in  lacking  a 
quality  which  O  has,  or  in  possessing  a  quality  which  O  lacks,  or 
in  possessing  a  quality  in  a  degree  or  manner  differing  from  that 
ofO. 

22. — The  ideal  center  constructed  in  the  preceding  paragraph 
has,  of  course,  no  perceptual  existence.  But  suppose  that  after 
the  construction  of  such  an  ideal  center  it  should  be  found  that 
one  of  the  concrete  objects  to  be  classified  shows  no  variation 
from  it,  that,  e.  g.,  o^  =  O. 

In  this  case  it  is  plain  that  we  might  discard  O  altogether  and 
substitute  o^  in  its  place.  This  concrete  object  o^  would  then 
appear  in  a  double  rdle,  v\z.  first  as  one  of  the  many  concrete 
objects  forming  the  series  0^0^ . .  •  ^r,  and  second  as  ideal  center 
or  type  of  this  series. 

iCf.Wundt.  Logik,  II47. 


Digitized  by^ 


430  AMERICAN  JOURNAL    OF  PHILOLOGY. 

And  in  ihis  case  the  ideal  center  really  does  possess  perceptual 
existence,  and  we  distinguished  it  from  the  ideal  type  of  §21  by 
calling  it  concrete  type. 

23. — Whenever,  therefore,  we  have  to  deal  with  generic  terms 
or  types  we  must  examine  in  each  case  whether  we  have  to  do 
with  an  ideal  or  a  concrete  type.  This  examination  will  be  our 
next  task,  and  by  it  I  hope  to  show  that  a  dialect-form  differs 
from  a  language-form  in  that  the  former  is  a  concrete  type,  while 
the  latter  is  an  ideal  one. 

24. — For  this  purpose  it  is  necessary  to  investigate  those  indi- 
vidual objects  which,  in  the  manner  discussed  in  §21,  II,  are  fused 
into  the  generic  concept  of  a  dialect.  These  elements  are,  of 
course,  the  speech-forms  of  the  various  members  of  the  dialectal 
unit,  which  may  be  designated  as  U^U^. . .  U^.  But  'speech- 
form  of  a  member  of  a  dialectal  unit*  is  itself  a  generic  concept. 
It  is  based  on  the  sum  of  momentary  utterances  (UiU^ . . .  2<.)  of 
each  member,  and  our  attention  must  therefore  be  first  directed 
toward  these  momentary  utterances. 

25. — The  basis  for  any  given  momentary  utterance  («)  of  an 
individual  is  a  certain  psycho-physical  condition  or  diathesis  A.^ 
In  this  respect  language  does  not  differ  from  any  other  move- 
ment. As  the  expressive  movement  of  a  gesture  affects  our 
sight,  so  the  expressive  movement  which  gives  rise  to  the  spoken 
word  affects  our  hearing.  As  a  repeated  gesture  is  not  the  same 
as  the  first  original  gesture,  so  the  repeated  utterance  is  not  the 
same  as  the  first  original  utterance.  Neither  the  gesture  nor 
utterance  has  a  latent  existence  during  the  interval;  but  the 
original  gesture  or  utterance  on  the  one  hand,  and  the  repeated 
gesture  or  utterance  on  the  other  hand,  are  linked  together,  not 
directly,  but  indirectly  by  the  psycho-physical  diathesis  of  which 
they  are  respectively  the  results.  This  diathesis  remains;  its 
results  absolutely  vanish.  Consequently  the  repeated  momentary 
utterances  do  not  exist  independently  of  each  other,  but  as  effects 
of  their  respective  psycho-physical  diathesis;  so  that  as  long 
as  this  diathesis  remains  the  same,  the  utterances  will  remain  so 
also. 

^  Paul,  I  believe,  was  the  first  to  introduce  these  psychological  and  physio- 
logical considerations  into  linguistic  literature. 


Digitized  by  VjOOQ IC 


INFERRED  PARENT  LANGUAGES,  43 1 

But  while  we  may  thus  speak  of  repeated  utterances  as  results 
of  a  given  diathesis,  it  is  conversely  true  that  this  diathesis  itself 
is  in  turn  the  product  of  all  the  speech  movements  which  have 
gone  before.  For  the  strength  of  the  diathesis  depends  on 
practice.  The  constancy  of  a  diathesis  is  proportionate  to  the 
number  of  repetitions  of  the  movement,  and  the  probability  that 
a  given  movement  will  be  performed  in  a  given  way  is  the 
stronger  the  oftener  such  movement  has  been  so  performed. 

The  first  utterance,  ^^  creates  a  weak  diathesis,  A,  on  account 
of  which  a  second  utterancej  «„  will  be  similar  to  u^  \  but,  like 
every  subsequent  utterance,  «,  will  react  on  A  and  strengthen  it. 
In  the  adult,  therefore,  the  diathesis,  under  normal  conditions, 
must  be  constant  and  the  utterances  belonging  to  it  alike. 


A 

(Because  A  is  of  increasing  stability,  fi^  =  ti,  =  .  .  .  m..) 

26. — It  must  be  borne  in  mind  that  in  these  observations  like- 
ness and  unlikeness  are  determined  by  purely  subjective  methods. 
It  is  not  denied  that  variations  may  exist  and  could  be  discovered 
by  an  objective  examination.  All  that  is  claimed  is,  that  if  such 
variations  exist  they  are  not  perceived  as  such,  partly  because 
our  senses  are  not  keen  enough,  partly  because  our  attention  is 
not  directed  to  them, /ar//^  because  we  compare  sensations  which 
do  not  follow  each  other  in  quick  succession,  but  the  memory  of 
one  sensation  is  separated  from  the  next  similar  sensation  by  a 
longer  or  shorter  interval,  and  partly  for  the  following  reason : 
Suppose  that  we  have  n  variations  (v{v^ .  .  •  z^J  grouped  around 
the  type  or  mean  V.  Suppose,  further,  that  of  these  n  variations 
a  few  lying  close  to  Vi  and  v,  (i.  e.  close  to  either  extreme)  are 
sensibly  perceptible ;  but  that  those  variations  lying  between  v, 

X 9k K * * 

and  z/y  are  not  perceptibly  different  from  the  type  V.  The  Law 
of  the  Frequency  of  Error  teaches,  then,  that  of  all  n  variations 
the  greatest  number  is  bunched  closely  around  V.  That  is,  by 
far  the  greater  percentage  of  these  n  variations  must  fall  between 


Digitized  by  VjOOQ IC 


432  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Vj,  and  z/„  and  is  therefore  sensibly  perceived  as  equal  to  V. 
Consequently  the  effect  of  n  variations  on  the  diathesis  is  not 
alike.  For  a  large  number  of  these  n  variations,  being  perceived 
as  F,  strengthens  the  diathesis,  and  the  number  of  variations 
whose  difference  from  V  is  perceived  is  in  many  cases  far  too 
weak  to  act  as  a  disturbing  element. 

27. — We  have  seen  in  §25  that  repeated  momentary  utterances 
of  the  same  individual  are  subjectively  perceived  as  alike.  If  we 
now  form  the  type  U  of  the  whole  series  of  these  momentary 
utterances  («i . .  .  ««),  we  may,  under  these  conditions,  take  any 
u  as  such  a  type,  and  we  thus  obtain  a  concrete  type : 

We  may,  in  other  words,  take  a  given  momentary  utterance  of  an 
individual,  say  «„  as  representative  of  his  average  utterance  U^ 
because  there  is  an  overwhelming  probability  that  the  diathesis 
which  gave  rise  to  «„  and  itself  was  the  product  of  the  whole 
series  u^  to  «„  will  produce  an  Uy  and  «.  which  will  be,  subjec- 
tively, like  «,. 

28. — Having  thus  determined  of  what  character  the  average 
utterance  of  an  individual  is,  we  must  now  compare  the  average 
utterances  i/,  C/, ...  6^  of  the  various  members  of  a  dialectal 
unit,  on  which,  as  we  saw  above,  our  concept  of  a  dialect  is 
founded. 

Now,  at  the  time  when  the  concept  of  a  dialect  was  first  formed 
they  must  have  been  subjectively  alike,  for  this  very  likeness  was 
the  cause  for  combining  them  into  a  class.  And  if  all  C/'s  were 
alike,  their  bases,  viz.  the  respective  diatheses  (AjA, .  . .  A«)  of  the 
various  members  of  the  dialectal  group,  must  have  been  very 
similar. 

Again,  in  the  case  of  normal  accessions  to  the  dialectal  group» 
viz.  by  birth,  the  diathesis  of  each  child  was  formed  by  the  sum- 
total  of  its  utterances.  And  these  utterances  being,  consciously 
and  unconsciously,  fashioned  after  the  utterances  of  its  surround- 
ings, would  naturally  produce  in  each  child  a  diathesis  similar  to 
the  diatheses  around  it. 

But  suppose  that,  for  reasons  which  need  not  be  discussed 
here,  new  members  of  a  dialectal  group  should  perceptibly  differ 
from  the  rest.     In  this  case  we  may  plainly  see  how  the  admixture 


Digitized  by  VjOOQ IC 


INFERRED  PARENT  LANGUAGES. 


433 


of  the  heterogeneous  ethnological  elements  (cf.  end  of  §i6)  will 
tend  to  vitiate  the  very  connotation  which  the  concept  originally 
possessed.  For  we  have  seen  that  it  was  devised  to  denote  a 
likeness^  to  unite  in  one  class  the  speech  of  individuals  talking 
alike.  To  group  together  a  number  of  individuals  in  a  dialectal 
group  when  their  speech  differs  is  plainly  a  contradiction  in 
terms.  And  such  grouping  does,  in  fact,  not  rest  on  the  basis  on 
which  the  original  concept  of  the  dialect  was  formed,  but  on  an 
entirely  different,  heterogeneous  basis,  viz.  sameness  of  origin  or 
naHonalUy.  The  introduction  of  this  double  standard  is  the 
source  of  vagueness  and  ambiguity,  to  which  reference  was  made 
in  §i6.  And  for  scientific  purposes  it  is  certainly  essential  to 
remove  from  the  concept  of  a  dialect  these  heterogeneous,  ethno- 
logical elements  and  confine  it  most  strictly  to  its  original  sense. 
Suppose,  then,  that  new  members  added  to  the  political  group 
which,  up  to  that  time,  had  continued  to  be  identical  with  the 
dialectal  group,  do  perceptibly  differ  in  speech  fi'om  the  rest.  It 
will  simply  mean  that  this  identity  of  political  group  and  dialectal 
group  has  ceased,  and  that  we  now  have  two  (or  more)  dialectal 
groups  within  the  same  political  group. 

From  these  observations  it  will  be  apparent  that  from  the  very 
definition  of  a  dialectal  group  we  must  assume  all  U'%  of  its 
members  to  be  subjectively  alike,  from  which  a  corresponding 
similarity  of  the  respective  diatheses  (Aj  . . .  A.)  may  be  inferred. 

The  diagram  below  may  serve  to  represent  the  relations  to  the 
dialect  D  of  the  diatheses  a^a^.  of  the  various  members  of  a 
dialectal  unit ;  of  the  average  speech  of  each  member,  U^  U^  Z/, ; 
and  of  the  momentary  utterances  of  three  such  members,  viz. 
u^ii^  and  tJxti<^u\  and  ti{t£iu\ : 


/N 


A. 


Now,  if 
then 


A, 
V 

A. 

A,  =  A,  =  A„ 


ttj  =  tt,  =  tt.  =  tti  =  tti  =  «1  =  jtf  =  «;  =  «;,- 


Digitized  by  VjOOQ IC 


434  AMERICAN  JOURNAL    OF  PHILOLOGY. 

also  U,=zU,  =  U.\ 

also  U,  =  U^  =  U^  =  D. 

And  because  17^  =  u^,  therefore  D  also  =  u^. 

Or,  in  other  words,  any  momentary  utterance  (u)  of  any  member 
of  a  dialectal  unit  may  be  taken  as  a  type  of  the  dialectal  utter- 
ance. And  because  any  t^  is  a  concrete,  perceptual  entity,  all 
dialectal  utterances  are  cancrete,  perceptual  entities. 

29. — We  may  therefore  now  define  thus : — A  dialect  is  the  sum 
of  all  dialectal  utterances.  A  dialectal  utterance  is  the  type  of  the 
average  (typical)  utterances  of  the  members  of  a  dialectal  unit. 
This  average  utterance  is  subjectively  equal  to  any  one  momen- 
tary utterance.  The  type  referred  to  is  therefore  concrete,  and 
any  one  momentary  utterance  of  a  member  of  a  dialectal  group 
may  be  taken  as  representing  a  dialectal  utterance.  A  dialectal 
unit  is  constituted  by  the  speech  of  all  those  persons  in  whose 
utterances  variations  are  not  sensibly  perceived  or  attended  to. 
A  dialectal  unit,  especially  at  first,  may  coincide  with  an  ethno- 
logical unit,  but  such  coincidences  grow  rarer  as  development 
continues. 

30. — There  finally  remains  to  be  examined  the  term  *  perceptible 
variation'  which  has  been  used  throughout,  and  which  we  have 
found  to  be  the  one  criterion  according  to  which  a  dialectal  group 
must  be  determined.  The  more  exactly  we  can,  therefore,  draw 
the  line  between  those  variations  which  are  subjectively  perceived 
and  those  which  are  not  so  perceived,  the  more  sharply  shall  we 
be  able  to  distinguish  what  lies  within  a  dialectal  group  from  that 
which  lies  without. 

There  appears  to  be  but  one  method  of  ascertaining  whether 
two  utterances  are  subjectively  felt  to  agree  or  to  differ,  and  that 
is  to  take  the  testimony  of  the  persons  whose  sensations  form  the 
subject  of  our  inquiry.  There  is  indeed  no  other  way  of  deter- 
mining a  dialectal  group  than  to  take  the  testimony  of  the  men 
who  are  to  compose  it.  For  the  very  reason  that  the  concept  of 
a  dialect  is  formed  wholly  on  a  subjective  basis,  all  objective  tests 
are  barred  out. 

The  question  whether  the  inhabitants  of  two  villages,  A  and 
B,  belong  to  the  same  dialectal  group  can  only  be  answered  on 
the  testimony  of  the  villagers  as  to  whether  they  believe  they 


Digitized  by  VjOOQ IC 


INFERRED  PARENT  LANGUAGES.  435 

Speak  alike.  They  are  the  court  of  last  resort,  from  which  there 
is  no  appeal.  And  occasionally  a  nickname  or  a  jest  will  be 
prima  facie  evidence  of  the  fact  that  the  villagers  of  A  clearly 
feel  the  difference  of  their  speech  from  that  of  B. 

But  he  who  is  unwilling  to  operate  with  so  uncertain  and 
variable  a  quantity  as  a  dialect  thus  appears  to  be,  will  be  forced 
to  discard  this  subjective  attempt  at  classification  and  substitute 
for  it  some  new  objective  arrangement. 

31. — In  §26  we  grouped  n  objective  variations  around  a  center 
Fand  assumed  that  the  variations  lying  between  v^  and  v^  were 
not  sensibly  perceptible,  while  those  which  lie  between  v^  and  v^ 
and  between  v^  and  v^  were  sensibly  perceptible : 

• 1 ^ ■ • 1 

And  we  have  seen  in  the  preceding  paragraph  how  the  two 
points,  v^  and  v^^  can  be  experimentally  fixed.  Whatever  lies 
between  them  is  intra-dialectal ;  what  lies  outside  is  extra-dialectal. 
These  extra-dialectal  variations  (between  v^  and  Vy  and  between 
v^  and  z/,)  are  alike  in  that  they  are  always  sensibly  perceived ; 
they  differ  in  the  degree  in  which  they  affect  the  ease  and  clear- 
ness with  which  a  given  word  may  be  understood.  For,  as  con- 
veyance of  ideas  is  the  chief  aim  of  language,  everything  which 
stands  in  the  way  of  an  utterance  being  understood  is  of  the 
greatest  moment.  But  it  is  plain  that  the  nearer  the  center  a 
perceptible  variation  lies,  the  more  easily  will  the  utterance  con- 
taining it  be  understood ;  the  farther  away  from  the  center  it  lies, 
the  more  will  it  interfere  with  the  understanding  of  the  utterancei 
until  it  absolutely  prevents  the  utterance  from  being  understood. 

32. — If  we  now  continue  our  classification  of  speech  along  the 
same  lines  which  led  to  the  formation  of  the  concept  of  a  dialect, 
we  may  proceed  to  unite  two  or  more  dialects  into  a  dialect- 
family  ;  two  or  more  dialect-families  into  a  language ;  and  two  or 
more  languages  into  a  language-family.  But  whereas  a  dialect- 
form,  as  we  have  seen  (§28,  end),  is  a  concrete  type  and  hence  a 
perceptual  entity,  the  speech-forms  of  the  types  enumerated  here 
are  ideal  types  and  have  no  perceptual  existence.  No  one 
concrete  utterance  belonging  to  any  of  these  three  classes  may, 
as  in  the  case  of  a  dialect,  be  taken  as  concrete  type,  because  in 


Digitized  by  VjOOQ IC 


436  AMERICAN  JOURNAL    OF  PHILOLOGY. 

each  such  concrete  utterance  one  or  more  variable  elements  per- 
ceptibly  vary.  The  typical  utterance  is  here  similar  to  all  con- 
crete utterances,  but  like  none.  Remove  the  perceptible  variation^ 
and  these  classes  revert  into  the  dialect  whose  distinguishing 
mark  is  the  imperceptible  variation  of  its  variables.  The  whole 
may  be  illustrated  by  the  following  diagram,  in  which  the  classes 
are  represented  as  (logical)  functions  of  a  series  of  constant  (Latin 
letters)  and  variable  (Greek  letters)  elements.  The  variable 
elements  whose  variations  are  not  subjectively  perceived  are 
enclosed  in  brackets. 

Variablb  Elbm  bnts. 


CONSTAir 

r 
t. 

t 
t 
t 
t 

Variations 
perceived. 

a    p    y 
a    /3    y 
a    b    y 
a    b    c 

Variations 
not  perceired. 

6. 
5. 
4. 
3. 

2. 

A  Language-family,        Z/  =  /"(q 
A  particular  Language,  L   =  -^  (q 
A  Dialect-family.           iy=F{q 
A  Dialect,                       £>=F{q 
Average  Utterance  of  a 
member  of  a  dialectal 

r    s 

r    s 
r    s 
r    s 

W   W) 
[-n    W) 
ra  W) 
W    W) 

I. 

unit,                                C/:=F{q 
Momentary    Utterance 
of  a  member  of  a  dia- 

r   s 

t 

a 

b 

c 

[d]    W) 

lectal  unit,                    u  r=F{q 

r    s 

t 

a 

b 

c 

[d]    W) 

33. — If  I  have  succeeded  in  showing  that  an  insurmountable 
bar  thus  separates  language-forms  from  dialectal  forms,  that  the 
latter  are  perceptual  objects  while  the  former  are  imaginary,  it 
follows  that  from  the  very  nature  of  the  case  the  perceptual 
reconstruction  of  language-forms  is  impossible.  When  a  given 
/orm  is  said  to  be  German  we  thereby  mean  that  it  is  a  member 
of  a  large  mass  which,  for  convenience  sake,  we  have  accustomed 
ourselves  to  group  together  on  account  of  certain  resemblances 
exhibited  by  all  members.  But  while  we  may  thus  classify  a 
given  form  as  German  or  Greek,  just  as  we  might  classify  a  given 
animal  as  a  bird  or  a  fish,  it  is  as  impossible  to  construct,  inferen- 
tially,  a  German  form  or  a  Greek  form  as  it  is  to  construct  a  fish 
or  a  bird.  Or,  to  be  more  exact,  the  result  of  these  constructions 
in  either  case  will  be  an  ideal  type  for  which  only  an  illogical 
hypostasis  could  claim  a  past  reality. 

*  Perceptibility  is  the  gewisse  maass  of  Paul  (p.  36)  which  variations  must 
reach  in  order  to  result  in  dialectspaltung. 


Digitized  by  VjOOQIC 


INFERRED  PARENT  LANGUAGES,  A17 

34. — It  would  be  very  rash  to  deny  the  value  of  constructive 
parent  forms  because  perceptual  reality  cannot  be  claimed  for 
them.  Their  distinct  value  lies,  however,  as  indicated  above,  in 
the  fact  that  they  are  the  means  by  which  we  classify  and  arrange 
a  given  number  of  existing  forms.  The  posited  Indo-European 
gen-  signifies  that  Latin  gen-^  Avestan  zan-y  Sanskrit  jan-,  etc., 
belong  together.  To  claim  more  means  losing  one's  self  in  a 
maze  of  speculative  possibilities.  So  it  is,  of  course,  perfectly 
proper  to  say  that  the  I.E.  possessed  the  vowels  a,  Cy  Oy  if  it  is 
borne  in  mind  that,  in  doing  so,  we  simply  maintain  that  there  is 
sufficient  evidence  to  show  that  all  prehistoric  I.E.  dialects  (which 
form  the  bases  of  the  historical  I.E.  languages)  possessed  these 
vowels  at  stages  of  their  development  antedating  the  historical 
epochs.  This  evidence,  however,  is  not  sufficient  to  warrant  a 
claim  that  these  three  vowels  belonged  to  the  whole  prehistoric 
period  of  I.E.  speech,  or  to  settle  the  question  as  to  their  original 
independence.  It  is  impossible  to  prove  or  to  disprove  on  such 
evidence  any  of  the  points  involved  in  the  controversy  outlined 
by  Bechtel,  Hauptprobleme,  p.  63  f. 

The  sum  total  of  inferred  forms  does  not  give  us  a  true  picture 
of  any  language  ever  spoken ;  nay,  even  the  single  forms  cannot 
lay  claim  to  being  representatives,  true  in  every  detail,  of  words 
ever  in  actual  use.  Yet  it  is  only  by  reducing  the  results  of  our 
investigations  to  such  formulae  that  they  become  convenient 
enough  to  be  easily  handled  and  permit  a  clear  arrangement  of 
the  facts  of  a  language.  It  is  a  significant  fact  and  a  sign  of  clear 
logic  that  Schleicher's  great  successor,  Brugmann,  in  the  Grund- 
riss,  does  not  follow  his  predecessor  in  placing  on  the  title-page 
an  'Indo-European  parent  language'  alongside  of  the  historical 
languages. 

The  danger  increases  if,  after  infusing  life  into  a  parent  lan- 
guage constructed  by  our  own  hands  after  the  defective  method 
examined  above,  we  proceed  to  erect  upon  it  as  a  basis  a  lofty 
superstructure  of  mythological  or  sociological  inferences.  Inves- 
tigations of  this  character,  which  are  beset  by  enormous  diffi- 
culties even  when  carried  on  under  the  most  favorable  conditions, 
must  necessarily  see  their  ends  defeated,  if  based  upon  material 
so  unfit  because  designed  for  entirely  different  purposes.  The 
method  of  Usener  in  his  Gotternamen  (1896),  and  Wernicke 
(Pauly's  Real-Encyclop.,  2d  ed.,  vol.  II,  *  Apollo')   is  a  most 


Digitized  by  VjOOQ IC 


438  AMERICAN  JOURNAL    OF  PHILOLOGY. 

healthy  reaction,  which  will  result  in  placing  positive  results  in 
the  place  of  unprovable  and  often  improbable  hypotheses. 

There  are  certain  limitations  which  are  inherent  in,  and  common 
to,  all  historical  sciences.  Their  objects  must  be  clearly  defined 
in  space  and  in  time.  They  all  start  where  tradition,  in  one  form 
or  another,  begins.  It  is  true  that  inferences  may  be  permitted 
as  to  what  lies  beyond  this  boundary  line  which  divides  the 
historic  from  the  prehistoric.  But  these  inferences  must  be 
confined  to  the  period  immediately  preceding  the  beginnings  of 
tradition.  The  farther  they  depart  from  it,  the  more  shadowy, 
general,  and  unreal  they  become,  because  the  data  of  time  and 
space  are  wanting,  and  without  them  historical  investigation 
becomes  impossible. 

YalbUhivbrsity.  HaNNS  OERTEL. 


Digitized  by  VjOOQ IC 


IV.— CONCLUDING  NOTES  ON  THE  ORIGIN  OF  THE 
GERUND  AND  GERUNDIVE. 

In  the  A.  J.  P.,  vol.  XV,  part  2,  July,  1894,  pp.  194-216,  there 
was  published  my  first  essay  on  *  The  Origin  of  the  Gerund  and 
Gerundive,'  and  in  a  subsequent  number  (A.J.  P.  XVI,  July,  1895, 
p.  217,  n.  i)  were  given  some  'Addenda  et  Corrigenda'  thereto. 

The  first  essay  was  supplemented,  in  A.  J.  P.  XVI,  July,  1895, 
pp.  217-222,  by  'Further  Notes  on  the  Origin  of  the  Gerund  and 
Gerundive.' 

The  present  paper  consists  of  what  may  fidy  be  called  '  Con- 
cluding Notes  on  the  Origin  of  the  Gerund  and  Gerundive,' 
together  with  a  'Precise  Statement  of  My  View.' 

Before  turning  attention  to  the  further  consideration^  of  the 
main  propositions  which  I  sought  to  establish  in  the  two  papers 
already  published,'  I  should  like  to  give,  concerning  the  said 
papers  themselves,  the  following 

Addenda  ei  Corrigenda : — 

A.J.  P.  XV  198, 1.  13:  For  "This  suffix  -do  may,  so  far  as 
Latin  alone  is  concerned,  represent,"  etc.,  read:  "This  suffix  -do 
might,  so  far  as  Latin  alone  is  concerned  (that  is  to  say,  if  there 
were  no  Osc-  Umbr.  forms  of  the  Gerundive  to  be  taken  into 
account"),  represent,"  etc' 

*  See  infra,  pp.  444  sqq.  '  Namely,  A.  J.  P.  XV  194-216,  XVI  217-222. 

'There  are  two  other  remarks  on  this  page,  viz.  198,  which  may  require 
slight  amendment : 

11.  15-16:  For  "  J^dhi'  'place,*  or  rather  'make,'  since  the  Idg.  ^dhi-  has 
lost  its  meaning  '  place '  in  Italic,  and  retained  only  that  of  '  to  make,* "  read  :• 
"  J^dhi'  *  to  make'  or  *  to  place.' " — The  meaning  *  to  place'  is  preserved  appa- 
rently in  (a)  con-do  (see  Brugmann,  Grundr.  I,  §370,  p.  282),  ab-dd  (Brugmann, 
op.  cit.  I,  §507,  Rem.),  cridd  (see  Brugmann,  op.  cit.  I,  §507,  Rem.,  and  II, 
§160,  I,  p.  479;  and  also  A.  J.  P.  XV  207),  and  in  {b)  prae-ficio  (whence  prae- 
fectus)  and  in-fuid  (the  original  meaning  of  which  appears  to  have  been  *  to 
place,  put,  or  dip  something  in  something  else'). 

11.  31-33 :  For  ••  He  shews  that  in  the  middle  of  a  word  in  some  cases  we 
find  Idg.  dh  represented  in  Latin  by  ^,"  read :  "  He  shews  that  medial  Idg.  dh, 
unless  followed  by  r,  /,  or  preceded  by  r,  u  (ju)  and  perhaps  m,  is  represented 
in  Latin  by  d," 


Digitized  by  VjOOQ IC 


440  AMERICAN  JOURNAL    OF  PHILOLOGY. 

XV  199, 1.  33 :  After  the  word  ^^ dhiyam-dhd-'^  omit  the  words 
"  quoted  above." — On  a  Greek  parallel  to  this  word  see  further 
infra,  p.  441,  s.  v.  "XV  205,  11.  21-26." 

XV  201, 1.  6  and  note  2,  1.  i :  It  is  here  observed  that  "the 
suffixes  'dhc'  and  -do-  are  not  frequent  in  forming  Greek  nouns 
like  Lat.  rubidus^  etc."  I  ventured,  however,  to  hazard  one 
Greek  example  of  -do-y  namely,  /ccJpu-do-ff  'crested  lark*  (with 
which,  in  order  to  shew  that  there  was  no  objection  to  the  view 
that  its  diy-  might  come  from  a  verb,  I  compared  Gk.  jcepacr-^/xx 
K€po'(t>6pos  and  Lat.  comi-fer  comi-ger). 

Beside  the  suffix  -do-  thus  seen  in  K6pv'do-s,  we  may  possibly  see 
the  suffix  'dhO'  in  the  parallel  bird-name  KSpv-Bo-s  (if  rightly  thus 
analysed)  *  crested  trochilus.' 

In  a  recent  article  in  Bezzenberger*s  Beitrage,  XXII,  pp.  128- 
130,  Dr.  W.  Prellwitz  furnishes  us  with  a  very  interesting  example 
shewing  -dho-  (given  infra,  p.  441). 

From  the  concluding  remarks  of  the  article  in  question  (quoted 
infra,  p.  446)  we  are  led  to  hope  that  a  forthcoming  treatise  by 
the  same  scholar  will  give  us  further  examples  of  Greek  nouns 
formed  with  the  suffixes  -do-  and  -dAo-. 

XV  201,  note  2, 1.  22 :  After  "This  explanation  of  -dhi"  add : 
"namely,  that  it  is  derived  from  the  Idg.  ^ dhi-J* 

XV  201,  at  the  end  of  note  2,  add  :  "P.  S. — It  may  be  equally 
possible  to  regard  this  -dki  as  a  case-form  belonging  to  the  same 
group  as  Gk.  '$€  -6tv  -3a  -Bai.  -6tjv,  etc.  [cf.  the  groups  no  ne  neti  mi 
tiPi  notf  etc.  (Per  Persson  in  Idg.  Forsch.,  vol.  II,  pp.  199  sqq.), 
'Ki  K(  Ktp  Koi  fca,  Be  8tj  dai  -da(m)  -do  de],  these  and  such  like  primi- 
tive particles  being  generally  considered  to  shew  forms  with 
varying  vowel  representing  various  case-forms. 

But,  whether  the  i  of  -dhi  be  explained  in  this  way  or  as  in  the 
actual  text  of  the  note  (i.  e.  A.  J.  P.  XV,  p.  201,  note  2),  my 
derivation  of  -dhi  as  from  the  Idg.  V  dhi-  (a  derivation  in  which 
it  is  gratifying  to  note  that  Prof.  E.  W.  Fay  concurs^)  is  nowise 
affected." 

XV  204,  1.  5:  For  ** ahumer^nc^'  read  ^' ahUmer'c^  or 
ahumer*nc^'^  See  Brugmann,  Grundr.  II,  §§27  and  163  (as 
amended  by  A.  J.  P.  XV  204,  n.  i). 

204, 1.  6 :     For  "  ahu-mer^c  ^ "  read  "  ahu-mer^c'^ 
^A.  J.  p.  XVI,  1895.  p.  2. 


Digitized  by  VjOOQ IC 


ORIGIN  OF  THE  GERUND  AND  GERUNDIVE,         44 1 

iccpo- 


VKIUIN  UJ*    THE    GEKUND   AND    GERUNDIVE, 

XV  205,  1.  I :    Read  "  Ktpav^pos  beside  Kcparo^por  and 


XV  205, 11. 21-26 :  The  explanation  here  given  of  Lat.  vindex 
and  index,  as  compounds  wherein  the  prior  member  is  an  accu- 
saHve  case  governed  by  the  second  member — (a  class  of  compounds 
which,  in  A.  J.  P.  XV  203  sqq.  and  XVI  217  sqq.,  I  have  illus- 
trated from  Sanskrit,  Avestic,  Greek,  Italic  and  Lithuanian) — is 
supported  also  by  W.  M.  Lindsay,  The  Lat  Lang.,  1894,  c^i^- 
III,  §16,  VIII,  §95,  p.  544,  and  by  P.  Giles,  Short  Manual  of 
Comparative  Philology,  1895,  §284,  p.  215. 

So  also  W.  Prellwitz  (in  Bezzenberger's  Beitrage,  XXII,  pp. 
128,  129),  who  there  adds  from  Greek  the  following  very  inter- 
esting example  of  such  composition : 

*Bm'Bo"  or  *3ta-3(ri) — the  base  of  "^Bia-B^os  whence  Blaaos  "*eine 
versammlung,  die  einer  gottheit  zu  ehren  opfer,  chore,  aufziige 
u.  dgl.  veranstaltet' "  (Prellwitz,  1.  c.) — is  to  be  equated  with  Skr. 
dhiyam'dh{a)  "*das  anschauen  worauf  richtend*  (jdhd-  *setzen*), 
daher  von  menscben  'andachtig,'  von  gottern  *achtsam' "  (a  word 
which  I  had  already  given  in  A.  J.  P.  XV,  pp.  200,  203,  as  an 
instance  of  a  compound  the  prior  member  of  which  is  an  accu- 
sative case  governed  by  the  second  member),  *Bia^  being  an 
accus.  sing,  (identical  with  Skr.  dhiyam)  governed  by  the  second 
member  of  the  compound. 

[Several  more  examples  of  such  compounds,  which  apparently 
have  escaped  the  notice  of  Prellwitz,  are  to  be  found  (as  above 
mentioned)  in  A.  J.  P.  XVI,  pp.  217  sqq.] 

Further  Latin  instances : — 

iocundus  [with  d  due  to  the  U  of  iUcundus  (from  which  it  is 
etymologically  to  be  separated*)]  idcundus  [the  latter  only  in  late 
Latin] — the  ioc-  whereof  is  generally  identified  with  the  idc-  of 
idcus  idcare^ — is  probably  to  be  regarded  as  a  compound  of  this 
class,  representing  originally  *idcum-dO'S  'giving  merriment, 
giving  pleasure.' 

rdiundus  should  also  be  added  here,  if  the  explanation  of  it 
offered  by  Dr.  Prellwitz  (cf.  infra,  p.  443,  and  see  infra,  p.  446)  be 
correct. 

'  Seen  also  (according  to  Prellwitz,  1.  c.)  in  Bia'y6v£q,  whose  literal  meaning 
is  held  by  Prellwitz  to  be  *  huld-zeuger.' 
'  See  R.  S.  Conway  in  the  Class.  Rev.,  vol.  V,  1891,  p.  300  ^,  ad  fin. 
'So,  for  instance,  Conway,  1.  c. 


Digitized  by  VjOOQ IC 


442  AMERICAN  JOURNAL    OF  PHILOLOGY. 

XV  206 :  On  Lariscolus  and  legiscrepa  see  further  A.  J.  P. 
XVI,  p.  220. 

XV*  208,  209:  The  derivation  here  offered  of  venundo  and 
pessumdo  finds  an  advocate  also  in  W.  M.  Lindsay,  Class.  Rev., 
vol.  VII  (1893),  p.  106,  n.  I,'  and  The  Lat.  Lang.  (1894),  ch. 
VIII,  §95,  p.  544. 

XV  212, 1.  22:  It  was  here  observed  that  Osc.  eehiianastim 
was  "of  uncertain  meaning."  But  I  now  think  so  no  longer.  In 
an  article  on  'The  Italic  Verb  eehiia-  ehia-^  published  in  the 
Class.  Rev.,  vol.  X,  May,  1896,  I  venture  to  think  that  I  have 
offered  the  correct  explanation  of  this  Oscan  word. 

XV  212  ad  fin.:  **Hmi'dus  :  iimen-dU'S^^  etc — Pairs  of  such 
compounds  shewing  in  the  one  case  a  bare  stem  as  prior  member, 
in  the  other  a  full  case-form,  are  not  rare  in  the  Idg.  languages. 
Cf.,  e.  g.,  the  following : 

Ved.-Skr.: 

dhanadd'^  :       dkanam-jayd-*' 

AvESTic : 
ahumer'C'^  ahUmer*C',  standing  for  ^ahnm'tner'C'  • 

virajan-  (:  Skr. 

virahdn-y  :       virenjan-^  standing  for  ^virem-jan-^ 

Greek:' 


jrvp'(f>6pos'^ 

{iroSa-viirr^p  * 
nobd'Viirrpov 
fiv<r'<j)6vos 
fii0Xia'ypd<f>og  '^ 
/3t/3Xia-0opoff' 


^  XV  208 :  Concerning  the  explanation  given  of  palam-facio  in  XV  208,  I 
would  add  the  following  note :  *'  Even  granting  that  palam  was  originally  a  • 
fem.  accus.  sing.,  it  is,  however,  quite  possible  that  at  the  time  of  its  compo- 
sition  with  fcuio  the  purely  adverbial  meaning  of  the  word  had  become 
established,  and  the  true  explanation  of  the  word  (as  ace.  fem.  sing,  of  a 
substantive)  forgotten.  Of  course,  if  this  be  ^o^  palam-facio  is  not  here  a  case 
in  point.'* 

'  Referred  to  again,  infra,  p.  445. 

»  A.  J.  P.  XV  199.  *  A.  J.  P.  XV  203.  »  A.  J.  P.  XV  204. 

•  A.  J.  P.  XV  204,  and  herein,  supra,  p.  440. 

^In  addition  to  the  examples  here  cited,  see,  further,  the  letter  of  Dr. 
Prellwitz  cited  infra,  p.  446. 


Digitized  by  VjOOQ IC 


ORIGIN  OF  THE  GERUND  AND  GERUNDIVE. 


443 


<fiaea<f>6pos^  1 

<f>C»TO(f>dpOS^  } 

Kepo<f>6pof*      1 

K€paTO<f)6pOt*  J 

^i(l>o<l>6pos* 
TeparotrKdrrot^ 


0vod6Kos'^ 


Latin  : 
iuridtcus^ 
muricidus'^ 
Larifuga^ 
legirupa} 

(Skr.  bhaga-dheya-^) 
lepidus 


<f>ma'<f>6pos 

K€paa-<f>6pos 

^i<f>ri-<l>6pos*' 

T€pafTK6iroSi  Standing  for  ^rtpawKonot* 

BvffKOOS 

OvrfiroXos 

index,  standing  for  ^ius-dex^ 

muscipula^ 

Laris{=  Lares)'Colus^ 

legis{=  leges)'Crepa^ 

merenda^ 

iocundus? 


XV  213, 1.  i:  *' If  from  ^  dhe^r  A  reference  to  XV  202,  11. 
14-16,  will  explain  the  point  of  this  "if." 

XV  215,  U.  6,  7 :  A  still  simpler  explanation  of  roiundus  is 
now  offered  by  Dr.  Prellwitz  (cf.  supra,  p.  441,  and  see  infra,  p. 
446). 

XVI  218, 1.  14 :  Add :  With  respect  to  the  second  of  the  two 
suggested  explanations  of  Lesb.  diKaaKdnog,  compare  the  following 
remark  of  Brugmann,  Grundr.,  vol.  II,  §161,  p.  487 :  "No  doubt 
diK-ri  has  replaced  an  older  *dif  =  Skr.  diS"  And,  for  ^diK-a  : 
diK'tif  compare  KpSic-a  :  KpSK-ri  and  oXk-I  :  aXie-17. 

XVI  218, 1.  30:  On  the  vocalism  of  Lat.  cdvid  see  now  the 
essay  on  'The  Establishment  and  Extension  of  the  Law  of 
Thurneysen  and  Havet,'  A.  J.  P.  XVI,  Dec.  1895,  P-  448. 

XVI  222,  n.  I,  U.  13-15:  For  "Umbr.  pane''  read  "Umbr. 
pane." 


>  A.  J.  P.  XV  204. 

*A.J.  P.  XVI218. 

^A.  J.  P.  XV  206. 

*  Given  herein,  supra,  p.  441 


•A.  J.  P.  XV  205. 
*A.J.  P.  XVI219. 


'  Herein,  p. 
•A.  J.  P.  XV  205. 
•A.  J.  P.  XVI  219. 


Digitized  by  VjOOQ IC 


444  AMERICAN  JOURNAL    OF  PHILOLOGY. 

We  may  now  turn  our  attention  to  the  further  consideration  of 
the  main  propositions  which  it  was  the  object  of  my  two  first 
papers  to  establish : — 

(i)  That  the  Italic  Gerundive  developed  itself  an  Italic  soily  i,  e. 
in  other  words ^  wa^  purely  an  Italic  development} 

Of  this  view  I  need  say  no  more  here  than  that  it  has  the 
advantage  of  being  supported  by  "the  influence  of  a  great 
name"' — the  name  of  Brugmann.*  I  believe  the  view  to  be 
correct  as  stated.* 

(2)  That  the  Gerundive  arose  before  the  Gerund^  the  loiter 
being  a  development  from  the  former} 

In  support  of  this  proposition — advocated  not  only  by  myself 
in  the  two  papers  on  the  *  Origin  of  the  Gerund  and  Gerundive,* 
A.  J.  P.  XV  194-216,  XVI  217-222,  but  also  by  such  distinguished 
philologists  as  Dr.  Joseph  Weisweiler  in  his  *Das  Lateinische 
Participium  Futuri  Passivi  in  seiner  Bedeutung  und  Syntaktischen 
Verwendung*  (Paderborn,  1890),  Professor  Karl  Brugmann  in 
his  Grundriss,  vol.  IV,  1892,  §1103,  Rem.,  p.  1424,  and  Mr.  W. 
M.  Lindsay,  The  Latin  Language,  1894,  ch.  VIII,  §94,  p.  543 — 
Mr.  F.  W.  Thomas,  Fellow  of  Trinity  College,  Cambridge,  sends 
me  the  following  communication : 

"You  may  be  interested  to  note  as  confirmatory  of  your  and 
Weisweiler's"  and  (he  might  have  added  also)  Bnigmann*s  and 
Lindsay's  "  opinion  as  to  the  priority  of  the  adjective  in  Latin, 
that,  beside  the  Greek,  the  Sanscrit  shews  an  Impersonal  Gerun- 
dive, viz.  in  -yam  and  -lyam^  derived  from  the  adjective.  (This 
will  be  found  discussed  by  Bohtlingk  in  a  short  article  *  Ueber  d. 
impersonalen  Gebrauch  d.  Participia  necess.  im  Sanskrit'  in  the 
Zeitschrift  d.  Deutschen  Morgenlandischen  Gesellschaft,  vol.  42, 
1888,  pp.  366-9.)" 

*  See  A.  J.  P.  XV  195,  202,  212,  and  XVI  222,  n.  i. 

*See  Prof.  E.  W.  Fay  in  the  Proc.  of  the  Amer.  Philolog.  Assoc,  vol.  26, 
1895,  p.  Ixv. 

'Brugmann,  Grundr.,  vol.  IV,  §1103,  3,  Rem. 

^**' Suppose  for  a  moment  that  my  first  proposition  is  incorrect,  and  that  the 
Italic  Gerundive  took  its  rise  (not,  as  I  have  held  and  believe,  in  Prim,  Italic, 
but)  in  Idg,  itself.  My  main  propositions  concerning  the  Origin  of  the  Gerund 
and  Gerundive  would,  notwithstanding,  remain  practically  unaffected."  See 
Postscript  (i),  infra,  p.  450. 

*See  A.  J.  P.  XV  195,  202,  215,  2i6  ad  fin. 


Digitized  by  VjOOQ IC 


ORIGIN  OF  THE  GERUND  AND  GERUNDIVE,         44$ 

(3)  That  the  Gerundive  itself  was  a  compound,  wherein  the 
prior  member^  consisting  of  the  Prim.  ItaL  accusative  infiniiive 
in  -m}  was  governed  as  object  by  the  second  member,  the  verbal 
suffix  dO'^ 

It  was  pointed  out  in  A.  J.  P.  XVI,  p.  217,  that  this  view  is 
supported  by  Mr.  W.  M.  Lindsay  in  his  work  on  The  Latin 
Language,  1894,  ^^  VIII,  §§94-5,  and  also  in  his  *  Addenda  et 
Corrigenda*  thereto  (p.  660).  I  find,  too,  that  Mr.  Lindsay,  in 
the  Classical  Review,  vol.  VII,  1893,  p.  106,  note  i,  instancing 
laudandus  as  an  example  of  the  Gerundive,  wrote  "  I  would  make 
'dus  a  Verbal  Adjective  'giving  praise.*  Cf.  ruborem-do^  venum- 
doJ*  With  Mr.  Lindsay's  permission,  I  now  venture  to  supple- 
ment this  with  the  following  extract  from  a  letter  which,  in  Oct. 
1894,  he  wrote  to  me  (then  personally  quite  unknown  to  him) : 
"I  have  just  seen  your  excellent  article  on  the  'Origin  of  the 
Gerund  and  Gerundive*  in  the  Amer.  Journ.  of  Philology.  I 
have  long  believed  that  the  Gerundive  was  a  Compound  of  a 
Verbal  Noun,  in  Accusative  and  not  in  Stem-form,  with  the  root 
DO,  and  am  delighted  to  read  your  thorough-going  justification 
of  the  theory.** 

My  view  is  supported  also  by  Dr.  Prellwitz  (Joint-Editor  of 
Bezzenberger*s  Beitrage)  in  a  letter  to  me  (dated  i  Feb.  1896), 
which  runs  as  follows : 

"Ihre  Ansicht  Uber  das  Gerundium  ist  mir  sehr  interessant 
gewesen.  Aehnliche  Gedanken  haben  mich  lange  beschaftigt 
und  sind  dann  durch  andere  Thatigkeit  zuriickgedrangt.  Dem- 
nachst  hofFe  ich  aber  Ihnen  einige  Sachen  iibersenden  zu  konnen, 
die  Ihrer  Ansicht  iiber  das  Gerundium,  die  ich  voUig  billige, 
ziemlich  verwandt  sind.*' 

This  letter  Dr.  Prellwitz  has,  more  recently  (28  Sept.  1896), 
supplemented  by  a  second,  in  which,  (after  courteously  granting 
me  permission  to  publish  the  foregoing)  he  gives  me  a  first 
instalment  of  the  promised  *  Sachen.*     He  writes  as  follows : 

"Die  Ihrer  Gerundiverklarung  ziemlich  verwandten  Aufsatze 
habe  ich  noch  nicht  zum  Druck  geben  konnen. 

''Ein  Vorlaufer  aber  davon  ist  die  Erklarung  Blavos  :  ai.  dhiyam- 
dhd  (B.  B.  22,  128  ff.).**— See  above,  p.  441. 

^See  A.  J.  P.  XV  196, 198.  202. 

'See  A.  J.  P.  XV  198,  202,  203  sqq.,  212,  213;  XVI  218  sqq.,  and  herein, 
pp.  441,442,  446,447. 


Digitized  by  VjOOQ IC 


446  AMERICAN  JOURNAL    OF  PHILOLOGY. 

"  Wenn  Bka  Accusativ  +ftoff  von  i^  dhi-  ist,  so  ist  es  auch  erlaubt 
die  andern  Worter  auf  a-^o;  so  zu  trennen ;  ferner  die  Worter  auf 
KV'Boiy  vv'Bos,  u.  s.  w.,  als  Compost/a  aus  Accusaiiven  auf  i»,  «», 
•\-dhos  von  ij  dhe-  zu  betrachten.  Die  Nebenformen  auf  i-^or  v^- 
ohne  V  sind  dagegen  mit  dem  reinen  Stamm  auf -i  -v  zusammen- 
geselzt." 

[**  Endlich  ist  es  klar,"  he  writes  at  the  conclusion  of  his  above- 
mentioned  article  in  Bezz.  Beitr.  22,  *'dass  sich  aus  obiger  Deutung 
von  Blacrot  u.  s.  w.  wichtige  Folgerungen  f iir  die  Worter  auf  -aBw 
und  -ad-,  ja  auch  fiir  die  auf -v^or  ergeben."] 

"So  ist  rotundusy^  he  continues  in  his  letter,  "  sehr  viel  einfacher 
zu  erkl'aren  als  Brugmann  und  auch  Sie  meinen.  Es  ist  ^roihom-- 
dos  (resp.  -dkos)  *  Umdrehung  gebend  (oder  machend) ' — *rdiho5 
'Umdrehung*  (mit  ih  wegen  ai.  raiha-s  und  gr.  imppoOoi),  neben 
*roihtL  (lat.  rota)  *Rad'  ist  ganz  regelrecht  angesetzt. 

So,"  he  concludes  his  letter,  "erklare  ich  noch  eine  grosse 
Anzahl  von  Wortern."  * 

Numerous  examples  of  this  kind  of  composition  (which  is  seen 
in  the  Italic  Gerundive) — composition,  that  is,  wherein  the  prior 
member  of  tJie  compound  is  an  accusative  case  governed  as  object 
by  the  second  member — will  have  been  found  collected  in  A.  J.  P. 
XV  203  sqq.,  XVI  218  sqq.,  and  herein,  supra,  pp.  442  and  443, 
and  the  present  page. 

And  now  we  come  to  the  remaining  proposition  which  I  sought 
to  establish,  namely : — 

(4)  Thaty  unless  it  be  assumed  that  the  Umbr.'Osc,  Gerundive 
was  borrowed  from  Latin,  its  formation  {assuming^  the  latter  to 
be  identical  with  that  of  the  Latin  Gerundive)  should  compel  us 
to  regard  the  said  suffix  do-  cts  the  representative  {not  of  Idg.  dhd- 
from  Idg.  »J  dhi',  but)  of  Idg.  do- from  Idg.  ij  dd-? 

In  A.  J.  P.  XVI,  July,  1895,  P-  222  (text),  at  the  conclusion  of 
the  second  of  the  two  papers  on  *The  Origin  of  the  Gerund  and 

^  P.  S. — See  further,  now,  a  **  Sonderabdnick  aus  *  Wochenschrift  fflr  klas- 
sische  Philologie'  (Berlin),  1897,"  wherein  Dr.  Prellwitz,  again  adverting  to 
the  subject,  writes  (on  p.  ii):  *'  Ich  hoffe  bald  Zeit  zu  finden,  urn  ausfiihrlich 
zu  zeigen,  dass  -vBoq  oder  vielmehr  avdo^,  ivdog^  wdo^,  aSoc  auf  Accusative  Sing. 
auf  -av,  £v,  w,  0  +  ^  d.  i.  dhi-^os  'machend'  zuriickgehen." 

*See  A.  J.  P.  XV  198,  202  (text  and  note  3) ;  XVI  217,  note  i  (ad  fin.),  222 
(text  and  note  i),  and  herein,  supra,  p.  439. 


Digitized  by  VjOOQ IC 


ORIGIN  OF  THE   GERUND  AND  GERUNDIVE.         44/ 

Gerundive'* — in  which  I  had  endeavoured  to  prove  that  the 
Gerundive  [which  (in  common  with  Weisweiler,  Brugmann,  and 
Lindsay*)  I  held  to  be  the  earlier  formation  of  the  two']  was  a 
Prim.  Italic  compound  wherein  the  prior  member  consisted  of  the 
Prim.  Ital.  accus.  infin.  in  -nis  governed  as  object  by  the  second 
member,  namely,  the  sufHx  do-^  and  that  (as  said  above)  "unless 
it  be  assumed^  that  the  Umbr.-Osc,  Gerundive  was  borrowed  from 
Latin,  its  formation  (assuming  the  latter  to  be  identical  with  that 
of  the  Latin  Gerundive)  should  compel  us  to  regard  the  said 
sufHx  dd-  as  the  representative  (not  of  Idg.  d^  from  Idg.  V  dhe-^ 
but)  of  Idg.  do-  from  Idg.  nj do-** — occasion  wcls  taken  to  toiuh  on 
Prof.  E,  W.  Fay's  theory  of  *  The  Latin  Gerundive  -^ndo-J 

The  latter  had  appeared  in  the  same  number  of  the  A.  J.  P.  as 
my  own  first  paper  on  *  The  Origin  of  the  Gerund  and  Gerundive,' 
namely,  A.  J.  P.  XV,  July,  1894,  pp.  217-222. 

I  pointed  out  (XVI  222,  text)  that,  inasmuch  as  Prof.  Fay's 
theory,  explaining  the  d  of  *The  Latin  Gerundive  -^ndo-*  as  the 
epresentative  of  an  Idg.  dh  from  the  Idg.  nj  dhe-  (thus  h^X.ferend- 
from  Idg.  *bherndh-) — an  explanation  to  which,  so  far  as  the 
Latin  forms  alone  were  concerned,  no  objection  need  be  raised — 
took  no  account  whatever  of  the  Umbr.-Osc.  forms  of  the  Gerun- 
dive, and  inasmuch  as  (according  to  my  view*)  the  latter  could 
not  possibly  represent  Idg.  dh,  it  followed  that  Prof.  Fay's  theory 
(with  respect  to  which  we  need  here  concern  ourselves  only  with 
his  views  on  the  phonetic  development  of  Idg,  dh  in  Italic)  must 
therefore  "  inevitably  fall  to  the  ground." 

With  this  comment  on  Prof.  Fay's  theory  I  (originally)  con- 
cluded my  *  Further  Notes  on  the  Origin  of  the  Gerund  and 
Gerundive ' ;  and  it  was  not  till  after  I  had  remitted  them  to  Prof. 
Gildersleeve  for  the  A.  J.  P.  that  I  received  my  copy  of  A.  J.  P. 
XVI,  part  I,  April,  1895  (containing  Prof.  Fay's  attempt  to  escape 
from  the  difficulty  confronting  him). 

Having  observed  that  my  remarks  concerning  the  representation 
of  Idg.  dh  in  Italic  would,  unless  they  could  be  proved  to  be 
incorrect,  tell  distinctly  against  his  own  explanation  of 'The  Latin 
Gerundive  -^ndo-*  (as  given  in  A.J.  P.  XV  217-222),  Prof.  Fay  in 
A.  J.  P.  XVI,  April,  1895,  p.  I,  n.  i,  brought  forward  certain 

*  A.  J.  P.  XV  194-216,  XVI  217-222.  'See  herein,  supra,  p.  444. 
'See  A.  J.  P.  XV  193,  202,  215,  216  ad  fin.,  and  herein,  supra,  p.  444. 

*  An  assumption  '*  which  does  not  seem  very  probable,"  A.  J.  P.  XVI  222. 
*See  A.  J.  P.  XV  198,  202  (text  and  note  3) ;  XVI  217,  note  i  ad  fin..  222. 


Digitized  by  VjOOQ IC 


448  AMERICAN  JOURNAL    OF  PHILOLOGY. 

forms  whereby  he  hoped  to  shew  that  Idg.  "  -ndh-  gave  Osc* 
Umbr.  -«!>-> -«^->-««-  II  -»-,"  i.  e.  in  other  words,  that  the 
-««-  II  -«-  of  the  Osc.-Umbr.  Gerundives  could  represent  Idg» 
-ndh'  as  well  as  Idg.  -nd-. 

This  led  to  my  adding  at  the  end  of  the  *  Further  Notes  on  the 
Origin  of  the  Gerund  and  Gerundive*  (A,  J.  P.  XVI,  July,  1895, 
pp.  217-222)  a  Postscript  (namely,  the  note  on  p.  222)  wherein  I 
endeavoured  to  prove  that  "the  examples  adduced  by  Prof.  Fay 
do  not  seem  at  all  convincing"  and  that  we  must  "therefore  feel 
compelled  to  regard  Prof.  Fay's  suggestion,  that  Idg.  -ndh-  gave 
Osc.-Umbr.  -«)>-> -«flr->-««- 1|  -«-,  as  unproven." 

In  A.  J.  P.  XVI,  Dec.  1895,  pp.  491-5,  Prof.  Fay  has  returned 
once  more  to  the  charge.  He  is,  however,  unable  to  adduce  in 
support  of  his  suggestion — ^that  Osc-Umbr.  -««- 1|  -«-  can  repre- 
sent Idg.  -ndh'  as  well  as  Idg.  -nd-  — any  further  examples  than 
those  which  he  had  previously  given  (A.J.  P.  XVI  i,  n.  i),  the 
unconvincing  nature  of  which  I  had  exposed  (A.  J.  P.  XVI  222, 
n.  iV 

To  my  own  view  concerning  the  representation  of  Idg.  dh  in 
Italic — expressed  in  A.  J.  P.  XV  198,  202  (text  and  note  3) ;  XVI 
217,  note  I  (ad  fin.),  222  (text  and  note  i),  and  identical  with  that 
of  Brugmann,  Grundr.  I,  §370 — I  still  feel  bound  to  adhere,* 
whether  the  particular  Oscan  word  Ana/riss  (to  which  Prof.  Fay 
devotes  three  pages')  be*  or  be  not*  ultimately  considered  as  an 
example  of  Idg.  dh. 

In  conclusion  let  me  quote  in  support  of  my  view  the  opinion 
of  Prof.  C.  D.  Buck,  as  expressed  in  his  paper  on  *  The  Oscan- 
Umbrian  Verb-System,*  published  in  The  University  of  Chicago 
Studies  in  Classical  Philology,  vol.  I,  1895.  Speaking  of  the 
origin  of  the  Latin  Gerundive,  he  there  says  (on  p.  184)  as 
follows : 

''The  Oscan- Umbrian  forms*'  of  the  Gerundive  ''furnish,  in 
my  opinion,  conclusive  evidence  for  original  (n)d  rather  than 
(jn)dhr 

'"  The  comparison  of  in-de  with  iv-dnf'^ — in  the  validity  of  which  Prof,  Fay 
says  that  he  does  ''not  believe''  (A.  J.  P.  XVI  491) — was,  it  may  be  observed  in 
passing,  no  suggestion  of  mine.    See  A.  J.  P.  XVI  222,  note  I. 

'  For  the  present,  at  all  events ;  pending  the  discovery  of  further  and  much 
stronger  evidence  than  any  which  Prof.  Fay  has  yet  been  able  to  adduce  in 
support  of  his  suggestion. 

"  A.  J.  P.  XVI  492-5.  *  The  view  held  by  one  set  of  scholars. 

6  The  view  held  by  Prof.  Fay.  A.  J.  P.  XVI  492-5. 


Digitized  by  VjOOQ IC 


ORIGIN  OF  TITE  GERUND  AND  GERUNDIVE.         449 

In  bringing  to  a  close  these  'Concluding  Notes/  it  would 
perhaps  be  well  for  me  to  take  the  opportunity  of  summarising, 
in  precise  terms,  my  view  concerning  *  the  Origin  of  the  Gerund 
and  Gerundive*  (as  expressed  in  A.  J.  P.  XV  194-216,  XVI  217- 
222,  and  in  the  foregoing  pages  herein). 

Precise  Statement  of  My  View 
{with  references). 

I  regard  the  Italic  (i.  e.  Latin- Oscan-Umbrian)  Gerundive  as 
having  '^developed  itself  an  Italic  soiV^  {XV  igs)^  or^  in  other 
words,  as  being  '* purely  an  Italic  development  (XVI 222,  n.  /).* 

Of  the  Gerundive  and  Gerund^which  latter  indeed  "  does  not 
appear  in  the  Umbr.'Samn.  monumeTits"  {XV  1^4) — I  consider 
that  ^'the  Gerundive  was  the  earlier  formation  of  th^  two^^  {XV 
2i6*)f  and  that "  the  Gerund  wcls  developed  from  the  Gerundive  " 

The  Gerundive  itself,  held  {as  aforesaid)  to  have  arisen  on 
Italic  soil,  I  explain  as**  a  compound,  wherein  the  prior  member, 
consisting  of  the  Prim.  ItaL  Accusative  Infinitive  in  -w*,"  is 
** governed  as  object  by  the  second  member,  the  verbal  suffix  dd-  * " 
(jupra,p.445% 

And,  Icutly,  I  consider  that  **  unless  it  be  assumed— an  assump- 
tion *  which  does  not  seem  very  probable'  {XVI  222) — that  the 
Umbr.'Osc.  Gerundive  was  borrowed  from  Laiin'^,  its  formation 
(assuming  the  loiter  to  be  identical  with  that  of  the  Latin  Gerun- 
dive) should  compel  us  to  regard  the  said  suffix  dd-  as  the  repre- 
sentative {notofldg.  dhO'from  Idg.  ij  dhi-,  but)  of  Idg.  dd-from 
Idg.  li/dd'**  {supra,  p.  446%^ 

^  A.  J.  P.  XV  195,  202,  212 ;  XVI  222,  note  i,  and  herein,  p.  444* 
'A.  J.  P.  XV  195,  202,  215,  216,  and  herein,  p.  444. 

*  A.  J.  P.  XV  195,  215.  and  herein,  p.  444. 

*See  A.  J.  P.  XV  196, 198,  202,  and  herein,  p.  445. 

'A.  J.  P.  XV  198,  202,  203,  212,  213,  and  herein,  p.  445. 

'For  examples  of  such  composition  see  A.  J.  P.  XV  203  sqq.,  XVI  218  sqq., 
and  herein,  pp.  442,  443,  446. 

'A.  J.  P.  XV  198,  202  (text  and  note  3) ;  XVI  217,  note  i  (ad  fin.),  222,  and 
herein,  pp.  439i  447- 

*  A.  J.  P.  XV  198,  202  (text  and  note  3);  XVI  217,  note  i  (ad  fin.),  222  (text 
and  note  i) ;  and  herein,  pp.  439,  447  sqq. 

*P.  S. — Indeed,  assuming  the  correctness  of  my  first  proposition  (namely, 
that  (he  Italic  Gerundive  took  its  rise,  not  in  Idg.,  but  in  Prim.  Italic),  it 
would  appear  that  the  suffix  even  of  the  Latin  Gerundive  can  hardly  be  held  to 


Digitized  by  VjOOQ IC 


450  AMERICAN  JOURNAL    OF  PHILOLOGY, 

Postscript. 

(i)  Suppose  for  a  moment  that  my  first  proposition  is  incorrect^  and  that  the  ItaKc 
Gerundive  took  its  rise  (not,  as  I  have  held  and  believe^  in  Prim.  Italic^,  btii)  in 
Idg.  iUelf: 

My  main  propositions  concerning  the  Origin  of  the  Gerund  and  Gerundive 
would,  notwithstanding,  remain  practically  unaffected,  (a)  The  Gerundive 
would  still  have  arisen  before  the  Gerund,  the  latter  being  a  development 
from  the  former',  [b)  The  Gerundive  would  still  have  been  a  compound, 
wherein  the  prior  member  was  an  accusative  infinitive  in  -m  governed  as 
object  by  the  second  member  the  verbal  suffix  -do-'^.  And  (c)  this  latter — 
*' unless  it  be  assumed"  (etc.,  as  above ^) — must  still  be  regarded  as  the  repre- 
sentative (not  of  Idg.  dhd'  from  Idg.  j^dhe-^  but)  of  Idg.  dd-  from  Idg.  y'^fe-*. 
In  fact,  the  only  difference  arising  would  be  that  the  said  "accus.  infin.  in 
-m"  would  have  to  be  regarded  (not  as  a  Prim.  Ital,,  but)  as  an  Idg,  infinitive 
in  -m  (concerning  which  latter,  reference  may  be  made  to  Brugmann,  Grundr. 
IV.  §900,  p.  1268,  §1088,  p.  1414,  §1094,  p.  1419,  §1103,  3,  Rem.,  p.  1425). 

[If  *i^mdlk7-j  is  rightly  to  be  regarded  as  the  Idg.  groundform  of  Lat. 
/wmbfi-J  (see  Brugmann,  Grundr.  I,  §370 ;  Stolz,  Lat.  Gr.',  §55,  p.  295  ;  Osthoff, 
Zur  Gesch.  d.  Perf.,  533  sq.),  it  would  appear  to  be  matter  of  doubt  whether 
Idg.  -mdh'  could  even  in  Latin  have  normally  given  rise  to  -nd-^.     If,  there- 

represent  Idg,  dhd-  save  on  the  assumption  that  the  Getundival  suffix  was  taken 
direct  from  the  old-established  Latin  adjectives  already  containing  the  verbal  suffix 
'dd'S  representative  medially  of  Idg,  -dhd-s  (from  Idg,  j^dhe-),  [Concerning 
these  adjectives  see  below,  Postscript  2,  n,  7.] 

But  in  Umbrian  and  Oscan  themselves,  as  already  observed,  there  could 
have  been  no  such  adjectives  shewing  -dd-s  from  Idg.  -dhd-s ;  all  such  must  in 
Umbrian  and  Oscan  have  shewn  not  -dd-^  h}itfd-\  hence,  if  we  were  to  insist 
on  seeing  Idg.  -dhd-  in  the  -dd-  of  the  Latin  Gerundive,  we  should  (as  said 
above)  be  compelled  to  accede  to  the  improbable  view  that  the  Umbr.-Osc. 
Gerundive  was  borrowed  from  Latin. 

All  our  difficulties  vanish  the  moment  we  decide  that  the  suffix  of  the  Italic 
Gerundive  is  the  representative  (not  of  Idg.  dhd-  from  Idg.  t^dhe-^  but)  of  Idg* 
dd'  from  Idg.  y^o-. 

^See  herein,  p.  449,  and  the  passages  there  referred  to. 

'See  herein,  p.  449,  and  the  passages  there  referred  to. 

'  See  herein,  p.  449,  and  the  passages  there  referred  to. 

^  See  herein,  p.  449. 

^See  herein,  p.  449,  and  the  passages  there  referred  to. 

'  How  far  Lat.  condo  could  be  regarded  a  case  in  point  is  uncertain.  It  has 
been  thought  by  some  that  the  -do  of  Lat.  condo  comes  direct  from  Idg.  /^dhi- 
(i.  e.  cofUl'  from  *kom-dh').  It  appears,  however,  that  it  may  legitimately  be 
considered  doubtful  (i)  at  what  exact  date  Lat.  condo  came  into  being, 
/2)  whether  it  may  not  in  fact  have  arisen  in  Latin  itself,  simply  on  the 
analogy  of  certain  other  Latin  verbs  which  have  been  held  to  preserve  Idg. 
^dhi-,  such  as  addo  and  abdd.  Should  this  latter  suggestion  be  correct,  the 
-nd'  of  Lat.  condd  would  afford  no  argument  *to  counterbalance  that  afforded 
on  the  other  side  by  Lat.  lu$nbu^  from  Idg.  *lomdho^. 


Digitized  by  VjOOQ IC 


ORIGIN  OF  THE   GERUND  AND  GERUNDIVE,         45 1 

fore,  the  three  scholars  jast  named  are  right  in  their  said  derivation,  they 
furnish  us  with  an  additional  argument — this  time  from  Zo/i'if— in  favour  of 
the  contention  that,  even  assuming  the  Italic  Gerundive  to  have  taken  its  vise  in 
Idg,  itself t  it  must  nevertheless  be  regarded  as  representing,  not  Idg.  -m-^dhO' 
(from  Idg.  ^dhi^\  but  Idg.  ~m  +  dd-  (from  Idg.  V</<J-)-] 

(3)  Assuming  again  for  the  moment  that  the  Gerundive  may  have  taken  its  rise 
(not,  as  I  myself  believe  and  have  said,  in  Prim.  Italic^,  bui)  in  Idg.  itself— in 
which  case,  as  said  above,  the  accus.  infin.  in  -m,  whereof  the  prior  member  of 
the  gemndival  compound  consisted,  would  have  to  be  regarded  (not  as  a 
Prim,  Ital,,  but)  as  an  Idg,  in6nitive' — 

And  thereupon  assuming  further  (a)  the  change  in  Idg.  of  -mdh-  to  -ndh-  (an 
assumption  which  in  itself  would  appear  to  be  somewhat  doubtful'),  ifi)  that 
an  Idg.  -mdh'  could  normally  have  given  rise  even  in  Latin  to  -nd-  (again  a 
somewhat  doubtful  assumption^),  and,  lastly,  also  {y)  that  it  had  been  proved 
(and — with  all  respect  to  so  brilliant  a  scholar  as  Prpf.  Fay  be  it  said — ^it 
certainly  has  not  been  proved  as  yet')  that  Idg.  "-ik/A-  gave  Osc.-Umbr. 
-«!>->-«</->-»«- 1  -»-,"  in  other  words  that  the  -nn-  \  -»-  of  the  Osc.-Umbr. 
Gerundives  can  be  the  representatives  of  an  Idg.  -ndh-  as  well  as  of  Idg.  -if^-'; 

Assuming  all  the  foregoing  hypotheses,  I  should  not  perhaps  be  so  anxious  to 
quarrel  as  to  which  exactly  of  the  two  suffixes  Idg.  dd-  (from  Idg.  j^dd-)  and 
Idg.  dhd'  (from  Idg.  t^dhe-)  is  represented  in  the  much-discussed  suffix  of  the 
Italic  Gerundive— both,  on  the  given  premises,  being  equally  admissible. 
Indeed,  on  the  given  premises,  I  should  feel  almost  inclined  to  suggest  that  the 
suffix  of  the  Italic  Gerundive  might  very  probably  represent  both  Idg.  dd-  and 
Idg.  dhd', 

I  say  **  both  Idg.  dd-  and  Idg.  dhd-^*  and  I  say  it  not  unadvisedly.  There 
would  be  no  inherent  improbability  in  such  a  view.  Indeed,  it  could  aptly  be 
paralleled  by  the  following  pairs  of  Sanskrit  compounds  '.-^garbha-da-s : 
garbha-dhd'S,  a-doma-dd-s  :  a-doma-dhd'S,  jani-dd-s  :  jani-dhd-s,  sahasra-dd^  : 
sahasra'dk&  (all  of  which  I  cited  in  A.  J.  P.  XV  199).^ 

*  See  herein,  p.  449,  and  the  passages  there  referred  to. 

*See  above,  Postscript  (i). 

'See  Postscript  (i)  concerning  Idg.  *lomdho-s  as  the  groundform  of  Lat. 
lumbu-s. 

^Doubtful,  that  is  to  say,  in  view  of  the  derivation  of  Lat.  lumbu-s  from 
Idg.  *lomdho-s.    See  above.  Postscript  (i). 

'See  A.  J.  P.  XVI  222  (text  and  note  i),  and  herein,  supra,  pp.  447  sqq. 

•The  view  of  Prof.  Fay  in  A.  J.  P.  XVI,  p.  i. 

^  In  A.  J.  P.  XV  202  I  said  that  on  account  of  a  certain  Umbrian  word  there 
quoted  "it  is  more  probable  that  we  should  trace  ^do-  than  ^dhe*  in  the  Latin 
adjectives  in  -du^  (given  above)."  But  I  doubt  whether  the  conclusion  was 
sufficiently  warranted  by  the  premise.  One  Umbrian  word  can  hardly,  in  such 
a  case,  be  allowed  to  'legislate'  for  all  the  Latin  adjectives  in  -du-s,  I  now 
think  that  while  many  of  the  latter  (along  with  the  particular  Umbrian  word) 
doubtless  represent  Idg.  dd-  (from  Idg.  ^dd-),  others  may  nevertheless  repre- 
sent Idg.  dhd'  (from  Idg.  ^dhe-),  and  some  (in  all  probability)  both  Idg.  dd- 
and  Idg.  dhd-. 


Digitized  by  VjOOQ IC 


452  AMERICAN  JOURNAL    OF  PHILOLOGY. 

So  far  as  the  original  meaning  and  the  syntax  of  the  Gerundive  are 
concerned:  a  glance  at  §5  of  my  first  paper  (A.  J.  P.  XV  213-215)  will  at  once 
shew  that  the  acceptability  of  my  explanation  concerning  these  points  woald 
be  in  nowise  affected  (unless,  indeed,  in  the  direction  of  still  greater  accept- 
ability) by  the  advent  of  Idg.  y<M/-  upon  the  stage. 

"  The  root  dhi-  in  agglutinative  groups*'  is,  too,  of  such  frequent  occurrence 
(see  Prof.  Fay's  interesting  paper  in  A.  J.  P.  XVI,  pp.  i  sqq.)  that  one  would 
gladly  welcome  its  appearance  (if  possible)  in  one's  old  friend,  the  Italic 
(gerundive. 

But  {as  already  indUaUd)  the  hypotheses^  on  which  alone  such  appearance  comid 
be  held  to  come  even  ** within  the  sphere  of  practical  politics**  remain  as  yei 
•* unproveny     They  remain,  in  a  word, — ** hypotheses** 

Lionel  Horton-Smith. 

53  Queen's  Gardens,  Lancaster  Gate,  London,  W.,  England, 
23  Sept.,  X897. 


Digitized  by  VjOOQ IC 


v.— NEGATIVE  FUTURES  IN  THE  GREEK  NEW 
TESTAMENT. 

It  has  long  been  the  common  doctrine  of  New  Testament 
grammarians  and  commentators  that  the  double  negative  ov  fi^, 
which  is  used  with  the  aorist  subjunctive,  and  more  rarely  with 
the  future  indicative,  in  denials  referring  to  the  future,  is  an 
4mphaHc  negative.  It  is  the  purpose  of  this  paper  to  show  that 
the  facts  of  New  Testament  and  Septuagint  usage  prove  that  ov 
|i7  was  not  regarded  by  Hellenistic  writers  as  an  emphatic  nega- 
tive, but  that,  on  the  contrary,  the  aorist  subjunctive  with  ov  y^ 
was  the  more  common  way  of  expressing  a  negative  future.  In 
modern  Greek  all  futures  are  expressed  by  the  subjunctive,  the 
word  ^d,  abbreviated  for  ^An  ira,  being  placed  before  it,  to  show 
that  the  future  is  intended.  In  Hellenistic  times  the  change  to 
the  use  of  the  subjunctive  for  the  future  had  so  far  advanced  that 
in  the  majority  of  cases  where  the  prediction  was  negative  the 
subjunctive  was  used,  the  peculiar  double  negative,  whatever  its 
origin,  having  come  to  indicate  the  iense^force  intended,  not  the 
quality  of  the  negation. 

It  is  not  our  purpose  to  discuss  now  the  origin  of  the  remark- 
able combination  ov  /i^,  or  the  question  how  far  it  is  really  emphatic 
in  classic  usage.  Our  contention  is  only  that,  however  it  origin- 
ated and  however  it  may  have  been  previously  used,  it  was  not 
•emphatic  to  Hellenistic  writers. 

There  are  two  ways  in  which  a  negative  future  may  be  expressed : 
a  negative  word  may  be  affirmed,  as,  ''No  man  shall  set  on  thee 
to  harm  thee,"  or  the  verb  may  be  denied,  as,  "  He  shall  not  lose 
his  reward."  Of  course,  we  have  in  mind  in  this  paper  only  this 
latter  kind  of  negative  futures.  These  may  be  subdivided  again 
into  Prohibitions  and  Predictions.  The  prohibitions  are  expressed 
in  the  New  Testament  by  ov  with  the  future  indicative;  for 
example,  ** Thou  shalt  not  kill"  (Mat  v.  21). 

Negative  predictions  in  the  New  Testament  are  expressed  either 
by  ov  with  the  future  indicative  or  by  oh  yai  with  the  aorist  subjunc- 
tive, or  rarely  with  the  future  indicative.  A  pretty  careful  exam- 
ination shows  that  oh  with  the  future  indicative  occurs  about 


Digitized  by  VjOOQ IC 


454  AMERICAN  JOURNAL    OF  PHILOLOGY. 

eighty  times,  while  ov  \u]  occurs  about  ninety  times,  in  eighty  of 
which  cases  Westcott  and  Hort  give  with  it  the  aorist  subjunctive. 
Westcott  and  Hort  give  in  ten  instances  the  future  indicative 
with  ov  fii},  but  on  account  of  the  variations  of  the  manuscripts, 
and  the  indecisiveness  of  the  spelling  even  if  there  were  na 
variations,  only  one  of  these  is  certain,  Mat.  xvi.  22,  "  This  shall 
never  be  unto  thee."  It  seems  most  likely  that  when  the  future 
indicative  was  used  with  o£>  fu^,  as  often  in  the  Septuagint,  it  was 
from  a  confusion  of  the  two  original  idioms,  and  that  no  different 
meaning  was  intended.  Our  question  is  simply  whether  in  the 
ninety-odd  cases  with  ov  /117  the  negation  is  emphatic,  and  in  the 
eighty  cases  with  06  unemphatic.  The  fact  that  the  common 
doctrine  requires  the  recognition  of  special  emphasis  in  the 
majority  of  the  instances  raises  a  doubt  at  the  outset. 

Let  us  see  now  how  ov  \ui  has  been  treated  in  our  English 
versions.  Out  of  ninety-three  cases  in  which  ow  /iij  stood  in  the 
Textus  Receptus,  the  Authorized  Version  rendered  as  emphatic 
only  the  following  seventeen : 

Mat.  V.  18.  one  jot  or  one  tittle  shall  in  no  wise 
20.  ye  shall  in  no  case  enter  into. 
26.  Thou  shalt  by  no  means  come  out. 

X.  42.  he  shall  in  no  wise  lose  his  reward. 

xxiv.  21.  no,  nor  ever  shall  be. 
Mk.  xiv.  31.  I  will  not  deny  thee  in  any  wise. 
Luke  x.  19.  nothing  shall  by  any  means  hurt  you. 

xviii.  17.  shall  in  no  wise  enter  therein. 
John  vi.  37.  I  will  in  no  wise  cast  out. 
Acts  xiii.  41.  which  ye  shall  in  no  wise  believe. 
Rev.  xviii.  14.  thou  shalt  find  them  no  more  at  all 

21.  and  shall  be  found  no  more  atalL 

22.  shall  be  heard  no  more  at  all  in  thee  (bis) 

23.  shall  shine  no  more  at  all  in  thee 
23.  shall  be  heard  no  more  at  all  in  thee 

xxi.  25.  shall  not  be  shut  at  ail  by  day. 

27.  there  shall  in  no  wise  enter  into  it. 

It  is  evident  that  the  selection  of  these  seventeen  cases  for 
emphasis,  and  the  omission  of  emphasis  in  the  other  seventy-six 
cases,  was  a  matter  of  pure  accident,  since  no  one  would  contend 
for  a  moment  that  these  denials  are  in  themselves  more  vehement 
than  many  of  the  others. 

The  authors  of  the  Revised  Version  added  emphasis  to  the 
translation  of  o&  iiq  in  twenty-two  additional  cases : 


Digitized  by  VjOOQ IC 


NEGATIVE  FUTURES  IN  THE  GREEK  N,  T,  455 

Mat.  xiii.  14.  shall  in  no  wise  understand, 
shall  in  no  wise  perceive. 
xvi.  22.  this  shall  never  be  unto  thee 

28.  which  shall  in  no  wise  taste  of  death. 
xviii.  3.  shall  in  no  wise  enter  into  the  kingdom. 
Mk.  ix.  I.  shall  in  no  wise  taste  of  death. 

41.  shall  in  no  wise  lose  his  reward. 
X.  15.  shall  in  no  wise  enter  therein, 
xiii.  19.  and  never  shall  be 
xvi.  18.  it  shall  in  no  wise  hurt  them. 
Luke  ix.  27.  which  shall  in  no  wise  taste  of  death. 

xii.  59.  Thou  shalt  by  no  means  come  out  thence. 
John  iv.  48.  ye  will  in  no  wise  believe. 
Acts  xxviii.  26.  shall  in  no  wise  understand 

shall  in  no  wise  perceive 
z  Thes.  iv.  15.  shall  in  no  wise  precede  them  that  are  fallen 

V.  3.  they  shall  in  no  tuise  escape. 
Heb.  xiii.  5.  I  will  in  no  wise  fail  thee 

neither  will  I  in  any  wise  forsake  thee. 
Rev.  iii.  5.  I  will  in  no  wise  blot  his  name, 
xviii.  7.  shall  in  no  wise  see  mourning. 

22.  shall  be  found  any  more  at  all  in  thee. 

In  two  instances  the  Revisers  have  weakened  what  was  emphatic 
in  the  Authorized  Version : 

Mk.  xiv.  31.  But  he  spake  exceeding  vehemently,  If  I  must  die  with  thee,  I 

will  not  deny  thee.    (A.  V.:  I  will  not  deny  thee  in  any  wise,) 
John  vi.  35.  he  that  cometh  to  me  shall  not  hunger  (A.  V.:  shall  never  hunger). 

These  changes  were  not  made  on  account  of  any  change  of 
reading  in  the  Greek  text.  The  first  may  have  come  from  the 
feeling  that  the  briefer  rendering  would  be  more  forceful.  In 
Rev.  ix.  6  a  change  of  reading  has  introduced  ov  fu;  into  the  text, 
and  the  Revisers  have  rendered  it  emphatically,  "  shall  in  no  wise 
find  it." 

We  have  therefore  in  the  Revised  Version  thirty-eight  cases  in 
which  ov  lui  is  rendered  emphatically  and  about  fifty-three  in  which 
it  is  not.  We  say  'about/  because  there  are  differences  of  read- 
ings and  we  may  have  overlooked  a  few  cases.  About  forty-two 
per  cent,  of  the  cases  are  now  rendered  as  emphatic,  and  fifty- 
eight  per  cent,  are  not  so  rendered.  But  the  division  seems  again 
to  have  been  made  by  accident,  as  will  appear  from  the  following 
list  of  the  cases  which  still  stand  unemphasized  in  the  Revised 
Version  : 

Mat.  X.  23.  Ye  shall  not  have  gone  through. 
XV.  6.  he  shall  not  honor  his  father. 


Digitized  by  VjOOQ IC 


456  AMERICAN  JOURNAL    OF  PHILOLOGY, 

Mat.  xxiii.  39.  Ye  shall  not  see  me  henceforth. 

xxiv.  2.  There  shall  not  be  left  here  one  stone. 

34.  This  generation  shall  not  pass  away. 

35.  my  words  shall  not  pass  away, 
xxvi.  29.  I  will  not  drink  henceforth. 

Mark  xiii.  2.  there  shall  not  be  left  here  one  stone  upon  another  which  shall 
not  be  thrown  down. 

30.  This  generation  shall  not  pass  away. 
xiv.  25.  I  will  no  more  drink. 

31.  I  will  not  deny  thee. 
Luke  1. 15.  sliall  drink  no  wine. 

vi.  37.  ye  shall  not  be  judged 

ye  shall  not  be  condemned, 
viii.  17.  that  shall  not  be  known, 
xiii.  35.  Ye  shall  not  see  me. 
xviii.  7.  And  shall  not  God  avenge. 

30.  who  shall  not  receive  manifold  more, 
xxi.  18.  And  not  a  hair  of  your  head  shall  perish. 

32.  This  generation  shall  not  pass. 

33.  my  words  shall  not  pass  away, 
xxii.  16.  I  will  not  eat  it 

18.  I  will  not  drink  from  henceforth. 

67.  ye  will  not  believe. 

68.  ye  will  not  answer. 
John  iv.  14.  shall  never  thirst. 

vi.  35.  he  that  cometh  to  me  shall  not  hunger,  and  he  that  believeth  on 

me  shall  never  thirst, 
viii.  12.  shall  not  walk  in  the  darkness. 

51.  shall  never  see  death 

52.  he  shall  never  taste  of  death. 

X.  5.  And  a  stranger  will  they  not  follow. 

28.  and  they  shall  never  perish, 
xi.  26.  whosoever  liveth  and  believeth  on  me  shall  never  die. 

56.  That  he  will  not  come  to  the  feast  ? 
xiii.  8.  Thou  shalt  never  wash  my  feet. 

38.  The  cock  shall  not  crow 
XX.  25.  I  will  not  believe. 
Rom.  iv.  8.  to  whom  the  Lord  will  not  reckon  sin. 
I  Cor.  viii.  13.  I  will  eat  no  flesh  for  evermore. 
Gal.  iv.  30.  shall  not  inherit  with  the  son  of  the  freewoman. 

V.  16.  ye  shall  not  fulfil  the  lust. 
Heb.  viii.  11.  And  they  shall  not  teach. 

12.  their  sins  will  I  remember  no  more. 
X.  17.  their  iniquities  will  I  remember  no  more. 

1  Pet.  ii.  6.  shall  not  be  put  to  shame. 

2  Pet.  i.  10.  ye  shall  never  stumble. 

Rev.  ii.  II.  shall  not  be  hurt  of  the  second  death, 
iii.  3.  thou  shalt  not  know  what  hour. 


Digitized  by  VjOOQ IC 


NEGATIVE  FUTURES  IN  THE  GREEK  N.  T.  457 

Rev.  iii.  12.  he  shall  go  out  thence  no  more. 

vii.  16.  neither  shall  the  sun  strike  upon  them. 
XT.  4.  Who  shall  not  fear,  O  Lord. 

There  is  nothing  to  indicate  that  the  line  of  division  between 
the  cases  in  which  o^  fi^  is  translated  emphatically  in  the  Revised 
Version  and  those  in  which  it  is  not  translated  emphatically  is 
other  than  an  accidental  one.  What  but  inadvertence  can  have 
been  the  reason  that  Peter's  vehement  protestation,  *'I  will  not 
deny  thee"  (Mat.  xiv.  31),  and  Thomas's  obstinate  refusal,  "I  will 
not  believe*'  (John  xx.  25),  and  similar  passages  were  left  unem- 
phatic  by  men  who  rendered  "I  will  in  no  wise  fail  thee"  (Heb. 
xiii.  5)  and  "ye  will  in  no  wise  believe"  (John  iv.  48)?  Some- 
times the  Revisers  seem  to  have  seen  the  ov  /i^,  and  then  to  have 
emphasized  the  negation,  and  sometimes  they  seem  to  have  over- 
looked it.  It  is  curious  that  the  quotation  from  the  Septuagint  in 
Mat.  xiii.  14  and  Acts  xxviii.  26  has  been  made  emphatic,  so  that 
we  now  read  "shall  in  no  wise  understand"  and  "shall  in  no  wise 
perceive,"  instead  of  "shall  not  understand"  and  "shall  not  per- 
ceive," while  the  very  same  idiom  in  quotations  from  the  Septua- 
gint is  left  "shall  not  inherit"  in  Gal.  iv.  30,  "they  shall  not 
teach"  in  Heb.  viii.  11,  and  "shall  not  be  put  to  shame"  in  i  Pet. 
ii.  6. 

It  is  next  in  order  to  inquire  whether  those  negative  predictions 
which  are  expressed  by  the  future  indicative  with  ov  are  less 
emphatic  in  their  meaning  than  those  which  are  expressed  by  the 
aorist  subjunctive  with  w  fi^.  Twenty  of  these  taken  at  random 
will  serve  a3  specimens : — 

Mat.  X.  29.  not  one  of  them  shall  fall  on  the  ground  without  your  Father, 
xii.  19.  He  shall  not  strive  nor  cry  aloud : 

20.  A  bruised  reed  shall  he  not  break,  And  smoking  flax  shall  he  not 
quench 

31.  but  the  blasphemy  against  the  Spirit  shall  not  be  forgiven 

32.  it  shall  not  be  foigiven  him. 

39.  there  shall  no  sign  be  given  to  it. 

xvi.  18.  the  gates  of  Hades  shall  not  prevail  against  it. 
Mark  iii.  25.  that  house  will  not  be  able  to  stand. 

xiii.  24.  the  moon  shall  not  give  her  light. 
31.  my  words  shall  not  pass  away. 
Luke  X.  42.  which  shall  not  be  taken  from  her 

xvi.  31.  neither  will  they  be  persuaded  if  one  rise  from  the  dead. 

xix.  44.  shall  not  leave  in  thee  one  stone  upon  another. 
John  vii.  34.  Ye  shall  seek  me,  and  shall  not  find  me. 

xix.  36.  A  bone  of  him  shall  not  be  broken. 


Digitized  by  VjOOQ IC 


458  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Rom.  iii.  20.  by  the  works  of  the  law  shall  no  flesh  be  justifled. 

ix.  33.  shall  not  be  put  to  shame. 
Heb.  i.  12.  And  thy  years  shall  not  fail. 
Rey.  xxii.  3.  And  there  shall  be  no  curse  any  more. 
5.  And  there  shall  be  night  no  more. 

Are  these  passages,  in  their  spirit  and  purpose,  less  emphatic 
than  those  with  ov  \ai  ?  Let  this  list  be  read  over  to  some  intelli- 
gent person  familiar  only  with  the  English  version,  and  then  let 
the  list  of  passages  in  which  ov  ^  has  not  been  emphatically 
translated  be  read  to  him,  and  let  it  be  seen  whether  such  a 
person  can  guess  which  list  of  passages  has  the  stronger  nega- 
tives in  the  Greek.  Our  point  is  that  if  one  should  approach  the 
New  Testament  with  the  notion  that  simple  ov  is  an  emphatic 
negative,  he  would  find  in  usage  as  much  to  support  that  gratu- 
itous theory  as  he  can  find  to  support  the  doctrine  that  ov  \ii\  is 
emphatic. 

Even  the  most  hasty  reading  of  the  Septuagint  is  sufficient  to 
convince  any  one  that  the  authors  of  that  version  regarded  ov  ^ 
with  the  subjunctive  or  future  indicative  as  the  natural  translation 
of  any  Hebrew  negative  future  without  regard  to  its  character  for 
emphasis.  The  Hebrew  imperfect  with  fe6,  which  is  not  an 
emphatic  negative,  is  freely  so  rendered.  Often  in  the  same 
connection  and  even  in  the  same  verse  where  two  or  more 
Hebrew  imperfects  stand,  each  with  fe6,  the  Septuagint  changes 
from  ov  with  the  future  to  ov  f<^  with  the  subjunctive,  obviously  in 
unconsciousness,  or  merely  for  variety.  For  example,  we  read 
in  Isaiah  ii.  4,  ''nation  shall  not  lift  up  (ov  XiT^crat)  sword  against 
nation,  neither  shall  they  learn  (ov  fi^  fidBaviv)  war  any  more." 
Isaiah  v.  27  is  an  instructive  example,  "They  shall  not  hunger  (ofr 
with  fut.  ind.)  nor  grow  weary  (ov  with  fut.  ind.)  nor  slumber  (ov 
with  fut.  ind.)  nor  sleep  (ov  with  fut.  ind.),  neither  shall  they  loose 
(ov  fi^  with  fut.  ind.)  the  girdle  of  their  loins,  nor  shall  the  latchets 
of  their  shoes  be  broken  (ov  fi^  with  aor.  subj.)."  The  negative 
in  the  Hebrew  is  simple  k^  every  time.  In  the  two  cases  just 
given  ov  fjof  comes  last  and  might  therefore  be  thought  to  be 
climacteric ;  but  it  is  liable  to  come  first,  as  in  Isa.  vii.  7,  "  It  shall 
not  stand  (ov  firj  with  aor.  subj.),  neither  shall  it  come  to  pass  (ov 
with  fut.  ind.).''  In  commenting  upon  Paul's  quotation  in  Rom. 
iv.  8  from  Psalm  xxxii.  2,  "  Blessed  is  the  man  to  whom  the  Lord 
will  not  reckon  sin,"  Dr.  Shedd  says :  "  The  double  negative  is 
noticeable :  the  fact  that  there  is  certainly  no  imputation  of  sin 


Digitized  by  VjOOQ IC 


NEGATIVE  FUTURES  IN  THE  GREEK  N.  T.  459 

must  first  be  established,  before  there  can  be  felicitation."  But 
there  is  no  double  negative  in  the  original  Hebrew,  and  it  does 
not  seem  likely  that  the  Septuagint  meant  to  put  into  that  verse 
more  negation  than  the  Psalmist  had  put  into  it,  or  that  St.  Paul 
meant  to  attribute  to  David  more  negation  than  he  actually  used. 

To  sum  up  the  argument:  We  do  not  regard  ov  fii  as  an 
emphatic  negative  in  the  New  Testament — 

1.  Because  it  is  freely  used  in  the  Septuagint  to  render  unem- 
phatic  kS  in  the  original  Hebrew. 

2.  Because  in  the  Septuagint  oO  luj  with  the  aorist  subjunctive, 
or  future  indicative,  is  mingled  in  the  same  sentences  with  ov  and 
the  future  indicative  without  discernible  distinction. 

3.  Because  the  majority  of  negative  predictions  in  the  New 
Testament  have  06  fi^,  which  is  inexplicable  if  it  is  emphatic. 

4.  Because  the  negative  predictions  in  which  ov  fuj  occurs  would 
not,  on  other  grounds,  be  regarded  as  more  emphatic  than  those 
which  have  simple  ov. 

5.  Because  ov  f4  is  not  used  in  the  New  Testament  in  future 
prohibitions  (except,  possibly,  in  one  or  two  doubtful  cases), 
although  in  these  an  emphatic  negative  would  be  peculiarly 
appropriate. 

6.  Because  ov  /i^  is  used  in  relative  clauses,  and  questions  which 
amount  to  positive  assertions,  connections  in  which  an  emphatic 
negative  is  wholly  out  of  place;  for  example,  "there  shall  not  be 
left  here  one  stone  upon  another,  which  shall  not  (ov  ^)  be  thrown 
down"  (Mark  xiii.  2);  and  "shall  I  not  (ov  fuf)  drink  it?"  (John 
viii.  11). 

7.  Because  the  makers  of  our  English  versions,  although  holding 
to  the  doctrine  that  ov  fuj  is  an  emphatic  negative,  have  not  seen 
fit  to  apply  that  doctrine  in  the  majority  of  the  instances,  and  we 
fail  to  discover  any  principle  by  which  they  were  guided. 

The  foregoing  considerations  are  sufficient,  in  our  judgment,  to 
decide  the  matter  inductively  for  the  New  Testament.  If,  how- 
ever, any  one  remains  unconvinced,  we  should  like  to  ask  him  one 
question:  Would  you  approve  of  inserting  the  phrase  'in  no 
wise,'  or  an  equivalent  phrase,  fifty-three  more  times  in  the  Revised 
Version  ?  and  if  not,  why  not  ? 

W.  G.  Ballantine. 


Digitized  by  VjOOQ IC 


460  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Note  by  the  Editor. 

Professor  Ballantine  is  right  in  saying  that  oh  /aij  is  commonly 
set  down  as  an  emphatic  or  strong  negative.  True,  Blass  varies 
the  phraseology  somewhat  and  says  that  ov  fir^  is  *'die  bestimm- 
teste  Form  der  verneinenden  Aussage  iiber  Zukiinftiges."  But 
no  great  chasm  divides  'emphatic'  from  'decided.'  The  tone  of 
personal  interest  which  I  have  claimed  for  firj  ov  in  certain  combi- 
nations (A.  J.  P.  VII  170)  is  recognized  for  ov  fu?  by  Ewald 
(320  a),  cited  in  my  Justin  Martyr,  Apol.  I  38,  9.  Ewald  says : 
"  Nur  selten  ist  ^^  bei  blossen  Aussagesatzen,  driickt  dann  aber 
doch  stets  eine  innigere  Theilnahme  wie  ov  fti}."  *  Innigere  Theil- 
nahme'  is  perhaps  better  than  'emphatic'  or  'strong,'  but  of  the 
passages  cited  by  Ewald,  in  only  one  (Ps.  34,  6)  is  ^K  rendered  in 
the  LXX  by  ov  foj.  Professor  Ballantine's  demonstration  that  the 
'emphasis,'  'strength,'  'interest,'  whatever  we  may  call  it,  is  not 
very  palpable  in  the  N.  T.  is  in  accordance  with  the  blunting  of  a 
great  number  of  pointed  idioms  in  the  transfer  from  classic 
Greek.  Nothing  is  more  natural  than  exaggeration  and  emphasis 
in  the  use  of  an  adopted  language  (see  J.  M.,  Apol.  I  16,  6).  Of 
this  there  are  many  instances  in  Hellenistic  Greek,  not  merely  in 
the  vocabulary,  but  also  in  grammatical  construction.  Instead  of 
saying  with  ov,  the  Hellenist  swears  with  fiif  (cf.  A.  J.  P.  I  50)* 
Instead  of  using  the  quieter  present  imperative,  he  is  prone  to 
employ  out  of  all  proportion  the  more  pungent  aorist  imperative* 
Comp.  Justin  Martyr,  Apol.  I  16,  6,  with  Professor  Miller's  statis- 
tics, A.  J.  P.  XIII  425.  Josephus  has  a  tendency  to  overdo  the 
participle  (A.  J.  P.  IX  154).  The  articular  infinitive,  which 
belongs  to  argument,  is  made  to  figure  conspicuously  in  narrative 
(A.  J.  P.  VIII  337).  There  is  no  just  sense  of  sphere,  of  propor- 
tion. So  here,  ov  fu/,  however  explained,  belongs  to  the  dramatic 
domain  of  classical  Greek  (A.  J.  P.  Ill  202).  It  has  very  little 
scope  outside  of  dialogue,  and,  in  my  judgment,  can  only  be 
accounted  for  by  a  certain  amount  of  passion.  But  it  would  be 
hard  to  see  any  special  passion  in  many  of  the  examples  that 
Professor  Ballantine  cites,  and  we  must  suppose  that  the  stress 
has  been  lost  by  over-familiarity.  Swearing  has  no  place  in 
narrative,  as  Lucian  has  remarked  emphatically  (Quomodo  his- 
toria,  II,  II  19  R.),  but  the  ordinary  Hellenist  is  not  troubled  by 
that  and  uses  his  1^  Ala  freely.  Nothing  seems  to  be  more 
likely  than  that  ov  ^  has  found  its  way  into  the  LXX  and  the  N.  T. 


Digitized  by  VjOOQ IC 


NEGATIVE  FUTURES  lAT  THE  GREEK  N,  T  461 

from  the  range  of  everyday  speech.  The  Hellenist  picked  up 
ov  ^  as  the  Scythian  archer  picked  it  up  in  Ar.  Thesm.  1108 :  ovki 
fui  \a\fjat  tnf>  It  is  not  to  be  emended  into  the  text  of  Pindar, 
O*  3>  45»  and  if  it  occurs  in  the  oracular  Parmenides  (A.  J.  P.  Ill 
203),  it  is  only  one  more  evidence  of  the  essentially  unpoetic 
character  of  his  philosophic  epic.  Nor  does  it  seem  to  me  that 
Professor  Jannaris,  in  his  recently  published  Historical  Greek 
Grammar  (§1828),  has  cleared  the  matter  up  at  all  by  his 
advocacy  of  the  theory  that  ov  ^117  is  a  corruption  of  w  firjv,  the 
very  combination,  by  the  way,  which  has  been  used  by  scholars 
to  correct  unruly  ov  firf%  as  in  Ar.  Ranae,  508.'  ov  firjp,  of  which  o^ 
fi^  is  supposed  to  be  a  corruption,  Professor  Jannaris  calls  a 
negative  form  of  5  fujv.  "Hence,"  he  adds,  **in  the  N.  T.  ov  fui(p) 
or  rather  ovfi^v  and  dfjJiv  or  rather  ^/xi)v  hold  the  balance,  95  and  77 
times  respectively."  I  have  not  been  careful  to  count  after  Pro- 
fessor Jannaris,  but  I  submit  that  it  ought  to  have  given  him 
pause  to  find  that,  according  to  his  own  reckoning,  the  N.  T. 
holds  more  ov  fi^'s  than  are  to  be  found  in  the  whole  range  of 
classic  Greek.  Professor  Jannaris's  ov  hti{¥)  theory  carries  with 
it,  of  course,  the  survival  of  the  futural  subjunctive  in  classical 
Greek,  and  it  is  a  striking  illustration  of  the  fact  that  in  historical 
syntax  we  cannot  afford  to  neglect  aesthetic  syntax  or  the  history 
of  the  department.  Professor  Jannaris  cites  among  his  arguments 
the  non-occurrence  of  ov  fjJi  in  inscriptions.  But  who  would  look 
for  ov  fjJi  in  inscriptions?  Our  Greek  syntax  would  be  much 
simplified  if  we  restricted  it  to  what  may  be  found  in  Meisterhans. 
Not  less  striking  is  the  illustration  of  the  danger  that  besets  the 
Greek  of  to-day  in  dealing  with  recondite  problems  of  Greek 
syntax.  The  vernacular  is  apt  to  prevail  over  the  historical 
sense,  and  many  things  besides  the  futural  subjunctive  give  no 
shock  to  the  consciousness  (A.  J.  P.  I  242).  As  Ba  <f>vyiif  must 
deaden  the  sensibilities  to  the  shock  of  ov  fujv  <l>vyjj£,  so  the  impos- 
sible aphaeresis  postulated  by  ov  fAtjp  in  certain  passages  might  be 
welcome  to  a  man  whose  mother-tongue  is  full  of  similar  prod- 
elisions.'  But  the  whole  theory  refutes  itself  and  breaks  down  at 
the  first  serious  application.' 

*  Tyrrell  on  Eur.  Bacch.  852. 

'So.  Tr.  978:  ov  fif^  *fc7'c/)e2f,  Ar.  Pax  1302:  ov  fjoj  *niX6dif^  both  cited  by  a 
reviewer  in  Th€  Nation^  Jan.  20, 1898. 
'  E.  g.  Xen.  An.  4,  8, 13 :  ow5c«f  fUiKkri  fuivy. 


Digitized  by 


Googl 


NOTE. 
On  Latin  «/A//' naught,  not/ 

The  etymology  of  Latin  nihil  seems  to  me  to  be  still  in  need  ot 
explanation.  Brugmann  keeps  a  dead  silence  as  to  its  origin,  in 
the  Grundriss.  Victor  Henry  hazards  nothing  as  to  its  formation, 
in  his  Grammaire  compar6e.  All  that  Stolz  says  (Iw.  Miiller's 
Hdbch.',  p.  315)  is  *^  nihil  neben  nihilumJ'^  Wharton  (Eiym. 
Lat,  s.  V.)  makes  this  entry :  **«f  not,  see  «f,  +adj.-ending  -Old-. 
Ovid  has  nihil  through  a  popular  connection  with  hllum*^'  The 
entries  in  Lewis  and  Short's  lexicon,  s.  v.  hllum,  do  not  vindicate 
the  actuality  of  any  such  word.  When  Ennius  divides  ne-que  .  •  • 
hllum,  the  skeptically  minded  will  bethink  themselves  of  his 
method  of  tmesis  in  *saxo  cere-  comminuit  -drum,*  I  am  of  the 
opinion  that  hllum  is  the  veriest  of  ghost-words,  although  F.  H. 
Fowler,  in  what  is  probabjy  the  last  treatment  of  the  word  (*  The 
Negatives  of  the  Indo-European  Languages,'  Univ.  of  Chicago 
dissertation),  allows  himself  to  write  "^ne-hllum. 

To  make  a  new  explanation  of  the  word,  I  start  with  *nihllumy 
with  a  by-form  nihll^  which  originated  before  vowels,  and  was 
shortened  in  its  (pen-)ultimate  quantity  by  the  operation  of  the 
iambic  law :  from  nihil  the  adverb  nihllo  got  its  quantity.  Still, 
the  quantity  oinihilo  may  be  due  to  shortening  in  composition,  a 
phenomenon  we  cannot  yet  precisely  limit,  but  may  not  forthright 
deny. 

I  propose  to  divide  *nihllum  into  ne-k-hi-^-elum:  -hi-  is  the 
particle  affixed  to  Aryan  *«^  (or  a  compound  of  *«^-)  in  Sk.  nahi 
and  Lith.  neigl^  affixed  in  Greek  to  a  new  negative  particle  in 

We  have  next  to  explain  ^-elum.  We  may  define  nihil  very 
exactly  by  Eng.  naught 'no  any  whit,'  and  its  by-form  not 'non.* 
It  is  to  be  noted  also  that  Lat.  non^  the  ordinary  negative,  is  a 
compound  of  ne-\-unum\  and  compound  negatives  meet  us  in 
French  ne— pointy  ItaX.  non^punto  'not  at  all.'  We  may  seek, 
therefore,  in  *-elufn  for  the  meaning  *whit,  bit.' 


Digitized  by  VjOOQ IC 


NOTE.  463 

For  this  we  can  find  plenty  of  cognates :  to  begin  with,  eU- 
,fnenium  'atom.'  I  am  aware  of  the  explanation  of  eiemenium 
from  the  names  of  the  letters  ei-\-em'hen,  but»  ingenious  as  it  is, 
it  has  never  seemed  to  me  plausible.  I  lean,  for  my  own  part,  to 
the  comparison  of  eiemenium  with  Sk.  ar^tl '  fine,  thin ' — as  noun 
'  atom ' — afjdmdn  '  the  finest  particles  of  an  object.' 

Here,  too,  we  may  refer  dtakam  (R.V.)  *to  no  purpose,'  Lat. 
*parum.' 

Greek  is  also  not  without  cognates,  for  we  may  well  put  here 
^Xi-yor  *  short,'  which  G.  Meyer  explains  as  o-Xiyos  (with  prothetic 
^-),  comparing  Lith.  ligH  'illness'  and  noting  the  Hesychian 
gloss  Xifoy*  IXarroy;  while  Prellwitz  in  his  dictionary  suggests, 
with  a  query,  a  connection  with  Homeric  Xiy^fiv  (adv.)  'grazing.' 
As  to  its  formation,  l\lyos  is  a  quasi  rhyme-word  with  boKixos 
Mong.' 

Another  Greek  cognate  is  ikaxvt  'small.'  Here  again  the  c-  has 
met  its  explanation  as  a  prothetic  vowel,  because  of  Sk.  iaghii^ 
Lat  levis.  But  we  have  to  do  here,  I  suggest,  either  with  the 
problem  of  dissyllabic  gradation,  or  with  a  case  of  contamination 
of  two  stems  meaning  'small.' 

It  is  easier  to  declare  for  the  latter  alternative,  and  we  have  a 
parallel  case  to  our  hands  in  Gk.  €pvBp6s  'red,'  in  which  we  should 
see  not  only  a  cognate  of  Sk.  rudhird,  but  also  a  cognate  of  Sk. 
aru-ndy  both  meaning  *  red.'    Why  not  ? 

This  evidence  seems  to  me  enough  to  warrant  us  in  positing  an 
Aryan  base  el-  'small,  a  bit,  whit.'  A  trace  of  this  el-  I  would 
see  in  Latin  nihil '  naught,  not.' 

lbxihctok,  va.,  Ocl  18, 1897.  Edwin  W.  Fay. 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES. 

A  History  of  English  Poetry.  By  W.  J.  Courthopb,  C.  B.,  M.  A..  D.  Litt.^ 
Professor  of  Poetry  in  the  University  of  Oxford.  Vols.  I  and  II.  Mac- 
millan  &  Co.,  New  York  and  London,  1895, 1897.    $2.50  each. 

These  volumes  form  the  opening  instalments  of  Professor  Courthope's  work,, 
intended  to  be  the  first  compute  history  of  English  poetry  that  we  have  had 
since  Warton,  or  rather  that  we  have  ever  had,  and  we  trust  that  the  fates  may 
be  more  propitious  to  him  than  to  ten  Brink  and  Morley,  cut  off  before  they 
had  more  than  half  completed  their  great  designs. 

That  a  complete  history  of  English  poetry  is  wanted  will  not  be  questioned^ 
for  in  the  multiplicity  of  monographs  and  partial  works,  of  single  volumes 
treating  certain  periods,  and  of  text-books  of  English  literature,  we  have  no 
work  that  gives  a  thorough  and  scholarly  treatment  of  all  our  literature,  or 
even  of  the  poetical  literature  alone.  It  is  to  be  regretted  that  the  plan  of  the 
work  did  not  include  a  history  of  the  prose  literature  also,  as  in  the  works  of 
ten  Brink  and  Morley,  for  literature  is  individual  as  well  as  national,  and  the 
separation  of  a  man*s  work  in  poetry  from  his  work  in  prose  looks  like  depriving 
him  of  a  part  of  his  individuality,  dividing  his  mind  in  two,  as  it  were,  and 
giving  but  a  partial  view  of  the  man  himself — but  it  is  not  worth  while  to 
quarrel  with  what  we  have. 

The  first  volume  covers  the  period  included  in  "The  Middle  Ages:  influ* 
ence  of  the  Roman  empire ;  the  encyclopaedic  education  of  the  Church,  and 
the  feudal  system " ;  the  second,  that  marked  by  **  the  Renaissance  and  the 
Reformation :  influence  of  the  Court  and  the  Universities.*'  It  will  thus  be 
seen  that  the  first  volume  corresponds  generally  with  Part  I  of  Jusserand's 
*'  Literary  History  of  the  English  People/'  published  in  the  same  year  in 
English,  although  the  French  edition  appeared  the  year  before  (1894).  Prof* 
Courthope's  preface  gives  his  point  of  view.  He  quotes  Pope's  scheme,  which 
remained  but  a  scheme,  and  Gray's  design,  which  he  abandoned  on  learning 
of  Warton's  work,  but,  with  his  well-known  procrastination,  it  is  doubtful 
whether  he  would  ever  have  executed  it,  even  if  Warton  had  not  been  engaged 
on  a  similar  work.  Courthope  regrets  that  Warton  set  about  his  work — which 
was  never  completed — *'in  the  spirit  of  an  antiquary"  rather  than  in  that  of  a. 
literary  critic,  and  thinks  that  he  was  better  fitted  for  the  latter  than  for  the 
former,  hence  the  deficiencies  of  Warton's  history.  He  then  explains  the 
principles  of  his  own  work.  Taking  warning  from  the  experience  of  Warton, 
he  concludes  that  the  design  of  the  historian  of  English  poetry  must  possess 
unity ;  Gray*s  design  fulfilled  this  condition,  but  his  classification  did  not  cor- 
respond with  the  facts ;  Courthope  aims  **  to  treat  poetry  as  an  expression  of 
the  imagination,  not  simply  of  the  individual  poet,  but  of  the  English  people.'* 
He  thinks  too  that  a  historical  treatment  of  poetry  "  must  exhibit  the  principle 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES,  465 

of  its  growth  and  movement,"  and  finds  fault,  justly,  with  Taine  because  he 
looks  with  disdain  upon  the  minor  poets  of  the  fourteenth  century  and  finds 
little  of  interest  in  them,  which  causes  blunders  with  regard  to  them.  But 
"  The  Elizabethan  and  Jacobean  dramatists  regarded  themselves  as  the  lineal 
descendants  of  the  poets  of  the  fourteenth  century,"  and  so  "  we  must  examine 
the  foundations  on  which  they  built'*  While  rightly  regarding  Taine's 
criticism  of  our  earlier  poets  as  too  depreciatory,  Courthope  takes  exception  to 
Pater's  and  Symohds's  views  of  the  Renaissance  "  as  a  sudden  and  isolated 
movement  of  the  human  mind,  which  cannot  be  explained  by  the  ordinary 
methods  of  historic  investigation.*'  He  thinks  rather  that  **  the  business  of 
historical  criticism  is  to  trace  the  stream  of  thought  that  connects  age  with 
age,  and  the  almost  imperceptible  gradations  which  mark  the  advance  of 
language  and  metrical  harmony."  This  causes  him  too  to  deal  "  not  only  with 
the  progress  of  poetical  invention,  but  with  the  more  technical  question  of  the 
development  of  metrical  harmony."  For  taking  up  this  subject  he  apologizes 
in  advance  both  to  the  philologist  and  to  the  general  reader.  No  apology  is 
necessary  to  the  former  if  he  gives  the  correct  results  as  ascertained  by  philo- 
logists. 

After  this  sketch  of  his  object  and  principles  in  the  preface,  Prof.  Courthope 
announces  his  intention  of  tracing  '*  the  history  of  the  art  of  English  poetry 
from  the  time  of  Chaucer  to  the  time  of  Scott."  He  thus  disclaims  any  inten- 
tion of  giving  a  history  of  poetry  anterior  to  Chaucer,  and  justifies  himself  for 
abandoning  the  method  of  ten  Brink  and  Jusserand,  on  the  ground  that 
**  between  the  poetry  produced  in  England  before  the  Norman  Conquest  and 
the  poetry  of  Chaucer  there  is  absolutely  no  connection."  But  half  of  the 
volume  is  taken  up  with  a  consideration  of  poetry  in  England  antecedent  to 
Chaucer,  and  chapters  III  and  IV  treat  specially  **  the  poetry  of  the  Anglo- 
Saxons,"  and  "Anglo-Norman  poetry";  so  he  himself  feels  that  this  period 
cannot  be  altogether  neglected,  and  however  distinct  the  alliteration  of  Lang- 
land  is  from  that  of  Layamon  and  from  the  earlier  normal  scheme  of  the 
Anglo-Saxons,  it  is  scarcely  probable  that  Langland  would  have  originated  the 
metrical  scheme  of  his  great  poem  if  he  had  not  had  earlier  models,  and  the 
history  of  "  the  development  of  metrical  harmony  "  alone  would  have  necessi- 
tated an  investigation  of  the  form  and  style,  and  hence  the  contents,  of  these 
earlier  models,  so  that  the  historical  connection  cannot  be  overlooked. 

Renouncing  then  any  "  attempt  to  derive  the  originals  of  Chaucer  from  the 
cradles  of  the  Anglo-Saxon  race,"  the  author  seeks  *'  to  trace  his  imagination 
through  its  immediate  literary  sources,"  to  **  connect  it  with  the  poetry  of  races 
of  partial  Latin  descent,"  judging  that  in  this  way  cause  and  effect  may  be 
linked  together.  These  peoples,  while  differing  in  language  and  race,  '*  are 
united  by  a  common  system  of  faith,  education,  and  military  institution,"  and 
their  writers  deal  with  similar  problems  of  thought,  which  take  their  rise  in  a 
more  ancient  system  of  civilization,  **but  not  joined  to  the  life  of  Europe  in 
the  Middle  Ages  by  any  apparently  continuous  stream  of  literature."  It  is  the 
course  of  this  stream  that  Prof.  Courthope  proposes  to  trace  in  order  *'  to  arrive 
at  the  primal  fountains  of  mediaeval  poetry."  Next  will  come  '*  the  progressive 
stages  in  the  formation  of  the  mediaeval  stream  of  thought,  which  feeds  the 
literatures  of  England,  France  and  Italy,"  and  its  connection  with  **  the  great 


Digitized  by  VjOOQ IC 


466  AMERICAN  JOURNAL    OF  PHILOLOGY. 

system  of  Graeco- Roman  culture,  which  seems — ^bnt  only  seems — to  disappear 
from  the  world  after  the  death  of  Boethius."  Then  the  course  of  the  national 
language  must  be  explored  "in  order  to  observe  the  changes  produced  by 
Saxon  and  Norman  influences  on  the  art  of  metrical  expression  before  it 
received  the  developments  of  Chaucer."  Finally,  the  meaning  of  the  word 
"  Renaissance"  must  be  examined,  and  *'  the  early  effects  of  the  movement  on 
the  literature  of  Europe.'*  We  can  then  appreciate  the  evolution  of  poetical 
thought  and  language  that  characterized  the  art  of  succeeding  poets. 

This  comprehensive  method  is  analogous  to  that  of  Prof.  Freeman  in  the 
field  of  history,  for  it  seeks  to  connect  the  modem  with  the  ancient,  to  evolve 
the  former  out  of  the  latter,  to  show  that  the  modem  is  the  legitimate  descend- 
ant of  the  ancient,  and  is  not  separated  by  any  hard-and-fast  line.  Like  Free- 
man's unity  of  history — which  he  enforced  on  all  occasions — it  emphasizes 
the  principle  of  the  unity  of  literature.  This  work  needed  to  be  done,  for,  so 
far  as  we  know,  it  had  not  yet  been  done  for  English  literature,  historians  of 
literature  apparently  thinking  it  necessary  only  to  begin  with  Chaucer  as  an 
isolated  phenomenon,  as  if  he  too  were  not  the  child  of  his  age,  and  often  to 
overlook  the  antecedent  literature. 

After  this  preliminary  sketch  of  his  plan,  showing  the  philosophic  method 
.  which  he  intends  to  pursue.  Prof.  Courthope  examines  "  the  character  and 
sources  of  mediaeval  poetry,"  his  object  being,  for  a  right  understanding  of 
the  character  of  English  poetry,  '*  to  appreciate  the  nature  of  the  vast  change 
in  the  life  of  imagination  effected  during  the  decline  of  the  Roman  Empire 
and  the  gradual  formation  of  the  mediaeval  system  of  Europe."  The  subject 
is  treated  at  some  length  under  the  four  heads :  *'  (i)  The  decline  of  the  civic 
spirit  under  the  Roman  Empire,  and  the  corresponding  decay  of  classical 
taste ;  (2)  the  transformation  of  the  system  of  imperial  education  by  the  Latin 
Church  ;  (3)  the  rise  of  a  new  mythology  among  the  nations  embraced  within 
the  system  of  Latin  Christianity;  and  (4)  the  influence  of  feudal  institutions, 
of  the  scholastic  logic,  and  of  Oriental  culture"  (p.  14). 

Our  limits  will  not  permit  a  detailed  examination  of  each  of  these  points. 
Suffice  it  to  say  that,  after  a  brief  discussion  of  the  first  two.  Prof.  Courthope 
concludes  "  that,  in  this  continuous  stream  of  education  ...  we  find  the  con- 
trolling force  which  has,  in  one  form  or  another,  guided  the  imagination  and 
judgment  of  every  generation  of  poets  from  the  days  of  Augustus  down  to  our 
own  era."  The  third  point  is  treated  at  greater  length,  and  the  replacing  of 
the  pagan  mythology  by  the  Christian,  and  its  effects  on  poetry,  are  discussed, 
especially  when  accompanied  by  the  growth  of  heroic  legends,  as  those  of 
Troy  and  Alexander  from  the  ancient  world,  of  Charlemagne  and  King  Arthur 
from  the  modern.  This  cause,  it  seems  to  us,  had  the  greatest  infiuence  on 
the  progress  of  mediaeval  poetry,  as  a  similar  cause  affected  ancient  poetry. 
The  fourth  point  is  next  treated,  and  the  common  bond  of  minstrelsy  is  found 
to  connect  the  Teutonic  Scop  (Prof.  Courthope  prefers  the  older  form  Scbp) 
and  the  Romance  TVnTttv^r/ and  Troubcuiour\  the  (^iSf^man  becomes  the yiMi^ZrMr. 
"  Teutonic,  as  well  as  Celtic,  poetry  is,  in  its  origin,  an  embodiment  of  the 
imagination  of  the  Tribe,  not  of  the  State  "  (p.  60),  but  the  character  of  this 
minstrelsy  changed,  especially  under  the  influences  that  emanated  from  Charle- 
magne, "  the  last  great  figure  of  Teutonic  epic  song,"  and  '*  the  decline  in  the 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES,  467 

spirit  of  minstrelsy  led  to  a  great  variety  of  style  in  metrical  composition.** 
The  result  was  shorter  compositions  and  a  great  variety  of  tales;  society 
became  more  settled,  and  **  the  Prankish  intellect,  coming  under  the  influence 
of  an  old  civilization,  began  to  aim  at  new  artistic  ideals."  The  scholastic 
logic  too  affected  literature,  as  in  the  case  of  Dante  and  Chaucer ;  so,  while 
**  the  motive  power  of  Christian  European  poetry  springs  from  the  oral  min- 
strelsy of  the  Teutonic  [i.  e.  Germanic]  and  Scandinavian  tribes,"  this  was 
''profoundly  modified  by  contact  with  Latin  civilization,"  and  the  resulting 
effects. 

Inquiry  is  next  directed  to  "  the  origin  of  the  metrical  forms  and  literary 
models  adopted  by  the  early  poets  of  France  and  Italy,  who  gave  the  first 
examples  of  composition  to  the  fathers  of  English  verse"  (p.  69).  The 
author  shows  how  the  prosody  of  Latin  verse  was  modified  by  the  prevalence 
of  the  principle  of  accent  over  that  of  quantity,  referring  to  Prudentius,  who 
treated  mathesis  as  having  the  second  syllable  short,  and  Diomedes,  who 
treated  armatus  as  an  amphibrach.  So  in  popular  verses  the  laws  of  quantity 
were  soon  completely  disregarded,  and  this  was  done  deliberately  in  the 
h3rmns  of  the  early  Christian  fathers,  the  Hymn  of  St.  Ambrose,  cited  by  Bede 
{De  MetrUa  RaHone\  being  given  as  an  illustration.  Other  illustrations  of 
this  point  follow,  and  **  The  sum  of  what  has  been  said  as  to  the  history  of 
modem  European  metres  is,  that  many  of  the  Greek  metres  were  imported 
into  the  Latin  language  by  the  literary  Roman  poets ;  that  some  of  them  were 
afterwards  modified,  by  the  disregard  of  quantity,  to  suit  the  requirements  of 
the  popular  ear;  and  that,  still  later,  by  some  obvious  retrenchments,  they  . 
were  accommodated  to  the  changed  character  of  the  Romance  languages  which 
grew  up  out  of  the  rustic  Latin."  But  as  this  was  not  sufficient  to  explain 
*'  the  rise  of  the  new  system  of  rhyming  architecture,"  the  author  traces  it  to 
the  Arabs,  from  whom  the  poets  derived  their  models  of  harmony.  The  Italian 
poets  took  them  from  the  Arabs  of  Sicily  and  the  French  from  those  of  Spain, 
and  from  the  Arabs  came  not  only  the  Italian  and  Provencal  metres,  **  but 
even  the  poetical  conventions  observed  by  Petrarch  and  the  troubadours." 

This  is  but  a  brief  summary  of  Prof.  Courthope*s  interesting  chapter  on  the 
mediaeval  poetry  of  Europe,  but  it  has  been  thought  advisable  to  give  it,  as 
the  subject  is  usually  omitted  in  histories  of  English  literature,  and  as  it  illus- 
trates his  attempt  **  to  bridge  over  in  various  directions  the  gulf  that  seems  to 
separate  the  civilization  of  the  ancient  world  from  the  thought  and  imagina-, 
tion  of  the  community  of  Europe  in  the  Middle  Ages,  at  the  time  when  the 
rising  nations  were  beginning  to  make  use  of  the  vulgar  tongues  for  the  pur- 
poses of  poetical  composition  "  (p.  78). 

He  now  travels  back  to  England  and  traces  the  fusion  of  the  elements  of 
the  English  language  from  Anglo-Saxon  times  to  Chaucer.  The  chapter  on 
Anglo-Saxon  poetry  need  not  detain  us  long.  This  has  been  much  better 
done  elsewhere.  Prof.  Courthope  is  evidently  not  so  much  of  a  philologist  as 
of  a  literary  critic,  and  forgets  the  old  saw,  ne  suior  supra  crepidam.  A  summary 
of  the  Beowulf  \%  given,  based  on  Arnold  and  Earle,  and  a  notice  of  the  Byrhi- 
moth  and  of  the  Metrical  Paraphrase^  *'  once  ascribed  to  Caedmon."  We  should 
be  glad  to  have  the  grounds  of  the  "  certainly"  in  the  enumeration  of  the  works 
of  Cynewulfs  composition.    There  is  a  reference  to  Kemble's  Codex  Vercel- 


Digitized  by^ 


468  AMERICAN  JOURNAL    OF  PHILOLOGY. 

lensis  for  Cynewulf's  date,  which  date  has  been  long  since  abandoned  \xf 
scholars,  a  short  extract  from  Weymouth's  translation  of  the  EUne^  a  reference 
to  Stopford  Brooke's  History  to  combat  his  view  that  Cynewulf  was  equal  to 
Caedmon,  and  a  short  account  of  the  Crista  with  reference  to  Gollancz's  edition. 
A  few  lines  from  Beowulf^  in  which  there  are  several  misprints,  ore  given  **  as 
a  good  sample  of  the  style,"  and  a  few  pages  follow  (in  which  there  ar« 
several  errors)  intended  to  show  how  Anglo-Saxon  changed  to  English.  The 
point  of  view  may  be  given  from  the  following  quotation  of  one  "change**: 
"  5.  The  gradual  substitution  of  the  termination  ing  for  that  of  and  in  the 
present  participle.  This  change  at  first  sight  seems  anomalous,  considering 
that  the  Norman-French  ant  resembled  the  Saxon  ending.  The  latter,  how- 
ever, in  the  southern  part  of  the  country,  was  replaced  by  the  variation  ind^ 
and  it  may  be  conjectured  that  the  final  labial  \nc\  of  this  ending,  under  the 
influence  of  the  Normans,  with  whom  the  /  of  the  present  participle  was  mute, 
gave  way  to  the  guttural  ^"  (pp.  109,  no).  It  would  have  been  better  to  omit 
this  whole  antiquated  linguistic  discussion.  In  its  attempt  to  turn  Anglo- 
Saxon  into  English  with  a  wave  of  the  hand,  it  reminds  us  of  Jefferson's  Essay 
towards  facilitating  instructi<m  in  the  Anglo-Saxon  language^  but  that  was  one 
hundred  years  ago.  It  is  not  well  to  make  an  excursus  into  language  in  a 
purely  literary  work. 

The  chapters  on  "Anglo-Norman  Poetry"  and  on  **  The  Early  Renaissance'* 
are  better  done,  and  here  we  traverse  the  ground  of  Jusserand's  Book  II.  The 
genius  of  the  Normans  is  contrasted  with  that  of  the  Saxons ;  their  poetical 
activity  includes  three  stages,  that  of  Wace  and  Benoit  de  Ste.  More  in  the 
twelfth  century,  the  romantic  poetry  (and  prose)  of  the  chansons  de  geste  and 
the  King  Arthur  cycle,  in  the  twelfth  and  thirteenth  centuries,  and  the  lays 
and  fables  of  Marie  de  France,  who  is  placed  by  Courthope  in  the  reign  of 
Henry  III,  although  Jusserand  says  she  "lived  in  the  time  of  Henry  II." 
Orm  and  Layamon  are  touched  upon,  but  the  Owl  and  Nightingale  receives 
the  fullest  attention.  The  author  places  it  in  the  reign  of  Edward  I,  rather 
later  than  it  is  usually  assigned,  and  thinks  that  it  shows  "  the  influence  of 
French  models "  and  '*  a  careful  study  of  the  style  of  Marie."  The  Cursor 
Mundi  shows  the  writer  "  to  be  a  genuine  descendant  of  Caedmon,  though 
breathing  the. atmosphere  of  the  Middle  Ages,*'  If  so,  this  helps  the  connec- 
tion between  Caedmon  and  Chaucer.  A  notice  of  the  works  of  Robert  of 
Brunne  and  Robert  of  Gloucester  closes  the  chapter.  The  King  Horn  and 
Haveloh  the  Dane  are  barely  mentioned,  but  they  still  further  aid  the  historical 
connection  and  might  have  been  described. 

The  early  Renaissance  is  treated  as  affecting  Italy,  France,  and  England. 
"  Up  to  the  middle  of  the  thirteenth  century  European  poetry  may  be  said  to 
possess  a  universal  character,'*  and  that, "  because  it  reflects  the  image  of  a 
society  which  still  preserves  many  of  the  essential  features  of  the  universal 
Roman  Empire."  "  The  Renaissance  "  is  called  a  phrase  at  once  misleading 
and  obscure,  and  exception  is  taken  to  the  usual  definitions,  "for  on  the  one 
hand  the  pioneers  of  the  movement  were  the  schoolmen,  .  .  .  and  on  the  other, 
the  stream  of  classical  culture  . .  .  had  never  entirely  ceased  to  flow."  It  was 
*"  a  tendency  inherent  in  the  condition  of  things,  and  it  was  promoted  from 
different  quarters  by  the  independent  action  of  all  the  greatest  minds  of  the 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES,  469 

thirteenth  and  fourteenth  centuries."  This  seems  like  saying  it  was  so  because 
it  was  so,  and  fails  to  give  any  adequate  cause.  Until  this  great  movement 
affected  each  country  in  turn  there  was  no  national  literature,  and  the  pro- 
•duction  of  such  was  one  of  its  greatest  effects.  The  effects  in  Italy,  as  seen 
in  the  works  of  Dante,  Petrarch,  and  Boccaccio,  and  in  France,  as  represented 
by  the  Roman  de  la  Rose^  of  which  a  full  analysis  is  given,  are  traced,  and  then 
we  pass  to  England,  where  **  the  idea  of  national  life  and  of  the  just  relations 
between  Church  and  State,"  had  been  conceived  more  clearly  than  in  any 
other  country  of  Europe.  This  development  of  political  liberty  had  its  foun- 
dation in  the  institutions  of  the  Saxon  race,  but  *'  had  the  Saxon  race  remained 
in  complete  isolation,  a  certain  slowness  of  temperament,  which  is  apt  to  dis- 
^ise  its  more  heroic  qualities,  might  have  sunk  it  in  torpor  and  decay."  The 
Danes  *'  Infused  new  blood  and  energy  into  the  northern  part  of  the  island," 
and  '*  the  Normans  from  the  south  communicated  a  fresh  shock  to  the  national 
life  by  the  introduction  of  feudal  institutions,  and  of  a  ruling  race  possessed 
of  all  the  qualities  in  which  the  exhausted  Saxon  dynasty  was  deficient" 
(p.  186).  As  these  results  did  occur,  we  cannot  speculate  on  what  might  have 
happened  if  the  Danes  and  Normans  had  not  invaded  England,  but  from  the 
dogged  energy  and  persistency  of  the  Saxon  race,  which  have  ever  character- 
ized the  English  people,  and  from  the  conspicuous  lack  of  such  qualities  in 
the  French  people,  the  nearest  of  kin  to  the  Normans,  we  may  reasonably  sup- 
pose that  the  basic  qualities  of  the  English  would  have  re-asserted  themselves 
in  the  course  of  time  even  if  there  had  been  no  such  cataclysm  as  the  Norman 
invasion,  which  more  or  less  repressed  the  native  race  for  many  years,  and 
even  if  the  result  had  shown  somewhat  less  elasticity  of  temperament  and 
brilliancy  of  imagination.  English  =  three-fourths  Saxon -|- one-fourth  Nor- 
man. This  spirit  of  political  libierty  is  seen  in  the  political  songs  of  the 
thirteenth  and  fourteenth  centuries  (see  Wright's  Political  Songs  in  the  Rolls 
Series),  several  of  which,  in  English,  French  and  Latin,  and  some  of  the 
macaronic  kind  (French  and  Latin),  are  commented  on.  Abuses  of  all  sorts, 
and  especially  the  universal  corruption  of  justice  in  both  the  ecclesiastical  and 
the  king's  courts,  are  ridiculed  and  denounced.  "  Non  est  lex  sana  quod  regi 
sit  mea  lana,"  illustrates  the  tenor  of  one  of  this  class.    So  also, 

"  Sum  cum  justitiario 
Qui  te  modo  vario 
Possum  adjuvare, 
Si  vis  impetrare 
Per  suum  subsidium. 
Da  michi  dimidium 
Et  te  volo  juvare"  (p.  193). 

The  song  after  the  battle  of  Lewes — which  has  also  been  separately  edited — 
states  the  respective  positions  of  the  king  and  the  barons.  The  patriotic 
poetry  of  Laurence  Minot  (1333-1352)  illustrates  another  side  of  the  national 
spirit.  The  effect  of  the  Renaissance  in  England  was  then  to  awaken  this 
spirit  of  political  liberty,  to  *'  reveal  a  consciousness  of  united  purpose  and 
-corporate  pride  in  the  nation,  for  which  no  contemporary  parallel  can  be 
found  in  any  other  country  of  Europe."    **  The  time  had  not  yet  come  for 


Digitized  by  VjOOQ IC 


470  AMERICAN  JOURNAL   OF  PHILOLOGY, 

England  when  the  masterpieces  of  ancient  literature  could  exercise  a  refining 
influence  on  the  efforts  of  her  native  genius "  (p.  198).  This,  however,  was 
to  come  a  little  later. 

Prof.  Courthope  calls  Langland  the  Naevius  and  Chaucer  the  Ennius  of 
English  poetry,  and  the  two  following  chapters  are  devoted  respectively  to 
a  study  of  each  of  these  great  poets.  This  order  is  preferable  to  that  of  M. 
Jusserand,  who  treats  Chaucer  and  Gower  before  Langland,  whereas  the  first 
form  of  Langland's  work  was  written  before  Chaucer  had  even  translated  the 
Romauntof  the  Rose,  Langland*s  Vision  of  Piers  Plowman  is  rightly  called  "  a 
classic  work  in  English  literature " ;  his  "  vigorous  satire,  vivid  powers  of 
description,  strong  sense  of  justice,  so  faithfully  reflect  the  conscience  of  the 
English  people,  that  his  Vision  often  seems  to  be  projecting  its  light  upon  the 
ethical  problems  of  our  own  day."  The  author  discusses  "  the  two  great 
principles  on  which  society  in  the  Middle  Ages  rested,  Catholicism  and 
Chivalry,"  and  shows  that  they  **  reached  their  grand  climacteric,  and  sank 
into  rapid  decay."  Even  before  Wycliff  the  religious  sense  of  the  time 
embodied  itself  in  Langland's  great  work.  It  is  analysed  very  fully,  and  a 
parallel  is  drawn  between  Dante  and  Langland:  "both  poets  present  an 
image  of  the  ideal  or  spiritual  order  of  nature  and  human  society,  in  striking 
contrast  with  the  actual  course  of  the  world  ";  '*  but  Dante's  conception  was 
based  on  the  metaphysical  side  of  Catholic  Christianity,  Langland's  on  the 
ethical  and  practical  side '';  this  gives  the  keynote  of  the  criticism. 

A  brief  extract  from  the  Brunanburh^  with  many  misprints,  and  one  from 
Piers  Plowman^  are  given  as  illustrative  of  the  metre,  but  Prof.  Courthope  does 
not  analyse  the  structure  of  the  Anglo-Saxon  verse.  The  two  do  not  admit 
of  strict  comparison.  Langland  merely  employs,  though  with  great  skill,  the 
old-fashioned  alliteration  as  an  ornament  to  his  verse,  for  he  has  been  to 
some  extent  affected  by  the  Renaissance  in  his  metrical  form,  and  does  not 
attempt  to  reproduce  accurately  the  original  rhythm,  but  he  uses  the  same 
metrical  principle.  His  metre  is  sometimes  very  regular,  but  at  others  very 
irregular,  and  at  variance  with  the  older  scheme. 

An  interesting  chapter  on  Chaucer  and  his  works  follows  that  on  Langland. 
It  is  confessedly  based  on  Prof.  Skeat's  edition,  but  unfortunately  Courthope 
does  not  always  follow  Skeat  in  his  quotations,  and  hence  his  text  is  some- 
times bad  and  needs  emendation.  We  note  in  passing  on  p.  252  the  common 
misprint  of  Village  for  Visage,  As  to  the  burning  question  of  the  final  v, 
Courthope  is  inclined  to  follow  Payne  in  Essays  on  Chaucer  (Chaucer  Society, 
IV,  pp.  84-154)  vs.  Skeat.  He  thinks  no  positive  answer  can  be  returned  to 
the  question  '*  whether  or  not  it  was  pronounced  at  the  end  and  in  the  caesura 
of  an  English  verse,  in  words  where  it  had  a  grammatical  significance"  (p.  256). 
He  is  willing  to  grant  that  '*  strong  arguments  are  forthcoming  on  both  sides 
of  the  question,"  but  he  leaves  each  reader  to  follow  his  own  taste.  This 
view  would  now  be  regarded  as  antiquated,  and  most  scholars  would  certainly 
prefer  to  follow  Morris  and  Skeat  and  pronounce  the  final  -^  notwithstanding 
the  **  feminine  rhymes."  The  existing  translation  of  the  Romaunt  of  the  Rose 
is  regarded  as  Chaucer's,  and  we  are  not  even  informed  that  there  is  any  doubt 
about  it.  An  extract  from  it  is  given  as  showing  that  Chaucer  "furnished 
the  English  language  with  a  new  standard  of  versification  which  no  poet 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES,  47 1 

henceforth  could  afford  to  disregard";  and  it  is  **not  only  remarkable  as 
making  a  landmark  in  the  refinement  of  our  Tersification,"  but  **  it  marks  with 
equal  significance  the  rise  of  a  new  spirit  in  English  poetry,  the  importation 
of  thoughts  and  themes  from  the  Continent,  announcing  the  approach  of  the 
Renaissance  *'  (p.  258).  While  this  work  and,  at  a  farther  remove,  Troilus  and 
Criseyde^  are  treated  as  showing  Chaucer's  powers  as  a  translator,  the  Book  of  the 
Duchess^  the  ParUment  of  FouUs^  the  House  of  Fame^  and  the  Legend  of  Good 
fVomen,  show  his  powers  as  an  imitator;  and  the  Canterhtry  Tales^  as  an 
inventor.  They  are  "  the  full  harvest  of  the  art  of  the  trouvire,"  who  *•  was 
the  lineal  literary  descendant  of  the  tribal  gleeman  "  (p.  279). 

The  trouvire  was  dependent  upon  the  Ftibles  of  Bidpai  and  the  History  of 
the  Seven  Wise  Masters^  of  Hindoo  and  Persian  origin,  which  provided  him 
with  his  models.  Thus  originated  mediaeval  story-telling.  **  The  main  object 
of  the  literary  trouvire  was  to  collect  appropriate  subjects,  and  Chaucer,  with 
his  habits  of  encyclopaedic  study  and  omnivorous  reading,  had  amassed  a 
supply  of  stories,  not  indeed  so  numerous  as  those  collected  by  Boccaccio,  but 
covering  a  wider  range  of  tastes  and  interests  "  (p.  288).  His  framework  was 
his  own,  and  who  but  Chaucer  could  have  given  us  such  inimitable  pictures  of 
the  various  characters  in  English  society  ?  A  table  of  the  Tales^  with  their 
respective  sources,  is  given,  and  a  summary  of  the  time  spent  on  the  journey, 
after  Skeat.  A  criticism  of  Chaucer's  excellences  closes  the  chapter,  and  he 
is  pronounced  *'  the  first  national  poet  of  England."  Chaucer  emancipated 
poetry  from  the  trammels  of  *'  Metaphysics,  Allegory,  and  Theology,  and  from 
the  deductive  methods  of  thought  encouraged  by  encyclopaedic  science,*'  and 
"  reanimated  it  by  the  old  classical  principle  of  the  direct  imitation  of  nature." 
Others  developed  this  principle,  Ariosto,  Cervantes,  Molidre,  "  but  to  Chaucer 
must  be  assigned  the  honor  of  having  led  the  way."  Thus  the  movement  of 
the  fourteenth  century  from  the  mediaeval  to  the  modern  had  its  pioneer  in 
Chaucer. 

Our  limits  permit  but  a  mention  of  the  succeeding  chapters.  One  follows 
on  "The  Epical  School  of  Chaucer — Gower,  Lydgate,  Occleve,"  in  which 
these  poets  are  much  more  fully  treated  than  by  Jusserand.  The  Progress  of 
Allegory  in  English  Poetry  is  next  considered,  as  illustrated  in  the  Pearl  znd. 
the  Vision  of  Piers  the  Plowman^  which  exemplify  two  opposite  modes  of 
treatment,  the  contemplative  and  the  active,  in  the  Temple  of  Glass,  which 
follows  the  rules  of  the  fashionable  Love-allegory,  in  the  Omrt  of  Love,  of 
mach  later  date,  and  in  the  Kin^s  Quair,  which  stands  midway  between  the 
other  two,  and  **  forms  a  landmark  in  the  history  of  allegorical  poetry." 
These  are  followed  by  the  works  of  Henryson,  Dunbar,  and  Douglas,  in 
Scotland;  Hawes,  Skelton,  and  Barclay,  in  England.  The  last  has  been 
osually  known  only  from  his  translation  of  Brandt's  SMp  of  Fools,  but  his 
Eclogues,  in  which  "the  bucolic  style  is  adopted  merely  as  a  vehicle  of  a 
moral  allegory,"  are  more  fully  treated  by  Prof.  Conrthope.  Allegory  was 
a  most  popular  form  of  composition,  and  it  seems  that  scarcely  any  writer 
could  avoid  using  it  as  a  vehicle  for  moral  instruction.  It  is,  indeed,  as  old  as 
Esop,  and  the  Faerie  Queen  and  the  Pilgrim*s  Progress  are  doubtless  the  most 
brilliant  examples  of  it  in  English  literature,  but  the  taste  for  it  has  declined, 
and  it  is  now  considered  heavy  and  antiquated. 


Digitized  by  VjOOQ IC 


472  AMERICAN  JOURNAL    Ot  PHILOLOGY, 

The  Rise  of  the  Drama  in  England  is  next  treated  in  the  familiar  forms  of 
the  Mysteries  and  the  Moralities,  but  we  find  nothing  new, — a  mere  synopsis 
of  the  well-known  authorities. 

A  chapter  on  the  Decay  of  English  Minstrelsy  succeeds,  and  here  the 
deficiency  in  Jusserand  is  supplied.  Percy's  and  Ritson's  views  are  dis- 
cussed ;  the  author  considers  Percy  **  amply  warranted  in  concluding  that '  the 
minstrels  seem  to  have  been  the  genuine  successors  of  the  ancient  bards.'" 
His  very  free  '*  editing  "  of  his  MS.  is  not,  however,  endorsed.  The  changes 
in  the  art  of  minstrelsy  are  traced,  **  as  illustrated  by  the  progress  of  society 
from  the  tribal  to  the  civil  state,  by  the  transition  from  oral  to  written  poetry, 
and  by  the  character  of  the  ballad."  The  ballad  is  defined,  its  origin  and 
development  are  treated,  and  some  illustrations  are  given,  particularly  the 
Mary  Hamilton^  as  showing  that  the  ballad  "  was  a  type  of  poem  adapted  by 
the  professors  of  the  declining  art  of  minstrelsy  from  the  romances  once  in 
favor  with  the  educated  classes." 

A  Retrospect  of  some  half-dozen  pagest  giving  a  summary  of  the  History, 
closes  the  volume.  With  the  exception  of  the  third  chapter  on  the  poetry  of 
the  Anglo-Saxons,  Prof.  Courthope  has  given  us  in  this  volume  an  interesting 
contribution  to  a  complete  history  of  English  poetry. 

In  the  second  volume  the  same  design  as  that  noted  in  the  first  volume,  of 
tracing  the  course  of  English  poetry  by  the  stream  of  the  national  thought  and 
imagination,  and  by  its  European  relations,  is  continued,  and  it  is  now  all  the 
more  important  because  **  the  sixteenth  century  is  the  great  age  of  transition  from 
mediaeval  to  modern  times ;  the  chief  poets  of  the  period  work  from  the  basis  of 
culture  provided  for  them  by  the  Middle  Ages,  but  they  are  alive  to  all  the  inflv- 
ences  of  their  own  age ;  and,  like  their  ancestor  Chaucer,  they  avail  themselves 
of  ideas  and  feelings  flowing  in  upon  them  from  a  foreign  source."  A  sketch  is 
first  given  of  the  religious  and  political  system  of  Europe  in  the  early  sixteenth 
century,  "  in  the  still  Catholic  European  community,"  as  shown  in  the  Diet  of 
Augsburg  (i  51 8).  The  idea  of  the  modern  state  arose  out  of  the  decaying  fabric 
of  the  Christian  Republic,  and  "  Spain,  France,  and  England  began  to  display  a 
clearly  marked  individuality  in  all  matters  relating  to  religion,  art,  literature 
and  manners."  This  is  seen  in  the  works  of  the  great  European  writers.  The 
Courtier  of  Castiglione  and  the  Discourses  and  the  Prince  of  Machiavelli,  "  par- 
ticularly impressed  the  minds  of  knightly  poets  and  scholarly  dramatists  in 
England."  Mr.  Courthope  accounts  for  Machiavelli  by  the  circumstances  by 
which  he  was  surrounded,  and  thinks  that  his  works  exerted  an  all-powerful 
influence.  The  Colloquies  of  Erasmus  too  exerted  a  strong  influence  "  on  the 
more  reflective  part  of  European  society  in  the  sixteenth  century,  by  educating 
public  opinion  indirectly  in  a  more  rational  scheme  of  manners  and  conduct"; 
and  Luther's  treatise  on  Christian  Liberty  exerted  a  similar  influence  in  the 
religious  sphere.  The  three  countries  of  Spain,  England,  and  France  were 
sufficiently  organized  to  receive  the  influences  of  the  Renaissance  and  the 
Reformation.  Spain  resisted  them ;  More's  Utopia  showed  how  **  the  unity  of 
Christendom  might  be  expanded  to  satisfy  modem  requirements,"  and  it  was 
destined  to  bear  fruit  hereafter;  in  France  power  was  concentrated  in  the 
hands  of  the  king,  who  was  absolute ;  the  writers  reflected  the  ideas  of  the 
Court,  and  lacked  the  ideas  of  rational  liberty  and  toleration  seen  in  the 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES,  473 

Utopia,  The  author  traces  these  influences  so  as  to  give  a  view  of  '*  the  col- 
lective forces  acting  on  the  imagination  of  Europe  at  the  beginning  of  the  six- 
teenth century."  Especially  *'  the  idea  of  liberty  of  thought  and  action,  in 
the  constitution  both  of  the  State  and  the  Individual/'  arises,  and  all  of  these 
influences  affiect  the  English  imagination,  and  "begin  to  break  up  the  solid 
structure  of  traditional  belief  and  ancient  chivalry."  Prof.  Courthope's 
method  is  thus  seen  to  be  very  different  from  that  of  Prof.  Morley,  more 
philosophic,  and  directed  to  tracing  the  history  of  ideas  and  how  these 
European  ideas  affected  English  literature. 

After  this  sketch  of  "intellectual  conflict  in  Europe  in  the  sixteenth 
century,"  the  following  chapters  are  devoted  to  showing  its  effect  in  England 
in  the  poems  of  Wyatt  and  Surrey,  the  pioneers  of  the  English  Renaissance. 
In  the  poems  of  Wyatt  we  see  the  first  results,  since  Chaucer,  of  the  study  of 
Italian  literature.  He  is  distinguished  from  the  preceding  poets  by  "  the  indi- 
vidual energy  of  his  thought  and  his  persistent  imitation  of  foreign  miodels." 
To  him  is  due  the  credit  of  the  introduction  of  the  sonnet  into  English  litera- 
ture ;  but,  although  he  followed  the  Petrarchan  model,  from  his  unfortunate 
lack  of  ear  he  was  unable  to  make  the  improvement  in  English  versification 
which  characterised  the  poems  of  Surrey.  In  his  satires  he  imitated  Alamanni 
in  his  use  of  the  tena  rima^  and  he  also  combined  the  Alexandrine  with  the 
Septenar,  so  that  he  struck  out  new  paths  in  English  verse,  even  if  he  did  not 
possess  the  skill  to  handle  his  instrument  very  successfully.  **  Wyatt  is  a 
noble  figure  in  English  poetry.  His  strength,  his  ardor,  his  manliness,  his 
complete  freedom  from  affectation,  make  him  a  type  of  what  is  finest  in  the 
national  character,  and  there  is  little  exaggeration  in  the  very  fine  epitaph 
written  on  him  by  his  great  contemporary,  Surrey" — which  epitaph  closes 
the  chapter. 

Surrey  was  a  man  of  more  ardent  disposition  than  Wyatt,  and  not  so  grave 
a  character.  He  lacked  Wyatt's  **  vehement  individuality,"  "  but  he  succeeds 
where  Wyatt  failed  in  naturalizing  the  ideas  he  borrows  by  the  beauty  of  his 
style,"  and  to  it  **he  owes  his  great  position  in  the  history  of  English  poetry." 
His  unfortunate  death,  or  murder  rather,  on  a  trumped-up  charge  by  the 
advisers  of  the  king,  when  Henry  was  on  his  death-bed,  was  a  great  loss  to 
English  letters.  Like  Sir  Philip  Sidney  later,  he  was  the  flower  of  chivalry, 
and  both  by  birth  and  character  was  a  fine  representative  of  English  nobility. 
His  "  Fair  Geraldine  "  was  an  idealized  lady-love,  and  his  love-poems  were, 
like  Wyatt's,  modeled  after  Petrarch;  but  he  possessed  higher  gifts  than 
Wyatt  of  "  terseness,  sweetness,  purity,  and  facility  of  style."  His  reform  of 
English  versification  is  traced  under  several  heads,  showing  that  he  grounded 
himself  on  Chaucer,  but  it  is  unfortunate  that  Prof.  Courthope  should  use  such 
a  poor  text,  with  so  many  misprints,  as  that  given  on  p.  87  as  a  specimen  of 
the  beginning  of  the  Prologue ;  Prof.  Skeat  might  have  supplied  him  with  a 
better  one.  Surrey's  sonnet-stanza  is  that  used  afterwards  by  Shakspere  and 
is  not  the  Petrarchan  form.  He  too  first  used  blank  verse  in  his  translation  of 
Virgil ;  and  "  he  was  also  the  first  to  refine  the  system  of  poetical  diction  so 
as  to  adapt  it  to  the  reformed  versification." 

But  while  Wyatt  and  Surrey  were  thus  refining  the  style  and  versification  of 
English  poetry  after  Italian  models,  political  poetry  was  being  developed  in 


Digitized  by^ 


474  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Scotland  in  the  writings  of  Sir  David  Lyndsay,  and  a  little  later  in  England  in 
the  huge  collection  of  political  tragedies  known  as  the  Mirror  far  Magistrates^ 
which  owes  its  claim  to  consideration  as  poetry  to  Sackville's  Induction  and 
his  Complaint  of  the  Duke  of  Buckingham,  Lyndsay  was  a  follower  of  Donglas 
and  Dunbar  in  his  use  of  dream  and  allegory,  but  he  was  more.  He  used  his 
allegory  to  illustrate  the  political  condition  of  the  country.  Lyndsay's  manner 
is  still  mediaeval :  it  points  backward,  but  his  matter  points  forward,  and,  in 
its  *' union  of  Lutheran  piety,  political  philosophy,  and  classical  imagery ,*** 
reflects  the  mind  of  the  Scottish  aristocracy  on  the  eve  of  the  Reformation. 
A  full  account  is  given  of  the  composition  of  the  Mirror  for  Magistrates,  and 
its  language  and  versification  are  pronounced  to  be  "  full  of  instruction  for  the 
student  who  seeks  to  trace  systematically  the  growth  of  the  art  of  English 
poetry"  (p.  17).  The  transition  in  style  from  the  rude  and  archaic  of  the- 
early  sixteenth  century  to  the  finished  manner  of  the  Elizabethan  writers  is- 
here  seen.  It  culminates  in  Sackville's  Induction,  for  Sackville  showed  the 
beneficial  influence  of  Surrey.  *'  Of  the  epic  poets  of  England,  if  Chaucer  is 
the  first  to  exhibit  the  genuinely  classic  spirit,  Sackville  is  the  first  to  write  in 
the  genuinely  classic  manner." 

A  chapter  follows  on  the  translations  of  the  classics,  due  to  the  establish- 
ment of  the  new  learning  in  the  universities,  and  the  love  of  learning  shown 
by  Elizabeth  herself.  **  Wolsey  promoted  the  study  of  Greek  by  the  founda- 
tion of  Christ  Church  [Oxford].  Colet  and  Grocyn  lectured  on  the  Greek 
orators  and  poets  in  the  same  university ;  and  Cheke  and  Ascham  familiarized 
their  scholars  at  Cambridge  with  the  dialogues  of  Plato,  the  philosophy  most 
highly  approved  by  the  reformers  of  the  Continent.  Here  was  an  influence 
that  could  not  fail  to  be  felt  in  literature.  But  Elizabeth,  according  to  Roger 
Ascham,  knew  Latin,  Greek,  French  and  Italian,  and  could  speak  with  facility 
all  of  those  languages.  Ascham  was  her  preceptor  in  Latin  and  Greek  for  two 
years,  so  he  ought  to  know.  This  learned  influence  communicated  itself  to 
the  Court,  the  Universities,  and  the  writers.  A  translation  of  the  Aeneid  had 
been  made  by  Douglas  in  151 3,  Surrey  had  translated  Books  II  and  IV  into 
blank  verse,  the  first  in  English,  and  now  Thos.  Phaer,  1 555-1560,  and  Thos. 
Twine,  1562,  translated  the  whole  of  it  into  the  iambic  septenar.  Jasper 
Hey  wood,  1559-61,  translated  three  of  the  tragedies  attributed  to  Seneca  into 
the  same  ballad  metre ;  and  Arthur  Golding,  1565-67,  turned  into  the  same 
verse  Ovid's  Metamorphoses;  this  was  the  translation  which  Shakspere  used. 
All  of  these  translations  testify  to  a  desire  for  a  knowledge  of  the  new  learning 
in  English  form. 

Now  came  many  imitators  and  followers  of  Wyatt  and  Surrey  in  the 
numerous  Miscellanies  of  Elizabeth's  reign,  which  were  preceded  in  1557  by 
the  most  important  of  all,  TotteVs  Miscellany,  containing  the  poems  of  Wyatt 
and  Surrey,  of  Thomas,  Lord  Vaux,  Grimald,  and  others.  Grimald  led  in 
the  pedantic  and  conceited  style  of  the  later  school  of  Cowley,  dubbed  by  Dr. 
Johnson  "  metaphysical,"  doubtless  merely  because  it  was  unnatural.  He  was 
the  first  to  imitate  Surrey  in  the  use  of  blank  verse,  in  which  he  showed  his 
good  sense,  writing  in  that  verse  a  poem  on  the  Death  of  Cicero,  Googe  fol- 
lowed with  his  Eglogues,  Epytaphes,  and  Sonettes,  and  Turbervile,  with  his  Songs^ 
Epitaphs^  and  Epigrams,  Googe,  who  translated  Thi  Zodiac  of  Life^  of  Marcel lus 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES,  475 

Palingenitts,  and  Turbervile,  Ovid's  HetoUal  Epistles  and  Mantuan's  Echgius, 
Chnrchyard,  who  had  contributed  "a  tragedy  called  Shores  IVi/e'*  to  the 
Mirror  for  Magistrates^  contributed  poems  also  to  these  Miscellanies^  and  lived 
to  be  called  old-fashioned  by  the  later  Elizabethans.  But  he  was  far  exceeded 
by  Gascoigne,  after  Sackville  the  greatest  name  in  English  letters  before 
Spenser.  Gascoigne  was  a  leader  in  many  ways  and  deserves  more  credit 
than  he  has  usually  received.  Prof.  Courthope  does  not  seem  to  be  aware  of 
Prof.  Schelling's  excellent  monograph  on  his  life  and  works.  He  wrote  son- 
nets, lyrics,  a  satire,  The  Steel  Glass,  and  other  Posies^  and  was  a  critic  of 
English  verse — but  we  shall  have  to  return  to  him  under  the  drama. 

Gascoigne  died  in  1577,  and  we  now  reach  the  middle  of  Elizabeth's  reign, 
the  beginning  of  the  Elizabethan  efflorescence  in  literature.  The  preceding 
thirty  or  forty  years  had  been  preparation  and  now  we  have  fulfillment. 
Three  chapters  follow,  treating  Court  Dialect,  as  seen  in  Lyly,  Court  Romance, 
in  Sidney,  and  Court  Allegory,  in  Spenser.  Euphuism  is  treated  as  a  move- 
ment towards  refinement  in  language  which  affected  every  literature  in  Europe. 
Although  the  Euphues  and  the  Arcadia  are  written  in  prose,  the  author  thinks 
they  are  '*  so  closely  associated  with  metrical  composition  and  with  the  pro- 
'  gress  of  English  taste  that  it  would  be  unphtlosophical  to  regard  them  as 
beyond  the  limits  of  a  history  of  English  poetry."  This  illustrates  the  dis- 
advantage of  treating  poetry  and  prose  separately  in  a  history  of  English 
literature ;  they  mutually  act  upon  each  other,  and  both  must  be  considered 
in  any  complete  view  of  the  national  thought  and  imagination.  Passing  over 
what  is  said  of  Italy,  Spain  and  France,  we  find  the  problem  in  England  of 
forming  a  standard  of  literary  composition  more  difficult  because  of  the  mix- 
ture of  races  and  languages,  but,  fortunately  for  the  development  of  style, 
"French  influence  so  far  prevailed  that  the  order  of  words  in  a  sentence 
follows  the  logical  order  of  the  thought.*'  This  is  an  advantage,  it  may  be 
noted,  that  the  Germans  have  never  attained,  and  hence  their  involved  and 
cumbrous  prose  style.  The  introduction  of  Latinized  words  was  carried 
so  far  that  many  never  took  root.  Witness  Douglas's  dulcorate  and  facund^ 
Lyndsay's  prepotent  and  celsitude^  and  Wilson's  specimen  (p.  183),  which  he 
says  is  no  caricature,  containing  such  words  as  reifoltUing,  ingenia,  accessited, 
adjuvate,  obtestate,  contignate,  inidgUate,  &c.,  &c.,  so  that  it  requires  that  one 
should  know  Latin  in  order  to  know  English.  It  may  be  remarked  in  pass- 
ing that  the  date  here  given  for  Wilson's  Art  of  Rhetoric^  the  first  treatise  of 
the  kind  in  English,  is  1562,  whereas  on  p.  289  it  is  given  as  1553.  and  this 
was  not  the  first  edition.  The  date  of  this  work  is  of  importance  as  bearing 
upon  the  date  of  our  first  comedy,  Ralph  Roister  Doister.  The  history  of 
Euphuism  is  traced,  and  it  is  carried  back,  as  usual,  to  Guevara.  If  Prof. 
Courthope  knows  of  Landmann's  study  of  the  subject,  he  does  not  mention  it. 
The  common  characteristics  of  this  stylistic  fad,  as  seen  in  Lyly's  work, 
natural  history  metaphors,  antithesis  and  alliteration — **  transverse  allitera- 
tion," as  Landmann  calls  it — are  duly  noted,  and  its  great  influence  remarked. 
** Euphues -wtL^  as  much  esteemed  by  polite  society  as  by  the  critics.  It  was 
accepted  with  the  Arcadia  as  fixing  the  standard  of  eloquence  at  Court." 
Euphues  is  called  "  an  example  of  rhetoric  in  the  language  of  love  composed 
to  suit  the  taste  of  the  Court,"  hence  its  flattery,  its  logic  and  its  illustration. 


Digitized  by  VjOOQ IC 


476  AMERICAN  JOURNAL    OF  PHILOLOGY. 

'*  The  metaphorical  style  in  the  love-poetry  of  the  Euphuists  is  a  natural 
growth  of  the  classical  Renaissance:  it  marks  the  decay  of  the  allegorical 
interpretation  of  nature,  which  itself  largely  accounts  for  the  abundant  use  of 
metaphor  in  the  poetry  of  the  Middle  Ages."  The  influence  of  Euphues  lasted 
for  a  hundred  years,  and  Prof.  Courthope  considers  Lyly's  discovery  as  of 
permanent  value,  for  **  he  perceived  the  advantage  of  clearness,  correctness  and 
precision,  in  the  arrangement  of  words."  For  this  he  should  receive  due 
credit,  and  his  style,  when  purged  of  its  unnecessary  adjuncts,  left  a  residuum 
of  value  in  the  history  of  English  prose  style.  "Addison  and  Steele  .  .  » 
learned  from  Lyly  how  to  present  genuine  thoughts  in  an  artistic  form ;  and 
Burke,  Johnson,  and  Macaulay,  .  .  .  followed  his  example  in  working  up- 
sentences  and  periods  to  the  climax  required  for  the  just  and  forcible  pre- 
sentation  of  the  argument "  ^  (pp.  aoi-a). 

Sidney  was  at  the  head  of  the  school  opposed  to  the  Euphuists,  and  he  criti- 
cises them  in  one  of  his  sonnets.  Sidney,  Dyer,  Harvey,  and  Spenser  for  a 
time,  until  his  good  sense  predominated,  wished  to  reform  English  versification 
after  the  Latin,  and  perpetrated  certain  barbarous  hexameters.  Edward  Vere*. 
Earl  of  Oxford,  was  at  the  head  of  the  other  party,  and  was  a  great  favorer  of 
the  Euphuists ;  he  and  Sidney  quarreled  and  it  came  near  resulting  in  a  dueL 
On  Sidney's  retirement  from  the  Court  in  consequence  of  his  bold  protest 
against  the  Anjou  marriage,  he  amused  himself  with  writing  the  Arcadia  for  hi» 
sister  Mary, "  Sidney's  sister,  Pembroke's  mother,*'  intermingling  poetry  with  hi» 
prose  romance.  This  work  is  treated  at  some  length  and  its  literary  defects 
are  commented  on.  **  The  action  is  wanting  in  human  interest,  the  characters 
are  conventional,  the  structure  of  the  story  is  confused  and  irregular."  The 
style  too  is  criticised :  **  Nothing  is  said  plainly;  commonplace  is  disguised  by 
metaphor ;  style  is  mechanically  elevated  by  a  tricky  arrangement  of  words.*" 
But,  "  regarded  historically,  as  a  mirror  of  the  feelings  of  Sidney  and  the  best 
of  his  contemporaries,  and,  as  a  work  of  fiction  contributing  to  the  develop- 
ment of  the  English  drama,  the  Arcadia  is  a  most  interesting  monument.'*' 
The  element  which  affected  contemporary  taste  was  derived  from  the  study  of 
Montemayor,and  consisted  in  '*  concentrating  the  main  interest  of  his  narrative 
in  the  complications  of  the  love-plots."  The  dramatists  were  indebted  to  the 
Arcadia  for  sentiment  and  landscape,  for  development  of  action  and  character^ 
and  for  the  complications  arising  from  the  disguise  of  sex. 

The  Astrophel  and  Stella  sonnets  are  also  treated  with  critical  insight,  and 
Prof.  Courthope  differs  from  several  other  critics.  Lamb,  Trench,  and  Symonds, 
in  his  interpretation.  He  thinks  that  their  "theory  of  a  profound  and  all- 
pervading  passion  is  contradicted  by  the  facts  of  the  case,  by  the  character  of 
Sidney,  by  the  character  of  the  sonnets  themselves."  Each  of  these  points  is 
developed,  and  I  must  say  that  I  am  inclined  to  agree  with  Prof.  Courthope. 
This  attempt  of  critics  to  find  in  ideal  love-poems  some  personal  reference, 
which  has  been  "  run  into  the  ground "  in  the  case  of  Shakspere's  Sonnets, 
seems  to  me  far-fetched  and  mistaken.    Penelope  Devereux  was  a  young  girl 

II  may  be  permitted  to  remark  just  here  that  in  my  "Selections  in  English  Prose  from 
Elizabeth  to  Victoria  "  (1891)  I  purposely  beg^n  with  Lyly  as  the  pioneer  in  the  formation  of 
English  prose  style,  and  I  fully  concur  with  Prof.  Courthope  in  giving  credit  to  Lyly  for  his 
contribution  to  this  object. 


Digitized  by  VjOOQ IC 


REVIEIVS  AND  BOOK  NOTICES,  477 

who  attracted  Sidney's  attention  when  she  was  nnder  fifteen,  and  her  father 
was  anxious  for  the  match.  Some  cause*  now  unknown,  prevented  it,  and  after 
she  was  married  to  Lord  Rich  at  the  aj^e  of  nineteen,  she  became  Sydney's 
"  Laura.'*  and  received  his  ideal  adoration.  **  Artistic  opposition  to  the 
Euphuists  "  also  inspired  these  Sonnets,  and  "  sonnet  after  sonnet  sounds  the 
note  that  love  alone  is  an  adequate  source  of  inspiration,  without  the  artificial 
supplement  of  science  and  learning."  This  is  a  much  more  reasonable  theory 
than  the  personal  one. 

The  chapter  on  Court  All^ory  is  a  very  full  criticism  of  Spenser.  As  the 
greatest  genius  since  Chaucer,  and  the  writer  in  whom  the  influences  of  the 
Elizabethan  age  culminated,  he  receives  the  fullest  share  of  critical  attention. 
'*  He  wanted  no  quality  required  to  place  him  in  the  same  class  with  Homer, 
Virgil,  Dante,  Shakespeare,  Milton,  and  perhaps  I  may  add  Chaucer,  but  that 
supreme  gift  of  insight  and  invention  which  enables  the  poet  to  blend  con- 
flicting ideas  into  an  organic  form."  Prof.  Courthope  does  not  agree  either 
with  those  who  regard  Spenser  **  primarily  as  a  poetical  philosopher^*  or  with 
Lowell,  who  thinks  '*  the  true  use  of  Spenser  is  as  a  gallery  of  pictures,*'  and 
who  compares  the  moral  to  "  a  bit  of  gravel  in  a  dish  of  strawberries  and 
cream.**  The  sense  is,  for  Courthope,  a  characteristic  part  of  his  work,  but 
the  allegory  is  mainly  interesting  in  so  far  as  it  serves  the  purposes  of  poetry » 
not  then  as  a  vehicle  of  moral  truth.  The  designs  of  the  poems  are  separately 
examined,  and  artistic  unity  is  found  even  in  The  ShephercTs  Calendar,  These 
poems  were  his  experiments  in  poetical  diction.  The  design  of  the  Faery 
Qtieen  is  found  in  the  letter  to  Raleigh,  not  in  the  poem  itself.  **  As  he  sought 
in  The  Shepherd's  Calendar  to  treat  the  Eclogue  in  a  new  style,  so  in  the  Faery 
Queen  he  aimed  at  producing  a  variety  of  the  Romantic  Epic  of  the  Italians.'* 
As  is  seen  in  Harvey's  correspondence,  he  sought  to  **  overgo  Ariosto." 

His  works  are  examined  to  see  "  how  far  his  conceptions  were  formed  in 
harmony  with  the  laws  of  his  art,  and  how  far  his  execution  did  justice  to  his 
subjects  as  he  conceived  them."  In  the  examination  of  The  Shepherd's  Calen" 
dar  from  a  metrical  point  of  view,  Mr.  Courthope  thinks  that  "  the  metres  of 
several  of  his  Eclogues  are  founded  on  what  he  erroneously  believed  to  be  the 
metre  of  Chaucer's  Canterbury  Tales**  and  that  he  read  the  Prologue  with  four 
accents  instead  of  five,  as  follows: 

*'  Whdnne  that  A  |  prfle  with  his  |  sh6wres  |  s6te 
The  dr6ught  |  of  Mdrch  |  had  perc'd  |  to  the  r6te,"  &c. 

This  is  a  remarkable  mixture  of  dactylic-trochaic  and  iambic*anapaestic 
rhythm,  the  latter  predominating  in  the  following  lines,  but  from  Spenser's 
ignorance  of  Chaucer's  language,  it  is  barely  possible  that  he  so  read  the  Pro- 
logue, for  he  does  employ  this  "composite  style"  in  this  poem.  It  is  pro- 
nounced, on  the  whole,  **a  truly  beautiful  and  graceful,  if  somewhat  artificial 
composition." 

The  Faery  Queen  is  also  examined  at  length  and  compared  with  Ariosto. 
*'  Ariosto's  word-painting  is  unequalled  for  brilliancy  and  distinctness  of  color, 
but  Spenser  surpasses  him  in  depth  of  imagination."  Beside  being  a  great 
picture-gallery,  the  Faery  Queen  *'is  also  a  vast  experiment  in  English 
metrical  composition,"  and  Spenser's  treatment  of  his  stanza  is  judged  to  be 


Digitized  by  VjOOQ IC 


478  AMERICAN  JOURNAL    OF  PHILOLOGY. 

*'a  triumph  of  art."  Spenser  was  the  poet  of  chivalry  and  of  mediaeval 
allegory.  '*He  composed  his  poems  in  the  spirit  of  a  great  painter,  a  great 
musician/*  and  his  Faery  Queen  is  "  a  thing  of  beauty,  a  joy  forever." 

The  next  chapter  on  the  growth  of  criticism  and  its  effect  on  poetry,  with 
notice  of  the  poetical  euphuists,  traces  the  advance  of  national  taste  as  seen  in 
the  later  Elizabethan  Miscellanies  and  the  consequent  development  of  poetical 
criticism.  Gascoigne  was  here  a  pioneer,  for  his  Notes  of  Instruction  concern* 
ing  the  Making  of  Verse  is  the  first  critical  treatise  that  we  have  on  the  verse- 
practice  of  the  time.  Campion,  Webbe,  Puttenham,  and  Sidney  receive  brief 
notice,  and  the  four  main  groups  of  metrical  composers  are  successively 
described,  the  university  scholars,  Harvey  and  Fraunce,  who  wished  **to 
reform  the  national  poetry  on  classical  lines";  the  sonnet  writers,  Watson, 
Constable,  Lodge,  Daniel,  and  others,  who  imitated  Petrarch  and  the  Italians 
until  they  ran  out  in  the  nonsense  of  Barnes ;  the  court  poets,  Sidney,  Dyer, 
Essex,  Raleigh,  Oxford,  who  have  written  some  of  the  best  lyrics  of  the  age ; 
and  finally,  the  men  of  letters.  *'  who  embodied  the  spirit  of  the  Renaissance 
in  poetical  romance  or  classical  mythology,"  such  men  as  Breton,  Barnfield, 
Greene,  Lodge,  and  Marlowe.  These  last  have  left  the  most  permanent 
impress  on  the  poetry  of  the  time,  and  in  Marlowe's  Hero  and  Leander^^  have 
the  culmination  of  the  poetry  of  passion.  **  Though  his  style  is  colored  with 
the  conceits  and  mannerism  of  the  period,  yet,  as  compared  with  the  diction  of 
contemporary  Euphuistic  writers,  it  has  a  fiery  strength  and  vigor  not  to  be 
found  in  any  other  man." 

We  now  reach  that  form  of  literature  for  which  the  Elizabethan  age  was 
distinguished  par  excellence^  and  the  last  two  chapters  of  the  volume  are  devoted 
to  the  development  of  the  drama.  Schlegel's  theory  that  Shakspere  *'  owed 
hardly  anything  to  his  predecessors  "  is  rightly  excepted  to  ;  such  a  theory  is 
**  the  height  of  critical  superstition,"  and  Prof.  Courthope  devotes  his  efforts  to 
showing  the  historical  evolution  of  the  drama,  **  the  transition  from  pageant 
to  theatre,  from  interlude  to  tragedy,  comedy,  and  history."  The  professed 
historians  of  the  drama,  Collier,  Ward,  and  Symonds,  have  done  much  to 
elucidate  its  history,  but  the  author  thinks  that  **  something  still  remains  to  be 
done,"  and  he  devotes  this  chapter  to  tracing  **  the  slow  gradations  by  which 
the  dramatic  art  passed  out  of  the  rudimentary  conditions,  peculiar  to  it  in  the 
Middle  Ages,  into  the  hands  of  those  who  brought  the  form  of  the  romantic 
drama  to  its  full  perfection."  Without  going  into  details,  which  would  take 
more  space  than  we  have  at  command,  we  may  say  that  the  facts- fully  justify 
Mr.  Courthope's  position.  He  investigates  the  progress  of  the  stage  from  the 
Miracle  Play  to  the  Morality;  the  influence  exercised  upon  the  stage  by  the 
Court,  the  Universities,  and  the  Inns  of  Court ;  the  opposition  of  the  Puritans 
and  the  effect  of  the  building  of  theatres  outside  of  the  municipal  jurisdiction 
of  London ;  to  all  which  were  added  the  improvements  in  dramatic  art  made 
in  the  course  of  time.  Prof.  Courthope  analyses  many  of  these  Moralities,  and 
shows  the  gradual  progression  from  allegorical  personification  to  individual 
action,  from  mere  dialogue  to  development  of  a  complicated  plot.  Like  Will 
to  Like  (1568)  illustrates  the  manners  of  the  time,  and  here  we  have  a  mixture 
of  allegorical  and  individual  characters,  and  the  personal  Vice,  Nichol  New- 
fangle.     In  The  Three  Ladies  of  London  (1584),  **  the  genius  of  the  old  Morality 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES,  479 

probably  reached  the  highest  level  of  which  it  was  capable  ":  it  is  too  a  dramatic 
^satire  on  the  manners  of  the  day.  These  were,  however,  very  late  examples 
of  the  Moralities.  Full  credit  is  given  to  John  Heywood  for  the  steps  taken 
in  advance  by  him,  and  under  the  influence  of  the  study  of  Plautus  and 
Terence,  is  finally  reached  the  first  regular  comedy,  Udall's  Ralph  Rcisttr 
Doistfr  {tihoui  1550).  Progress  is  made  in  Gascoigne*s  translation  of  Ariosto's 
Supposes  and  in  his  Glass  of  Government^  but  Prof.  Courthope  omits  to  note 
that  this  last  is  based  on  the  Latin  Acolastus^  founded  on  the  parable  of  the 
Prodigal  Son.  It  was  no  more  original  with  Gascoigne  than  The  Supposes, 
Gviscoxgnt^s  Jocasta  too  was  but  a  translation,  from  the  Italian  of  Dolce,  of  the 
Phoenissae,  Lyly's  prose  Court  comedies  mark  the  highest  development  in 
this  direction.  Prof.  Courthope  does  not  notice  the  supposed  allegorical  sig- 
nification of  the  Endymion^  which  has  been  brought  out  in  the  Introduction 
to  Mr.  Baker*s  edition  of  that  play.  The  dramatic  movement  from  Interlude 
to  Comedy  is  summed  up,  from  didactic  allegory  to  imitation  of  manners, 
thence  to  action  of  human  personages  in  fuller  plots  and  with  greater  refine- 
ment of  dialogue. 

So  too  was  Tragedy  evolved  from  Interlude,  and  here  exception  is  taken  to 
Symonds's  view  of  two  types  of  tragedy,  one  modelled  after  Seneca  and  the 
other  after  the  Italian  plays,  the  latter  finally  prevailing.  The  tragedies  and 
tragi-comedies  *'  all  have  a  close  affinity  with  the  Interlude/'  and  there  was  no 
**  conflict  between  the  type  of  tragedy  favored  by  the  Court  and  that  dear  to 
the  people."  As  the  Mirror  for  Magistrates  presented  tragedy  in  epic  form, 
so  the  plays  presented  it  in  dramatic  form.  The  influence  of  Seneca  was 
plainly  seen  in  our  first  regular  tragedy,  Ferrex  and  Porrex,  or  Gorboduc  (1562), 
which  Mr.  Courthope  speaks  of  only  as  Sackville's,  ignoring  Norton  altogether, 
although  he  wrote  more  than  half  of  it.  Notwithstanding  Schlegel's  opinion, 
it  is  characterized  as  '*  a  work  of  great  merit,"  and  so  it  was  if  we  consider  its 
time,  but  it  was  well  for  English  tragedy  that  it  did  not  develop  on  the  lines 
of  Gorboduc,  Hughes  imitated  it  in  his  Misfortunes  of  Arthur  (1587),  but  this 
was  the  year  of  Marlowe's  Tamdurlaine,  and  no  one  thought  of  Seneca  after 
that.  Bale,  with  his  Ring  fohan,  *'  probably  written  during  the  reign  of 
Edward  VI/'  began  the  evolution  of  Chronicle  History  out  of  the  Interlude. 
So  throughout  the  sixteenth  century  there  was  a  gradual  progress  in  dramatic 
development.  The  Moralities  and  Interludes,  themselves  developed  from  the 
old  Miracle- Plays,  gradually  passed  into  Tragedy,  Comedy,  and  Chronicle 
History,  under  the  influence  of  the  Renaissance. 

The  last  chapter  is  occupied  with  a  study  of  the  infancy  of  the  romantic 
drama  as  seen  in  the  works  of  Greene,  Peele,  Marlowe,  and  Kyd.  Now  we 
find  a  conflict  in  England  between  the  principles  of  the  Renaissance  and  the 
Reformation.  Puritanism  showed  itself  in  the  attacks  made  upon  the  stage. 
Between  1570  and  1587  there  were  no  less  than  six  violent  attacks  in  pamphlet 
form  made  upon  the  stage.  Prof.  Arber  has  given  us  a  summary  of  these  in  his 
edition  of  Gosson's  School  of  Abuse  (1579),  which  was  an  invective  against 
"Poets,  Pipers,  Players,  and  such-like  Caterpillars  of  a  Commonwealth.'* 
Being  dedicated  to  Sir  Philip  Sidney,  it  produced  his  Apologie  for  Poetrie^  as 
Sidney  did  not  relish  the  inclusion  of  Poets  in  this  general  onslaught.  The 
Puritans  had  ample  justification  for  their  attacks.  Englishmen  had  become 
Italiematedf  and  here  too  Gascoigne  had  led  the  way,  but  he  repented,  so  that 


Digitized  by^ 


480  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Whetstone,  his  friend,  could  indite  funereal  verses  on  The  Well-employed  Life 
and  Godly  End  of  G,  Gascoigne,  Esquire,  which  elegy  the  curious  reader  will  find 
prefixed  to  Prof.  Arber*s  edition  of  The  Steel  Glass,  Greene,  however,  went 
further  than  Gascoigne  in  this  process  of  Italianisation,  and  if  what  he  states 
in  his  Repentance  and  his  Groats-worth  of  Wit  is  true,  he  was  veritably  **a  hard 
case."  Courthope  sees  no  reason  to  question  the  authenticity  of  his  auto- 
biography. Greene's  dramas,  all  of  which  except  Friar  Baeon  and  Friar  Bun- 
gay zx^  thought  to  show  Marlowe's  influence,  are  successively  noticed,  and 
'*what  is  best  and  most  characteristic  ...  is  the  poetry  of  his  pastoral  land- 
scape, and  his  representation  of  the  characters  of  women;  in  both  of  these 
respects  he  exercised  an  unmistakable  influence  on  the  genius  of  Shakes- 
peare." Greene  was  the  true  predecessor  of  Shakspere  in  comedy,  as  Mar- 
lowe was  in  tragedy.  Peele  is  thought  by  Prof.  Courthope  to  have  **a  finer 
range  of  imagination "  than  Greene,  and  he  is  given  a  higher  rank  than  that 
usually  assigned  him  by  the  historians  of  the  drama.  His  two  best  plays.  The 
Arraignment  of  Paris^  which  inspired  Shakspere  in  A  Midsummer  Nighfs 
Dream,  and  David  and  Bethsabe,  are  alone  noticed,  and  that  briefly.  The  bulk 
of  the  chapter  is  rightly  reserved  for  Marlowe,  '*  the  great  genius  who  may 
justly  be  called  the  founder  of  English  poetic  drama."  In  Marlowe  "the 
rupture  with  the  Puritanic  element  of  the  nation  was  absolute  and  compleie."^ 
Marlowe  wrote  *'  freed  from  the  restraints  of  Conscience  and  Law.'*  He 
believed  in  "  the  freedom  of  the  human  will,"  and  he  wrote  accordingly.  The 
incarnation  of  absolute  power  is  seen  in  Tamburlaine^  of  knowledge  in  Faustus.. 
"  Tamburlaine  is  the  type  of  resistless  force ;  Faustus  represents  the  resolute 
pursuit  of  knowledge  as  an  instrument  of  material  power."  Faustus  is  pro- 
nounced *' unquestionably  Marlowe's  greatest  play;  one  of  the  greatest  plays 
that  the  world  possesses."  Others  have  given  the  palm  to  Edward  //.  They 
are  both  great  in  difl'erent  ways,  but  I  must  concur  with  the  estimate  of  the 
final  soliloquy  of  Faustus,  '*  which,  as  a  representation  of  mental  agony  and 
despair,  is  only  equalled,  in  the  whole  range  of  the  world's  poetry,  by  the 
speech  of  Satan  to  the  Sun  in  Paradise  Lost.^*  The  other  plays  are  briefly 
noticed,  and  some  excellent  criticism  follows.  Marlowe's  violence  and  exagge- 
ration are  recognized  ;  also,  the  ill-construction  of  his  plots.  His  **  theory  of 
dramatic  action  is  contrary  to  the  constitution  of  human  nature,"  for  **  it  elimi- 
nates the  factor  of  Conscience " ;  and  "  the  narrowness  of  his  conception  of 
man  and  nature  is  seen  in  his  representations  of  female  character."  As  has 
been  recognized  by  all  critics,  Marlowe  could  not  paint  a  female  character. 

The  chapter  closes  with  a  notice  of  Kyd,  but  he  was  only  a  disciple  of  Mar- 
lowe who  exaggerated  Marlowe's  faults  ;  rant  and  bloodshed  are  his  predomi- 
nant  characteristics.  Shakspere  rescued  the  drama  "  by  restoring  to  tragedy 
the  elementsof  conscience,  religion,  and  chivalry,  which  Marlowe  had  expelled 
from  it." 

This  volume  carries  forward  the  history  of  English  poetry  to  the  time  of 
Shakspere,  and  we  await  with  interest  its  successors.  Prof.  Courthope's  work 
traces  with  philosophic  judgment,  critical  taste,  and  literary  skill,  the  course 
of  English  poetry,  and  is  a  useful  addition  to  its  history,  notwithstanding 
some  defects.  To  each  volume  a  very  full  analytical  table  of  contents  is 
prefixed,  but  indexes  are  wanting. 

James  M.  Garnett. 


Digitized  by  VjOOQ IC 


REVIEiVS  AND  BOOK  NOTICES,  48 1 

T.  Lucretius  Carus  de  Rerum  Natura.     Buch  III  erklaert  von  Richard 
Heinze.    Teubner,  Leipzig,  1897.     Pp.  vi  +  206. 

The  edition  of  the  third  book  of  Lucretius  which  we  have  before  us  is 
the  second  volume  in  the  new  series  of  Wissenschaftliche  Commentare  pub- 
lished under  the  editorship  of  Professor  Kaibel  of  Strassburg.  It  is  a 
worthy  successor  to  the  inaugural  volume  of  the  series,  the  Electra  of 
Sophocles,  edited  by  Professor  Kaibel  himself.  Like  that  work,  it  not  only 
undertakes  to  present  a  new  and  more  thoroughgoing  interpretation  of  the 
text  chosen,  but  it  also  stands  for  a  theory  of  interpretation  as  yet  but 
scantily  represented.  If  I  understand  the  purpose  of  the  commentary 
aright,  it  assumes  that  there  is  much  more  in  an  ideal  interpretation  than 
an  explanation  of  the  difficulties  of  thought,  language  or  text.  It  would 
seem  to  aim  at  something  further — at  illustration  of  the  background  of 
thought  and  the  habit  of  language  out  of  which  the  poet's  work  has  pro- 
ceeded. Its  effort  is  not  only  to  explain  difficulties,  but  in  a  manner  to 
reproduce  the  creative  atmosphere  in  which  the  poet  wrought.  It  would 
substitute  for  mere  explanation  a  background  of  consciousness.  It  may  be 
that  this  is  to  put  more  into  the  editor's  work  than  he  himself  felt,  but  if  so, 
it  is  under  the  influence  of  the  agreeable  feeling  that  in  this  work  we  have 
a  real  approximation  to  a  true  interpretation,  infinitely  removed  from  the 
vast  bulk  of  editions  **with  notes." 

In  the  brief  but  instructive  preface  the  editor  calls  attention  to  the 
main  directions  of  his  effort.  The  task  of  the  editor,  he  points  out,  consists 
in  an  explanation  of  the  relation  of  the  poet  to  his  material,  since  the  poet 
is  only  the  interpreter  of  the  teachings  of  another,  and  of  the  transforma- 
tion of  this  material  from  scientific  prose  to  verse.  The  content  of  Lucre- 
tian  verse  therefore  requires  attention  first  of  all.  In  this  consists,  I 
believe,  the  most  original  and  positive  contribution  of  the  editor  to  the 
interpretation  of  Lucretius.  The  vast  mass  of  scattered  and  fragmentary 
material  relating  to  the  philosophy  of  Epicurus  has  been  brought  to  bear 
upon  this  portion  of  the  de  rerum  natura  with  an  insight  before  which  many 
an  obscurity  has  disappeared,  and  with  a  sureness  of  touch  that  reveals 
the  master  in  the  field  of  Greek  philosophy,  to  one  phase  of  which  an 
earlier  work  of  the  editor  was  devoted  (Xenocrates,  Lpz.  1891).  This 
material  is  presented  not  only  current  with  the  text,  but  in  the  introduc- 
tion to  the  commentary  the  psychology  of  Epicurus  is  presented  briefly, 
but  with  great  clearness,  and  in  such  a  manner  as  to  cast  much  light  on  a 
field  in  which  the  obscurity  is  not  wholly  the  fault  of  a  scanty  tradition. 
In  the  whole  matter  of  the  relation  of  Lucretius  to  his  sources,  Heinze 
seems  to  have  penetrated  much  further  than  his  predecessors  (notably,  for 
example,  in  the  arguments  for  the  mortality  of  the  soul),  as  was  no  more,  to 
be  sure,  than  could  fairly  be  demanded  with  the  results  of  much  recent 
investigation  in  this  field  at  hand  (e.  g.  Usener's  Epicurea),  but  still  with 
an  originality  and  breadth  of  grasp  that  deserve  the  fullest  recognition. 

Closely  connected  with  this  subject  is  the  question  of  terminology,  to 
which  the  editor  also  calls  attention  in  his  preface.  Here  it  has  been  his 
effort  to  ascertain  in  every  case  the  equivalence  of  the  terms  chosen,  and 


Digitized  by  VjOOQ IC 


482  AMERICAN  JOURNAL    OF  PHILOLOGY. 

when  the  exactness  of  the  Latin  word  employed  might  be  qnestioned,  the 
possibilities  of  expression  at  the  disposal  of  the  poet  have  been  weighed. 
In  this  connection  Munro's  defence  of  the  poet  against  his  own  complaint 
of  the  poverty  of  the  Latin  tongue  and  of  the  difficulty  of  giving  expression 
to  obscure  ideas  in  a  field  of  thought  never  before  trodden  by  Roman  bard, 
will  be  remembered  (cf.  Munro,  Int.  p.  11).  Certainly  in  this  well  known 
passage  Munro  has  given  expression  to  the  feelings  of  Lucretian  scholars 
since  Lachmann  restored  a  legible  text.  For  there  is  a  confidence  and 
sureness  of  movement  in  the  language  of  the  poet  that  does  not  leave 
room  for  much  consciousness  of  the  inadequacy  of  the  language  to  the 
theme.  Bat  Heinze  in  his  note  on  vss.  258-261,  on  the  nature  of  the 
admixture  of  the  elements  of  the  soul,  observes  that  no  portion  of  the 
poem  is  more  obscure,  and  furthermore  that  it  is  the  only  portion  of  the 
Epicurean  doctrine  which  Lucretius  greatly  abridged.  The  causes  of  this 
he  holds  are  therefore  not  only  the  obscurity  of  the  subject-matter,  but  he 
believes  that  we  must  also  give  credence  to  the  poet,  whose  complaint 
here  is  reiterated,  that  the  language  did  not  permit  him  to  say  what  he 
gladly  would. 

In  the  matter  of  the  text  the  editor  has  been  quite  conservative.  His 
own  conjectures  are  not  numerous,  nor  do  they  extend  to  changes  that  have 
a  radical  efiEect  upon  the  thought.  In  vs.  58  (eripitur  persona  manare)^ 
where  the  Itali  read  manet  res^  he  suggests  mala  re^  but  does  not  introduce 
the  reading  into  his  text,  and  wisely,  since  from  no  point  of  view  does  it 
seem  so  satisfactory  as  the  correction  of  the  It.,  which  in  turn  we  may  still 
grant  is  not  convincing.  In  vs.  194  constat  for  extat^  a  rather  doubtful 
change  in  the  interest  of  conventional  phraseology.  In  vs.  337  praeterea 
is  changed  to  propterea,  as  it  seems  to  me  correctly  in  view  of  the  argu- 
ment. In  vs.  394  quam  sis  =  suis^  attractive.  In  vs.  433  feruntur  seems 
correctly  restored  iot  geruntur.  That  vs.  493  is  hopelessly  corrupt  is  not 
made  convincing  to  me.  In  the  matter  of  transposition  and  rearrange- 
ment (apart  from  single  verses),  which  has  been  a  favorite  field  for  the  dis- 
play of  editorial  ingenuity,  the  editor  is  very  conservative,  and  has  shown 
very  clearly  that  most  difficulties  of  this  kind  are  to  be  removed  by  inter- 
pretation. Thus  at  vss.  417  and  526,  the  apprehension  of  the  true  dispo- 
sitio  makes  transposition  quite  superfluous.  Indeed  the  editor's  grasp  of 
thought  and  arrangement  reminds  me  not  infrequently  of  the  keen  sense 
for  psychological  suggestion  in  explanation  of  transitions  which  Kiessling 
displays  and  was  the  first  to  apply  with  discernment  to  the  interpretation 
of  Horace.  When  it  is  remembered  that  Heinze  has  assumed  the  respon- 
sibility of  revision  for  Kiessling's  Horace,  it  is  not  unlikely  that  we  have  a 
clue  to  the  source  of  the  training  which  marks  much  of  the  characteristic 
quality  of  this  work. 

It  is  a  satisfaction  to  find  so  pervading  a  sympathy  with  all  the  moods 
and  themes  of  the  poet  as  the  editor  reveals,  and  one  is  pleased  with  his 
expression  of  the  feeling  that  Lucretius  has  in  most  cases  elevated  the 
prosaic  parts  of  his  theme  to  the  rank  of  true  poetry.  It  used  to  be,  and  I 
think  still  is,  commonly  said  (e.  g.  by  Teuflfel)  that  Lucretius  was  a  great  poet 


Digitized  by  VjOOQ IC 


REVIEWS  AND  BOOK  NOTICES.  483 

sadly  astray  in  the  choice  of  a  subject  I  am  sure,  however,  that  many  a 
devotee  of  Lucretius  will  join  with  me  in  protest  against  this  utterance. 
For  when  we  consider  the  sort  of  work  that  was  possible  or  that  was  likely 
to  have  challenged  the  attention  of  a  Roman  poet  in  the  first  century  B.  C, 
we  cannot,  I  believe,  conceive  of  any  theme  that  we  should  willingly 
exchange  for  the  de  rerum  natura.  What  were  an  epic  of  any  theme, 
mythological  or  national,  or  the  Alexandrine  sources  of  inspiration  of  his 
contemporaries,  in  comparison  with  a  subject-matter  which  called  forth  a 
passionate  intensity  of  feeling  and  devotion  that  we  miss  in  all  other 
Roman  poetry?  Of  refined  workmanship  and  rhetorical  vigor  there  is  no 
lack  elsewhere  in  the  higher  poetry  of  Rome,  but  of  feeling,  verging  at 
times  to  an  almost  unhealthy  fervor,  there  is  no  other  grandly  sustained 
example,  and  let  us  not  therefore  complain  of  a  subject-matter  which  was 
its  inspiration.  I  know  not  if  there  is  a  statelier  or  simpler  example  of 
intense  dramatic  conception  than  the  long  cumulative  enumeration  of  the 
considerations  which  show  the  mortality  of  the  soul,  summed  up  in  those 
triumphant  verses  beginning  Nil  igitur  ad  nos  mors  est  neque  pertinet 
hilum.  To  the  sagacious  and  sympathetic  interpreter  of  this  culminating 
book  of  Lucretius,  scholars  who  have  leisure  to  peruse  his  work  will  feel  a 

sense  of  personal  obligation. 

G.  L.  Hbndrickson. 


Digitized  by  VjOOQ IC 


REPORTS. 

Romania,  Vol.  XXIV  (1895).    Second  half. 

Juillet. 

Ferdinand  Lot.  Celtica.  17  pages.  A  series  of  interesting  observations 
throwing  light  upon  various  obscure  points  in  the  relations  of  the  romances  of 
the  Round  Table  to  their  Breton  prototypes.  Thus,  Mabonagrain  in  the  tale 
of  Erec  et  Enide  (the  Geraint  et  Enid  of  the  Mabinogi)  is  a  fusion  of  the  juxta- 
posed names  of  two  magicians,  Mubon  and  Evraifiy  the  latter  name  being  in 
turn  a  corruption  of  Euuain  (Owen),  due  not  to  phonetic  change,  but  to  an 
error  of  reading  in  written  transmission.  Again,  the  "chateau  de  Lis"  is  a 
mere  tautology,  llys  signifying  *  castle'  in  Breton. 

A.  Thomas.  Les  noms  composes  et  la  derivation  en  frangais  et  en  proven^al. 
18  pages.  In  his  treatise  on  the  FormaHon  des  mots  composis^  Arsine  Darme- 
steter  was  able  to  draw  up  a  list  of  only  66  derivatives  from  French  compound 
words,  a  fact  which  he  offered  in  support  of  his  statement  that  **  la  derivation, 
richement  d^veloppee  chez  nous,  s'exerce  cependant  avec  difficulte  sur  les 
composes.*'  M.  Thomas  begins  by  eliminating  from  this  list  a  certain  number 
of  derivatives  from  compound  proper  names  (such  as  saint-cyrun^  ttrreneuvier), 
"car  Ton  pent  dire  que  tous  les  noms  propres,  composes  ou  non,  sont  suscep- 
tibles  de  foumir  des  derives"  (Le  Chat-Noir^  chatnoiresque) ;  and  such  words  as 
vaurienne,  while  citing  in  addition,  in  this  category,  fainAmte^  "  qui  est  re^u 
partout,  et  proprarienne^  qui  est  encore  confine  dans  les  bas-fonds  parisiens 
(c'est  I&  de  la  flexion  et  non  de  la  derivation).'*  The  author  then  proceeds  to 
supplement  very  considerably  the  list  given  by  Darmesteter  and  to  furnish 
one  of  his  own  for  the  Provencal.  In  conclusion  he  points  out  a  species  of 
parasynthetic  not  before  signalized.  "  La  pomme  dite  blandureau  tire  mani- 
festement  son  nom  de  ce  qji*elle  est  blanche  et  dure :  on  ne  voit  pas  que  Tadj. 
dur  ait  produit  de  deriv^  dureau  en  dehors  de  ce  mot  compose."  The  words  of 
this  class  are  comparatively  numerous. 

P.  Meyer.  La  descente  de  Saint  Paul  en  enfer:  po^me  frangais  compose  en 
Anglcterre.  Beginning  at  least  as  early  as  the  fourth  century,  there  have 
been  numerous  versions  of  this  legend  in  various  languages.  M.  Meyer  here 
publishes,  face  to  face,  a  Latin  prose  abridgment  of  the  legend  and  an  Old 
French  rhymed  version  of  282  verses,  founded  upon  it  and  contained  in  a 
manuscript  of  Toulouse. 

Paget  Toynbee.  Dante's  references  to  Pythagoras.— Dante's  obligations  to 
Orosius.— Some  unacknowledged  obligations  of  Dante  to  Albcrtus  Magnus. — 
Dante's  obligations  to  Alfraganus  in  the  Vita  Nuava  and  Omvivio,  55  pages. 
This  series  of  articles  constitutes  an  original  and  valuable  contribution  to  our 
knowledge  of  the  sources  and  references  of  Dante's  writings.    Speaking  of 


Digitized  by  VjOOQ IC 


REPORTS,  485 

Alberlus  Magnus,  the  author  says :  **  It  is  singular  that,  though  he  has  made 
such  liberal  use  of  the  works  of  Albertus  Magnus,  Dante  does  not  mention 
him  by  name  more  than  four  times,  viz.  Convito,  III  5  (where  he  is  called 
•Alberto  della  Magna'),  Conviio,  III  7  and  IV  23  (in  both  of  which  he  is 
called  simply  'Alberto*),  and  Paradiso,  X  98  (where  he  is  called  'Alberto  di 
Cologna').**  Apropos  of  this  paragraph  the  present  writer  pointed  out,  in  a 
paper  read  before  the  philological  section  of  the  N.  Y.  Acad,  of  Sciences,  that 
the  modern  appellation  of  *the  Great,*  as  applied  to  Albert,  appears  to  have 
been  the  outgrowth  of  a  misapprehension.  The  epithet  Magnus  was  pretty 
•certainly  given  originally  to  Albertus  simply  to  designate  him  as* the  Tall,* 
but  may  also  have  been  used  as  a  mistaken  latinization  of  the  *  della  Magna' 
{'of  Germany')  which  we  find  used  by  Dante.  It  is  noteworthy  that,  so  far  as 
appears,  no  other  scholar  or  writer  in  the  world's  history  ever  acquired  the 
appellation  of  *  the  Great.* 

Melanges.  A.  MussaBa.  Francese  vals,  vali^  imlent;  salt,  salt;  chult^  chali, 
4  pages.  Latin  va/rf,  valet^  valeni  should  have  given  regularly  in  O.Fr.  vels, 
^elt^  velent^  but  the  latter  forms  are  nowhere  found.  Most  scholars  have 
regarded  the  universally  occurring  forms  with  a  {vais,  valt^  valerU)  as  due  to 
analogy  with  vahns^  valez^  where  the  a  is  regular,  being  pretonic.  Mussafia 
here  supports  an  explanation  long  since  advanced  by  Comu,  that  in  this  verb 
the  a  of  the  forms  in  question  was  preserved  chiefly  by  the  influence  of  the  a 
regularly  occurring  in  a  majority  of  the  stem-tonic  forms,  "forme  rizotoniche ** 
ivailU^  etc.). — E.  Langlois.  Interpolations  du  jeu  de  Robin  et  Marion.  In- 
geniously surmises  and  unmasks  two  important  interpolations. — G.  Raynaud. 
Le  dit  du  Cheval  4  vendre,  public  d'apris  un  manuscrit  du  chateau  de 
-Chantilly.     A  fragment  of  only  51  verses,  carefully  edited. 

Comptes  rendus.  Abhandlungen  Herrn  Prof.  Dr.  Adolf  Tobler,  zur  Feier 
seiner  fUnfundzwanzigjfihrigen  Th&tigkeit  als  ordentlicher  Professor  an  der 
Universitat  Berlin  von  dankbaren  SchQlem  in  Ehrerbietung  dargebracht  (G. 
Paris).  10  pages.  The  work  reviewed  consists  of  22  articles,  "  tons  interes- 
sants,  quelques-uns  de  tris  grande  valeur,**  covering  over  500  pages. — W. 
Meyer-LUbke.  Zur  Geschichte  des  Infinitivs  im  Rum&nischen.  "Dans  cette 
savante  et  pen^trante  ^tude,  M.  Meyer-Ltlbke  nous  donne  un  avant-goClt  de 
ce  que  sera  le  troisidme  volume  de  sa  Grammaire^  consacre  k  la  syntaxe.*' — J. 
Vising.  Qtumtodo  in  den  romanischen  Sprachen.  "  Montre  que,  dans  diverses 
Ungues  romanes,  a  cdte  de  com  ou  como<\9X,  wxlg.  quomo<quomodo  il  existe 
ou  a  exists  une  forme  coma^  qui  s'explique  par  quorno  ad^  et  une  forme  come^  qui 
s*explique  par  quomo  et , , ,  Presque  partout  les  formes  se  sont  confondues. 
En  fran^ais,  c'est  come  (comme)  qui  a  absorbe  comj'* — A.  WallenskSld.  Zur 
Ldsung  der  Lautgesetzfrage  (G.  P.).  "  Je  suis  depuis  longtemps  4  peu  pris  de 
I'avis  de  mon  savant  ami  sur  le  traitement  exceptionnel  aUquel  sont  sujets  cer- 
tains mots  tr^s  usites  .  .  .  Cette  negligence  tient  k  ce  qu*on  sait  qu*il  n*est  pas 
neccssairc  de  les  prononcer  pleinement  pour  que  I'auditeur  les  comprenne, 
quelques-uns  d'entre  eux,  surtout  parmi  les  formules  d'allocution, — comms 
prov.  crtf  na,  esp.  (/stcd, — arrivent  4  n*6tre  en  reality  que  de  simples  allusions 
vocales."— K.  Breul.  Le  Dit  de  Robert  le  Diable.  "  L*edilion  du  Dit  de 
Robert  le  Diable^  faite  avec  autant  de  soin  que  d'intelligence,  est  assurement 


Digitized  by  VjOOQ IC 


486  AMERICAN  JOURNAL    OF  PHILOLOGY. 

une  des  contributions  qui  ajoutent  le  plus  de  prix  4  ce  beau  recueil." — £. 
Gorra.  Delle  origini  della  poesia  lirica  del  medio  evo  (A.  Jeanroy).  **La 
proluHone  de  M.  Gorra  est  aussi  habilement  composee  qu'elegamment  ccrite." 
— H.  Springer.  Das  altprovenzalische  Klagelied  (A.  Jeanroy).  "Conscien- 
cieuse  dtude." — G.  A.  Cesareo.  La  poesia  siciliana  sotto  gli  Svevi  (A.  Jean- 
roy). 8  pages.  "  II  y  a  dans  ce  livre  des  pages  excellentes  et  qui  resteront ; 
mais  I'ensemble  est  gUte  par  des  erreurs,  des  &  peu  pr^s,  une  tendance  au 
paradoxe." — E.  Wechssler.  Ueber  die  verschiedenen  Redaktionen  des  Robert 
von  Borron  zugeschriebenen  Graal-Lancelot-Cyclus  (G.  Paris).  "L'ordre  et 
la  lumi^re  pendtrent  peu  &  peu  dans  ce  chaos  indigeste,  dans  cette  selva  oscura 
des  romans  en  prose  de  la  Table  Ronde  ...  Le  memoire  de  M.  Wechssler 
est  une  des  plus  importantes  contributions  qui  aient  ete  apport^es  &  cette 
oeuvre  de  reconnaissance  et  de  dechiffrement  qui  ne  sera  pas  de  longtemps 
achevee." 

Chronique. 

P^riodiques. 

Livres  annonc^s  sommairement.  40  titles.  W.  H.  Schofield.  The  Source 
and  History  of  the  Seventh  Novel  of  the  Seventh  Day  of  the  Decameron. 
"  Excellente  petite  etude."— James  D.  Bruner.  The  Phonology  of  the  Pisto- 
jese  Dialect.  "Ce  travail  porte  sur  l*etat  ancien  aussi  bien  que  snr  T^tat 
moderne  du  parler  de  Pistoja." — Rene  de  Poyen-Bellisle.  Les  sens  et  les 
formes  du  cr^ole  dans  les  Antilles.  "  L*auteur  de  ce  petit  livre  est  intelligent 
et  sufHsamment  au  courant  de  la  philologie  romane." — G.  Lanson.  Histoire 
de  la  litt^rature  frangaise.  Paris,  Hachette,  pp.  xvi,  T158.  5  francs.  *'Dans 
ce  trds  remarquable  ouvrage,  qui  conduit  I'histoire  de  notre  litterature  de  ses 
premieres  origines  jusqu'aux  oeuvres  les  plus  r^centes,  le  moyen  age  occupe 
une  place  justement  proportionnee  (216  pages).  Cette  place  est  extr6mement 
bien  remplie." 

Octobre. 

F.  Lot.  Etudes  sur  la  provenance  du  cycle  arthurien.  32  pages.  I.  Le 
sens  du  mot  breion  au  XI I^  si^cle.  In  the  nth  and  12th  centuries,  for  the 
French,  the  Normans  and  the  English,  the  Bretons  were  the  inhabitants  of 
ancient  Armorica,  "la  Bretagne."  The  insular  descendants  of  the  ancient 
Bretons  were  called  Gualeis,  Gallois  (Welsh),  and  their  country,  to  the  west  of 
the  Severn,  is  GuaUs  (Wales).  They  called  themselves  Cymri  (Latin  Cambri). 
— II.  De  la  provenance  des  lais  dits  bretons.  The  author  analyzes  critically 
all  the  evidence,  and  concludes  that  at  least  half  of  the  iais  that  have  come 
down  to  us  derive,  not  from  Armorica,  but  from  southern  Great  Britain. 

If.  Meyer.  Cet^  suivis  d*a  en  provengal:  ^tude  de  geographie  linguistique 
(avec  carte).  47  pages.  It  has  been  well  known  that  in  the  southern  portion 
of  the  regions  speaking  the  langue  d^oc^  Latin  c  (and  similarly  ^).  initial,  or 
second  consonant  of  a  group,  preserved  its  Latin  sound  virtually  intact  before 
a,  9,  u  (camp,  castel^  galina)^  while  in  the  more  northern  parts  c  and  ^,  in  like 
case,  take  on  a  complex  sound  commonly  represented  by  ch  and  /  (champ, 
chastely  jalina).    The  present  elaborate  memoir  undertakes  to  establish  the 


Digitized  by  VjOOQ IC 


REPORTS,  487 

geographical  line  of  demarcation  between  the  sounds  ca^ga  and  the  sounds 
cha^ja,  "  Si  on  instituait  une  recherche  semblable  pour  la  zone  od,  dans  le 
nord  de  France,  cha  passe  4  CA,  les  resultats,  autant  que  j'en  puis  juger  par 
quelques  etudes  qui  je  continuerai  peut-6tre  un  jour,  seraient  moins  satisfai- 
sants  h.  tous  egards.  Pourquoi?  Pour  diverses  causes  qui  se  reduisent  en 
somme  4  une  seule :  parce  que  la  limite  de  ch  et  de  c^  de/  et  de  gu  (Chastel  et 
Castely  Jouy  et  Gouy)  passe  trop  prds  de  Paris  . . .  Les  parlers  locaux  sont 
actuellement  ou  eteints  ou  trop  impregnes  de  frangais  pour  donner  des 
indications  nettes  et  precises." 

Melanges.  F.  Bonnardot.  A  qui  Jacques  de  Longuyon  a-t-il  d^die  le 
po^me  des  **  Voeux  du  Paon  "  ?  The  great  vogue  of  this  poem  during  the  latter 
part  of  the  Middle  Ages  lends  some  interest  to  the  discovery  of  the  identity 
of  the  personage  to  whom  it  was  dedicated,  viz.  Thiebaut,  bishop  of  Lidge 
from  1303  to  1 312,  son  of  Thiebaut  II,  comte  de  Bar.— A.  Thomas.  Etymo- 
logies fran^aises.  Chevhte  [sort  of  fish  resembling  the  whitebait].  From 
*capit{nem,  doublet  of  capitonem  (cf.  archaic  turbonem  for  turblnem). — Hanse 
[shaft  of  a  pin].  Probably  an  alteration  of  hante  (O.Fr.  hanste\  due  to  influ- 
ence of  anse. — Haque,  Harengs  h  la  haque  are  herrings  prepared  and  salted  to 
be  used  as  bait.  The  phrase  h  la  haque  is  doubtless  for  h  Vactque^  from  the 
verb  aeschier  *  to  bait,'  from  Latin  esca. — OrpailUur  [gatherer  of  gold-dust].  A 
folk-etymological  modification  of  harpailleur  (influenced  by  or  *  gold '),  through 
harpaiUer^  pejorative  of  harper  *to  seize.' — Routs  [Mod.  Prov.  for  •  bush,  brier']. 
From  *rusteum  for  rustum, — O.  Densusianu.  O.Fr,  baufan  [a  dappled  horse]. 
For  baucenc,  from  Lat.  balteum,  with  substitution  of  Germanic  suffix  -ing. — G. 
A.  Nauta.  La  Danse  Macabre.  '*  La  danse  de  la  Mort  ^tait  nomme  au  XV (^ 
si^cle  dans  les  Pays-Bas  Makkabetis  dansP—V,  Meyer.  La  Descente  de  Saint 
Paul  en  Enfer :  po^me  fran^ais  compose  en  Angleterre  (note  complementaire). 
— A.  Morel-Fatio.  Espagnol  yogar.  Among  the  orders  pronounced  by  the 
immortal  Sancho  Panza  in  the  government  of  his  Island  appears  the  following: 
*'  Procurad  que  no  os  venga  en  voluntad  de  yogar  con  nadie."  The  etymolo- 
gists have  hitherto  failed  to  distinguish  between  the  familiar  verb  yogar  '  to 
sport,'  homjoeare^  and  the  archaic  word  yogar,  meaning  *to  lie  (with).'  Latin 
jtuuit  gave  regularly  the  strong  preterit  yogo  (pronounced  ydgd).  This  form 
came  to  be  confused  with  the  weak  preterit  yogd  z=.jocavit^  and  the  confusion 
extended  to  other  forms  of  the  two  '^^xh^  yacer  ?iXi^  yogar, 

Comptes  rendus.  P.  Marchot.  Les  gloses  de  Cassel.  Les  gloses  de  Vienne 
(G.  Paris).  M.  Marchot  considers  these  two  glossaries  as  both  belonging  to 
one  of  the  Raeto- Romance  dialects.  Holtzmann  long  ago  assigned  the  Cassel 
glossary  to  a  Romance  idiom  spoken  in  Bavaria  in  the  8th  century,  and  Gaston 
Paris  formerly  adhered  to  this  opinion,  but  now  admits  that  Marchot  has 
rendered  his  view  very  probable.  The  Vienna  glossary  seems  to  be  correctly 
assigned  to  the  dialect  of  Friuli. — Kritischer  Jahresbericht  liber  die  Fort- 
schritte  der  romanischen  Philologie.  Unter  Mitwirkung  von  115  Fachge- 
nossen  herausgegeben  von  Karl  Vollm5lIer  und  Richard  Otto.  I.  Jahrgang 
(G.  Paris).  *'  C'est  en  fait  le  GrundrUs  de  GrOber  indefiniment  continue  et 
mis  au  courant:  je  ne  saurais  mieux  en  faire  comprendre  et  le  m^rite  et 
I'utilit^."— L.  Willems.    Etude  sur  I'Ysengrinus  (L.  Sudre).     Even  after  the 


Digitized  by  VjOOQ IC 


488  AMERICAN  JOURNAL    OF  PHILOLOGY, 

masterly  introduction  prefixed  by  Voigt  to  his  edition  of  the  Ysengrimus  in 
1884.  the  present  work  is  valuable  as  throwing  light  on  a  certain  number  of 
obscure  points.  The  orthography  of  the  title,  Ysengrinus  (with  an  »),  is 
defended  by  serious  arguments. 

Periodiques.  A  report  is  given  of  vol.  I  (1894)  of  the  Revue  Hispanique: 
recueil  consacre  k  I'etude  des  langues,  des  litteratures  et  de  Thisioire  des  pays 
castillans,  Catalans  et  portugais,  public  par  R.  Foulche-Delbosc. 

Chronique.  M.  Anatole  de  Montaiglon,  professor  at  the  Ecole  des  Chartes 
in  Paris,  died  at  Tours  on  September  i,  1895,  in  his  72d  year.  He  was  a 
scholar  of  varied  attainments,  a  prolific  writer,  and  the  editor  of  numerous 
works  in  the  domain  of  Romance  philology. — There  has  been  founded  at 
Madrid  a  Revista  critica  de  historia  y  literatura  espaflolas,  which  is  published 
monthly. 

Livres  annonces  sommairement.  27  titles.  Guernsey:  Its  People  and 
Dialect,  by  Edwin  Seelye  Lewis.  Johns  Hopkins  University  dissertation. 
*•  II  est  k  souhaiter  que  M.  Lewis  reprenne  et  complete  son  etude,  dont  Tobjet 
est  des  plus  int^ressants,  non  seulement  pour  la  dialectologie  moderne,  mais 
pour  I'histoire  de  revolution  du  fran^ais." 

H.  A.  Todd. 


Rheinisches  Museum  fOr  Philologie,  Vol.  LI  I,  parts  3,  4. 

Pp.  305-32.  Lateinische  Uebersetzungen  aus  der  Aratusliteratur.  M.  Man- 
itius.  A  critical  edition  of  the  "  Arati  ea  quae  uidentur"  (=  *Ap&Tov  <^v6fieva) 
of  the  Dresden  MS  Dc  183.  This  is  a  translation  from  the  Greek  into  an 
extremely  barbarous  Merovingian  Latin.  The  MS  belongs  to  the  9th  century, 
the  translation  may  have  been  made  about  the  beginning  of  the  6th. 

Pp.  333-7.  Die  Exostra  des  griechischen  Theaters.  A.  KOrte.  In  the 
fifth  century  B.  C.  k^doTpa  and  tmcvKXriua  were  different  names  for  the  same 
thing.  At  a  later  period  k^uarpa  meant  a  balcony.  From  the  accounts  of  the 
temple  of  Delos  for  the  year  B.  C.  274  (B.  C.  H.  XVIII  163)  it  is  clear  that 
there  were  several  exostrae.  This  inscription  confirms  the  statement  of 
Pollux,  IV  127  Kcu  xp^  TOVTO  voelodoi  Koff  kudorrfv  dhpaVf  oiovel  Koff  kK&aTrjv  oiidav, 
Reisch  has  exaggerated  the  difficulty  of  working  such  a  machine  (Das  grie- 
chische  Theater,  pp.  244,  246). 

Pp.  338-47.  Antiker  Volksglaube.  W.  Kroll.  I.  The  ancient  popular 
belief  that  the  soul  of  man  dwells  in  the  air,  whence  it  is  breathed  into  the 
body,  is  found  in  some  Orphic  verses  quoted  by  Vettius  Valens  (cod.  Oxon. 
Selden.  22,  saec.  XVII):  Kadbg  koL  6  detdrarog  *Opi^evc  Xeyei'  rjwx^  <J'  avOpCmoujtv 
OTT*  aWipog  tppi^uTai'  koI  AAA6)f  dipa  (T  ehcovreg  iffvxv^  deiav  dpeizdfuada.  II. 
The  KwdvdpLmoc  or  hjKdvdpuiroc  v6aoc  described  by  Marcellus  of  Sida  was  not 
a  religious  hallucination,  as  W.  H.  Roscher  has  recently  maintained  (Abh.  d. 
s&chs.  Ges.  XVII  3).  Kroll  himself  assumes,  without  any  obvious  warrant, 
that  the  disease  of  lycanthropy  presupposes  a  belief  in  werwolves.  III.  The 
command  of  the  anonymous  dialogue  Hermippus  to  change  the  names  of  the 


Digitized  by  VjOOQ IC 


REPORTS,  489 

dead  in  order  that  they  may  escape  from  evil  spirits  is  to  be  compared  with 
the  belief  that  the  dead  may  be  recalled  to  life  by  thrice  pronouncing  their 
names. 

Pp.  348-76.  Lucubrationum  Posidonianarum  Spec.  II  (cf.  Comm.  in  hon. 
Wachsmuthii  scr.,  1897,  pp.  13  sqq.).  £.  Martini.  A  study  of  Cleomedes, 
Cycl.  Theor.,  lib.  I.  cc.  2-7. 

Pp.  377-90.  Lebte  Erasistratos  in  Alexandreia  ?  R.  Fuchs.  It  is  probable 
that  Erasistratus  was  settled  in  Alexandria  towards  the  close  of  his  life,  and 
was  closely  connected  with  the  royal  court. 

Pp.  391-8.  Altes  Latein  (Fortsetzung  von  Band  LI,  S.  471 ;  see  A.  J.  P. 
XVIII  114).  F.  BUcheler.  XXI.  Favere  is  a  later  form  of  fovtre.  An 
inscription  which  apparently  belongs  to  the  latter  half  of  the  3d  century  B.  C. 
runs:  FOVE  L.  CORNELIAI  L.  F.  The  gloss  fovet:  nuirit  studet  (IV,  p. 
239,  21  G)  may  be  explained  without  assuming  a  confusion.  XXII.  Lexical 
notes  on  the  tesserae  described  by  HUlsen,  Mitth.  des  r5m.  arch&ol.  Instituts, 
1896,  pp.  228-37.  XXIII.  Aplopodite,  C.  I.  L.  XII  6025,  is  for  (h)aplopotide{m). 
For  the  word  anTuoTZfrriQ  cf.  Goetz,  Corp.  Gloss.  Ill,  p.  219,  23  dos  aplopotin: 
da  filiolam  (i.  e.  fiolatn  •=.phialatn),  XXIV.  The  legal  praestat^  eyyvdrcu,  was 
formed  by  combining  prms  and  stat,  about  the  time  of  Sulla.  Cf.  Mon.  ant. 
dei  Lincei,  VI  (1895),  p.  411,  7  ff.  quel  pro  se  praes  stat. 

Pp.  399-411.  Buphonien.  P.  Stengel  rejects  the  view  of  H.  von  Prott,  pp. 
187  ff.,  that  this  represented  an  earlier  human  sacrifice. 

Pp.  412-24.  Studien  zur  Geschichte  der  griechischen  Rhetorik.  L.  Rader- 
macher.  I.  TimSlus  und  die  Ueberlieferung  Qber  den  Ursprung  der  Rhetorik. 
II.  Plutarchs  Schrift  de  se  ipso  citra  invidiam  laudando. 

Pp.  425-34.  Zur  lateinischen  Wortbildungslehre.  M.  Pokrowskij.  I.  S^e- 
nus  originally  meant  '  dry,*  rather  than  *  bright.'  It  is  perhaps  formed  from 
*sirhre^  inchoat.  serescere.  Crudus  :  ^crude-o  :  cradi-li-s  ::  /<</<oc  :  fUfxe-oficu  : 
fUfirf-X6-<:,  11.  Difraudit^  Petron.  69  \  fraudire  : :  olunt^  Petron.  50  :  olere.  III. 
Die  mit  x»-  negativum  zusammengesetzten  Verba.  Fateor  and  infitear  are 
derived  from  ^fa-to-s  (=  0tf-r<5-f),  *infl-to-s.  With  *infiios^  infiUor^  infUiae  cf. 
hnioToq^  airiOTiiJ,  aTrurria. 

Pp.  435-45.    Zu  Pseudokallisthenes  und  Julius  Valerius.    I.    Ad.  Ausfeld. 

Miscellen. — Pp.  446-7.  W.  Schmid.  Zwei  Vermuthungen  zu  der  Schrift 
irepl  infwvi  (fldpov^  for  pdOov^,  II  i ;  owapoMi  for  owdpoi,  XLIV  5). — Pp.  447-9. 
R.  Schneider.  Zu  dem  Lexicon  Messanense  de  iota  ascripto.  Textual  notes. 
— P.  449.  J.  Ziehen.  Zu  Cicero  ad  fam.  VIII  17,  2.  Read  vos  inviios  inncere 
coegero  astuHa!  num  me  Catonemf^Y^,  449-50.  A.  Frederking.  Zu  Horat. 
carm.  II  6.  Lasso^  v.  7,  may  refer  to  Septimius,  not  to  Horace. — Pp.  450-54« 
J.  Ziehen.  Eine  Zeitbeziehung  in  der  ersten  Maoenaselegie. — Pp.  454-7> 
M.  Ihm.  Nemesians  Ixeutica. — Pp.  458-9.  A.  Zimmermann.  Ueber  Ent- 
stehung  von  neuen  Verwandtschaftsnamen  aus  alten  im  Latein.  Opiier  is 
derived  from  the  vocative  ave  pater. — Pp.  459-61.  M.  Ihm.  Mars  Mullo, 
Mars  Vicinnus  und  drei  pagi  der  Redones. — Pp.  461-2.    C.  Wachsmath.    Ein 


Digitized  by  VjOOQ IC 


490  AMERICAN  JOURNAL    OF  PHILOLOGY. 

inschriftliches  Beispiel  von  Kolometrie.  A  marble  tablet  inscribed  with  the 
Septuagint  version  of  the  15th  Psalm,  recently  found  at  Lapethus  in  Cyprus. 
It  was  probably  used  in  public  worship,  not  later  than  the  4th  century. — Pp. 
463-4.    W.  Schwarz.    Eigennamen  in  griechischen  Inschriften. 

Pp.  465-504.  Kritische  und  exegetische  Bemerkungen  za  Philo.  I.  P. 
Wendland. 

Pp.  505-8.  Zur  lateinischen  und  griechischen  Etymologie.  M.  Nieder- 
mann.  I.  The  -per  of  the  Latin  adverbs  nuper^  parumper^  paulisper^  semper^ 
topper^  etc.,  is  to  be  identified,  not  with  -pert  of  the  Oscan  peHropert  (=  quatef)^ 
but  with  the  Greek  particle  ffcp.  Ctparumper  and  pauxilUsper,  e.  g.,  with  the 
Homeric  fuvwOd  irep,  II.  'BeX^po-^vrriq  is  the  Lycian"  form  of  the  hero's 
name,  'lnn6vmK  the  corresponding  Greek  name.  Cf.  *AM'pav6oc,  the  Carian 
equivalent  of  ^iKirdifUDOC, 

Pp.  509-18.  Die  Composition  der  Chorlieder  Senecas.  F.  Leo.  The 
essential  difference  between  the  ardaifjui  of  Euripides  or  Sopholcles  and  the 
choruses  of  Seneca  is  that  the  latter  have  neither  strophe  nor  antistrophe. 
Like  the  monodies  of  Euripides  and  the  cantica  of  Plautus,  they  exhibit  a 
systematic  arrangement  of  metrical  periods  corresponding  to  divisions  of  the 
subject.  Cf.  especially  Oed.  403-508.  The  majority  of  the  lyric  metres  are 
taken  from  Horace;  the  general  form  of  the  chorus  represents  the  actual 
dramatic  practice  of  Seneca's  time. 

Pp.  519-56.  Der  korinthische  Bund.  J.  Kaerst.  The  Hellenic  confederacy 
formed  by  Philip  of  Macedon  after  the  battle  of  Chaeronea,  and  its  influence 
upon  the  subsequent  political  development  of  Hellas. 

Pp.  557-68.    Zu  Pseudokallisthenes  und  Julius  Valerius.    II.    A.  Ausfeld. 

Pp.  569-90.  Das  afrikanische  Latein.  W.  Kroll.  There  may  have  been 
in  the  time  of  Apuleius  and  Tertullian  a  tendency  toward  a  special  develop- 
ment of  the  Latin  spoken  in  Africa,  but  from  the  material  which  has  come 
down  to  us  we  cannot  learn  more  of  these  dialectic  variations  than  a  few 
uncertain  details.  The  archaisms  of  the  so-called  African  Latin  are  due  to  a 
general  tendency  of  the  time,  a  tendency  which  has  its  parallel  in  the  Greek 
of  the  same  period.  The  Grecisms  are  mainly  due  to  translation.  The 
tendency  of  the  period  is  distinctly  rhetorical,  but  this  concerns  the  literary 
language,  not  the  everyday  Latin  of  popular  speech.  Very  few  of  the  supposed 
provincialisms  which  are  usually  cited  as  characteristic  of  vulgar  Latin  are 
peculiar  to  African  writers. 

Pp.  591-623.  Ueber  die  Schriftstellerei  des  Klaudios  Galenos.  IV.  J. 
Ilberg. 

Miscellen. — Pp.  624-8.  L.  Radermacher.  Varia.  Notes  on  Plant.  Stich. 
270-1;  Varr.  Sat.  Menipp.,  fr.  384;  Propert.  IV  i,  7;  Aetna  Carm.;  Gratt. 
Cyneg. — Pp.  628-32.  K.  Laddecke.  Ueber  Beziehungen  zwischen  Isokrates' 
Lobrede  auf  Helena  und  Platons  Syraposion. — Pp.  632-3.  A.  Brinkmann. 
Bin  neues  Axiochoscitat. — P.  633.     M.  Ihm.     Probi  de  nomine  excerpta. — 


Digitized  by  VjOOQ IC 


REPORTS.  491 

Pp.  633-4.     R.  Fuchs.    dau  vorn,  l^«  hinten  (in  medical  Greek). — Pp.  634-5. 
L.  Radermacher.    airdpuna^  birUfuOev, 

A  third  section  of  the  Beitrige  zur  lateinischen  Grammatik,  by  Th.  Birt 
(see  vol.  LI,  pp.  70-108,  240-72 ;  A.  J.  P.  XVIII  108-9,  m).  is  printed  as  an 
ErganznngshefL  This  forms  a  book  of  218  pages,  entitled  *  Sprach  man  avrum 
Oder  aurumV  It  is  provided  with  a  table  of  contents,  two  indexes,  three 
Anh&nge  and  eight  pages  of  Nachtr&ge  und  Berichtigungen.  The  author 
collects  the  evidence  for  the  pronunciation  avrum, 

Havbkporo  Collbcb.  WiLFRKD  P.  MUSTARD. 


Digitized  by 


Gooi 


BRIEF  MENTION. 

The  long-expected  edition  of  Bacchylides  was  received  at  the  office  of  the 
Journal  from  the  authorities  of  the  British  Museum  after  the  present  number 
had  been  made  up,  and  there  is  no  space  for  a  full  account  of  this  priceless 
addition  to  the  Golden  Treasury  of  Greek  Lyric  Poetry.  The  editor  is  the 
distinguished  scholar  F.  G.  Kbnyon,  whose  mastery  was  evinced  by  his 
*A$7fiHiiuv  iro^reia  and  his  Herondas^  and  there  is  no  one  who  will  begrudge 
him  the  privilege  of  another  ediiio  princeps.  This  time  Mr.  Kenyon's  rare 
palaec^raphic  ability  was  not  put  to  so  severe  a  test  and  he  has  had  the 
advantage  of  help  and  counsel  from  eminent  Hellenists,  from  Jbbb,  with  his 
faultless  taste  and  his  unique  faculty  for  Greek  verse-composition,  from  the 
lamented  Palmbr,  from  Blass,  the  skilled  palaeographer,  from  Sandys,  with 
his  wide  command  of  the  whole  Greek  domain.  But  it  is  Mr.  Krnyon's 
edition  after  all,  and  to  him  the  gratitude  of  scholars  is  first  due.  Twenty 
poems,  some  of  them  entire,  have  been  brought  to  light  and  a  new  chapter  in 
the  history  of  Greek  literature  has  to  be  written.  Before  many  days  the 
philological  world  will  be  flooded  with  literature  on  the  subject,  with  emen- 
dations, restorations,  characteristics.  The  happy  hours  of  the  first  possession 
will  be  succeeded  by  weeks  muggy  with  extemporized  learning — extemporized^ 
for  comparatively  few  are  the  scholars  who  have  earned  the  right  to  speak 
authoritatively  by  reason  of  special  studies  in  this  too  much  neglected  domain 
of  Greek  poetry;  and  in  the  dense  air  which  is  about  to  envelop  Bacchylides, 
the  memory  of  these  three  or  four  undisturbed  days  will  come  back  with  the 
sigh,  fulsere  vere  candidi  tibi  soles.  True,  every  one  knew  in  advance  from 
the  old  fragments  what  manner  of  poet  we  were  to  expect ;  no  oid^puv  irirpa 
like  Aischylos,  no  raviirrepoq  cuerd^  like  the  Theban  singer,  but  a  clear  and 
fluent  and  brilliant  master  of  his  art,  one  who  well  deserves  the  title  by 
which  he  calls  himself  in  one  of  his  Hieronic  odes,  *  a  honey-tongued  KeYan 
nightingale.'  Still,  Bacchylides  has  given  us  much  more  than  we  could 
have  dreamed  of,  combinations  that  no  one  could  have  anticipated,  dramatic 
effects  which  theorists  had  denied  to  lyric  poetry,  and,  like  Cortes'  men^ 
scholars  are  looking  at  each  other  with  a  wild  surmise.  But  amid  all  the  joy 
over  the  new  treasure  and  the  endeavor  to  master  the  new  points  of  view,  the 
lover  of  Pindar  may  be  pardoned  for  thinking  chiefly  of  the  important  acces- 
sion to  the  Pindaric  apparatus  that  has  come  to  us  through  the  discovery  of 
Bacchylides.  Here  he  welcomes  confirmsrtion  of  previous  judgments,  there  he 
yields  with  what  grace  he  may  to  the  contravention  of  cherished  views.  Four- 
teen of  the  poems  are  epinikian  odes,  and  enough  of  these  are  sufficiently 
well  preserved  to  show  that  they  are  built  on  the  same  lines  as  those  of 
Pindar's  Songs  of  Victory.  The  type  is  older  than  Pindar.  It  is  in  the 
handling  of  the  type  that  the  differences  come  out.  Praise,  myth,  praise  are 
found  as  in  Pindar,  and  those  who  believe  with  Drachmann  that  the  myths 


Digitized  by  VjOOQ IC 


BRIEF  MENTION.  493 

are  mere  inpi^futra  will  doubtless  point  to  Bacchylides  with  triumph.  But 
others  will  maintain  that  Pindar  has  put  a  deeper  meaning  into  the  conven- 
tional adornment,  and  that  Bacchylides  was  satisfied  with  the  mere  embellish- 
ment, and  has  given  us  a  Euripidean  as  over  against  an  Aischylean  choral. 
Those  who  have  made  so  much  of  recurrent  words  in  Pindar  will  find  that 
Bacchylides  lends  scarcely  any  handle  to  repetition  as  a  reOfid^  of  lyric  poetry, 
as  an  indication  of  the  various  members  of  the  Terpandrian  udfio^.  No  such 
toying  iteration  is  to  be  found  in  Bacchylides  as  we  have,  for  instance,  in 
Pindar*s  Sixth  Olympian,  and  those  who  are  not  willing  to  concede  that 
Pindar,  like  other  strong  natures  such  as  Samson  and  Aias,  delighted  in  the 
play  on  words,  will  have  to  set  up  the  theory  that  Bacchylides  deliberately 
abandoned  the  technique  of  repetition  and  paronomasia  just  as  alliteration 
was  abandoned  in  English  poetry.  The  short  line  which  reigned  in  the  days 
before  Boeckh  will  probably  be  brought  to  honor  again  by  this  MS  of  Bacchy- 
lides, but  to  those  who  are  familiar  with  the  principles  of  kolometry  the  short 
line  is  merely  a  matter  of  convenience  to  the  eye.  Those  who  have  learned 
to  recognize  the  importance  of  the  literary  sphere  for  syntax  and  of  syntax  for 
the  literary  sphere  will  be  interested  to  find  that  different  as  Pindar  and 
Bacchylides  are  in  race,  in  gifts,  in  temperament,  the  lyric  law  keeps  them  to 
the  same  range.  One  has  not  much  new  syntax  to  learn  in  passing  from 
Pindar  to  Bacchylides. 

But  while  the  lover  of  Pindar  may  be  prone  to  dwell  on  the  resemblances 
and  differences  of  the  two  rival  poets,  the  student  of  Greek  literature  in 
general  will  be  most  interested  in  the  '*  lyrical  idylls/'  as  the  editor  calls 
them.  One  of  them  tells  of  the  demand  for  the  surrender  of  Helen,  another 
of  DeYaneira's  gift  to  Herakles,  yet  another  and  a  most  spirited  poem  of  the 
contest  between  Minos  and  Theseus,  in  which  Theseus  trusts  himself  to  his 
father  Poseidon  with  all  the  unreserve  of  the  divers  in  the  Bay  of  Naples. 
Most  remarkable  of  all  is  a  lyrical  dialogue  between  Aigeus,  king  of  Athens, 
and  Medeia,  his  queen,  which  not  only  increases  our  repertory  by  a  fine 
poem,  but  constitutes,  as  Mr.  Kenyon  says,  "  a  striking  and,  in  some  respects, 
unique  addition  to  our  knowledge  of  Greek  lyrical  composition." 

Some  of  the  poems  are  much  mutilated  and  tempt  the  restoring  hand. 
Would  that  a  thorough  study  of  the  odes  that  remain  entire  might  precede  the 
ready  fancy  of  the  multitudinous  guessers !  But  the  wish  is  vain.  As  I  lay 
down  my  pen  I  catch  sight  of  strings  of  unconvincing  restorations  and,  which 
is  worse,  hear  a  critic's  voice  declaring  Pindar  a  landlubber  and  Bacchylides  a  . 
seasoned  sailor.  It  is  true  that  Pindar*s  fellow-Boeotian,  Hesiod,  was  a  land- 
lubber oirre  ri  vavriTuTK  atao^tofitvog  obre  ti  vifuv  (O.  et  D.  649),  but  even  he  had 
to  make  space  for  navigation,  because  his  brother  Perses  might  take  to  the  sea. 
**  The  Boeotians,*'  says  Mr.  Roberts  the  apologist,  **  never  made  use  of  the  sea, 
favourably  situated  as  they  were,  to  the  same  extent  as  the  Dutchmen,"  ^  with 
whom  he  parallels  them ;  but  for  all  that  the  land  that  had  been  the  abode  of 
the  Minyan  vikings,  that  headed  the  catalogue  of  the  ships  in  Homer,  that 
had  Aulis  for  a  harbor,  can  hardly  be  classed  as  the  home  of  the  landlubber, 
and  I  may  not  have  blundered  so  much  after  all  in  calling  attention  to  the 
sea-air  in  Pindar.' 

1  The  Ancient  Boeotians,  p.  59.  *  Introd.  Essay,  xliii. 


Digitized  by  VjOOQ IC 


494  AMERICAN  JOURNAL    OF  PHILOLOGY. 

Mr.  BussELL  has  written  a  book  entitled  The  School  of  Plato  (London, 
Methuen  ;  New  York,  Macmillan),  which  is  readable  in  spite  of  its  preciosity 
and  suggestive  in  spite  of  its  oracular  tone.  The  late  Mr.  Pater  is  his  cyno- 
sure, which  will  hardly  be  an  unqualified  recommendation  in  the  eyes  of  some 
people,  and  Mr.  Bussell  has  taken  that  alembicated  stylist's  Plato  and  Platon- 
ism  more  seriously  than  a  philologian  would  be  apt  to  do.  Paired  with  this 
admiration  of  Mr.  Pater's  powers,  one  finds  a  curious  neglect  of  a  somewhat 
more  conspicuous  thinker,  and  it  is  with  no  little  astonishment  that  one  reads 
the  candid  confession  that  the  author  **  did  not  read  Lotze's  Microcosm  until 
the  greater  part  of  the  work  was  in  the  press.**  One  might  forgive  Mr. 
Bussell  for  calling  a  hen  'the  solicitous  stepmother  of  the  farmyard,'  but 
Lotze  was  a  philologian  as  well  as  a  philosopher:  he  was  the  translator  of 
Antigone  into  Latin  verse  as  well  as  the  author  of  the  Mikrocosmos,  and  an 
editor  of  a  Journal  of  Philology — especially  one  who  knew  Lotze  in  the  flesh 
— cannot  help  taking  the  matter  to  heart.  But  really  the  work  hardly  enters 
into  the  range  of  this  periodical,  for,  when  Mr.  Bussell  says  that  he  is  afraid 
that  his  title  will  appear  somewhat  of  a  misnomer,  his  fear  is  fully  justified. 
The  aim  of  the  work  is  to  prepare  the  way  for  an  elaborate  defence  of  the 
Roman  Imperial  Age — that  age  which,  as  Mommsen  has  said,  is  **mehr 
geschm&ht  als  gekannt** — and  Sokrates  and  Plato  together  occupy  only  40 
pages  out  of  a  total  of  346. 


Mr.  G.  F.  Hill,  in  his  Sources  of  Greek  History  between  the  Persian  and 
Pehponnesian  War  (Oxford.  At  the  Clarendon  Press) — a  collection  of  docu- 
ments for  the  period  known  as  the  Trevr^/covraeria — has  primarily  had  in  view 
an  educational  object.  The  student  of  history  is  to  be  taught  to  study  it  in 
the  light  of  ancient  authorities,  not  in  the  reflections  of  modem  writers.  Such 
a  lesson,  it  is  superfluous  to  say,  is  not  needed  by  the  readers  of  this  Journal. 
But  the  secondary  object  is  not  less  praiseworthy,  and  many  advanced  students 
will  be  glad  to  be  spared  the  necessity  of  referring  to  the  originals  for  the 
verification  of  the  references  in  such  a  book  as  Busolt's  Pentekoniaetie,  Of 
course,  as  in  the  case  of  punctuation,  any  arrangement  is  ipso  facto  an  inter- 
pretation, and  an  excerpt  cannot  take  the  place  of  an  unbroken  context ;  but 
every  such  work  has  its  limitations.  The  book  is  divided  into  the  following 
chapters:  I.  Origin  and  Organization  of  the  Athenian  Confederacy;  IL  The 
Quota  Lists ;  IIL  External  History  of  Athens,  her  Allies  and  Colonies ;  IV. 
The  Athenian  City ;  V.  The  Athenian  Constitution ;  VI.  Biographical ;  VII. 
Sparta  and  Peloponnesus;  VIII.  The  Western  Greeks.  There  is  also  a  list 
of  Athenian  archons,  but,  except  here  and  in  the  quota  lists,  dates  are  a  rarity, 
belonging,  presumably,  to  the  reflections  of  modern  writers. 


Digitized  by  VjOOQ IC 


RECENT  PUBLICATIONS. 

Thanks  are  due  to  Messrs.  Lemcke  &  Buechner,  812  Broadway,  New 
York,  for  material  furnished. 

AMERICAN. 

Apuleius  (Lucius).  The  Story  of  Cupid  and  Psyche  done  out  of  the 
Latin  by  Walter  Pater.  Portland,  Me.,  T,  B,  Mosher^  1897.  i6mo,  pap., 
75  cts. 

Cicero  (M.  T.)  Selected  Letters ;  ed.,  with  introd.  and  notes,  by  F.  F. 
Abbott.     Boston,  Cr/ff/f  &>  C^.,  1897.     76  +  315  pp.     i2nio,  cl.,  ^1.35. 

Conway  (R.  S.)  The  Italic  Dialects;  ed.  with  a  grammar  and  glossary. 
V.  I,  pt.  I.  The  Records  of  Oscan,  Umbrian  and  the  Minor  Dialects.  V.  2, 
pt.  2.  Grammar  of  the  Dialects;  with  appendix,  indices  and  glossary. 
New  York,  The  Macmillan  Co,^  1897.  2  v.,  18  +  456,  6  +  457-686  pp.  8vo, 
cl.,  net  ^7.50. 

Horace  (Q.  H.  F.)  Works;  tr.  by  A.  Hamilton  Bryce.  New  York,  The 
Macmillan  Co,,  1897.     4  v.     i6mo,  cl.,  net,  ea.  30  cts. 

Lyttelton  (E.)  Are  we  to  go  on  with  Latin  Verses?  New  York,  Long- 
mans; Green  &*  Co.^  1897.     193  pp.     i2mo,  cl.,  ^1.25. 

Muller  (F.  Max).  The  Sacred  Books  of  the  East;  tr.  by  various  oriental 
scholars.  American  ed.  12  v.  V.  i.  The  Upanishads.  New  York,  The 
Christian  Literature  O.,  1897.  101  +  320+52  +  350  pp.  8vo,  subs.,  cl., 
I2.50. 

Plato.  The  Philebus  ;  ed.,  with  introd.,  notes  and  appendices,  by  R.  G. 
Bury.     New  York,  The  Macmillan  Co.,  1897.     87  +  223  pp.     8vo,  cl.,  net 

13.25. 

The  Republic ;  ed.,  with  critical  notes  and  an  introd.  on  the  text, 

by  Ja.  Adam.  New  York,  The  Macmillan  Co,,  1897.  21  +  329  pp.  i2mo, 
cl.,  net  ^1.25. 

Sophocles.  The  Text  of  the  Seven  Plays  ;  ed.  with  an  introd.  by  R.  C. 
Jebb.   New  York,  The  Macmillan  Co,,  1897.    14+364  pp.   8vo,cl.,net  ^1.40. 

Xenophon.  Works ;  tr.  into  English  by  H.  G.  Dakyns.  In  4  v.  V.  3, 
pt.  I.  The  Memorabilia,  and  Apology,  The  Economist,  The  Symposium, 
and  Hiero.  Pt.  2.  On  the  Duties  of  a  Cavalry  General,  On  Horsemanship, 
and  On  Hunting.  New  York,  The  Macmillan  Co.,  1897.  70+130  pp. 
i2mo,  cl.,  ^1.25. 

ENGLISH. 

Jannaris  (A.  N.)  Historical  Greek  Grammar,  chiefly  of  the  Attic  Dialect, 
as  written  and  spoken  from  Classical  Antiquity  to  the  Present  Time.  Lon- 
don, Macmillan  ^  Co,,  1897.     8vo,  776  pp.     £1  5s. 


Digitized  by  VjOOQ IC 


496  AMERICAN  JOURNAL   OF  PHILOLOGY. 

Jensen  (H.)  A  Classified  Collection  of  Tamil  Proverbs.  London,  1897* 
8vo.    88. 

Macalister  (R.  A.  S.)  Studies  in  Irish  Epigraphy.  Part  I.  London, 
1897.    8vo.     3s.  6d. 

Murray  (J.  A.  H.)  A  New  English  Dictionary.  Part  11.  Foisty-Frankish. 
London,  1897.     4to,  132  pp.     5s. 

Searle(W.  G«)  Onomasticon  Anglo-Saxonicnm.  A  list  of  Anglo-Saxon 
proper  names  from  the  time  of  Bede  to  that  of  King  John.  •  London,  1897.. 
8vo.    20S. 

FRENCH. 

de  Jnlleville  (P.  L.)  Histoire  de  la  langue  et  de  la  litt^rature  fran9aise. 
Tome  IV.  i7e  si^cle.     Paris,  1897.    8vo.     16  fr. 

Knms  (A  )    Les  choses  naturelles  dans  Hom^re.    Paris,  1897.    8vo.    5  fr» 

Ledrain  (E.)  Dictionnaire  de  la  langue  de  Tancienne  Chaldee.  Paris,. 
1897.    8vo,  600  pp.    50  fr. 

Michaelis  (H.)  et  Passy  (P.)  Dictionnaire  phon^tiqne  de  la  langue 
franfaise.     Paris,  1897.    8vo,  320  pp.     5  fr. 

Oischewsky  (S.)  La  langue  et  la  m^trique  d'H^rondas.  Leiden,  1897* 
8vo,  84  pp.    90  c 

GERMAN. 

Abhandlungen  f.  die  Kunde  des  Morgenlandes,  hrsg.  v.  der  D.  M.  GeselU 
schaft  unter  der  Red.  von  £.  Windisch.  X.  Bd.  Nr.  2  u.  4.  gr.  8.  L.,. 
F.  A,  Brockhaut*  Sort, — 2.  Avasyaka-Erzahlungen,  die.  Hrsg.  v.  £.  Leu- 
mann.  i.  Hft.  49  S.  m.  1.80. — 4.  Marath-Uebersetzung,  die,  der  Suka* 
saptati.     Marathi  u.  deutsch  v.  Rich.  Schmidt,    iii,  viii,  175  S.    m.  7.50. 

Acta  martyrum  et  sanctorum  (syriace),  ed.  Paulus  Bedjan,  Cong.  Miss.. 
Tom.  VII  vel  Paradisus  patrum.  gr.  8.  xii,  1019  S.  Parisiis.  L.,  O, 
Harrassowitt,     m.  28. 

Au8  der  Nibelungen  Not.  Fol.  8  Bl.  Fk8m.-Drucke.  St.  Gallen,  A. 
and  J.  Koppel     m.  6. 

Babrii  fabulae  Aesopeae,  recognovit,  prolegomenis  et  indicibus  instruxit 
Otto  Crusitts.     8.     cvi,  440  S.  m.  3  Taf .     L.,  B,  B,  Teubntr,    m.  8.40. 

Beitr&ge,  MQnchener,  zur  romanischen  u.  englischen  Philologie.  Hrsg.. 
v.  H.  Breymann  u.  J.  Schick.  XIII.  gr.  8.  L.,  A,  Deichert  Nachf,— 
XIII.  Fest  (O.)  Der  Miles  gloriosus  in  der  franzttsischen  KomOdie  von 
Beginn  der  Renaissance  bis  zu  Molidre.     xv,  123  S.     m.  2.80. 

.—  Wiener,  zur  englischen  Philologie,  unter  Mitwirkg.  v.  K.  Luick  u.. 
A.  Pogatscher  hrsg.  v.  J.  Schipper.  VL  gr.  8.  Wien,  W,  BraumUller,-^ 
VI.  Schmid  (D.)  William  Congreve,  sein  Leben  u.  seine  Lustspiele.  viii,. 
179  S.     m.  4. 

Bibliotheca  buddhica.  ^ikshasamuccaya.  A  compendium  of  Buddhistic 
teaching  compiled  by  ^antideva.  Ed.  by  C.  Bendall.  I.  gr.  8.  viii,  96  S. 
St.  Petersburg.     L.,  Vats'  Sort,    m.  2.50. 

Bibliothek,  altfranzj^sische.  15.  Bd.  8.  L.,  O,  R,  Reiiland,—\^,  Alls- 
cans,  m.  BerUcksicht.  v.  Wolframs  v.  Eschenbach  Willehalm  kritisch  hrsg. 
v.  Gust.  Rollin.    Ixix,  163  u.  132  S.     (1894.)     m.  6. 


Digitized  by  VjOOQ IC 


I 


RECENT  PUBLICATIONS.  497 

Bibliothek  der  angelsachsischen  Prosa,  begrttndet  v.  Chrn.  W.  M.  Grein, 
fortges.  V.  R.  P.  Wttlker.  4.  Bd.  i.  Halfte.  gr.  8.  L.,  G.  H,  Wigand.^ 
I.  Alfred's,  KOnig,  Uebersetzung  v.  Bedas  Kirchengeschicbte.  Hrsg.  ▼• 
Jac.  Scbipper.     i.  H&lfte.    ix,  272  S.    m.  15. 

Biasing  (F.  W.  ▼.)  Die  statistische  Tafel  v.  Karnak.  4.  xxxviii,  67  S. 
L.,/.  C.  Hinrieks'  Vtrl,    m.  15. 

Bolland  (G.  J.  P.  J.)  Die  althelleniscbe  Wortbetonung  im  Lichte  der 
Geschichte.  2.  Drock.  gr.  8.  loi  S.  Leiden.  Amsterdam,  y.  Muller* 
m.  2.50. 

Brockelmann  (Carl).  Geschichte  der  arabischen  Litteratur.  I.  Bd. 
I.  Hilfte.    gr.  8.     240  S.     Weimar,  £.  Filter,    m.  10. 

Brogmann  (Karl)  u.  Delbrtlck  (Berthold).  Grnndriss  der  vergleichenden 
Grammatik  der  indogermanischen  Sprachen.  2.  Bearbeitg.  i.  Bd.:  Ein* 
leitnng  u.  Lantlehre.  2.  H&lfte  (§695  bis  1084).  gr.  8.  ix  u.  S.  623*i098. 
Strassburg,  K,  /.  TrUbner  Vtrl.    m.  12. 

Catonis  (M.  Porci)  de  agri  cultura  liber,  M.  Terenti  Varronis  rerum  rus- 
ticarum  libri  III,  ex  rec.  Henr.  Keilii.  Vol.  III.  Fasc.  I.  Index  Terborum 
in  Catonis  de  re  rustica  librom  composuit  Rich.  Krnmbiegel.  gr.  8.  iii, 
82  S.     L.,  B.  G.  Teubner.    m.  3. 

Chwolson  (D.)  Syrisch-nestorianische  Grabinschriften  aus  Semirjet- 
schie.  Neae  Folge.  lirsg.  u.  erkUrt.  gr.  4.  62  S.  SL  Petersburg.  L., 
Vots*  Sort.     m.  6. 

Columellae  (L.)  opera  quae  exstant,  recensait  Vilh.  LandstrOm.  Fasc  I. 
Liber  de  arboribos  qai  Tocatur.  gr.  8.  xii,  43  S.  Upsaliae.  L.,  O.  liar' 
rassowitt.    m.  1.75. 

Firmici  Materni  (Inlii)  matheseos  libri  VIII,  edd.  W.  Kroll  et  F.  Skotsch. 
Fasc.  I.  Libros  IV.  priores  et  V.  prooemium  continens.  8.  xii,  280  S. 
L.,  B.  G.  Teubmr.    m.  4. 

Flensburg  (Nils).  Stndien  anf  dem  Gebiete  der  indogermanischen 
Wurzelbildnng.  I.  Die  einfacbe  Basis  ter-  im  Indogermanischen.  Lex.-8. 
xi,  115S.     'Lxxjidt  IIj.  Mdller's  Univ.'Buchh,     m.  2.60. 

GrOber  (Gust.)  Grundriss  der  romanischen  Pbilologie.  2.  Abtlg. 
4.  (Schlass-)Lfg.  gr.  8.  yIi  u.  S.  385-496.  Strassburg,  IC.  J.  TrObner 
Verl,     m.  2. 

Haeberlin  (C.)  Griechische  Papyri,  gr.  8.  131s.  \^^  O.  Harrassawitu 
m.  3.60. 

Hayyfig :  The  Weak  and  Geminative  Verbs  in  Hebrew.  The  Arabic 
text  now  published  for  the  first  time  by  Morris  Jastrow,  Jr.  gr.  8.  Ixxxv, 
271  S.     Leiden,  Buchh.  u,  Druckereivorm,  E.J.  Brill,     m.  10. 

Heberdey  (Rud.)  Opramoas,  Inschriften  vom  Heroon  zu  Rhodiapolis. 
Nea  bearb.     Lex. -8.     71  S.  m.  5  Taf.     Wien,  A.  Holder.     Kart.,  m.  5. 

Horatias  (Q.  F.)  Werke.  Hrsg.  v.  O.  Henke  u.  C.  Wagener.  i.  Bd.: 
Die  Oden  u.  Epoden.  Hrsg.  v.  Osk.  Henke.  Einleitung.  Text.  gr.  8. 
viii,  286  S.     Bremen,  M.  Heinsius  Nachf.     m.  3. 

Journal  de  la  Soci^te  finno-ougrienne.  XV.  gr.  8.  iii,  57,  64  u.  34  S. 
Helsingissft.     L.,  O.  Harrassowitz  in  Komm,     m.  6.40. 

Koehm  (los.)  Quaestiones  Plautinae  Terentianaeque.  gr.  8.  56  S. 
Giessen,  J.  Richer,    m.  1.20. 


Digitized  by  VjOOQ IC 


498  AMERICAN  JOURNAL    OF  PHILOLOGY. 

KOnig  (Fr.  Ed.)  Historisch-kritisches  Lehrgebaade  der  hebrftischen 
Sprache.  2.  Halfte.  2.  (Schla8s-)Tbl.  A.  u.  d.  T.:  Historiscb-compara- 
tivc  Syntax  der  bcbr.  Spracbe.  gr.  8.  iz,  721  S.  L.,y.  C.  Ilinrichs*  VerL 
m.  18. 

Kroehnert  (Otto).  Canonesne  poetanim  scriptorum  artificium  per  anti- 
quitatem  f uerunt  ?  Diss.  gr.  8.  iii,  67  S.  K5nigsberg,  Gr&fe  &*  Unter*s 
Sort.     m.  2. 

Lange  (Edm.)  Die  Arbeiten  zu  Thnkydides  seit  1890.  Eine  krit 
Uebcrsicht.  i.  H£t.  [Aus 'Philologus.']  gr.  8.  8.658-7x3.  I.,,  Dieterich, 
m.  1.20. 

Lindberg  (O.  £.)    Vergleichende  Grammatik  der  semitischen  Sprachen. 

1.  Lautlebre.     A.  Konsonantismus.     gr.  8.     xi,  160  S.     Gothenburg,  iVet' 
tergren  *•  Kerbtr.     m.  8.35. 

Lucani  (M.  Annaei)  Pharsalia.  Cum  commentario  critico  ed.  C.  M. 
Francken.  Vol.  II  continens  libros  VI-X.  gr.  8.  xiv,  276  S.  Leiden, 
A,  W,  Sijthoff,     m.  9.60. 

Lydi  (loannis)  liber  de  ostentis  et  calendaria  graeca  omnia,  iterum  edidit 
Cart.  Wachsmuth.    8.     Ixxii,  366  S.     L.,  B.  G.  Teubner.     m.  6. 

Martin  (E.)  u.  Lienhart  (H.)     WOrterbuch  der  els&ssischen  Mundarten. 

2.  Lfg.     gr.  8.     S.  i6x-304«     Strassburg,  JC,  J.  TrUbner  Verl,    m.  4. 
Mythographi  graeci.     Vol.  III.  Fasc.  I.     8.     L,jB.  G,  Teubner,     m.  1.20. 

—I.  Pseado-Eratosthenis  Catasterismi.recensuit  Alex.  Olivieri.   xviii,76S. 

Paulcke  (Max).  De  tabula  iliaca  quaestiones  Stesichoreae.  Diss.  gr.8. 
109  S.     Regimonti  Boruss.     L.,  G,  Fock,    m.  2.50. 

Paulson  (Jobs.)  Lukrezstudien.  I.  Die  fiussere  Form  des  lukretian. 
Hexameters,    gr.  8.     37  S.     Gothenburg,  Wettergren  ^  JCerber,     m.  — 85. 

Plini  Secundi  (C.)  naturalis  historiae  libri  XXXVII.  Ed.  Carol.  Mayhoff. 
Vol.  V.  Libri  XXXI-XXXVII.     8.     x,  512  S.     L.,  B,  G,  Teubner.    m.  6. 

Soltau  (Wilh.)  Livius'  Geschicbtswerk,  seine  Komposition  u.  seine 
Quellen.     gr.  8.     viii,  224  S.     L.,  Dieterich.    m.  6. 

Tallqvist  (Knut  L.)  Arabische  Sprichw5rter  u.  Spiele,  gesammelt  u. 
erkl&rt.     gr.8.     152  S.     Helsingfors.     \,.^0,  HarrassowitM,     m.  4. 

Texte,  Irische,  m.  Uebersetzungen  u.  WOrterbuch.  Hrsg.  ▼.  Wh.  Stokes 
u.  £.  Windisch.  3.  Serie.  2.  Heft.  gr.  8.  iii  u.  S.  285-596.  L..^*.  HirzeL 
m.  10. 

Theophrast's  Charaktere.  Hrsg.,  erkl&rt  u.  tibers.  ▼.  der  philolog.  Ge- 
sellschaft  zn  Leipzig,  gr.  8.  Ixiv,  276  S^m.  14  Abbildgn.  L.,  B,  G.  Teub- 
ner,   m.  6. 

Vatsyayana  :  Das  Kamasutram.  Nebst  dem  Commentare  des  Yagddhara. 
Aus  dem  Sanskrit  tibers.  u.  hrsg.  ▼.  Rich.  Schmidt  gr.  8.  7,478  S.  L., 
IV.  Friedrich.     m.  16. 

Witkowski  (Stanisl.)  Prodromns  grammaticae  papyrorum  graecomm 
aetatis  Lagidarum.  gr.  8.  65  S.  Krakau,  Buchh.  der  poln.  Verlagsgesell- 
schaft  in  Komm.    m.  3. 

Zimmern  (Heinr.)  Vergleichende  Grammatik  der  semitischen  Sprachen. 
Elemente  der  Laut-  u.  Formenlehre.  Mit  i  Schrifttaf.  ▼.  Jul.  Eating.  8. 
xi,  194  S.     B.,  Reuther  &*  Reichard.     m.  5.50. 


Digitized  by  VjOOQ IC 


RECENT  PUBLICATIONS.  499 

ITALIAN. 

Earipide.  Medea :  tragedia  con  introduzione,  commento  ed  appendice 
critica  di  G.  B.  Camozzi.     Imola.    8vo,  p.  i88.    L.  3. 

Novati  (Francesco).  LMnflosso  del  pensiero  latino  sopra  la  civilt^ 
italiana  del  medio  evo.    Milano.     i6nio,  p.  x8o.    L.  3. 

Pizzi  (Italo).  Grammatica  elementare  dell'  antico  iranico  (zendo  e 
persiano  antico),  con  antologia  e  relativo  vocabolario.  Torino.  8vo,  p.  92. 
L.  2.50. 

PulW  (L,  F.)     Studi  italiani  di  filologia  indo-iranica.    Pisa.    8vo.    L.  15. 

Sofocle.  Elettra,  con  note  di  Domenico  Bassi.  Torino.  8to,  p.  139. 
L.  1.80. 

Studi  italiani  di  filologia  classica.    Vol.  V.    Firenze.    8vo,  p.  518.    L.  20. 

SPANISH. 

Ximenez  de  Embnn  y  Val  (T.)  Lengua  espaftola  en  el  siglo  de  oro  de 
su  literatura.     Zaragoza,  1897.    8vo,  304  pp.    8  pes. 


Digitized  by  VjOOQ IC 


BOOKS  RECEIVED. 

Anzeiger  fttr  indogermanische  Sprache  u.  Altertamskunde.  Heraosg.  ▼. 
Wilhelm  Streitberg.    Achter  Band.   Drittes  Heft.     Freiburg  i.  d.  S.  1897. 

Archaeological  Institute  of  America.  Papers  of  the  American  School  of 
Classical  Studies  at  Athens.     Vol.  VI,  1890-97.     Boston,  Ginn  &>  Co.,  1897. 

Aristophanes  (The  Wasps  of).  With  introduction,  metrical  analysis, 
critical  notes  and  commentary.  By  W.  J.  M.  Starkie.  London,  Macmillan 
^  Co,     New  York,  The  Macmillan  Co.,  1897. 

Bacchylides,  The  Poems  of.  From  a  papyrus  in  the  British  Museum. 
Ed.  by  F.  G.  Kenyon.     London,  British  Museum,  1897. 

Bierma  (J.  W.)  Quaestiones  dc  Plautina  Pseudolo.  (Thesis.)  Gronin- 
gae,  Scholtens  ^  Zoon, 

Boiling  (Geo.  Melville).  The  Participle  in  Hesiod.  (J.  H.  U.  Doctoral 
Thesis.)  Washington,  D.  C,  Catholic  University  Bulletin,  vol.  Ill,  pp. 
421-71. 

Brugmann  (K.)  u.  Delbrttck  (B.)  Grundriss  der  vergleichenden  Gram- 
fnatik  der  indogermanischen  Sprachen.  Erster  Band.  Einleitung  u.  Laot- 
lehre  v.  Karl  Brugmann.  2.  Halfte.  2.  Bearbeitg.  Strassburg,  K,  J. 
2ratner,  1897.     X2  m. 

Buddisrao  per  P.  £.  Pavolino.  (Manuali  Hoeppli.)  Milano,  Ulrico 
Hoeppli,     I  1.  50  c. 

Cicero.  Laelius  de  Amicitia,  with  notes  by  C.  E.  Bennett.  (Students* 
Series.)     Boston,  Leach,  Shewell  &'  Sanborn,  1897. 

Selected  Letters  of.     Ed.  with  introduction  and  notes  by  F.  F. 

Abbott.     Boston,  Ginn  ^  Co,,  1897. 

Columbia  Bicycle  Calendar  for  1898.  Hartford,  Conn.,  Pope  Manufac- 
turing Co,     10  cts. 

Drakoules  (Platon  E.)  Neohellenic  Language  and  Literature.  Oxford, 
B,  H,  Blackwell,  1897.     is.  6d. 

Educational  Review.  Ed.  by  N.  M.  Butler.  January,  1898.  New  York, 
Henry  Holt  6f  Co,    I3  per  annum. 

Emery  (Annie  C.)  The  Historical  Present  in  Early  Latin.  (Bryn  Mawr 
Diss.)    Ellsworth,  Maine,  1897. 

English  Dictionary  (The  Oxford).  Ed.  by  J.  A.  H.  Murray.  Foistt- 
Prankish  (vol.  IV),  by  Henry  Bradley.  (Double  Section.)  Oxford, 
Clarendon  Press,     New  York,  Macmillan  &»  Co,,  1897.     5s. 

Forman  (L.  L.)  Index  Andocideus,  Lycurgeus,  Dinarcheus.  Oxouii, 
E  Prelo  Clarendoniano,  MDCCCXCVII. 

Grundriss  der  indo-arischen  Philologie.  Herausg.  von  Georg  Btthler. 
II.  Band.  3.  Heft,  B.  Inhalt :  Indian  Coins,  by  E.  J.  Kapson.  With  five 
plates.     Strassburg,  A^.  y.  TVtf^/f^r,  1898.     Subs.-pr.,  5  m.;  Einze]preis,6m. 

Harvard  Studies  and  Notes  on  Philology  and  Literature.  No.  V.  Child 
Memorial  Volume.     Boston,  Ginn  &>  Co.,  1896. 

Horton-Smith  (Lionel).  Two  Papers  on  the  Oscan  Word  Anasaket, 
London,  D,  Nuit,  1897. 


Digitized  by  VjOOQ IC 


BOOKS  RECEIVED.  501 

Isaiah,  The  Book  of  the  Prophet.  A  new  English  translation,  printed  in 
colors  exhibiting  the  composite  structure  of  the  book.  With  explanatory 
notes  and  pictorial  illustrations  by  T.  K.  Cheyne.  (Haupt's  Polychrome 
Bible.)     New  York,  Dodd,  Mead  £r*  Co.,  1897.     ^2.50. 

Jenkins  (O.  P.)  The  Passing  of  Plato.  Address.  Leiand  Stanford 
University,  1897. 

Kondratiew  (Sergius).  Index  ad  oratorem  Lycnrgum.  Mosquae,  0/m^ 
Tickomirovmm,  1897. 

Leemen  wagentje  (Het).  Indisch  toonenspiel  uit  Sanskrt  en  Pr&krt  in 
faet  Nederlandsch  Yertaald  door  J.  P.  Vogel.  Amsterdam,  Scheltema  en 
Hoikema,  1897. 

Livius'  Geschichtswerk,  seine  Komposition  a.  seine  Quellen.  Von 
Wilhelm  Soltau.     Leipzig,  Dieteriek'sche  Verlags^Buchhandlung^  1897. 

MacMechan  (A.)    Vergil.    A  lecture.     Halifax,  N.  S. 

Meyer  (P.)  Notice  sur  les  Corrogationes  Promethei  d* Alexandre  Neck- 
ham.     Paris,  Klincksieck,  1897. 

Modern  Language  Association  of  America,  Publications  of  the.  Ed.  by 
James  W.  Bright.  Vol.  XII,  No.  4.  N.  S.,  vol.  V,  No.  4.  Contents  :  XL 
Goebel  (J.)  On  the  Original  Form  uf  the  Legend  of  Sigfrid. — Proceedings. 
Baltimore,  yMff  Murphy  &»  Co,,  1897. 

Muret-Sanders  encyclopaedisches  WOrterbuch  der  englischen  u.  deot- 
-schen  Sprache.  Teil  II.  Deutsch-Englisch.  New  York,  The  International 
News  Co,,  1897.     24  Lfgn.     @  i  m.  50  pf. 

National  Educational  Association.  Journal  of  Proceedings  and  Addresses 
of  the  26th  annual  meeting,  held  at  Milwaukee,  July  6-9,  1897.  Chicago, 
University  of  Chicago  Press^  *897» 

Neue  JahrbOcher  fttr  das  klassische  Altertum,  Geschichte  u.  deutsche 
Litteratur  u.  fUr  Paedagogik.  Herausg.  v.  Johannes  Ilberg  a.  Richard 
Kichter.  Erster  Jahrgang,  1898.  I.  u.  II.  Band.  x.  Heft.  Leipzig,  B.  G, 
Teubner, 

Nolhac  (P.  de).  Le  Virgile  de  Vatican  et  ses  peintnres.  Paris,  Klinck* 
jieek,  1897. 

Old  Latin  Biblical  Texts.  No.  IV.  Ed.  by  Henry  J.  White.  Oxford, 
At  the  Clarendon  Press.    New  York,  American  Branchy  1897.     5s. 

Kevne  des  Humanit^s en  Belgique.  Publication  Trimestrielle.  ire  ann^e. 
Juillet-Octobre,  1897.     Anvers,  Hasselt,  1897.     5  fr.  par  an. 

Studia  Sinaitica.  No.  VI.  A  Palestinian  Syriac  Lectionary,  containing 
Lessons  from  The  Pentateuch,  Job,  Proverbs,  Prophets,  Acts,  and  the 
Epistles.  Ed.  by  Agnes  Smith  Lewis,  with  critical  notes  by  Eberhard 
Nestle  and  a  glossary  by  Margaret  D.  Gibson.     London,  C.  J.  Clay,  1897. 

128. 

Tannery  (P.)  Le  traite  du  quadrant  de  Maltre  Robert  Angles.  (Mont- 
pellier,  Xllle  siicle.)  Text  latin  et  ancienne  traduction  grecque.  Paris, 
Xlincksieck,  1897. 

Toutain  (J.)  L'Inscription  d'Henchir  Mettich.  Un  nonveau  document 
sur  la  propriety  agricole  dans  l*Afrique  romaine.  Academie  des  inscrip- 
tions,   i.serie.    Tome  XI.    ire  partie.    Paris,  AViWijiVfi,  1897.    3fr.  8oc. 


Digitized  by  VjOOQ IC 


INDEX  TO  VOL.  XVIII. 


Abinnius,  FlaTius,  Correspondence  of,  aaS,  899 
Accent,  Grave,  in  Greek,  xia 

Aeschylus,  Pers.  527-31,  328 

S.  c.  Theb.  961  foil.,  aaS 

Air,  the  seat  of  the  soul,  488 

Albertus  Magnus,  Dante's  obligations  to,  485 

his  name = della  Magna  ('  Germany '),  485 
Alexander  and  Ptolemy  worship  in  Egypt,  339 
Alkman's  Partheneion,  365 

Antiphon  and  Attic  criminal  law,  364,  365 
Antony,  Mark,  Parthian  expedition  of.  363 
Aristides,  Election  law  of,  Z13 

Aristid.  Quint.,  p.  96,  331 

Aristokritos  and  his  '  Theosophy,'  zx  x 

Aristotle,  *A9rivai»v  iroAireia,  54,  7,  xio 

Arrian's  Penplus,  zog 

Assyrian :  imittu  and  sattukku,  x  16 

Assyriologie,  Beitrage  zur,  X16-18 

Athenaeus,  530^  emended,  x  1  a 

Athenian  kings,  363 

Athens,  Primaeval,  109,  xza,  lao 

Attic  Dionysos-festivals,  340 

inscriptions  of  the  Vth  century,  303 

Augustin,  Confess.  8,  a,  3,  xi6 

Babrius.  LXI  75,  338 

Bacchylides,  Renyon's,  493-3 

Bacon-Shakespeare  question,  107 

Baixantinb,  W.  G.    Negative  Futures 

in  the  Greek  New  Testament,  453^9 

Beitrage  zur  Assyriologie,  z  16-1 8 

Benecke's  Antimachus  of  Colophon  and 

the    Position  of  Women   in   Greek 

Poetry  (noticed),  370 

Bloompibld,  Mauricb.    Review  of  Hil- 

lebrandt's  Ritual-Literatur,  350-3 

Boeotians  as  landlubbers,  493 

BoLLiNG,    Gborgb    Mblvillb.      Latin 

•astro'f  70-3 

Books  Received,  130-9,  353-4,  38«-a,  500-1 
Bomecque's  Ciceronis  Oratio  in  Verrem 

De  Signis  (noticed),  95 

Brial's  Essai  de  S^mantique  (noticed),  368 
Breton,  meaning   of  the   word    in    the 

Xllth  century,  486 

Brief  Mention,  XX9-34,  34X-^,  368-70,  493-5 
Bucket,  Kicking  the,  191 

Bussell's  School  of  Plato  (noticed),  494 

Byron  as  a  translator,  338 


Caecilius  of  Calacte, 
Caliban,  Etymology  of, 
Callimachus,  Epigrams  of, 
Cambyses,  Records  of  the  time  of. 
Carpus  (ap.  Petron.),  Etymology  of, 
Castilian,  Enclitic  and  proclitic 


3ox-ia 

337 
364 
XX7 
341 
in, 

357-60 

Catullus,  LVII  6,  X13 

Charisius,  Present  form  of,  X13 

Chaucer  characterized,  470,  471 

C.  1.  L.  X4679,  114 

Cicero,  Ep.  ad  Att.  II  30,  x;   IV  is,  a; 

XII  33,  3,  ZX5 

Ep.  ad  Brut.  14,5.  Z15 

Ad  Fam.  VIII  17,  3  (em.),  X90 

Ep.  ad  Qu.  Fr.  I  fx  ;  II  14,  3 ;  III  8,  x,  x  15 

III,  xxo 

Cicero's  Verrine  orations,  93-5 

influence,  94a,  343 

Cicotti's  Processo  di  Verre  (rev.),  93-5 

Clbmbnt,  Willako  K.     The   Use   of 

tntm  in  Plautus  and  Terence,         409-15 
Clement  of  Alexandria  and  Philo,  365-6 

Comedy,  Greek,  Superstitions  and  Popu- 
lar Beliefs  in,  189-304 
Commedia  dell'  arte,  359 
Cook,  Albbkt  S.    Report  of  Englische 

Studien,  96-108 

Courthope's  History  of  English  Poetry 

(rev.),  464-80 

Creole  dialect  of  the  Antilles,  486 

Ctesias  and  his  Assyrtaca,  340 

Cyriacus  of  Ancona  on  Strabo,  1x4 

Cyrus,  Records  of  the  time  of,  1x7 

Dahlerup's  Nekrolog  5ver  Karl  Vemer,  9X-3 
Dante's  obligations  to  earlier  literature,  489 
Definition  of  some  rhetorical  terms,  306-X3 
Delphi,  Conflagrations  of  the  temple  at,  X13 
Diodorus,XVIIl7, 5,  365 

Dionysus  Musagctes,  365-^ 

DoDGB,  Damibl  Kilham.  Review  of 
Dahlerup's  Nekrolog  Over  Karl  Ver» 
ner,  9X-3 

Donatus,  Commentum  Terentii,  MSS  of,  339 
Door,  Sacredness  of,  X9Z 

Ddrpfeld  and  the  Greek  Theatre,  x-98 

Dyroff's  Geschichte  des  Pronomen  Re* 
flexivum  (rev.),  3x4-37 


Digitized  by  VjOOQ IC 


INDEX. 


503 


Eclipses  caused  by  eril  spirits,  194 

Effyptj  Alexander  and  Ptolemy  worship 

in.  339 

Egyptian  worships,  364 

Elirabethan  drama,  335 

Embrt,  V.  J.    On  the  Definition  of  Some 

Rhetorical  Terms,  306-17 

Empedoclcs  emended,  366 

Englische  Studien,  Report  of,    giS-ioS,  333-8 

English,  Courthope's  History  of  English 

Poetry,  464-30 

drama  in  the  XVIth  century,  479 

Interchange  and  Tautology,  96 

Old  English  Poetry,  Expanded  Lines 

In,  337 

Quantity  of  vowel  before  'iti/in  ME.,  xox,  337 
Tautology,  96 

To  dart,  X03 

Enim  in  Plautus  and  Terence,  403-15 

EspagnoUe's  French  etymologies,  133 

Ethics  and  Amenities  of  Greek  Histori- 
ography, 355-74 
Etiam  in  Plautus  and  Terence,  36-43 
Euphuism,  475  foil. 
Euripides,  Danae  of,  109 
Troades,  Tyrrell's  (noticed),  340 
Europe  in  the  sixteenth  century,             473-3 

Fat,  Edwim  W.  Textual  Notes  and 
Queries  on  Plautus,  z68-88 

Noie  on  Latin  nihil,  463-3 

Fbrmald,  O.  M.  Review  of  Peck's 
Harper's  Dictionary  of  Classical 
Literature  and  Antiquities,  7^84 

Fescennine  verses,  txa 

Fictitious  sources  in  Greek  historiog- 
raphy, 363  foil. 

Folk-tales,  Orientalist  theory  of  dissemi- 
nation of,  359 

Forman  on  cri  w.  gen.  and  dat.,  zi8, 1x9 

French:  aochier,  ar tiller,  361;  baugan, 
487;  chevine,  487;  combre  and  its 
deriratives,  357;  cormoran,  358; 
d6me=:i.  Ut^A,  a.  domum,  361; 
girouette,  358;  goupillon,  361 ; 
hampe,  358;  harengs  ji  la  haque,  487 ; 
hauNse-col,  361;  macabre,  358,  487; 
orpailleur,  487 ;  penture,  rature,  361 ; 
rouls,  487 ;  vals,  valt,  valent,  485. 

French  compound  words,  484 

etymologies,  EspagnoUe's,  133 

Fttgner's  Lexicon  Livianum,  346 

Galen's  works,  zio 

Stichometry  in,  339 

Callaway's  Use  of  ^i)  with  participle  in 
classical  Greek,  3<S9-7o 

Gabubtt,  Jamss  M.  Review  of  Jusse- 
rand's  Literary  History  of  the  Eng- 
lish People,  343-8 


Review  of  Courthope's  History  of  Eng- 
lish Literatuie,  468-80 
Gascoigne  characterized,  473 
Germanic  Philology.  Journal  of  (rev.),   335-7 
Gerund  and  Gerundive,                           439-53 
GiLDBRSLBBVB,  B.  L.     Revlcw  of  Rai- 
bel's  Sophokles'  Elektra,                    353-6 
Note  on  ov  /m.i}  in  later  Greek,               460-z 
GoooBLL,  Thomas  Dwight.    DSrpfeld 

and  the  Greek  Theatre,  x-i8 

(Goodwin,  Chaki.es  J.    Review  of  Pun- 

tonl's  L'inno  Omerico  a  Demetra,      84-8 
Graeco-Syrian  literature,  zo8,  xia 

Gkandgbmt,  C.  H.  Review  of  Ram- 
beau  and  Patsy's  Chrestomathiefran- 
9aise,  934, 335 

Greece,  Modem,  Language-Standard  in,  19-35 
Greek:  BeAAepo^oKnic,  490;  0AcircJat- 
/M.MV,  195;  ^ov^di'ta,  340,  489;  Acio-i- 
Jai/M,oyia,  Z89 ;  Jixao-xbvtK,  443 ;  ci  m*> 
jia,  346 ;  jriavrdc  ^  anniversary,  367 ; 
cffi  w.  gen.  and  dat.,  ZZ8-19;  0iWoc, 
44X ;  9\m.v  and  0v«rtfat,  367 ;  xdpvSof , 
440;  /u.^  w.  part.,  369-70;  Mil  *»'  c. 
part. ^ opt.,  344;  K«/M.ciy  and  Wmc- 
a0ai,*76-7;  ov, 49  foil.;  ov  11^,453 foil.; 
WAai'oc,  366;  ffpdc,  Z09,  X30. 
Greek,  Archaeologists',  xx8 

Comedy,   Superstitions    and    Popular 

Beliefs  in,  x 89-904 

Historiography,  Ethics  and  Amenities 

of,  855-74 

and  Latin  in  the  XVIth  century,  474 

New  Tesument,  Negative  Futures  in, 

453-6« 
papyrus  MSS,  Catalogue  of,  333 

reflexive  pronoun,  314  foil, 

superstition,  488 

Theatre,  Ddrpfeld  and,  z-18 

theatre,  c{w<rTpa,  488 

GrCmbb,  Gustav.     Review  of  Journal 

of  Germanic  Philology,  335-7 

Godeman's  History  of  Classical  Philol- 
ogy (noticed),  370 

Harmodios  and  Aristogeiton,  xxo 

Harper's  Dictionary  of  Classical  Litera- 
ture and  Antiquities  (rev.),  78-84 
Hand,  Origin  of,                                        43-69 
Heinze*s  Lucretius  (rev.),                        48Z-3 
Hbm DKicKSOM,  G.  L.    Are  the  Letters  of 
Horace  Satires?                                313-34 
Review  of  Heinxe's  Lucretius,            481-3 
Heptanomis  of  Egypt   from  Hadrian's 

time,  zz6 

Hermes,  363-7 

Hermogenes  and  his  commentators,  xo8 

Hervleux's  Fabulistes  latins  (noticed),  36X 
Hillebrandt's  Ritual-Literatur  (rev.),  350-3 
Hill's  Sources  of  Greek  History  (noticed),  494 


Digitized  by  VjOOQ IC 


504 


INDEX, 


Hiitoriography,     Greek,     Ethics     and 

Amenities  of,  955-74 

Homeric  Hymn  to  Demeter,  84-8 

Hoplcins's  Religions  of  India  (rev.),       88-90 

Horace,  Carm.  I  27,  lai 

11 6,  489 

III  84,  31.  3a,  3»9-3i 

IV  3,  17-ao,  335,  397 
Sat.  I  X,  93-6,  331-4 

I  5,  7-  43.  5«.  77-«o,  334-6 

I  9,  XI.  aa,  336-8 

Are  his  letters  satires  T  3i3-«4 

Notes  on,  335-38 

Page,  Palmer  and  Wilkins*s  (noticed),  xax 
Hokton-Smxth,  Lionel.    The  Origin  of 
Latin  hand  and  Greek  ov  and  the 
extensions  of  the   originally   unex- 
tended  form,  43-69 

Concluding  Notes  on  the  Origin  of  the 
Gerund  and  Gerundive,  439-5a 

HvMPHKBTs,  Milton   W.     Report   of 
Revue  de  Philologie,  •a8-33 

India,  Religions  of,  88-9X 

Isocrates,  Order  of  the  orations  of,  in  the 
archetype,  108 

Helena,  490 

Jannaris,  Concise  Dictionary  of  the  Eng- 
lish and  Greek  Languages  (noticed),     ax 
On  ov  fi^.  461 

Jenkins's    L'Espurgatoire   Seint   Patrit 

(noticed),  36a 

Josephus,  Niese's  (noticed),  370 

Journal  of  Germanic  Philology  (rev.),    935-7 
Jutserand's  Literary  History  of  the  Eng- 
lish People  (rev.),  34*-8 

Kaibel's  Sophokles'  Elektrm  (rev.),  353-6 
Kalender  of  Shepherdes,  xoo 

Karsten's  Journal  of  (Germanic  Philology 

(rev.),  995-7 

Keau*  Youth  and  Early  Poems,  936 

Keelhoff,  J.,  on  ti  fii)  &«,  946 

Kenyon's  Bacchylides  (noticed),  493-3 

Kicking  the  bucket,  X9X 

Kirk,  William  Hamilton.    Etiam  in 

Plautus  and  Terence,  96-49 

viiiL*t,v  and  viiitoBoL,  76 

Knapp,  Chaklis.    Notes  on  Horace,  395-38 

Langland,  470 

LanguagC'Standardin  Modem  Greece,  19-95 
Latin  :  aplopodlta  ^  haplopotidem  =:  ar • 
AovoriJa,  489;  -astro-,  71-3;  aorum 
or  avrum?  491;  cuncti<cumque, -ti, 
Z09;  defraudit,  489;  enim,  409-15: 
etenim,  414;  etiam  (eti-jam)  and 
other  compounds  of  -jam,  109;  is 
Plautus  and  Terence,  96-49;  fateor 


and  infiteor,  489 ;  favere  and  fovere, 
489;  Kemelli  =  testiculi,  Z13;  haud, 
43  foil.;  iocundus,  441;  neque  with 
perf.  subj.,  193;  nihil.  469-3;  opiter, 
489;  opter,zx4;  -per  in  parumper  and 
the  like  =vcp,  490;  praestat=spraes 
sut,  489 ;  qu  in  liquldus,  liquor,  999, 
930;  rotundus,  446 :  serenus,  489. 
Latin,  African,  490 

correption  of  trochaic  words,  xxx 

Gerund  and  Gerundive,  439-53 

in  the  XVIth  century,  474 

Latin  rhetorical  terms:  adfictios=/ar#- 
nctHotia,  anticipatio,  vpoxaraAii^ic, 
articulus,  atyntUton,  9o6 ;  attenuatio, 
commiseratio,  conexum  ^  wfivAiMc^, 
contendere,  contentio  s=  antitktsist 
conversio,  a^ttropfu^  907;  conver- 
sum,  avrt9Tpo^^,  defectus,  cAAct^tv, 
denominatio=/arMitfiiMMM,  depre- 
catio,  detractatio  =  8taovpfi^,  908; 
disiunctum,  908, 909 ;  disparsum,  dis- 
solutio,  distributio,  909 ;  divisio  ^  tU- 
Umma,  evacuatio  ^  avoaacvi},  eape- 
dio,  exquIsitio=  j^crcur/ui^,  homoeon, 
9x0;  icon^xtmiVtf,  iniunctumi=v««- 
^wyikivov,  Interrogatum  ss  ipMTirfia, 
intemiptio=rap<»^<o-if,  permutatio 
3=  aKrt/ul«ra^oA^,  axx;  praeceptloss 
vpoxaraAii^if,  pronuntiatio=v«4jqM* 
o-t«,  regressio  =«-«AiAA«yui,  91a;  so- 
lutumss  asyndeton,  traductio,  tran- 
situs,  913. 
Latin  subjunctive  in  Independent  Clauses 

in  Plautus,  133-67,  ayS-sox,  389-401 

Leges  Anglorum,  99 

Linguistic  development.  Order  of,  934 

Longinus  (Ps.),  vcpl  v^ovc  emended,  489 

Lope  de  Vega,  Shakespeare  and,  104 

Lounsbury's  Chaucer,  xo6 

Lucian  and  Petronius,  1x4 

Nigrinus,  339-4^ 

Lucretitis,  Heinse's  (rev.),  481-3 

Lycanthropy,  488 

Lyly,  Importance  of,  476 

Lyndsay,  Sir  David,  characteriised,  474 

Macabre,  Danse  (Dance  MacabrOr     358,  487 
Marlowe  characterized,  478.  480 

Marston  as  a  dramatist,  105-6,  333-^4 

Merlin,  M.E.  Prose  Romance  of,  105 

Milittry  Colonies  of  Ocuvian  and  Caesar 

in  Gallia  Narboneasis,  36a 

MiLLBlt,  C.  W.  E.  Review  of  Dyroff's 
Geschichte  des  Pronomen  Reflexi- 
vum,  3x4-94 

Mitsodakis,  Praktische  Grammatik  der 
neugriechischen  Schriftsprache  (no- 
ticed), 9X 
Montgomerie,  Alexander  (b.  1545),           96-S 


Digitized  by  VjOOQ IC 


INDEX. 


505 


s,  B.  P.    SutguBCtive  in  Indepen- 
dent Sentences  in  Plautus, 

«33-«7f  a75-30i,  l^ysox 
Mottdlaria,  Textual  notes  and  queries, 

Z68-88 

MVSTARO,  WiLFRKD  P.    Report  of  Rhei- 

nisches  Museum,     zo8-i6,  •99-41,  488-91 

Review  of  Cicotti's  Processo  di  Verre,  93-4 

Notice  of  Bomecque's  Ciceronis  Oratio 

in  Verrem  Dt  Signit,  95 

Naiad,  Use  of,  75 

Name,  Power  of  the,  194,  295 

Change  of  name.  Superstitious,  489 

Necrology : 

AUen,  F.  D.,  247,  379-5 

Lane,  G.  M.,  247,  370-a 

Montaigion,  A.  de,  488 

Sundby,  Thor.,  359 

Vemer,  Karl,  91-3 

Zupitza,  Julius,  938 

Nereids  as  evil  demons,  ^9*~A 

Nero  and  the  Rhodians,  S31 

Niese's  Josephus  (noticed),  370 

Nigrinus,  Lucian's,  399~4> 

Nikokles  6  xttfop^tfr,  940 

Nonius,  p.  63  M,  aa8 

Nymph>names,  On  the  alleged  confusion 

of.  74-5 

Obsequens,  Julius,  ^39 
Obrtu.,  Hanks.     On  the  Character  of 

Inferred  Parent  Languages,  416-38 
Ophelia,  Irish,  not  Greek,  238 
Optative  and  Potential,  383  foil. 
Osgood,  Chaklbs  Grosvbnob,  Jk.  Re- 
port of  Englische  Studien,  s33-8 

Parent  Languages,  Inferred,  4x6-38 

Passy   and    Rambeau's    Chrestomathie 

fran^ise  (rev.),  294,  395 

Pbrrin,  B.    The  Ethics  and  Amenities 

of  Greek  Historiography,  >55'-74 

Petronius,  240 

and  Lucian,  114 

Phaedrus,  I  15.  z-9,  998 

Append.  z6,  7,  998 

IV  9, 9,  939 

Philo  and  Clement  of  Alexandria,  366 

Phonetic  laws  violated,  485 

Pindar  and  Women,  194 

Pisistratus,  Chronology  of,  215 

Piso  and  the  A.  U.  C,  365 

Pistqjese  dialect,  486 

Plagiarism  in  Greek  literature,  955  foil. 

Plautus,  Amphitr.  96,  930 

Trin.  540,  930 

The  subjunctive  in  independent  sen- 
tences in  Plautus,  133-^,  >75-9oi»  383^402 
Textual  notes  and  queries  in  Plautus,  168-88 


and  Terence,  Enim  in,  409-25 

Etiam  in,  96-^49 

Plutarch,  Scipio  Nmsica  as  a  source  of,      363 

vcpl  Aopyi|<riaf ,  Source  of,  365 

Polybios,  III  94,  224 

Notes  on,  366-7 

PosTGATB,  J.  p.    On  the  Alleged  Con- 
fusion of  Nymph-names,  74,  75 
Potential  and  Optative,                       383  foil. 
Propertius,  II 13  b,  1x4 

11x3.48.  X15 

Studies  in,  364,  365-6 

Provencal :  c  and  g  followed  by  «,  486 

compound  words,  484 

Pyrrhus,  War  of,  366 

Qnomodc  in  the  Romance  languages,        485 

Rambeau    and    Passy's   Chrestomathie 

fran^ise  (rev.),  994,  995 

Recent  Publications, 

295-9,  a4«-5a,  37^-8o.  495-9 

Reisch  on  the  Greek  Theatre,  5  foil. 

Renaissance  in  England,  469 

Reportt : 

Beitriige  xur  Assyriologie,  226-28 

Englische  Studien,  96-208,  933-8 

Hermes,  369-7 

Revue  de  Philologie,  998-42 

Rheinisches  Museum  far  Philologie, 

Z08-16,  939-42,  488-92 
Romania,  357-^*  484-* 

Reviews : 
Courthope's  History  of  English  Poetry, 

464-80 
Dahlerup's  Nekrolog  5ver  Karl  Vemer, 

91-3 
Heinxe's  Third  Book  of  Lucretius,  482-3 
Hillebrandt's  Ritual-Litermtur,  350-3 

Jusserand's    Literary  History  of  the 

English  People,  349-9 

Kaibel's  Sopholdes'  Elektra,  353-6 

Peck's  Harper's  Dictionary  of  Classical 

Literature  and  Antiquities,  78-84 

Puntoni's  Inno  Omerico  a  Demetra,      84-8 
Revue  de  Philologie,  998-33 

Rheinisches  Museum  fur  Philologie, 

208-26.  939-42,  488-92 
Rhetorical  Terms,  Definition  of  Some,  906-13 
Rhodians,  Nero  and  the,  932 

Ries,  Was  ist  Syntax  t  206 

RxBSS,  Ernst.    Superstitions  and  Popu- 
lar Beliefs  in  Greek  Comedy,         289-905 
Rizo-Rangab^'s    Practical    Method    in 

Modem  Greek  (noticed),  90 

RoBBRTS,  W.  Rhts.    Caecilius  of  Ca- 

lacte,  301-19 

Roman  colonial  administration,  363-4 

Romance:    OM0m0d0  in  Romance  lan- 
guages, 485 


— Digitized  by  VjOOQ IC 


So6 


INDEX, 


Romania,  357-6»f  484-8 

Romaunt  of  the  Rose,  104 

Rome,  founding  of.  Date  of,  365 

Rutherford's  Scholia  Aristophanica  (no* 
ticed),  345, 946 

Sackville's  Induction,  474 

Sandys'  ed.  of  First  Philippic  and  the 

Olynthiacs  (noticed),  345 

Scenery  of  Ajaz,  Eirene  and  Prometheus,  367 
Schneide win's   Antike    Humanitat   (no- 
ticed), 343 
Scipio  Nasica  as  a  source  of  Plutarch,      363 
Seneca's  Tragicus,  Choruses  in,                  490 
Shakespeare  and  the  Aubade,  99 
-Bacon  question,                                         107 
and  Lope  de  Vega,                                     X04 
Macbeth  and  Kyd's  Spanish  Tragedy,  337 
and  the  Pilgrimage  to  (Janossa,        xo3, 103 
Taming  of  the  Shrew,                        100,  xox 
Ships  in  antiquity,                                          364 
Sidney,  Sir  Philip,  characterized,               476 
Silence  as  to  sources  in  Greek  historiog- 
raphy,                                                     361 
Smitix,  Emily  Jambs.    Note  on  Lucian's 

Nigrinus,  339-4* 

Solinus  (Pseudo*),  ~  xis 

Soul  of  man  seated  in  the  air,  488 

Spanish:  yogar,  487 

Spenser  characterized,  477 

Statius,  Silrae,  Praef.  38.  339 

IV  X3, 19,  X16 

Thebais  and  the  Codex  Puteanus,  xoS 

Stedman's  Modem  Greek  Mastery  (no- 
ticed), .  30 
Stichometry  in  Galen,                                 339 
Strabo,  Cyriacus  of  Ancona  on,                   114 
Stratton,  a.  W.    Review  of  Hopkins's 

Religions  of  India,  84-7 

Subjunctive  in  Independent  Sentences  in 

Plautus,  «33-67#  a7S-3oi,  383-401 

Superstitions    and    Popular    Beliefs    in 

Greek  Comedy,  189-304 

Surrey  characterized,  473 

Tacitus,  Consulship  of,  x  X4 

Taine's  History  of  English  Literature,  465 
Telmessians,  Superstitious  practices  of,  193 
Terence,  Eum.  588,  333 

Enim  in,  40>-z5 


Eiiam  in,  36-43 

Tts*9raho*pitalUt  1x4 

Theatre,  Attic,  367 

Greek,  e^orpa,  488 

Three-legged  tables,  Why?  193 

Thucydides,  II 15,  X09,  xxa,  iso 

VI8x,5,  "44 

Marchant's  Book  VI  of,  344 

Steup's  Classen's  (noticed),  xaa 

Thumb's  Handbuch  der  neugriechischen 

Volkssprache  (noticed),  19 

TooD,  H.  A.     Report  of  Romania, 

357-«3,  483-8 
Tolman's  Taming  of  a  Shrew,  aoo,  xox 

Tracutus  de  diversis  historiis  Romano- 
rum,  98 
Triumvirate,  Second,                                    363 
Trochaic  words  in  Latin,  Shortening  of,    xxx 
Tyrrell's  Euripides'  Troadcs  (noticed),     345 


Vergil's  Eighth  Eclogue, 

x»5 

Fourth  Eclogue, 

"5 

Aen,  IV  4x. 

a39 

VI  58. 

231 

VI  616-30. 

"39 

Verner,  Necrology  of. 

9«-3 

Verrine  Orationea,  Cicero's, 

93-S 

Vestal  Virgins,  Habit  of. 

XXI 

Vocal  allusions. 

485 

Warton's  History  of  English  Poetry,         464 

Weil's  Etudes  sur  le  drame  antique  (no- 
ticed), 343,  344 

Whkblbx,  Benjamin  Ids.  The  Ques- 
tion of  Language-Standard  in  Modem 
Greece,  >9-«5 

Wied's  Praktischet  Lehrbuch  der  neu- 
griechischen Volkssprache  (noticed),    30 

Williams'  Questions  of  Good  English 
(noticed),  134 

Women  in  Pindar,  X34 

Wyatt  characterized,  473 


Xenophon,  Memorabilia,  Ub.  IV, 

IXO 

Cynegeticus, 

a39 

obelized, 

"5 

Zielinski's  Cicero  im  Wandel  der  Jahr- 
hunderte  (noticed),  S4S-3 


Digitized  by  VjOOQ IC 


Vol.  XVIII,  4  '  Whole  No.  72 


THE 


AMERICAN 


Journal  of  Philology 


EDITED  BV 

BASIL   L.   GILDERSLEEVE 

Professor  op  Gbbbk  in  thk  Johns  Hopkins  Univbrsity 


BALTIMORE:    THE  EDITOR 

New  York  and  London:   Macmillan  &  Co. 

Leipsic  :    F,  A.  Brockhaus 


December,  1897 


Digitized  by  VjOOQ IC 


CONTENTS. 

I. — The  Subjunctive  In  Independent  Sentences  in  Plautus.     III.     By 

E.  P.  Morris, ' 3S3 

H. — The  Use  oi enim  in  Plautus  and  Terence.    By  Willard  K.  Clkment,  402 
III. — On    the   Character  of    Inferred    Parent'  Languages.      By   Hanns 

Oertel,  . 416 

IV. — Concluding  Notes  on  the  Origin  of  the  Gerund  and  Gerundive.     By 

Lionel  Horton-Smith ■      .        .        .  439 

V^. — Negative  Futures  in  the  Greek  New  Testament.     By  W.  G.  Bal- 

LANTINE, 453 

Note  :  .        . ' 462 

On  Latin  nihil  *  naught,  not.'     By  Edvvln  W.  Fay. 
Reviews  AND  Book  Notices: 464 

Courtliope's  A  History  of  English  Poetry. — Heinze's  T.  Lucretius  Carus 
de  Rerum  Natura. 
Reports: 484 

Romania. — Rheinisches  Museum  ffir  f*hilologie. 

Brief  Mention,  492 

Recent  Publications, 495" 

Books  Received, 50c 

Index,  502 


Open  to  original  communications  in  all  departments  of  philology,  classical, 
comparative,  oriental,  modern  ;  condensed  reports  of  current  philological  work  ; 
summaries  of  chief  articles  in  the  leading  philological  journals  of  Europe  ; 
reviews  by  specialists ;  bibliographical  lists.'  Four  numbers  constitute  a  vol- 
ume, one  volume  each  year.  Subscription  j^rice  $3.00  a  year,  payable  to  the 
editor  in  advance;  single  numbers,  $1.00  each.  Suitable  advertisements  will 
be  inserted  at  the  following  rates: 


One  page 

Halt  page . 

Quarter  page., 
Eighth  paL;e. . 


I  TIME. 

2  T. 

3T. 

4  T. 

$^6   00 

#30  00 

;C40  00 

J50  C» 

8  00 

15  00 

20  CO 

25  c» 

4  00 

8  00 

12  00 

15  oc 

2  00 

4  00 

6  00 

b  00 

Bog-iiiiiiiig-  with  tlio  iK'xt  iiiiinbor  (78),  the  Ag:oiits  of  the  Journal 
ill  England  will  be  31exsrs.  Kegau  Paul,  Trench,  Triibner  k  Co., 
Paternoster  Honse,  Charing-  Cross  IJoad,  London,  W.  €'• 

Special  Notice. — A  few  sets  of  The  Amkkican  Journal  of  Philology 
have  i)een  made  up  of  late  by  purchase  and  excliaiigf.  and  will  be  sold  at  the 
regular  price,  ^54  for  the  eiLjliteen  volumes,  cash  to  accompany  the  ordei.  Of 
single  volumes  ($3)  Volumes  IV  aud  VI  cannot  be  furn'shed.  Of  single 
numijcrs  ($l)  13,  15,  21,  22,  23,  49,  50.  52  are  waiitiuL,'.  No.  40,  wliich  contains 
the  Index  to  Vols.  I-X,  is  not  furni>hcd  sepMalcIy,  I'Ul  a  small  edKion  of  the 
Index  itself  is  for  sale  at  $1.00  a  copy.     Address, 

H.  I..  (»lLDhKsLEF,VE,  li.ilt imore.  Md. 

I'uMished  (tu.irtcrly.  Ihree  dollars  a  year  ux-'^'taee  piici).  Kntcred  at  ihc  Po-«l(>fficc  of 
Bahiniore,  Md.,  as  sccund-cla^s  matter. 


Press  of    I  he    Fric.lcI.^\  rdd  Co. 
iI.«lttnior''.  Mil 


Digitized  by  VjOOQ IC 


Digitized  by  VjOOQ IC 


Digitized  by  VjOOQ IC 


"^       UNIVERSrrYOFMICHIQAN 

lllMlillili 

3  901506042  6006 


Replaced  with  Commercial  Microfbmi 
1993 


^ 


?:  •>  '  iT^^V  ^  ■<*  **"<> 


(*>.*••, 


^      f 


GO 


*#>€1AIJ 


r 


t 


ii\ 


Digitized  by  VjOOQ 


Digitized  by  VjOOQ IC