Skip to main content

Full text of "Studies of the Gods in Greece at certain sanctuaries recently excavated : being eight lectures given in 1890 at the Lowell Institute"

See other formats


aj) 


THE     GODS     IN     GREECE 


•AHY. 

STUDIES  OF  THE  GODS  IN  GREECE 

AT  CERTAIN   SANCTUARIES 

RECENTLY  EXCAVATED 

BEING    EIGHT    LECTURES    GIVEN    IN 
1890    AT    THE   LOWELL    INSTITUTE 

BY    LOUIS    DYER,    B.A.    OXON. 

LATE    ASSISTANT    PROFESSOR    IN    HARVARD   UNIVERSITY 


Better  stand  upon  the  fragments  of  antiquity  and  look  about  ns. 

\\.  S.  LANUOR. 


MACMILLAN   AND    CO. 

AND    NEW    YORK 
I  891 

All  rights  reserved 


COPYRIGHT 
1891 

BY 

LOUIS      DYER 


TO 

DEMETRIOS  BIKELAS 


PLANS 

PAGE 

Plan  of  the  Eleusinian  precinct  from  Proceedings  of  R.  I.  B.  A. 

to  face  page  198 

Outline-sketch  of  the  ground-plan  of  the  Hall  of  Ictinus       .         .  202 

Plan  of  the  temple  at  Old  Paphos  from  the  J.  H.  S.       to  face  page  310 


CONTENTS 


INTRODUCTION 

Latter-day  paganism — Greek  sanctuaries — Powers  of  Greek  gods — 
Roman  organisation— The  kingdom  of  Earth— The  Roman  order 
— Greek  individualism — Greek  religious  thinkers — Anomalies  of 
Greek  religion — Apollo  at  Delphi— Parnassus — Pater,  Mannhardt, 
and  Preller — Sanctions  of  Apollo's  power — Apollo's  temple  at 
Delphi — Delphian  worship — The  coming  of  Apollo  .  Pages  1-36 

APPENDIX  I — The  deification  of  Roman  emperors  .          .  37'45 


II 
DEMETER    AT    ELEUSIS    AND    CN1DUS 

The  home  goddess  of  grain — Two  Homeric  otherworlds — The  goddess 
of  enough  and  to  spare — The  hallowed  fruits  of  Demeter — The 
Homeric  Hymn  to  Demeter — The  wedlock  of  Persephone — The 
Iliad  and  the  Odyssean  stories — The  Eleusinian  legend  of  Demeter 
— Interlopers  in  the  story — Celeus  and  Demophoon — Woman's 
love  for  woman — Demeter's  love  for  Persephone — Demeter's  rules 
for  right  living — Demeter  of  Cnidus  ....  46-74 


viii  CONTENTS 

III 
DIONYSUS    IN    THRACE    AND    OLD    ATTICA 

The  crazed  god  of  Homer— Contradictions  of  Dionysus— Probations  of 
Athenian  Dionysus— Thracian  birth-marks  of  Dionysus— Asiatic 
Thracians— Brumalia  and  Rosalia— Modern  May-day  festivals— 
The  angel  of  the  darker  drink— Elements  of  wine,  fire,  water,  gold 
— Dionysus- Silenus  the  water  man— Bassarids,  Satyrs,  Corybantes, 
Curetes— The  savage  Pan  or  Aegipan— Dionysus  Omestes — Dionysus 
and  the  Muses— The  Icarian  legend — Icarius  entertains  the  god — 
The  shepherds  slay  Icarius— Legends  parallel  to  Icarian  story — 
Winter  festivals  on  Parnassus— Susarion  and  Thespis  at  Icaria — 
First  tragedies  at  Icaria Pages  75-117 

APPENDIX  II — Dionysus  Eleuthereus         ....        118-120 

IV 

DIONYSUS    AT    ATHENS 

Epimenides  of  Crete — The  Pisistratidae  and  Dionysus— Onomacritus  and 
Orphic  myth-making— Epimenides,  Pisistratus,  and  Onomacritus — 
Reformed  Athenian  festivals— The  people's  faith  in  Dionysus — The 
greater  Dionysia — The  Bacchanals  of  Euripides — Messianic  vision 
of  Euripides— Analysis  of  the  introduction — Entrance-song  of  the 
Bacchanals— Central  acts  and  conclusion— First  great  act— The  pro 
bation  of  Pentheus — Pentheus  is  deaf  to  reason — The  second  great  act 
— The  sin  of  Pentheus — Captive  good  attending  captain  ill — The 
perdition  of  Pentheus— Cadmus  cures  Agave's  madness  121-163 

APPENDIX  III— Second  birth  of  Dionysus— his  eastern  affinities    163-173 

V 

THE    GODS    AT    ELEUSIS 

The  coming  of  Dionysus  to  Eleusis — The  all -welcoming  Demeter — 
Xenophanes  and  Heraclitus — The  holy  silence  of  Eleusis — Dionysus 
an  enhancement  of  Demeter— The  pitiless  huntsman  Zagreus— 


CONTENTS  ix 

Description  of  Eleusis — The  only  church  of  antiquity — Eleusis 
fortified  in  early  days — The  present  condition  of  Eleusis — The 
Plutonian  precinct— The  Hall  of  Initiation— Halls  of  Pisistratus, 
Cimon,  Ictinus  —  Cyclopean  traces  —  Roman  restorations  —  The 
Hall  of  Ictinus  —  The  portion  of  Hades  —  Greater  and  Lesser 
Mysteries — The  Greater  Mysteries— Yearly  procession  of  the  Mystae 
— The  Frogs  of  Aristophanes— The  measure  led  by  the  Fates — 
Stations  of  the  yearly  pilgrimage  .  .  .  .  Pages  174-218 

VI 

AESCULAPIUS    AT    EPIDAURUS    AND    ATHENS 

The  god  Aesculapius  from  the  north — Hippocrates  and  Democedes — 
The  Coan  sanctuary — The  medicine  of  Homeric  days — The  debt 
of  Democedes  to  Homer — The  schooling  of  Aesculapius — Early 
history  of  Greek  medicine — Mind  and  body — Harmony  of  religion 
and  science — Aesculapius  a  friend  of  man — The  son  of  Apollo — 
The  Hieron  of  Epidaurus — Finding  of  the  infant  god — A  precinct 
of  Aesculapius — The  statue  of  ivory  and  gold — The  Tholos  of 
Polycletus — The  painting  of  Pausias — Apollonius  the  intercessor 

219-256 

APPENDIX  IV— Apollonius  of  Tyana          ....       257-266 
APPENDIX  V — The  status  of  modern  Greek  doctors    .          .       267-269 

VII 
APHRODITE    AT    PAPHOS 

Aphrodite  at  Greek  sanctuaries  -—  Eastern  characteristics  —  Three 
regions  of  Cyprus  —  Meeting  ground  of  East  and  West  —  No 
Cypriote  nationality — Phoenicians  in  Cyprus — Lack  of  Greek 
monuments  —  Agapenor  at  Paphos  and  in  Arcadia  —  Aphrodite 
always  Paphian — From  Limassol  to  Paphos — Cinyras  of  Paphos — 
The  ubiquity  of  Cinyras — The  temple  on  the  hill — Aphrodite's 
birth  at  Paphos — The  Greek  Aphrodite — The  Roman  and  Assyrian 
goddess— Aphrodite  and  Demeter  ,  270-304 


x  CONTENTS 

APPENDIX  VI— The  temple  at  Old  Paphos         .         .     Pages  305-314 
APPENDIX  VII— Aphrodite    of    the   Greeks,    Hittites,    and 

Phoenicians 3:5-323 

APPENDIX  VIII— The  Olympus  and  the  Bocarus  in  Cyprus 
— Hettore  Podocatharo  and  John 
Meursius 324'354 


VIII 
APOLLO    AT    DELOS 

Delian  nativity  of  Apollo  and  Artemis — Delos  a  wandering  island — 
The  consecration  of  Rhenea — The  exile  of  the  Delians — Archi- 
lochus  the  poet  of  the  Aegean — Apollo  a  King  Arthur  of  the  Greeks 
— Apollo  and  Cyrene — Self-discipline  of  Apollo — The  lonians  at 
Delos — Vicissitudes  of  Delian  festivals — From  Athens  to  Delos — 
The  procession  of  Nicias — The  earliest  Delian  festival — The  temple 
and  image  of  Leto — The  Cynthian  cave -temple  —  Latter-day 
worshippers  at  Delos — The  inheritance  of  Delos  .  .  355 -390 

APPENDIX  IX— The  Cyclades  and  Sporades       .          .         .  391-398 
APPENDIX  X— The  worship  of  Aphrodite  and  of  strange 

gods  at  Delos 399-4°3 

APPENDIX  XI — Photographs  referred  to  for  illustrations      .  404-413 

INDICES 414-457 


PREFACE 

ADEQUATELY  to  thank  all  whose  help  has  been 
lavished  upon  tJie  preparation  of  these  pages  is  not 
possible.  Although  I  can  give  no  catalogue  of  the 
names  of  benefactors,  my  gratitude  is  sincere ;  and 
this  expression  of  it  will,  I  Jiope,  reach  them  in  Greece, 
in  France,  in  Italy,  in  England,  and  in  America.  It 
is  but  fair,  however,  to  say  that  constant  criticism  and 
suggestion  from  my  wife  helped  the  present  work  to 
its  shape,  and  I  cannot  silence  the  particular  expres 
sion  of  my  thanks  to  Professor  Middleton  of  King's 
College,  Cambridge,  and  Professor  Ker  of  University 
College,  London,  for  invaluable  aid  given  most  un 
grudgingly  during  the  final  revision. 

Originally  prepared    as    lectures  for    the    Lowell 
Institute,    the   eight  chapters   here  given   are  printed 


xii  PREFACE 

with  corrections  and  notes,  the  fruit  of  a  years 
deliberation.  As  lectures  they  were  repeated  before 
various  Universities,  Colleges,  and  Societies  in  various 
parts  of  the  United  States.  A  lecture  on  the  Cyclades, 
given  before  Columbia  University,  forms  the  basis  of 
one  of  the  Appendices  which,  although  some  of  them 
are  unavoidably  technical,  have  seemed  necessary  to  the 
more  or  less  connected  presentation  of  Greek  religious 

thought  here  attempted. 

LOUIS  DYER 


SUNBURY  LODGE, 

OXFORD,  April  1891. 


I 

INTRODUCTION 

I  DO  not  mean  to  attempt  an  account  of  all  the  Greek 
gods  ;  eight  studies  would  be  insufficient,  even  if  my  investi 
gations  had  already  carried  me  over  the  whole  ground, 
which  is  a  vast  one.  I  propose  in  the  first  place  to  say 
what  I  can  of  Demeter  and  Persephone,  the  two  great 
goddesses  of  Eleusis  in  Attica.  Here  it  will  be  necessary  to 
consider  excavations  in  Asia  Minor  made  some  years  ago 
by  Sir  Charles  Newton,1  as  well  as  the  now  practically 
completed  diggings  of  the  Greek  Archaeological  Society  at 
Eleusis.2  In  the  second  place,  the  god  Dionysus — also 
worshipped  at  Eleusis — will  be  considered,  and  his  early 
cult  in  Attica  will  receive  illustration  from  recent  excava 
tions  made  by  the  American  School  of  Athens.3  The 

1  A  History  of  Discoveries  at  Halicarnassus,  Cnidus,  Branchidae.     By 
C.  T.  N.,  assisted  by  R.  P.  Pullan.     London,  1862,  folio  and  8vo. 

2  See  the  various  publications  of  the  Greek  Archaeological   Society, 
particularly  the  plan,  by  Dr.  Dorpfeld,  of  the  Eleusinian  Temple  in  the 
Praktika  of  1885,  and  ibid.  Dr.  Philios's  description,  as  well  as  his  subse 
quent  reports  on  inscriptions  and  discoveries  at  Eleusis.     Dr.  Dorpfeld  has 
kindly  allowed  a  reproduction  of  his  plan  to  appear  below  in  the  chapter 
on  "The  Gods  at  Eleusis,"  p.  202. 

3  See  the  Seventh  Annual  Report  of  the  American  School  at  Athens  in 
the  American  Journal  of  Archaeology  for  1889,  and  cf.  the  Nation  of 
22d  March  1888. 

B 


I 


2  INTRODUCTION  i 

third  topic  will  be  Aesculapius  and  his  worship,  more  par 
ticularly  at  Athens  and  at  Epidaurus,  as  known  through 
excavations  in  both  places.1  Fourthly,  a  consideration  of 
Aphrodite  and  her  worship  at  Old  Paphos  will  occupy  the 
seventh  of  my  chapters,  which  will  to  a  large  extent  be 
devoted  to  the  problems  which  have  been  raised  by  the 
recent  excavations  of  the  British  School  at  Couclia  in 
Cyprus.2  My  eighth  and  last  chapter  will  be  given  to  the 
holy  island  of  Delos,  and  to  Delian  Apollo.  The  French 
School  of  Athens,  chiefly  under  the  able  direction  of 
M.  Homolle,  has  uncovered  and  discovered  of  late  years 
all  manner  of  facts  about  Delos  and  Delian  Apollo;3  and 
with  these  I  shall  end  my  studies  of  five  of  the  greater 
gods  of  Greece  as  worshipped  in  their  recently  discovered 
sanctuaries. 

As  there  are  shrines  of  healing  and  sanctuaries  of  especial 
salvation  dedicated  to  immemorial  worship  by  the  medieval 
world  of  Christendom,  so  also  in  the  Hellenic  world  (much 
larger,  alas !  than  modern  Greece)  there  were  places  about 
which  lingered  through  many  centuries  a  dread  and  most 
religious  sanctity,  a  helpful  significance.  Of  such  spots 
in  Greece,  in  Cyprus,  in  Asia  Minor,  and  of  the  gods 
whose  presence  and  whose  help  was  sought  in  those 
holy  places,  I  am  to  speak.  I  am  to  speak  of  several 
sites  lately  investigated  where  the  beautiful  and  enno 
bling  religion,  first  of  Greece,  and  then— through  Greece 

1  Paul  Girard,  L' AscUpieion  d'Athenes,  d' aprh  de  recentes  dtcouvertes , 
Paris,  1881.      For  discoveries  at  Epidaurus  see  in  the  publications  of  the 
Greek  Archaeological  Society  various  reports  by  M.  Kabbadias. 

2  See  the  Journal  of  Hellenic  Studies  for  1889. 

3  See   Professor  Jebb's  article    "Delos"   in  the  first  number   of  the 
Journal  of  Hellenic  Studies  •  also  M.  J.  Albert  Lebegue's  Recherches  sur 
Delos  t  Paris,   1876,   8vo,  and  the  publications  of  the  French  School  at 
Athens,  particularly  reports  of  Delian  excavations  by  M.  Homolle  in  the 
Bulletin  de  Correspondance  HelUnique. 


i  LATTER-DAY  PAGANISM  3 

and  Rome— of  all  the  ancient  world,  had  its  growth;  of 
sanctuaries  where  that  old-time  worship  of  ideals,  by  some 
miscalled  idolatry,  grew  pure  and  yet  more  pure,  broad 
and  broader  still,  until  its  inner  significance  and  truth 
were  no  longer  to  be  confined  within  old  forms,  could  be 
fettered  no  longer  by  old  bonds ;  and  lo !  Christianity 
was  there  to  gather  in  a  heritage  of  high-born  thoughts  from 
Greece. 

In  the  latter  days  of  paganism  there  was  a  two-fold 
process  by  which  the  world  was  prepared  for  the  dawning 
of  an  endless  hope,  the  "dayspring  from  on  high."  There 
was  amelioration  and  purification,  as  well  as  a  growing 
superstition  and  gradual  decay.  No  less  distinguished 
an  authority  than  Professor  Jebb1  has  said  of  latter-day 
Greek  religion : — 

"  The  Greeks  were  a  people  peculiarly  sensitive  to  every 
thing  that  was  in  the  intellectual  air  of  the  time,  and  there 
was.  much  in  it  that  helped  Christianity  .  .  .  there  was  a 
tendency  to  take  refuge  from  polytheism  in  deism ;  and  in 
particular  there  was  a  spreading  belief,  half-mystic,  in  the 
resurrection  of  the  body, — a  belief  which  drew  many 
votaries  to  the  worship  of  the  Egyptian  Serapis,  and  was  in 
turn  strengthened  by  that  cult." 

It  has  unfortunately  been  habitual,  but  less  so  in  these 
latter  days  of  religious  tolerance,  to  accept  without  question 
the  estimate  of  paganism  made  in  the  heat  of  conflict  by 
the  early  fathers  of  the  Christian  Church.  What  those 
great  and  good  men  strove  to  do  has  been  in  the  fullest 
measure  accomplished.  Christianity  has  prevailed,  and  the 
tanglewood  of  ancient  mythology,  the  thickets  of  ancient 

1  Modern  .Greece,  two  lectures  delivered  before  the  Philosophical  Institu 
tion  of  Edinburgh  by  R.  C.  Jebb. 


4  INTRODUCTION  l 

ritual,  have  receded  from  the  broadening  pathway  of  our 
race.  But  yet  the  distant  view  of  it  remains,  its  influence 
is  real  to-day  though  more  remote.  Indeed  the  purer 
aspects  of  Greek  ritual  and  Greek  mythology  have  a 
counterpart  in  the  most  holy  Christian  places.  Surely 
there  is  no  lack  of  real  Christian  piety  in  feeling,  as  it  were, 
a  reminiscence  or  a  glorified  survival  of  the  ancient  worship 
of  Dionysus  and  Demeter  at  the  altar  where  the  bread  and 
wine  are  given.  It  was  no  fanciful  parallel  which  the 
Christian  author  of  Christus  Pattens  drew1  between  the 
yearly  passion  and  yearly  resurrection  of  Dionysus  in 
ancient  ritual  and  that  passion  and  resurrection  which 
Christians  yearly  celebrate  to-day. 

Consider  how  much— or  shall  it  rather  be  said  how 
little  ?_the  ideal  equality  of  all  sorts  and  conditions  of  men 
in  the  presence  of  God  is  to-day  maintained  by  Christian 
ritual  and  regulation.  And  consider  then  what  new  impulse 
might  perhaps  be  gained  from  a  careful  study  of  the  worship 
of  Aesculapius,  of  Dionysus,  or  from  a  reverential  under 
standing  of  the  Eleusinian  mysteries  of  Demeter.  In 
spring,  at  the  high  festival  of  Dionysus'  birthday,  one  of 
the  marked  features  in  the  celebration  was  a  free  wel 
come,  extended  more  than  in  words,  to  slaves  and  day- 
labourers.2  First,  sacrifices  were  made  with  solemnity, 

1  The  parallel  is  drawn  by  unmistakable  implication,  since  lines  from 
the   Bacchanals  of  Euripides  are  freely  applied  to  the  suffering  Christ, 
line  1344  of  the  Bacchanals,  e.g.,  "Dionysus,   we  are  thy  suppliants,  we 
have  sinned,"  becomes  "  O  Redeemer!  we  are,"  etc.      The  authorship  of 
Christ  Sii/ering  was  long  bestowed  on    Saint  Gregory  of  Nazianzum, 
surnamed  the  Theologian.     It  is  now  given  to  some  unknown  person 
belonging  to  a  later  day  of  the  Church,  when  the  worship  of  the  Virgin 
Mary  had  taken  the  more  definite  shape  implied  by  various  passages  in 
the  poem. 

2  The  rural  origin  of  the  festival  was  especially  marked  by  this  temporary 
obliteration  of  differences  in  station.    See  in  Hutchinson's  Northumberland, 


I  GREEK  SANCTUARIES  5 

and  then  all  alike  were  invited  to  come  to  the  banquet  of 
the  god,  and  partake  of  his  freshly  opened  wine.  So  also 
it  was  with  the  bread  of  Demeter.  All  manner  of  people 
were  free  to  come,  and  be  initiated  at  her  Eleusinian 
sanctuary,  excepting  only  those  polluted  in  some  incurable 
manner,  as  who  should  say,  those  who  had  committed  the 
sin  against  the  Holy  Ghost. 

Christianity  as  we  know  it,  Christianity  as  we  prize  it,  is 
not  solely  and  exclusively  a  gift  from  Israel.  It  is  time  to 
open  our  eyes  and  see  the  facts  new  and  old  that  stare  us 
in  the  face,  growing  more  clear  the  more  investigations  and 
excavations  on  Greek  soil  proceed.  To  the  religion  of 
Greece  and  Rome,  to  the  Eleusinian  mysteries,  to  the 
worship  of  Aesculapius  and  Apollo,  to  the  adoration  of 
Aphrodite  is  due  more  of  the  fulness  and  comforting 
power  of  the  Church  to-day  than  many  of  her  leaders  have 
as  yet  been  willing  to  allow. 

The  sanctuaries  of  Demeter  at  Eleusis  and  Cnidus,  the 
Icarian  demesne  of  Dionysus,  whither  he  came  to  meet  his 
earliest  Attic  worshippers ;  the  Delian  shrine  of  Apollo,  and 
the  temples  of  his  son,  the  healer  Aesculapius,  at  Athens 
and  at  Epidaurus, — these  and  the  Paphian  precinct  of 
sweetly  smiling  Aphrodite  should  be  well  known  to  him 
who  seeks  understanding  of  that  beautiful  religion  whose 
fifth  essence  and  nobler  quality  has  passed  into  our  own. 
Indeed  it  is  universally  true  that  to  understand  any  religion 
you  must  in  some  sense  come  under  the  spell  of  its 
sanctuaries,  in  some  way  you  must  visit  its  holy  places. 

vol.  ii.  end,  the  account  of  "  Mell  Supper":  "The  servants  having  per 
formed  the  most  valuable  part  of  their  labour,  are  entertained  by  their 
masters,  when  all  distinction  is  laid  aside.  This  feast  is  called  the  Mell 
Supper,  at  which  there  are  dancing,  masquing,  and  disguising,  and  all  other 
kinds  of  rural  mirth." 


6  INTRODUCTION  I 

This  has  been  most  beautifully  set  forth  in  the  most  recent 
work l  of  a  writer  whose  every  utterance  on  the  history  of 
beliefs  brings  the  greatest  help.  Speaking  of  the  "  Creed 
of  Heathen  Germany/'  and  of  gods  very  different  from  those 
of  Greece,  but  resembling  them  in  that  they  too  had  a  part 
in  the  shaping  of  Christianity,  Mr.  Keary  most  poetically 
and  truly  says  : — 

"  If  in  these  days  we  wish  to  feel  the  mystic  presence  of 
the  great  god  of  the  Germans,  we  must  do  as  our  worship 
ping  forefathers  did,  withdraw  from  the  concourse  of  men, 
find  out  some  forest  solitude,  and  wait  there.  Let  it  be,  if 
you  will,  in  one  of  the  great  stretches  of  woodland  which  are 
to  be  found  in  East  and  West  Prussia;  or,  better  still 
nowadays,  go  to  the  vast  primeval  forests  which  lie  upon 
the  upper  slopes  of  the  Scandinavian  peninsula,  far  away 
from  the  fjords  and  the  too  frequent  steps  of  tourists.  There 
you  will  feel,  as  you  should,  the  strange  and  awful  stillness 
which  from  time  to  time  reigns  in  pine  forests  such  as  these. 
Presently  the  quiet  is  broken,  first  by  a  sigh,  which  arises  as 
from  the  ground  itself,  and  breathes  throughout  the  wood. 
Anon,  from  a  distance  a  sound  is  heard  so  like  the  sound 
of  the  sea  that  you  might  swear  (had  you  never  been  in  such 
a  wood  before)  that  you  could  hear  the  waves  drawing 
backwards  over  a  pebbly  beach.  As  it  approaches  the 
sound  grows  into  a  roar ;  it  is  the  roar  of  the  tempest,  the 
coming  of  Wodin."  • 

1  The  Vikings  in  Western  Christendom,  by  C.  F.  Keary,  M.A.,  F.S.A. 
T.  Fisher  Unwin,  1891. 

2  To  quote  this  passage  without  Mr.  Keary's  justification  of  it  would 
not  be  a  great  harm,  since  it  bears  witness  in  a  great  measure  to  its  own 
truth,  but  what  Mr.  Keary  says  seems  to  me  of  a  very  universal  import 
ance,  and  to  bear  upon  the  right  treatment  of  Greek  as  well  as  of  German 
mythology.     Speaking  of  his  own  account  of  Wodin  in  the  forest,  he  says  : 
' '  It  will  be  said  by  some  that  this  description  is  purely  imaginary.      I 


I  GREEK  SANCTUARIES  7 

It  is  indeed  a  privilege  newly  and  exclusively  granted  to 
the  highest  moods  and  broadest  minds  of  to-day,  this 
enlightened  tolerance,  this  "genial  catholicity  of  apprecia 
tion,"  which  finds  even  in  paganism  a  message  from  the 
only  and  the  everlasting  God.  Now  at  last,  thanks  to  the 
painstaking  work  which  truly  scientific  men  have  done  in 
archaeology,  we  are  receiving  something  of  the  legacy  be 
queathed  us  by  those  who  lived  and  loved  and  prayed  of 
old  in  Athens  and  in  Rome.  Now  at  last  we  may  feel,  with 
no  petty  wish  to  carp  or  cavil,  the  sacredness  of  ancient 
sanctuaries,  and  know  them  for  ever  consecrated  to  "the 
sessions  of  sweet  silent  thought,"  where  we  summon  up  not 
only  "  remembrance  of  things  past,"  but  also  much  of  the 
sweet  usage  and  workaday  reality  in  things  now  present  for 
our  spiritual  aid. 

Let  this  new  privilege  console  somewhat  the  praisers  of 
the  past,  for  it  makes  up  for  and  takes  the  place  of  much 
that  modern  men  have  lost.  Let  malcontents  consider 
in  this  new-dawned  light  of  tolerance  the  early  worship  of 
both  Greece  and  Rome,  and  then  they  may  forget  to 
remember  how  their  lots  and  their  lives  are  cast  into  a 
world  so  filled  with  quantity,  so  choked  and  crammed  to 
bursting  with  millions  whether  of  men  or  money,  that 
quality  seems  lost,  or  even  to  be,  when  rarely  found, 

make  a  distinction  between  what  is  imaginative  and  what  is  imaginary.  If 
you  choose  not  to  go  into  the  study  of  mythology  or  of  beliefs  of  any  kind 
until  you  have  first  stripped  yourself  of  your  imagination,  you  will  travel 
indeed  lightly  burdened,  and  you  will  arrive  at  strange  results.  Because, 
as  belief  of  all  kinds  is  born  of  the  imagination,  and  Aberglaube  is,  as 
Goethe  says,  the  poetry  of  life,  you  will  have  taken  the  precaution  of  going 
into  the  dark  unprovided  with  a  lantern.  To  avoid  doing  this  you  are  not 
obliged,  however,  to  give  free  rein  to  your  fancy.  Nor  have  we  done  so  here." 
I  have  taken  the  liveliest  interest  in  seeing  how  near  to  my  own  point 
of  view  Mr.  Keary  arrived,  and  how  the  idea  of  studying  the  Pagan 
religion  in  its  sanctuaries  had  presented  itself  to  him  in  another  part  of  the 
field. 


8  INTRODUCTION  i 

neglected  and  misprized.  Let  us  then  exhort  the  pessimists 
of  this,  the  golden  hour  of  the  broadest  and  most  real 
Christianity  —  that  truest  consecration  of  democracy  — to 
look  not  backward  but  upward,  and  discern  in  the  broad 
humanity  and  strength,  and  above  all  in  the  toleration l  of 
latter-day  religion,  a  gleam  from  Olympus  of  the  Greeks. 
Their  religion,  so  far  as  it  was  true,  still  lives  and  shines  in 
the  light  of  to-day.  In  the  high  types  of  excellence  and 
beauty  which  Greek  religion  created,  and  Roman  practice 
made  more  all-embracing  and  enduring,  there  is  manifested 
a  mercy  whose  overruling  providence  leads  us  towards  the 
best. 

Indeed  the  quality  of  Greek  divinities  is  that  of  mercy. 
Demeter's  love  is  faithful,  although  the  heavens  seem  to  fall, 
and  though  earth  withholds  all  comfort  for  a  time ;  her 
chastened  joy  (when  at  last  it  comes)  is  divinely  pure  and 
gracious  in  the  fulness  of  its  perfect  peace.  Apollo,  that 
most  truly  Greek  of  all  divinities,  is  a  gloriously  dazzling 
exemplar  of  purity  and  light.  He,  the  sun-god,  is  mirrored 
like  the  sun  from  a  thousand  angles  of  refraction.  Understand 
this  god,  and  straightway  his  image,  shaped  this  way  or  that 
by  accident,  or  even  distorted  by  some  chance,  will  be  always 

1  It  is  a  fact  well  known  that  regular  offerings  were  made  on  behalf  of 
Augustus  at  the  Temple  in  Jerusalem.  This  was  but  one  of  many  small 
practices  which  grew  out  of  the  Theocrasia  or  commingling  of  gods,  a 
result  of  that  all-embracing  toleration  which  led  Greeks  and  Romans 
alike  to  treat  as  their  own  the  gods  of  other  nations.  See  the  opening 
plea  in  the  Octavius  of  Quintus  Minucius  Felix,  p.  6  B-E.  It  was  a  pious 
duty  for  travellers  in  remote  parts  of  the  empire  to  sacrifice  to  the  local 
gods  wherever  they  went.  When  the  absolute  antagonism  between 
paganism  and  Christianity  brought  persecution  to  pass,  there  was  still  a 
lack  of  thoroughgoing  intolerance  as  compared  with  that  of  the  Inquisition. 
A  comparison  of  the  numbers  put  to  death  by  Roman  and  Spanish 
intolerance  respectively  establishes  this  fact.  See  Friedlaender,  Dar- 
stellungen  aus  der  Sittengeschichte  Rom  s  in  der  Zeit  von  Augztst  bis  zum 
Ausgang  der  Antonine,  iii.  p.  586. 


I  POWERS  OF  GREEK  GODS  9 

returning  to  your  wandering  eye  from  every  corner  and 
surface  of  Greek  story  and  Greek  song.  Achilles  had  the 
swiftness  and  the  pure  white  heat  that  make  Apollo  known,1 
and  that  nobly  moulded  youth,  who  bursts,  divine  in 
righteous  anger,  on  our  view  upon  the  western  pediment  of 
the  temple  of  great  Zeus,2  is  a  very  incarnation  of  the  power 
of  Apollo.  Whether  he  be  Apollo,  or,  as  some  think, 
Pirithous,3  the  act  of  quenching  lust  and  foiling  brutal 
crime  is  most  pure  4  Apollo's  own. 

Dionysus,  the  dread,  the  deep,  the  darkly  irresistible,  a 
god  of  mystery  and  of  intensity,  the  all-possessor  of  men, 
and  even  of  beasts  and  things  upborne  by  onrush  and  inrush 
of  his  power — Dionysus  lives  on  to-day  in  the  fairy-land  of 
poetry,  mirrored  by  the  motley  throng  of  Orpheus  tales  and 
songs  of  wine  and  stones  of  overpowering  inspirations.  All 
and  each  of  the  greater  Greek  gods  still  live  their  charmed 
life,  and  even  to-day  each  one  in  some  sense  is  the  centre 
of  a  scheme  of  things,  a  universe  all  his  own. 

There  was,  in  fact,  little  or  no  thought  in  early  Greece  of 
how  one  god's  power  might  be  made  compatible  with  that 
attributed  to  and  exercised  by  another.  Each  Greek  state 
was,  according  to  Greek  theory,  absolutely  independent  of 
all  and  every  other.  Therefore  it  was  not  unnatural  for 
Greeks  to  think  of  each  one  of  their  great  divinities  as  in 
some  sense  partaking  of  an  absolute  independence.  Each 
great  god  had  been  worshipped,  no  doubt,  at  some  time, 

1  For  some  admirable  remarks  on  this  parallelism,  which  scarcely  needs 
to  be  pointed  out,  see  Keary's  Outlines  of  Primitive  Belief,  p.  192  :   "Each 
is  the  ideal  youth,   the   representative,  one  might  fairly  say,  of  '  young 
Greece,'  that  which  was  to  become  in  after  years  Hellas." 

2  See  Appendix  XI.  i.  141  and  143. 

3  Pausanias,  V.  x.  8. 

4  For  Apollo  as  the  god  of  purity  see  the  last  chapter  below  on  Apollo 
at  Delos. 


10 


INTRODUCTION 


somewhere  in  broad  and  various  Hellas,  as  the  supreme 
arbiter  of  destiny,  the  wielder  of  all  power,  and  the  hearer 
of  all  prayer.  Thus  we  have  glimpses  of  an  earliest  time  l 
when  Greek  religion  was,  if  the  word  be  insisted  upon, 
monotheistic.  At  each  centre  of  political  life  it  would  seem 
that  men  worshipped  a  god  whose  omnipotence  was  bounded 
by  the  boundaries  of  that  particular  state.  The  difference 
in  relative  importance  noticeable  between  the  gods  con 
nects  itself  with  the  history  of  the  chief  place  of  this 
and  that  god's  worship.  A  certain  early  importance  in 
matters  religious  belonged  to  Dodona  for  instance,  and  the 
national  character  borne  in  later  days  by  the  festival  of 
Olympian  Zeus  was  partly  the  cause  and  not  wholly  the 
effect  of  the  kingship  attributed  by  universal  consent  to  Zeus 
among  the  great  gods  on  Olympus  as  well  as  at  Olympia. 

Judge  Greek  religion  not  by  all  its  moods,  but  by  all  its 
highest  and  most  characteristic  ones.  Avoid  as  you  would 
the  very  spirit  of  untruth  a  judgment  of  Greek  political  and 
religious  ideas,  founded  on  notions  of  politics  and  religion 
that  only  came  into  being  in  modern  times,  and  then  you 
will  say  that  Greek  religion  was  a  polytheism  where  each  of 
many  great  gods  was  potentially  the  one  and  only  god  for 
every  Greek,  but  actually  and  more  particularly  in  one  place 
and  for  one  people  of  Greece.  The  Greek  religion  of  poly 
theism  was  more  monotheistic  than  monotheism  itself,  for 
the  Greeks  were  not  content  with  one  only  God  Almighty 
and  Supreme,  they  had  and  they  worshipped  many  such.2 

1  Of  this  earliest  day  we  can  have  only  glimpses,  and  it  must  be  to  a 
large  extent  ideal,  as  retrospects  of  the  kind  always  are.     In  reality  many 
disturbing  causes  came  in,  such  as  the  relics  of  an  utterly  barbarous  phase 
of  religion  and  a  certain  compatibility  between  local  all-importance  and 
ideal  subordination  in  a  rough  scheme  of  the  religious  world,  such  as  a 
peasant's  mind  could  form. 

2  Lehrs  (Populdre  Aufsdfze,  p.    130)  expresses  substantially  the  same 


I  ROMAN  ORGANISATION  11 

The  truth  which  this  paradox  contains  is  shown  by  the 
course  of  Greek  political  history — a  stream  which  ran 
curiously  and  closely  parallel  to  that  of  Greek  religion.  The 
whole  range  and  expanse  of  ancient  Greek  life  were  requisite 
before  the  Greeks  could  win  from  struggles  and  adversity 
the  lesson  of  political  subordination  and  national  unity. 
While  this  life-giving  knowledge  was  most  vivacious  within 
it,  the  mind  of  Greece  was  merged  with  its  territory  in  the 
body  politic  of  Rome,  and  Rome  was  livened  for  her  task 
by  Grecian  wit.1  Indeed,  humanity  has  possessed  the 
power  of  potent  organisation  and  broader  growth  only  since 
the  day  when  Hellas  died  as  a  separate  political  power, 
conscious  at  last,  though  late,  that  captious  independence 
overstrained  had  brought  destruction  in  its  train.  Rome 
—imperial  Rome — showed  forth  most  plain  the  moral  of 
Hellenic  failure,  and  Roman  success  made  way  for  the  con 
ception  of  something  higher  even  than  political  organisation 
of  the  broadest  kind.  After  and  because  of  the  heathen 
empire  came  the  Holy  Empire  which  crowned  all  others, 
the  empire  of  the  universal  Church. 

At  Marathon  and  Plataea,  Greece  defeated  the  patriarchal 
order  of  politics  and  religion.  There  was  a  moment  of 
universal  history  when  the  clan  was  succeeded  by  the  com 
monwealth,  when  the  spirit  of  humankind  required  new 
room  for  growth,  more  room  than  the  oriental  polity  of 

idea  by  drawing  a  distinction  which  eludes  translation.  "The  Greek," 
he  says,  ' '  could  perfectly  well  apprehend  (begreifen)  one  sole  God,  but 
his  mental  requirements  and  endowments  were  utterly  averse  to  any  real 
comprehension  (ergreifen).  He  always  fell  short  of  a  realising  sense,  and 
accordingly  never  lived  up  to,  never  acted  out  the  idea." 

1  ' '  The  Roman  conquest  of  Greece  was  a  welcome  event  to  the 
mass  of  the  European  Greeks.  The  popular  sentiment  at  the  time 
was  expressed  by  a  parody  of  the  saying  attributed  to  Themistocles  in 
exile  to  the  effect  that  his  ruin  had  made  his  fortune."  Jebb's  Modern 
Greece,  p.  3. 


12  INTRODUCTION  I 

Persia  could  afford  it.     To  the  old  Persian  order  succeeded 
the  new  era  of  Greece  and  Rome. 

Sir  Alfred  Lyall,  in  his  Asiatic  Studies^  most  acutely  says 
that  Greece  invented  "  political .  citizenship,  and  rules  of 
conduct  under  State  sanction.  Between  the  clans  and  the 
commonwealths  the  difference  is  not  so  much  between  law 
lessness  and  free  institutions  as  between  the  primitive  man, 
whose  social  and  political  customs  are  as  much  a  part  of 
his  species  as  the  inherited  habit  of  an  animal,  and  the 
highly  civilised  man  who  consciously  chooses  his  own 
laws  and  form  of  government  according  to  expediency  and 
logic."  The  props  of  tribe  and  caste  were  dispensed  with, 
that  the  wider  and  freer  political  organisation  of  our  day 
might  come  into  being.  For  the  evolution  of  so  vast  a 
system  the  ground  had  first  of  all  to  be  cleared.  This  took 
place  before  the  battle  of  Marathon,  on  the  very  restricted 
scale  which  Greece  allowed,  but  not  to  the  same  extent 
everywhere  in  Greece.  The  Spartan  institutions  of  Lycurgus 
(so  far  as  we  can  know  them)  are  a  curious  crystallisation  of 
primitive,  social,  and  political  habits,  hardly  in  advance  of 
what  to-day  excites  wonder  and  defies  a  casual  comprehen 
sion  in  India  and  Central  Asia.  Athens,  the  typically  free, 
the  truly  modern  Greek  state,  had  her  triumph  at  Marathon, 
Salamis,  and  Plataea.  For  a  time  this  new  way  of  living 
required  a  very  restricted  sphere  of  action,  these  new  states 
had  to  be  small.  Independence  rather  than  interdepend 
ence  was  the  watchword  of  this  dawning  period.  The 
time  was  not  yet  come  for  federations  or  confederations, 
and  hence  the  Athenian  Confederacy  failed.  Years  and 
years  of  political  experience  had  to  be  gathered  in  by 
Athenians  and  by  Romans  before  a  solidly  constituted  state 
of  this  new  type  became  possible  on  a  large  scale. 


i  THE  KINGDOM  OF  EARTH  13 

To  those  who  cannot  see  how  different  in  kind  from 
anything  known  before  was  the  small  Greek  Polis,  it  must 
seem  that  the  Greek  world  was  far  more  hopelessly  split  up 
after  the  Peloponnesian  war  than  it  had  been  in  the  days  of 
Homeric  song.  In  politics  the  new  principle  made  coali 
tion  impossible,  and  in  religion  it  left  to  local  divinities 
their  old-time  omnipotence.  The  loose  and  unsettled 
organisation  of  the  government  in  heaven  by  Zeus  was 
still  what  Homer  had  made  it. 

But  this  was  only  for  a  time.  When  at  last  the  world 
from  which  we  spring,  and  of  whose  life  we  are,  had  been 
qualified  for  wider  things,  and  had  outgrown  Greek  indi 
vidualism,  then  Greece  had  grown  into  Rome,  and  had 
allied  itself  with  that  spirit  of  subordination  and  self-dis 
cipline  by  which  alone  the  Romans  conquered  all  mankind. ' 
Imperial  Rome  at  its  best  has  seemed  a  realisation  of  Plato's 
dream,  a  state  where  philosophers  were  kings  and  kings  were 
philosophers.  This  was  truly  the  kingdom  of  Heaven  upon 
earth,  and  a  comparison  forced  itself  upon  religious  minds 
between  the  perfect  union  and  solidarity  of  all  functionaries 
of  the  empire  and  that  loose  government  of  the  spiritual 
world  on  Olympus  which  tradition  ascribed  to  Zeus. 
According  to  these  latter-day  notions  the  power  of  Zeus 
over  his  fellow-gods  and  subjects  was  contemptibly  small 
and  precarious.  A  desire  asserted  itself  gradually  that  the 
kingdom  of  earth  should  arrive  in  the  heavens.1  The 

1  This  desire  showed  itself  very  vigorously  even  in  the  lifetime  of 
Augustus,  having  already  found  expression  under  Julius  Caesar  in  a  very 
extreme  form.  Professor  Merriam  (see  the  next  note)  has  gathered 
most  interesting  proofs.  See  also  Giacomo  Lumbroso,  L Egitto  al  tempo 
dei  Greci  e  del  Romani.  Both  of  these  writers  correct  the  errors  of 
Mommsen,  who  maintains  that  the  Caesareum  at  Alexandria  was  dedicated 
to  Caesar  Appulsor.  This  epithet  is  a  mistranslation  ;  the  real  equivalent  of 
is  the  deity  to  whom  the  ^Tri/SaTijpta — sacrifices  at  embarkation 


i4  INTRODUCTION  i 

Roman  emperors  were  deified,  inter  atia,  as  the  represen 
tatives  of  a  more  logically  ordered  scheme  of  things  than 
that  presented  by  the  poetical  figures  that  ruled  Olympus. 
Nowhere  was  this  new  hierarchy  of  the  heavens  more 
sedulously  cultivated  than  by  the  Greek  members  of  the 
Roman  Empire.  Zeal  for  this  new  worship  has  earned  for 
Greece  much  slander  from  unreflecting  persons.  Servility 
and  base  flattery  are  attributed  to  men  who  really  were 
following  in  the  footsteps  of  their  forefathers,  and  seeking 
an  organised  religion  which  their  poetical  traditions  could  not 
give,  though  it  did  suggest  the  lack  of  it.  Indeed  Zeus  or 

and  disembarkation — were  made,  who  rules  and  protects  all  sailors.  For 
this  aspect  of  Imperial  divinity  see  Virgil,  Georgics,  i.  29-31,  and  Propertius, 
iii.  ii,  71.  Apollo  and  Zeus  were  the  two  other  divinities,  beside  Poseidon, 
with  whom  Augustus  was  identified.  Professor  Merriam's  array  of  in 
scriptions  is  particularly  interesting.  Philo  Judaeus,  Legatio  ad  Caium, 
describes  Augustus  as  ' '  the  source  of  worshipful  majesty  to  his  successors, 
the  defender  from  evil  ('AXe£t'/caKos).  The  entire  habitable  world  decreed 
honours  to  him  co-equal  with  those  of  the  Olympian  gods."  Caesar  (eTTi- 
/for7?pioy)  at  Alexandria  (navigantium  praeses]  was,  as  the  same  Philo  de 
scribes  him,  "  the  saving  hope  of  all  who  weigh  anchor  or  enter  within  its 
(the  harbour's)  shelter."  The  seriousness  of  Alexandrine  sailors  in  this  con 
ception  of  the  divinity  ot  Augustus  (the  Caesareum  of  Alexandria  was  not, 
as  Mommsen  says,  for  the  worship  of  Julius  but  for  that  of  Augustus  Caesar) 
is  picturesquely  shown  in  chapter  xcviii.  of  Suetonius'  Life.  Catilius  placed 
at  Egyptian  Philae  a  dedication  ' '  to  Caesar  ruler  of  the  sea,  a  mighty  Zeus 
swaying  limitless  regions,  son  of  Zeus  (Caesar)  the  Deliverer  .  .  .  star  of 
all  Hellas  that  rose  as  a  mighty  Zeus  the  Saviour."  At  the  temple  of  Isis 
at  Tenbyris  in  the  year  i  A.D. ,  Octavius,  of  the  emperor's  own  gens,  styles 
Augustus  Zeus  Eleutherios.  At  Herod's  Caesarea-by-the-sea  Augustus  was 
worshipped  under  the  aspect  of  the  well-known  statue  of  Zeus  at  Olympia. 
The  Athenian  temple  of  Olympian  Zeus  was  finally  dedicated  to  the 
genius  Augusti.  Of  Augustus  worshipped  under  the  guise  of  Apollo  Dr. 
Merriam  gives  many  instances  ;  his  Egyptian  style  is,  ' '  autocrator  Caesar, 
son  of  the  Sun,  King  of  Upper  and  Lower  Egypt."  Moreover,  upon  two 
Demotic  stelae  (in  the  British  Museum)  from  Memphis  are  interesting 
records  of  two  brothers  who  were  priests.  One  of  them  died,  and  his 
brother  was  appointed  in  his  place  by  Augustus  ' '  in  the  first  year  of  the 
god,  the  son  of  the  god,  the  great  foreign  god  Caesar  autocrator";  and, 
furthermore,  he  was  made  "  Prophet  of  Caesar."  Inscriptions  exist  dated 
during  Augustus's  lifetime  (qualifying  him— in  spite  of  his  prohibition — as  a 
god)  from  Apamea  (C.  /.  G.  3525),  Lesbos,  Delos. 


I  THE  ROMAN  ORDER  15 

Jupiter  could  not  very  long  remain  the  emperor  of  Olympus 
when  once  the  deifications  of  Roman  emperors  had  accus 
tomed  men  to  think  of  the  ordered  rule  of  heaven  as  a 
counterpart  in  some  way  of  the  ordered  and  most  plainly 
organised  rule  on  earth.1 

This  impulse  to  sanctify  the  secular  arm  of  government 
marked  the  reigns  of  many  Roman  princes,  and  made  all  good 
emperors,  and  some  even  whose  vices  earned  the  scorn  of  those 
near  by,  the  idols  of  the  provinces.  Hence  arose  that  consecra 
tion  of  a  universal  headship  covering  all  affairs,  both  spiritual 
and  temporal,  which  has  survived  in  the  Roman  Church  and 
in  the  Russian  part  of  what  to-day  still  bears  the  title  of 
the  Greek  Church.  This  glorification  of  imperial  attributes 
at  Rome  helped  to  undermine  men's  faith  in  the  established 
Greco -Roman  religion.  Each  of  the  careless  gods  on 
Greek  Olympus  trenched  too  palpably  upon  the  pperogatives 
of  various  of  his  fellows.  There  was  no  loophole  of  escape 
for  the  ingenuity  even  of  Roman  jurisprudence.  Zeus  had 
too  many  affinities  with  Dionysus,  who  resembled  Apollo  in 
his  powers  and  Aesculapius  in  his  story.  Apollo  the  healer 
was  more  than  Aesculapius'  father,  he  often  seemed  to  be 
all  there  was  of  the  godhead  of  Aesculapius.  Was  Apollo 
the  god  and  Aesculapius  but  a  man  ?  But  then  what 
became  of  the  miracles  and  divine  pretensions  of  the  latter  ? 
How  came  both  Dionysus  and  Aesculapius  to  birth  by  the 

1  Appendix  I.  below  deals  with  the  deification  of  the  emperors  (see 
Horace,  Od.  i.  2,  and  elsewhere),  and  on  the  general  subject  read  Gaston 
Boissier,  "  La  religion  romaine  d'Auguste  aux  Antonins,"  and  also  consult 
A.  E.  Egger,  ' '  Examen  critique  des  historiens  anciens  de  la  vie  et  du  regne 
d'Auguste."  An  interesting  account  of  The  Caesareum  and  the  worship  of 
Augustus  at  Alexandria  by  Professor  A.  C.  Merriam  will  be  found  in  the 
"Transactions  of  the  Am.  Philological  Association  for  1883."  The  con 
clusions  in  this  paper  are  the  more  interesting  because  arrived  at  without 
any  knowledge  of  Boissier's  views.  They  confirm  the  view  taken  in  this 
lecture  and  its  Appendix. 


16  INTRODUCTION  i 

same  miraculous  and  premature  intervention  of  fire  ?  These 
perplexities  and  others  remained  unanswered  even  by  the 
clumsy  hocus-pocus  of  venal  priests,  or  the  more  disinterested 
but  equally  unconvincing  lucubrations  of  Orphic  brother 
hoods.  Hence  arose  a  want  unsatisfied  and  universal ;  from 
the  natural  evolution  of  political  and  religious  life  in  an 
tiquity  arose  a  cry  to  which  then  came  the  answer  of 
Christianity  :  Illimitable  hope  has  new  made  heaven  and  earth^ 
and  order  has  supplanted  shapeless  and  unseemly  riot  in  the 
spiritual  realms  through  which  we  have  true  life.  Universal 
Christianity  was  called  into  being  to  meet  the  new  and 
Roman  order  of  things  where  organisation  was  everything. 
Then  the  turbulence  of  Greek  individualism  had  to  keep 
the  peace  or  cease  to  be. 

The  religion  of  Greece  as  such  was  guiltless  of  system  and 
wholly  devoid  of  method.  It  may  be  compared  to  a  way 
ward  prayer  poetically  prayed,  according  to  the  whimsies  of 
many  daring  flights  of  devotional  ecstasy,1  and  not  to  a  scheme 
of  the  ordered  universe  so  reasoned  out  and  so  systematised 
that  it  could  be  written  down  in  creeds  or  expressed  in 
articles.  But  now  arises  the  question,  How  could  worship 
or  prayer  of  any  kind  be  possible  unless  there  was  some 
definite  understanding  of  the  powers  and  provinces  of 
various  gods  ?  This  is  really  the  same  question  of  which 
we  tried  to  dispose  a  moment  since.  We  are  forced  to  ask 
it  in  this  day  of  clear-cut  creeds  because  of  all  the  history 
of  religious  ideas  between  us  and  early  Greece.  This 
question,  however,  no  really  Greek -minded  person  could 

1  ' '  We  can  reason  out  the  growth  of  a  belief ;  for  looked  at  over  a  wide 
area  and  followed  through  a  sufficient  period  of  time,  every  belief  has  a 
kind  of  reason  and  a  kind  of  reality.  But  to  each  individual  in  his  brief 
span  of  life  it  is  like  the  wind ;  he  cannot  tell  whence  it  cometh  or  whither 
it  goeth." — Keary,  The  Vikings,  p.  59. 


i  GREE  K  INDI VI D  UA  LISM  1 7 

have  understood,  let  alone  answering  it.  Indeed  the 
possibility  of  maintaining  the  old  ritual  and  of  worshipping 
the  old  divinities  depended  somewhat  upon  the  im 
possibility  of  asking  this  question.  From  this  it  is  evident 
that  in  considering  the  past,  and  more  especially  in  dealing 
with  a  bygone  religion,  we  must  perform  the  feat  of  leaving 
out  our  own  peculiar  selves,  and  all  the  ready-made  ideas 
with  which  our  minds  have  been  upholstered.  This 
involves  a  scrupulous  self-examination,  and  brings  before 
us  again  the  old  first  law  of  Delphian  Apollo — know 
thyself.  Think  yourself  away,  if  it  were  possible,  from 
all  this  workaday  world  of  business,  with  its  majority 
whose  thinking  is  more  or  less  at  second  hand — borrowed 
from  tradition  or  echoed  from  great  leaders'  lips — think 
that  you  have  neither  heard  nor  dreamed  of  a  large  state 
solidly  organised  as  we  know  states  to-day.  Think  of  a 
condition  of  mind  where  the  management  of  a  large  railway 
would  be  deemed  impossible ;  where  the  notion  of  policing 
a  town  of  half  a  million  had  not  dawned,  and  the  thought 
had  not  entered  in  of  combining  large  states  into  one 
political  aggregate  otherwise  than  by  the  lash  of  an  over 
powering  master, — of  the  Great  King,  as  the  Greeks  in  their 
most  independent  days  named  under  their  breath  the  ruler 
of  Persia. 

Incapable  as  the  Greeks  were  even  of  the  Persian 
counterfeit  of  political  organisation,  they  never  raised  nice 
questions  in  theology  about  the  prerogatives  of  their  gods. 
Apollo,  Dionysus,  Zeus,  Demeter,  and  many  others  were 
coexistent  and  all-powerful,  yet  there  was  room  for  the 
new  divinities  that  came  from  Egypt  and  the  East.  The 
abstract  question  of  subordination  among  the  gods  did  not 
dawn  even  upon  the  greatest  and  most  far-reaching  minds 

c 


i8  INTRODUCTION  I 

of  Greece.  Many  and  subtle  are  Plato's  arguments  about 
the  gods.  Difficulties  springing  from  the  various  and 
often  discreditable  tales  told  about  them  crowd  upon  his 
truly  pious  mind ;  but  he  is  never  distracted  by  the  desire 
to  fix  exact  limits  of  power  for  each.  Plato,  Pindar,  and 
Euripides,  Greek  minds  which  were  especially  pre-occupied 
by  religious  problems,  devoted  their  efforts  in  this  direction 
to  disentangling  the  wisdom  and  omnipotent  strength  of  God 
from  the  follies  and  frailties  of  man,  as  well  as  from  the 
more  than  human  infirmities  attributed  to  ancient  gods  in 
old-time  stories.  To  vindicate  the  poetic  purity  and  truth  of 
Apollo,  to  show  forth  the  uncalculating  and  tragical  intensity 
of  Dionysus,  was  their  chosen  task, — a  far  more  important 
one  at  that  time  than  the  elaborating  of  a  heavenly  hierarchy 
or  the  formulation  of  a  creed.  All  divinities,  says  Plato, 
are  false  if  they  are  not  spotless  and  free  from  imputations 
of  falsehood;  in  his  perfect  commonwealth  the  poets 
shall  be  forbidden  to  sing  of  gods  who  can  be  bribed 
like  unjust  men.  With  similar  intent  Pindar  piously 
remoulds  the  story  of  the  house  of  Tantalus,  since  it  is 
unlawful  to  attribute  evil  to  the  gods ;  while  almost  every 
page  of  Euripides  bears  the  impress  of  a  conflict  in  men's 
minds  between  the  noblest  ideal  conception  of  godhead 
and  the  popular  stories  and  superstitions  concerning  the 
gods. 

The  Greek  poets  and  philosophers  are  among  our 
intellectual  progenitors,  and  therefore  the  religion  of  to 
day  has  requirements  which  include  all  that  the  noblest 
Greek  could  dream  of — requirements  which  the  aspirations 
of  Israel  alone  could  not  satisfy.  Our  complex  life  had 
need  not  only  of  a  supreme  god  of  power,  universal  and 
irresistible,  of  a  jealous  god  beside  whom  there  was  no 


I  GREEK  RELIGIOUS  THINKERS  19 

other   god,    but   also   of  a   god   of    love   and   grace   and 
purity.      To  these  ideal   qualities  present  in    the   diviner 
godhead  of  the  Gospels  the  evolution  of  Greek  mythology 
brought  much  that  satisfies  our  hearts.     This  I  say  because 
the   purity   inculcated   in   the   religion   of   the   Jews   and 
enforced  by  penalties,  as  recited  in  various  episodes  of  the 
Old  Testament,  rarely  imposes  itself  by  the  inner  charm 
of  native  worth  and   loveliness.      It  comes  upon  us  fre 
quently  as  the  will  of  a  resistless  and  often  unrelenting 
God,    a   religious    point    of    view   transcended   by   Plato, 
Pindar,    and   Euripides.      Both   these   presentations    were 
doubtless  needed,  but  the  importance  of  this  latter  must 
not  blind  us  to  the  power  for  good  inherent  in  the  former. 
And  here  we  may  remember  the  quaint  and  solemn  words 
of  Henry  More  in    the  Mysterie   of  Godliness :    "  Christi 
anity  is  so  excellent  in  itself  that  we  need  not  phansy  any 
Religions  worse  then  they  are,   the  better  to  set  off  its 
eminency.     Besides  the  more  tolerable  sense  we  can  make 
of  the 'affairs  of  the  ancient  Pagans,  the  easier  Province  we 
shall  have  to  maintain  against  prophane  and  Atheistical  men, 
to  whom  if  you  would  grant  that  Providence  had  utterly 
neglected  for  so  many  ages  together  all  the  nations  of  the 
world  except  that  little  handfull-  of  the  Jews,  they  would, 
whether  you  would  or  no,  from  thence  infer  that  there  was 
no  Providence  over  them  neither,  and  consequently  no  God." 
All  these  considerations  certainly  arouse  a  feeling   of 
thankfulness    that    the   great   religious   leaders   of    Greek 
thought  should  before  all  things  have  occupied  themselves 
with  the  goodness  of  their  various  ideal  divinities.     Had 
their  preoccupation  been  to  show  that  one  god  was  more 
powerful  than  another,  rather  than  the  total  superiority  of 
gods  to  humankind,  then  the  charm  of  goodness  "  hearted  in 


20  INTR  OD  UCTION 

high  hearts"  would  never  have  ended  by  attaching  to  the 
best  man's  conception  of  the  best  god  in  Olympus.  It  is, 
in  fact,  difficult  to  see  how  the  spontaneous  charm  of 
Demeter's  love,  the  glorifying  efficacy  of  her  sorrows,  could 
have  been  set  before  the  human  mind,  could  have  been 
dramatised  otherwise  than  in  a  community  of  gods  no  one 
of  whom  had  an  absolute  and  omnipresent  supremacy.  It 
is  fortunate  perhaps  for  us  that  the  Greeks  were  poetical 
and  dramatic  rather  than  logical  and  literal-minded  in  their 
theology,— if  theology  we  choose  to  call  it,— for  in  the 
charmed  realm  of  their  great  gods  where  as  equals  they 
suffered  and  struggled,  hoped  and  helped,  loved  and  were 
loved,  the  ideal  character  of  the  perfect  god— a  man  divine, 
a  human  god— was  gradually  brought  to  be.1 

i  It  was  not  omnipotence  so  much  as  solitude,  lack  of  good  fellowship, 
of  susceptibility  to  comforts  and  delights,  that  Pagans  found  fault  with  in 
-the  Christian  ideal.  "Whence  comes,  who  is,  and  where  lives  their 
precious  god?"  asks  Caecilius,  and  then  he  gives  his  own  answer  by 
qualifying  the  god  as  "  Unicus,  solitarius,  destitutus,"—  without  a  fellow, 
solitary  -wholly  forsaken.  Forthefull  passage  see  the  Octavivsoi  Mmucius 
Felix  p  13  Possibly  the  mere  fact  that  the  Christians  boasted  of  the 
oneness 'of  their  god  enhanced  the  Pagan  appreciation  of  mere  multiplicity; 
and  yet  there  is  a  genuine  ring  in  the  saying  attributed  to  one  of  the  imperial 
defenders  of  Paganism  that  a  universe  emptied  of  numerous  divinities  was 
uninhabitable. 

Lehrs  has  given  the  best  account  of  the  matter  as  it  affected 
fundamentally  Greek  notions  of  religion.  "The  very  circles  long  accus 
tomed  to  a  view  of  myths  which  either  abandoned  or  explained  them 
away,  the  very  people  who  (like  the  influential  Stoics)  had  definitely  made 
up  their  minds  against  all  gods  in  human  shape,  entertained  side  by  side 
with  the  metaphysical  conception  of  one  highest  god  a  present  belief  m 
many  gods.  Nor  was  this  belief  a  mere  matter  of  formal  dogma  ;  it 
lived  and  glowed  with  a  power  that  influenced  men's  lives."  Again  Lehrs 
says  in  the  same  essay,  entitled,  Gott,  Goiter  und  Ddmonen,  ' '  When 
your  Greek  contemplated  nature  and  the  feebleness  and  dependence  of 
man,  there  arose  before  him  not  one  god  ...  but  there  was  a  spon 
taneous  outburst  of  the  fulness  of  life  divine.  He  saw  a  world  of  gods.' 
The  gifted  author  at  the  close  of  this  interesting  passage  finds  an  adequate 
though  untranslatable  phrase  for  the  lives  and  loves  ( of  the  gods  on 
Olympus,  "Dieses  vielgestaltige  Gotterineinanderleben."— Populare  Auf- 
sdtze,  p.  130. 


i  ANOMALIES  OF  GREEK  RELIGION  21 

In  searching  out  the  development  of  an  ideal  character, 
divine  and  human,  through  the  tangles  of  poetic  fiction  that 
served  at  once  to  hide  and  to  protect  it  till  its  growth  was 
strong,  we  must  be  ready  for  surprises.  We  must  forget 
that  Zeus  was  ruler  in  Olympus,  and  be  often  prepared 
to  treat  him  like  the  least  among  his  attendant  divinities. 
His  looks  and  even  his  attributes  are  given  sometimes  to 
Aesculapius,  one  of  the  latest  of  partakers  in  Olympian 
immortality.  Apollo  and  Dionysus  will  often  seem  almost 
convertible,  and  the  worship  of  Demeter  merges  into  that 
of  Rhea  on  the  one  hand,  and  on  the  other  into  that  of 
Cybele,  while  all  three  goddesses  are  continually  exercising 
powers,  giving  ear  to  prayers,  and  receiving  offerings 
which  might  be  equally  well  associated  with  the  name  of 
Aphrodite. 

In  fact,  the  most  profitable  state  of  mind  for  one  who 
would  learn  about  Greek  religion  treats  each  god  and 
goddess  in  turn  as  if  he  or  she  alone  existed,  and  at  the 
same  time  always  bears  in  vivid  mind  the  history  and 
attributes  of  all  and  several  of  the  other  gods.1 

There  was  noticeable  in  the*  last  days  of  Paganism  a 
breaking  down  of  barriers,  an  effacement  of  the  individual 
status  of  each  god.  This  process  began  much  earlier,  how 
ever,  in  the  case  of  some  gods  than  in  that  of  others.  In 
deed  one  of  the  greatest  Greek  divinities,  Dionysus,  seems 

1  In  dealing  with  this  difficulty  one  thing  chiefly  needful,  according  to 
Lehrs,  is  to  accustom  our  minds  to  the  notion  of  an  extended  sphere  of 
action  for  each  and  all  of  the  several  gods.  ' '  Every  god  had  his  own  peculiar 
and  appropriate  range  of  usefulness  and  activity,  and  yet  every  god  was 
besought  for  every  manner  of  help,  in  any  place  where  he  was  close  at 
hand,  where  he  was  propitious,  where  he  was  especially  worshipped. " — 
Populdre  Aufsdtze,  p.  138.  Indeed  a  very  close  parallel  to  this  over 
lapping  of  the  spheres  of  power  assigned  to  the  various  gods  may  be 
found  in  cases  of  appeal  made  to  patron  saints  in  the  Greek  and  in  the 
Roman  Church. 


22  INTRODUCTION  i 

never  to  have  been  to-day  what  yesterday  had  shown  him, 
and  on  every  morrow  he  was  changed  again.  He  began 
as  the  great  god  of  Thrace,  a  prophet-ruler  of  the  dead. 
Introduced  in  more  southerly  climes,  he  became  in  one 
place  a  god  of  clemency,  and  in  another  the  avenging  deity. 
So  far  as  the  religious  consciousness  of  Hellas  was  ever 
wholly  awakened,  just  so  far  was  there  an  attempt  to  frame 
the  universal  Dionysus  out  of  elements  drawn  from  all  this 
revelling  rout  of  fairy-tales.  In  one  direction  the  differ 
ences  fell  away  that  divided  Dionysus  from  his  father 
Zeus ;  on  another  side — and  this  is  a  vital  point  for  under 
standing  the  history  not  only  of  Greek  but  especially 
of  Attic  religion — a  close  affinity  showed  itself  between 
Dionysus1  and  Apollo,  more  especially  Delphian  Apollo. 

Delphi  was  the  seat  of  a  joint  worship  of  Dionysus  and 
Apollo.  Apollo  absented  himself  from  his  shrine  on  Par 
nassus  as  well  as  from  his  holy  island  of  Delos  during  the 
bitterer  winter  months.  At  Delphi  Dionysus  naturally 
ruled  supreme  while  Apollo  was  thus  absent,  since  he  was 
there  before  Apollo  came  at  all.  I  shall  now  speak  of 
Delphi,  not  only  because  of  this  interesting  coupling  of  two 
great  gods,  this  dwelling  together  in  unity  at  Delphi  of  two 
divine  and  blessed  brethren,  but  also  because  there  is  a 
certain  present  appropriateness  in  the  theme.  How  could 
I  more  suitably  close  an  introduction  to  lectures  that  are  to 
deal  with  excavations  already  achieved  than  by  talking  of  a 
far-famed  site  where  all  is  yet  to  be  done,  and  showing  what 

1  Dionysus  finally  reached  a  point  which  may  be  described  as  a  con 
fluence  of  epithets  derived  from  all  the  various  forms  of  his  own  story,  and 
also  from  the  closely  allied  worship  and  myths  of  Osiris  as  well  as  from 
other  sources  farther  away.  See  the  Plutarchian  tract  on  Isis  and  Osiris  ; 
and  for  the  mass  of  epithets  see  Orphic  Hymns,  No.  51,  which  gives  forty- 
five  epithets,  and  No.  30,  where  there  are  twenty-eight. 


i  APOLLO  AT  DELPHI  23 

vital  questions  may  be  answered  by  successful  diggings 
there  ? 

The  clearness  and  the  almost  intellectual  sparkle  of  the 
fountain  of  Castalia l  can  be  neither  overpraised  nor  over 
prized.  To  slake  the  thirst  at  this  bright  stream,  and  look 
from  Delphian  heights  downward  and  see  the  far-off 
glimmer  of  a  distant  sea  descried  from  aloft  and  afar  across 
the  Crissaean  plain, — memorable  for  the  sacred  wars  fought 
that  none  might  pollute  it  by  tillage  of  any  kind, — is  an 
experience  never  to  be  forgotten.  Then  turn  away  and  see 
the  sun-illumined  glories  of  those  high-heaved  bulwarks  on 
Parnassus'  side,  the  rocks  once  called  Phaedriades.  High 
above  the  ledge,  where  ancient  Delphi  rose,  are  reared 
these  sheer  walls  of  living  light ;  and  one  of  the  mysterious 
places  which  some  connect  with  the  Apolline  oracle  is  a 
seemingly  unmeasurable  rift  in  these  Phaedriades  that  may 
be  entered  by  adventurous  climbers  from  the  gathering- 
place  of  the  Castalian  spring. 

Here,  truly,  is  a  place  where  pilgrims  would  resort,  and 
at  Delphi  the  traveller  in  Greece  may  even  now  fitly  bring 
to  a  climax  all  those  feelings  of  wonder  and  exultation 
awakened  by  the  sight  of  Greece,  the  common  and  inalien 
able  fatherland  of  generous  souls.  If  two  friends  were 
shortly  to  be  parted,  and  each  to  see  the  other's  living  face  no 
more,  I  could  wish  for  them  no  more  solemn  place  for  their 
last  days  of  fellowship  than  Delphi, — Delphi  as  it  is  to-day. 

Here  they  could  read  together  that  most  solemn,  sweet, 
and  pious  play 2  where  Euripides  shows  forth  the  spirit  of 

1  Appendix  XI.  ii.  76.  * 

•  Such,  I  maintain,  is  the  character  of  the  play  which  has  recently 
been  described  by  a  distinguished  authority  as  an  "attack  upon  Delphi." 
My  reasons  for  dissenting  from  this  unusual  view  of  the  Ion  have  been 
fully  presented  in  the  Nation,  No.  1329  (i8th  December  1890). 


24  INTRODUCTION  i 

truth  and  noble  -  hearted  kindliness  that  inspired  the 
Delphian  worship  of  Apollo.  Above  the  actors  in  the 
play  of  Ion  towers  Parnassus ;  the  brighter,  purer  air  of 
its  twin  peaks  exhales  from  every  line  of  this  tragedy,  which 
may  after  all  be  deemed  no  tragedy,  since  it  comes  to  a 
happy  issue,  that  involves  neither  murder  nor  sudden  death. 
Both  of  these  twin  Parnassus  peaks  belonged  to  Apollo  and 
Dionysus,  as  Dante  remembered  in  his  invocation  of  Apollo 
at  the  beginning  of  his  Paradise ;  but,  so  far  as  Apollo  alone 
stood  for  the  highest  reach  of  the  poetic  spirit,  the  highest 
summit  was  peculiarly  his.  Hence  Dante  says  : 

Most  kind  Apollo,  for  my  final  task 

Make  thou  of  me  such  vessel  of  thy  grace 

As  with  thy  laurel-crown  thou  canst  reward. 

One  peak  thus  far  of  high  Parnassus'  twain 

I  found  enough  ;  but  now  must  have  them  both,1 

Or  enter  not  the  contest  that  impends. 

Now  enter  thou  my  breast,  inspire  me  thou  ! 


1  This  is  a  far  truer  and  more  effective  picture  than  that  of  Cervantes 
in  the  "Journey  to  Parnassus,"  where  Helicon  with  its  Hippocrene,  its 
Pegasus,  and  its  Aganippe  is  made  a  part  of  Parnassus.  Dante  only 
remembered  (as  Scartazzini  says)  the  beautiful  lines  of  Lucan  (Phars.  v. 
71-74)  describing  the  Parnassus  : 

"  Hesperio  tantum,  quantum  semotus  Eoo 
Cardine,  Parnassus  gemino  petit  aethera  colle, 
Mons  Phoebo,  Bromioque  sacer  :  cui  numine  mixto 
Delphica  Thebanae  referunt  trieterica  Bacchae. " 

It  is  interesting  to  see  that  Dante  was  not,  as  Cervantes  was,  appealed  to 
by  the  array  of  misguided  learning  which  he  might  easily  have  derived 
from  the  commentary  of  Servius  on  Virgil,  much  resorted  to  in  his  day. 
See  Servius  on  Georg.  iii.  43  ;  Eel.  vi.  29,  x.  u.  In  his  commentary  on 
the  Aeneid,  Servius  carries  this  confusion  farther  by  saying  on  Aeneid  vii. 
641  :  Parnassus,  mons  Thessaliae,  di•vidit^lr  in  Cithaeronem,  Liberi,  et 
Heliconem,  Apollinis,  cuius  sunt  Musae  ;  and  again  on  the  same  line 
repeated,  Aeneid  x.  163  :  Parnassus  mons  est  Thessaliae  iuxta  Boeotiam, 
qui  in  duo  finditur  iuga,  Cithaeronem  Liberi,  et  Heliconem  Apollinis  et 
Musarum.  The  origin  of  this  confusion  between  the  two  peaks  of  Par 
nassus  and  the  neighbouring  pair  of  Boeotian  peaks  is  probably  to  be  found 


I  PARNASSUS  25 

In  Dante's  mind  Apollo  stood  aloof  from  all  other 
exemplars  of  the  pure  poetic  spirit,  from  all  other  inspirers 
of  majestic  song.  The  Muses  and  Dionysus  were  enough 
for  him  while  he  but  sang  of  torments  and  of  earth,  but  for 
the  upward  winging  of  his  song  through  the  heavens, 
Apollo's  inspiration  was  required.  It  came  to  him  as  a 
crowning  consummation  and  a  grace  ineffable  from  God  to 
uplift  his  soul  and  transfigure  his  body  until  he  could  have 
a  perfect  vision  of  heaven,  the  wonderland  of  man's 
nativity,  the  fatherland  of  every  righteous  soul. 

This  true  insight  into  the  unperishable  function,  the 
indestructible  potency  of  Apollo,  was  possessed  not  by 
Dante  alone,  but  by  many  poets  ancient  and  modern.  It 
has  been  indeed  a  true  instinct,  an  unfaltering  flight  of 

in  the  vague  recollection  of  certain  details  in  a  quotation  from  Hermesianax 
of  Cyprus  made  in  the  Plutarchian  work  De  Fluviis  (II.  'lafj.rjvbs).  Par 
nassus  generically  includes  all  peaks  between  Mounts  Oeta  and  Corax  on 
the  one  hand,  and  Cirrha  and  Anticirrha  on  the  other  ;  specifically  the  name 
Parnassus  applies  to  the  two  highest  peaks  in  this  range  which  are  named 
Tithorea  ( Herod,  viii.  32,  Strabo,  p.  417,  and  Pausan.  X.  xxxii.  6) and  Lycorea 
(Pausan.  ibid,  and  Strabo,  p.  418).  For  the  Greek  poets  these  peaks 
were  inseparable,  and  were  associated  with  rites  more  frequently  connected 
with  Dionysus  than  with  Apollo  (Aeschylus,  Eum.  22  ff.  ;  Soph.  Ant.  1126 
ff.  and  1144  f . ;  Eurip.  /.  T.  1244,  Phoen.  205  ff. ,  226  ff.,  Ion,  713,  and 
Hypsipyle,  fr.  752).  Apollo  was  not  however  excluded,  but  his  presence  was 
involved  in  that  of  Dionysus.  Pausanias,  speaking  of  the  peaks  of  Par 
nassus  (X.  xxxii.  7),  says  :  TO,  8£  veQ&v  r£  CGTIV  cu/wrepw  TO,  &Kpa,  Kail  at 
6i>td§es  eTri  TOVTOIS  ry  Atovwry  /cat  T£  'A7r6XXwvt  /j.aivot>Tai.  Virgil  and 
Ovid  say  nothing  new  about  the  two  peaks.  Ovid  agrees  with  Pindar  in 
making  Parnassus  the  Mount  Ararat  of  Deucalion's  deluge.  Lucan  does 
not  exclude  either  god  from  either  peak  ;  but,  nevertheless,  Benvenuto  da 
Imola  has  rightly  interpreted  Dante's  meaning  here  by  saying  :  "  Unum 
iugum  Parnassi  deputatum  Baccho  suffecit  sibi  hucusque,  nunc  vero  et 
illud  et  aliud  consecratum  Apollini  est  sibi  necessarium.  Per  Bacchum 
autem  figuratur  scientia  naturalis  quae  haberi  potest  per  acquisitionem 
humanam,  sicut  physica  et  ethica."  Apollo  represents  metaphysica  or 
sacra  scientia  ;  Bacchus  stands  for  eloquence,  ' '  quae  hucusque  suffecit 
sibi " ;  but  Apollo  is  sapientia.  Then  he  maintains  that  Apollo  and 
Bacchus  represent  the  same  god  under  different  names,  quoting  Macrobius, 
and  "Orpheus  sacer  poeta."  Dante  himself  here  adopts  Orphic  views. 
See  Appendix  XL  ii.  86-89,  ar>d  Hi.  9- 


26  INTRODUCTION  I 

poetic  inspiration,  which  has  preserved  Apollo  more  than 
any  other  of  the  gods  in  Greece.  Let  us  then  see  at 
last  that  Apollo  rather  than  Zeus  was  governor  of 
Olympus,  that  the  only  real  discipline — if  such  a  word 
be  applicable  at  all — submitted  to  even  momentarily  by 
all  the  gods  in  Greece  emanated  from  Delphi  and  the 
far-sighted,  wide -minded  oracle  of  Apollo  at  that  holy 
place. 

Zeus  was  a  king  among  gods,  who  reigned  but  governed 
not.1  His  Premier  was  the  Delphian  god.  This  way  of 
stating  the  facts  is  new,  but  still  the  very  nature  of  Greek 
mythology  and  religion  warrants  us  in  adopting  it.  The 
god  of  purest  highest  poetry  alone  was  competent  rightly 
to  order  a  religion  which  was  pure  poesy.  Instinctively 
the  poets  Homer  and  Hesiod  shaped  Greek  religion,  and 
Herodotus  speaks  of  them  as  its  originators,  the  first  theo 
logians.  It  is  against  the  poets  and  their  poetical  theology 
that  Plato  makes  his  protest.  All  this,  together  with  the 
necessity  laid  upon  us,  even  to  this  present  hour,  of  going 
to  school,  to  the  great  Greek  poets,  when  we  seek  to  inform 
ourselves  about  the  Greek  gods  and  their  sanctuaries,  will 
prepare  us  for  one  of  the  many  exquisitely  true  utterances 
exquisitely  made  by  Mr.  Walter  Pater,  to  whose  various 

1  A  certain  latter-day  enhancement  of  the  supreme  power  of  Zeus  is 
one  of  the  interesting  differences  that  distinguish  Greco-Roman  from  early 
Greek  religion.  This  was  but  the  natural  result  of  the  political  preponder 
ance  of  Rome  and  the  Theocrasy  or  commingling  of  heterogeneous  gods 
taken  in  conjunction  with  the  new  place  made  for  imperial  ideas  in  the  reli 
gious  service  of  the  empire.  No  doubt  philosophy  and  the  clearly  thought- 
out  belief  in  one  supreme  power,  to  which  so  many  leaders  of  later  Pagan 
thought  gave  utterance,  also  played  its  part.  To  Jupiter  or  Zeus,  as  the 
titular  representatives  among  the  traditional  gods  of  this  supreme  maker 
and  orderer  of  the  universe,  universal  prayers  were  made.  It  must,  how 
ever,  be  remembered  just  here  that  we  are  prone  to  read  into  the  religion 
of  the  ancients  something  of  our  own  clear-cut  notion  about  an  indisputably 
supreme  author  of  all  being. 


i  PATER,  MANNHARDT,  AND  PRELLER  27 

essays  l  I  earnestly  refer  for  much  that  enlightened  me  in 
the  preparation  of  these  lectures.  To  him,  and  also  to  the 
well-known  book  of  Preller,  and  to  essays  by  William 
Mannhardt,2  that  deserve  to  be  better  known  than  they 
are,  I  desire  to  make  especial  acknowledgment. 

In  his  first  essay  on  the  Myth  of  Demeter  and  Perse 
phone,  Mr.  Pater  draws  to  his  close  with  words  for  which 
I  claim  a  wider  application  than  he  gives  to  them.  After 
truly  saying  that  "  there  is  a  certain  cynicism  in  that  over- 
positive  temper,  which  is  so  jealous  of  our  catching  any 
resemblance  in  the  earlier  world  to  the  thoughts  that  really 
occupy  our  minds,  and  which,  in  its  estimate  of  the  actual 
fragments  of  antiquity,  is  content  to  find  no  seal  of  human 
intelligence  upon  them,"  he  speaks  of  the  theory  of  com 
parative  mythology  and  of  the  specific  and  most  helpful 
doctrine  or  theory  of  animism.3  "  Only,"  he  adds,  in  the 

1  Two  essays  on  "The  Myth  of  Demeter  and  Proserpina"  in  the  Fort 
nightly  for  January  and  February  1876,  and  in  December  of  the  same  year, 
"A  Study  of  Dionysus."  This  last  is  completed  by  an  essay  on  "The 
Bacchanals  of  Euripides,"  published  in  Macmillari  s  Magazine  for  May  1889. 

-  Mythologische  Forschungen,  posthumously  published  by  H.  Patzig  in 
1874. 

3  Since  the  preparation  and  delivery  of  these  studies  as  lectures  this 
whole  subject  has  been  elaborated  in  Mr.  Frazer's  Golden  Bough,  which  is 
a  treasure-house  of  information  in  regard  to  primitive  religious  customs. 
Especial  attention  is  there  given  to  customs  and  stories  which  embody  this 
doctrine  of  animism.  As  a  matter  of  course  the  elements  of  especial  interest 
in  Greek  myths  as  such  reach  immeasurably  above  and  beyond  any  traces 
of  primitive  religion  or  fetichism  discernible  in  their  beginning.  Still, 
since  the  absence  of  a  right  account  means  almost  inevitably  a  wrong 
account  of  these  beginnings,  Mr.  PYazer's  book  is  relevant  to  the  study  of 
Greek  mythology  and  religion.  His  readers  are,  however,  in  serious 
danger  of  thinking  otherwise  because  the  centre  of  gravity  in  his  Golden 
Bough  falls  beyond  its  base.  The  picturesque  but  comparatively  unim 
portant  rite  of  the  Arician  Grove  is  no  proper  nucleus  for  the  important 
material  which  Mr.  Frazer  has  gathered  into  his  book. 

Much  light  is  thrown  upon  various  questions  discussed  below  in  another 
and  most  welcome  publication,  The  Monuments  and  Mythology  of  Ancient 
Athens,  by  Miss  Harrison  and  Mrs.  Verrall.  I  can  only  regret  that  I  had 
not  the  great  advantage  of  using  both  these  books  in  preparing  my  lectures. 


28  INTRODUCTION  i 

application  of  these  theories,  "the  critic  must  not  forget 
that  after  all  it  is  with  poetry  he  has  to  do.  The  abstract 
poet  of  that  first  period  of  mythology,  creating  in  this 
wholly  impersonal,  intensely  spiritual  way,  —  the  abstract 
spirit  of  poetry  itself,  rises  before  the  mind,  and  in  speaking 
of  this  poetical  age  the  critic  must  take  heed  before  all 
things  not  to  offend  the  poets." 

The  poets,  then,  and  Apollo,  or  the  personified  spirit  of 
poetry,  form  our  court  of  final  appeal  which  sits  upon  the 
loftier  peak  of  Parnassus,  and  judges  all  matters  of  vital 
concern  to  the  gods  in  Greece  and  to  Greek  religion. 
With  this  proviso  it  may  be  said  that  Apollo's  was  the  only 
authority  which  really  swayed  Olympus.  When,  however,  a 
more  extended  power  ove'r  all  the  other  gods  is  attributed 
to  Apollo,  the  fact  becomes  so  nearly  a  fact  of  poetry,  that 
the  statement  of  it  in  prose  almost  deprives  it  of  its  truth. 
Let  there  be,  then,  an  appeal  to  some  poet.  Hear  an  echo, 
a  translation  from  the  sweetest  strains  divine  of  poet 
Aristophanes : — 

"Come  to  me,  partner  mine,"  sings  the  hoopoe  to  the 
nightingale,  "cease  from  slumbers,  unloose  the  flights  of 
sacred  songs,  that  through  thy  lips  divine  dost  wail,  for 
mine  and  thine,  for  Itys  of  many  tears  trilling  and  shrilling 
in  the  liquid  melodies  of  thy  tawny  throat.  Pure  ascends 
through  the  greenwood  thicket  their  echoing  refrain  even 
unto  Zeus's  throne,  where  golden-haired  Phoebus,  giving 
attentive  ear  and  making  responsive  music  to  thy  mournful 
lays,  upon  his  lyre  of  ivory  wrought,  marshalls  the  dances 
of  the  gods.  Lo  !  from  deathless  lips  proceed  the  while 
concordant  with  thy  strains  most  heavenly  acclamations  from 
the  blessed  gods."  Here  was  no  place  for  father  Zeus  to 
interfere ;  like  all  the  other  gods,  he  too  obeyed  Apollo,  and 


I  SANCTIONS  OF  APOLLO'S  POWER  29 

followed  after  Phoebus,  leader  of  the  dance.  Delphian 
Apollo  was  mightier  in  song  and  in  prophetic  wisdom  than 
even  Zeus  himself.  The  poet's  poet-god  wielded  the  sceptre 
of  poetry  and  gave  his  law  to  all  the  gods  in  Greece.1 

After  all  is  said  and  done  such  rule  and  right  to  guide 
as  attached  to  Apollo  among  other  gods  belonged  to  him 
by  divine  right  of  righteousness,  and  has  the  final  sanction 
of  a  sense  of  tolerance  and  fair  dealing  conspicuous  in  the 
justice  of  Apollo's  acts  and  the  generosity  of  what  he 
abstained  from  doing.  His  rule  was  based  upon  a  truly 
poetical  sense  of  right  and  wrong.  Had  he  not  been 
generous  and  broadly  tolerant  of  powers  and  pretensions 
which  prosaic  minds  and  gods  of  prose  would  certainly 
have  resented  and  opposed,  he  never  could  have  prevailed 
at  Delphi.  It  was  this  supremely  poetical  quality  in  the 
Delphian  god,  his  possession,  so  to  speak,  of  imagination, 
which  enabled  him  serenely  to  contemplate  and  wisely  to 
further  the  welldoing  of  other  divinities  and  of  various 
worships  often  seemingly  the  rivals  of  his  own.  The  best 
instance  of  this  Delphian  tolerance  of  Apollo  is  in  the 
union  of  Apollo  and  Dionysus  at  Delphi  itself,  and  in  the 
cordial  and  useful  support  given  by  Apollo's  Delphian 
oracle  to  the  propagation  and  elaboration  of  Dionysus 
worship  elsewhere,  particularly  in  Attica.  Like  Apollo, 
Dionysus  was  a  poet-god  and  a  giver  of  oracles,  an  inspirer 
of  the  souls,  and  a  possessor  of  the  bodies  of  men.  And  yet 

1  This  is  a  very  different  primacy  from  that  primacy  of  fear  attributed 
to  Apollo  in  the  Homeric  Hymn.  The  difference  may  serve  as  a  measure 
of  the  advance  in  nobility  of  religious  thought  made  by  the  Greeks  under 
the  leadership  of  great  and  deeply  religious  thinkers  like  Euripides,  Plato, 
and  Pindar.  And  yet  something  of  the  later  strain  of  Apollo  is  heard  in 
the  prayer  of  Glaucus,  Iliad,  xvi.  514  ff.  :  "Hear,  O  Lord,  who  art 
somewhere  in  Lycia's  rich  land  or  in  Troy  ;  for  thou  canst  hear  in  every 
place  when  a  man  is  in  grief  such  as  now  is  the  grief  that  is  on  me." 


30  INTRODUCTION  i 

Dionysus  and  Apollo  went  hand  in  hand  through  all  the 
length  of  Hellas. 

Another  way  of  stating  the  case  would  be  to  say,  as  has 
recently  been  done,  and  most  truly,  that  one  great  reason 
for  the  prosperity  and  renown  of  the  oracle  and  temple  at 
Delphi  was  the  cleverness  shown  by  Apollo's  priests  in 
combining  and  maintaining  with  equal  hand  the  various 
cults  of  various  divinities  that  centred  there.  But  this  way 
of  counting  those  who  may  in  some  sense  have  been  wire 
pullers  as  wire-pullers  only,  of  counting  their  manoeuvres 
for  everything,  and  the  reality  of  the  cause  for  which  their 
work  was  done  for  nothing,  leads  nowhere.  It  is  profitless, 
because  falsehood  always  lurks  in  the  reasoning  of  those 
who,  from  the  heights  of  imagined  superiority,  look  down 
upon  the  great  religion  of  a  great  epoch  in  the  world's 
history,  deeming  it  a  sheer  delusion  through  and  through. 
Thank  Heaven  !  we  can  let  the  eighteenth  century  have  all 
for  its  own  that  canting  talk  about  the  "trickery  of  priests." 
No  vital  religious  fact  was  ever  materially  affected  by  the 
trickery  of  priests,  and  it  is  no  accident  therefore,  but  a 
deeply  significant  fact  in  the  course  of  Greek  mythology 
and  the  history  of  Greek  religion,  that  Apollo  and  Dionysus, 
the  sublime  and  the  intense,  dwelt  together  in  unity  before 
the  eyes  of  those  who  came  and  worshipped  at  the  moun 
tain  shrine  of  Delphi. 

"Apollo,  ivy-god  and  prophet  bacchanal"1  cries  Aeschylus, 
sublimest  of  the  singers  at  Dionysus'  Attic  theatre,  giving 
to  Apollo  the  characteristic  insignia  of  Dionysus.  "Lord 
Bacchus ,  lover  of  the  laurel  tree"*1  says  Euripides,  lending 

1  Fr.  394,  cf.  Macrobius,  Saturn.  18,  6. 

2  Macrobius,    ibid.  :    Euripides   in    Licymnio,   Apollinem  Liberumque 
unum    eundemque    deum    esse   significans,    scribit    "  dtcnrora    <j>i\6da.(pve 
Bd/c%e,  Ilcuaj'  "AirdXXov 


i  APOLLO'S  TEMPLE  AT  DELPHI  31 

Apollo's  sacred  laurel  bough  to  Dionysus  for  the  nonce. 
Traditions  kept  alive  in  far-away  places  show  the  brother 
hood  of  these  two  gods  of  poetry.  In  one  place  record 
was  preserved  of  it  by  a  worship  of  Apollo  under  the 
special  epithet  of  "one  sent  by  Dionysus."1  In  popular 
pictures,  such  as  decorated  vases,  the  ivy  often  crowns  not 
Dionysus  but  the  slenderer  Apollo.  The  Muses,  repre 
sented  as  Apollo's  attendants  upon  the  front  pediment  of 
his  great  Delphian  temple,  are  frequently  given  in  popular 
pictures  to  be  the  companions  of  Dionysus,  who  also  bor 
rows  very  often  his  brother  Apollo's  lyre.  To  close  these 
instances  with  the  strongest  proof  of  the  good  fellowship 
and  mutual  tolerance  between  them  as  conceived  by  their 
worshippers,  consider  the  western  pediment  or  gable  of 
the  great  temple  of  Apollo  just  before  mentioned.  To 
correspond  to  Apollo  and  the  Muses  of  the  other,  this 
pediment  presented  Dionysus  and  his  Thyades,  his  maenad 
band  of  bacchanalian  women.  The  temple  being  that  of 
Apollo,  Dionysus  still  could  be  made  most  prominent,  since, 
like  Apollo,  he  was  an  inspirer  of  song. 

Many  other  reasons,  but  especially  the  date  of  this 
Delphian  temple,  built  in  the  middle  of  the  sixth  century 
B.C.  by  Spintharus  of  Corinth,  indicate  how  early  this 
bond  of  brotherhood  between  Apollo  and  Dionysus 
received  conspicuous  sanction  from  the  Delphian  priests 
and  in  Greece  at  large.  Within  the  temple,  just  in  the 
Holy  of  Holies,  where  the  golden  statue  of  Apollo  stood, 
was  a  tomb  of  Chthonian  Dionysus,  not  far  from  the 
rounded  stone  that  marked  the  absolute  centre  or 
"navel"  of  the  earth.  This  last  was  flanked  by  two 
golden  eagles,2  for  it  was  well  known  that  Zeus  sent  forth 

1   At ovvffb Soros,  Pausanias,  I.  xxxi.  4.  2  Appendix  XI.  i.  126. 


32  INTRODUCTION  i 

two  of  his  own  imperial  birds — one  from  the  north,  the 
other  from  the  south, — and  the  fact  of  their  meeting  just 
in  this  spot,  and  perching  on  either  side  of  this  parti 
cular  stone,  witnesses  that  Delphi  is  at  the  centre  of  the 
world.  This  original  meeting  was  commemorated  by  the 
two  golden  eagles  set  up  upon  the  spot  and  sanctified 
to  Apollo.  Another  feature  of  this  shrine  that  goes 
to  prove  that  it  was  no  ordinary  sanctuary  of  Apollo, 
but  rather  a  meeting  ground  for  many  worshippers  of 
many  divinities,  was  an  altar  to  Poseidon,  the  shaker  of 
the  earth,  which  was  anciently  erected  and  always  main 
tained. 

Such  points  as  these,  and  others  to  be  gained  from 
Pausanias'  description  of  the  great  Delphian  temple,  show 
how  much  may  be  learned  from  excavations  on  this  site. 
To  make  excavations  at  Delphi  will  be  a  glorious  task  for 
any  to  whom  it  may  be  allotted,  and  would  indeed  be  a 
fitting  continuation  of  the  work  which  our  countrymen, 
inspired  and  directed  in  those  days  by  Dr.  Merriam,  a 
scholar  of  whose  great  attainments  and  sound  judgment 
America  is  proud,  have  already  done  among  the  Attic 
mountains  at  Icaria.  But  the  friends  of  the  American 
School  of  Classical  Studies  at  Athens  have  not  forgotten 
that  it  is  the  youngest  but  one  of  the  four  schools  there 
established.1  Therefore  they  will  not  sorrow  but  rejoice  if 
the  first  established  of  all  schools  at  Athens,  the  French 
School,  with  its  well  established  traditions  and  a  liberal 
grant  from  the  Government,  carries  out  work  so  well  begun 

1  Rumour  has  it  that  the  Italians  are  about  to  add  theirs  to  the  four 
established  already.  Those  who  are  familiar  with  the  organisation  of 
excavations  in  Italy  and  know  the  Italian  system  of  local  reports  will  under 
stand  the  gain  to  Greek  archaeology  which  an  Italian  School  at  Athens 
must  bring. 


I  HISTORY  OF  DELPHIAN  WORSHIP  33 

at  Delphi  by  M.  Homolle  and  M.  Foucart.1  What  could 
not  be  done  with  a  sufficient  grant  of  money  by  a  School 
that  accomplished  with  next  to  no  money  at  all  the  excava 
tions  and  investigations  at  Delos  which  have  made  us  all 
M.  Homolle's  debtors  ?  2 

A  disentangling  of  the  relations  between  Apollo  and  the 
other  Delphian  gods,  some  of  whom  seem  to  have  pre 
ceded  him  and  to  have  been  eclipsed  by  his  arrival,  will 
perhaps  be  possible  in  the  future.  But  this  can  only  be 
when  much  work  upon  the  site  shall  have  yielded  many 
new  inscriptions.  With  only  such  knowledge  as  is  now 
available,  contained,  be  it  said,  in  an  admirable  paper 
recently  published  by  Professor  Middleton  in  the  Journal 
of  Hellenic  Studies?  it  seems  possible  to  say  little  with 
positiveness.  It  is  not,  however,  rash  to  declare  even  now 
that  the  terms  upon  which  Apollo's  worship  finally  obtained 
supremacy  at  Delphi  are  likely  to  have  enlarged  the  final 
range  of  his  influence.  The  compromises  involved  in  his 
first  coming  no  doubt  begot  in  him  a  wide  and  tolerant 
strain. 

With  the  earlier  history  of  worship  at  Delphi  is  bound  up 
the  growth  and  increase  of  the  great  power  that  made  for 
order  in  Olympus  and  began  to  bring  into  the  religious 
ideas  of  Greece  a  spirit  of  reasonableness  if  not  of  logic. 
Just  as  the  highest  ideal  of  poetry,  the  work  of  a  poet's  poet 
like  Dante,  presents  the  universe  as  an  ordered  whole,  so 
the  highest  and  really  most  supreme  divinity  in  that  poetry 
of  poetry,  Greek  religion,  will  be  Apollo  on  Parnassus,  the 
poet's  god  of  poetry,  seeking  to  organise,  to  make  reason- 

1  Foucart,  Ruines  et  Histoire  de  Delphes,  1865. 

2  See  chap.  viii. 

3  See  the  Journal  for  1888,  vol.  ix.  p.  282. 

D 


L 


34  INTRODUCTION  i 

able,  and  justify  the  worship  and  the  ways  of  all  the  gods  in 
Greece,  and  to  present  the  world  of  Olympus  as  an  ordered 
whole. 

This  was  accomplished  chiefly  by  oracular  responses. 
A  constant  interchange  of  influence  is  perceptible  in  the 
relations  between  Delphi  and  Athens.  When  the  hitherto- 
despised  Dionysus -worship  was  brought  into  honour  at 
Athens  and  no  longer  hidden  in  the  country  denies,  the 
influence  of  the  Delphian  oracle  of  Apollo  was  one  of  the 
determining  forces  that  wrought  the  change.  It  is  certain 
that  the  Delphian  oracle  sanctioned  and  promoted  just  at 
this  time  an  additional  worship  of  Dionysus  not  known  of 
old  to  the  country  denies  of  Attica.  Under  this  new  aspect 
from  abroad,  Dionysus  was  known  as  the  god  of  Eleutherae, 
a  town  on  the  frontier  towards  Boeotia.  His  worship  was 
characterised  by  moderation,  and  Pegasus,  his  high  priest 
of  Eleutherae,  is  associated  with  the  practice  of  tempering 
the  strength  of  wine  with  water.  Accordingly  the  Dionysus 
of  Eleutherae  was  not  the  awful  Dionysus  of  the  nether 
world,  not  the  "angel  of  the  darker  drink,"  but  Dionysus 
the  Saviour,  who  came  to  show  men,  tired  and  dazed  by  his 
orgies,  how  they  might  make  themselves  clear -eyed  once 
more  and  have  untroubled  hearts  as  they  betook  themselves 
again  to  their  wonted  avocations. 

By  such  a  mitigation  of  the  more  outrageous  features  in 
the  rude  and  early  Attic  worship  of  Dionysus  did  Apollo 
repay  that  god  who  had  made  place  for  him  when  he  came 
to  Delphi.  Of  that  coming  Euripides  gives  a  beautiful 
picture  in  his  Taurian  Iphigenia  :  "  Sweet  was  the  babe  of 
Leto  born,  Phoebus,  a  god  with  golden  hair.  Borne  by 
Leto,  on  he  came  unto  Parnassus,  whose  peak  leaps  in  the 
bacchanalian  dance  that  honours  Dionysus.  There  of 


THE  COMING  OF  APOLLO  35 

mottled  hue  and  glance  wine -flashing  lurked  a  dragon, 
shaded  by  laurel  leafage ;  sheathed  as  in  brass,  Earth's 
monstrous  portent  guarded  the  seat  of  nether -world 
prophecy.  Him  didst  thou  slay,  a  mere  babe  though  thou 
wert,  Phoebus,  and  didst  enter  in  to  possess  it  the  seat 
of  oracles  most  divine,  and  now  thou  art  throned  on  thy 
tripod  all-golden,  even  thy  throne  unacquainted  with  false 
hood,  rendering  there  unto  mortals  thy  prophecies  that 
ascend  from  beneath  the  divine  Holy  of  Holies  close 
to  the  streams  of  Castaly,  in  thy  house  at  the  midpoint 
of  earth."  The  dragon  slain,  here  alluded  to,  is  the  Pythian 
monster.  Him  and  all  the  oracles  rendered  by  earth 
at  Delphi  Apollo  caused  to  disappear  by  the  irresistible 
power  of  his  coming.  All  that  was  antagonistic  to  Apollo 
Euripides  here  looks  upon  as  evil.  Perhaps  he  thought 
of  it  as  embodying  all  the  unpitying  relentlessness  of  the 
earlier  and  inhuman  phase  of  Greek  religion,  against  which 
his  own  poems  are  a  dramatised  protest.  The  good  that 
existed  in  local  rites  was  not  affrighted  by  Apollo's  coming. 
The  dawning  sun-god,  lately  born  of  Leto  on  that  miracle 
of  the  Grecian  seas,  the  holy  isle  of  Delos,  could  banish 
none  of  the  powers  of  light ;  only  darkness  fled  before  his 
rising. 

This  coming  of  Apollo  to  Delphi,  this  dawning  of  the 
light  in  which  we  see  revealed  the  highest  and  the  best 
that  worship  could  inspire  in  Greeks,  and  wherein  we  learn 
to  know  the  loftiest  characters  and  characteristics  among 
the  gods  in  Greece,  may  fitly  be  associated  with  another 
song  of  Euripides,  sung  by  Ion  at  the  open  door  of  his 
father  Apollo's  Delphian  temple  : 

"  Lo  from  his  gleaming  chariot  drawn  by  coursers  four 
the  sun  now  flashes  light  far  down  to  earth ;  the  stars  in 


36  INTRODUCTION  i 

flight  are  swiftly  plunged  into  holy  night  by  the  fires  of 
day.  Parnassus'  peaks  untrodden,  bathed  in  its  radiance, 
receive  for  men  below  this  wheel  of  day.  Meanwhile  the 
smoke  of  parched  frankincense  and  myrrh  wings  its  way 
upward  to  Phoebus'  roof.  Yea,  and  a  woman  on  the  thrice 
hallowed  tripod  is  sitting,  the  Delphian  one,  singing  forth 
such  sounds  for  Greece  even  as  Apollo's  voice  proclaims." 

This  tripod  at  Delphi  was  the  symbol  of  Apollo's  prim 
acy  on  earth ;  at  Athens  and  in  Attica  the  same  tripod  was 
awarded  as  the  victor's  prize  in  the  tragic  and  the  dithyram- 
bic 1  contest.  The  winner  always  consecrated  it  to  Dionysus. 
Thus  may  the  tripod,  so  constantly  present  on  the  Delphian 
coins  and  in  all  manner  of  Greek  religious  pictures,  stand 
for  one  of  the  most  vital  facts  in  the  Greek  world :  the 
unison  of  Apollo  and  Dionysus  in  concordant  rule  upon  the 
double  peak  of  Delphian  Parnassus. 

1  It  was  certainly  awarded  for  tragic  victory  at  Icaria,  and  as  certainly 
for  dithyrambs  at  Athens,  where  it  was  probably  also  given  for  tragedy. 


APPENDIX   I 
THE  DEIFICATION  OF  ROMAN  EMPERORS 

ROMAN  imperialism  has  not  usually  been  judged  upon  its 
merits.  Perhaps  this  would  have  been  otherwise  if  Julius 
Caesar,  the  first  and  in  many  ways  the  greatest  of  the 
emperors,  had  lived  longer.  But  his  heir  Augustus  was  a 
man  of  other  mould.  His  whole  effort  was  to  persuade 
Rome  and  the  Romans  that  their  worn-out  commonwealth 
and  all  its  antiquated  simplicity  of  religion  was  still  surviv 
ing.  He  wished  to  be  supreme  without  seeming  so,  to 
govern  but  not  to  reign.  This  masquerading  scheme  had 
a  marvellous  success,  and  here  is  one  reason  why  the  new 
religious  sanction  of  Roman  imperialism,  the  deification  of 
the  emperors,  has  not  as  yet  been  very  generally  under 
stood  as  it  deserves  to  be.  The  senate,  before  Julius 
Caesar  died,  ordered  the  institution  of  worship  in  his  honour  ; 
and,  if  the  report  of  Dio  and  Zonaras  were  considered 
more  than  a  misconception  of  Cicero's  mocking  allusion, 
they  styled  him  Jupiter  Julius.1  However  that  may  be, 

1  See  Dio  Cassius  (44,  6),  for  the  completes!  account  :  KCU  rAos  Ala 
e  avrbv  dvTiKpvs  'Iov\iov  irpoffi^ydpevaav,  Kal  vabv  ai/ry  T?7  T' 


avrov  Tefj.evKTdTJva.1  tyvwffw,  iep^a  ff<pt(ri  rbv  'AVT&VLOV,  ticrirep  nva.  did\iov, 
•rrpoxfi-piffdfjt.€voi  ;  Zonaras  (x.  ch.  12,  p.  492  A-C)  brings  in  as  a  climax 
to  his  long  list  of  honours  voted  and  given  to  Caesar  while  he  yet  lived 
Ata  re  avrbv  'lotfXioj/  Trpoff-rjybpevffav.  Cf.  Cicero  (Phil.  ii.  43,  no)  — 
Est  ergo  flamen  ut  Jovi,  ut  Marti,  ut  Quirino,  sic  Divo  Julio  M.  Antonius? 
Cf.  Phil.  xiii.  19,  41  —  Cujus,  homo  ingratissime,  flaminium  cur  reliquisti  ? 
See  Suetonius,  Caesar,  76.  Since  Leunclavius'  and  Fabricius"  notes  on 


38  APPENDIX  I 

Mark  Antony  was  nominated  to  be  his  flamen  or  master 
of  sacrifice.  But  then  came  Augustus  deprecating,  so  far 
as  he  was  personally  concerned,  the  establishment  of  temples 
for  the  new  imperial  worship,  all  but  forbidding  it  in  Rome 
and  barely  permitting  it  elsewhere.  He  deprecated  so  con 
spicuous  a  religious  innovation  in  the  full  glare  of  publicity 
at  Rome,  but  apparently  did  all  in  his  power  to  extend  a 
similar  worship  in  dark  corners  of  Rome  itself1  and  in 
various  parts  of  the  Roman  empire.  So  successful  was  this 
policy  of  artfully  dissimulating  the  new  and  artificially  reviv 
ing  the  old  cults  that  many  of  the  important  sources  for 
understanding  the  deification  of  the  emperors  are  outside 
of  the  known  literature  of  imperial  days.  Obscure  and 
fragmentary  inscriptions  have  to  be  appealed  to.  Many  of 
the  great  men  of  imperial  administration,  including  some  of 
the  emperors  themselves,  found  it  difficult  always  to  take 
the  new  religion  seriously;  it  is  therefore  not  surprising  that 
men  of  another  day  and  generation  should  pass  it  by  un 
heeding. 

And  yet,  if  a  close  connection  between  religion  and 
morality  can  be  taken  for  granted,  a  new  religion  was  required 
to  give  sanction  to  the  new  morality  of  imperial  days,  and 
this  religion  finds  expression  in  the  Augustan  poets.  In  a 

the  passage  in  Dio  above  cited,  the  fashion  has  been  to  ignore  it.  Scholars 
have  dispensed  themselves  of  the  trouble  required  to  sift  the  testimony  of 
Dio,  rejecting  it  summarily  as  coming  too  long  after  the  facts.  But  see 
Dr.  R.  Wilmans,  De  Dionis  Cassii  fontibus,  etc.,  Berlin,  1836,  pp.  24 
and  25.  Speaking  of  the  acta  publica,  Dr.  Wilmans  says:  "Ex  hoc 
igitur  fonte  multae  apud  Dionem  derivandae  sunt  narratiunculae. "  Dio 
took  his  point  about  Jupiter  Julius  from  the  acta  publica  no  doubt.  On 
such  a  point  Cicero  could  not  afford  to  be  explicit,  therefore  he  was 
ironical.  See  also  Hugo  Grohs,  Der  Wert  des  Geschichtswerkes  des  Cassius 
Dio,  etc.,  Ziillichan,  1884.  Livy  neglected  daily  events  that  happened  in 
Rome.  Dio's  merit  lies  in  his  account  of  these,  "  fiir  die  Internet  sah  er 
noch  (besides  what  Livy  notes)  die  Geschichte  Sueton's  und  die  acta  publica 
an. "  It  is  evident  that  his  account  of  honours  voted  to  Caesar  is  an  addi 
tional  proof  that  he  took  pains  about  daily  events  at  Rome. 

1  The  genius  of  Augustus  was  associated  with  the  Lares  Compitales. 
See  Marquardt  (Staatsverwaltung,  iii.  p.  199),  who  refers  to  Ovid  (F.  v. 
145)  and  Horace  (Od.  iv.  5,  34),  and  for  a  similar  worship  of  the  Genii  of 
later  emperors  to  inscriptions. 


THE  DEIFICATION  OF  ROMAN  EMPERORS         39 

lecture  before  the  Royal  Institution  on  Roman  Imperialism, 
Professor  Seeley  contrasts  the  Republican  and  the  Im 
perial  ideals  of  conduct  as  follows  :  "  Men  ceased  to  be 
adventurous,  patriotic,  just,  magnanimous ;  but,  on  the 
other  hand,  they  became  chaste,  tender-hearted,  loyal, 
religious,  and  capable  of  infinite  endurance  in  a  good 
cause."  They  cultivated  the  virtues  of  the  pious  Aeneas. 
Even  though  the  details  of  these  contrasted  catalogues  of 
virtue  may  not  be  to  everybody's  mind,  the  fact  of  a  changed 
standard  must  be  admitted ;  and  consequently  the  religious 
alterations,  the  distinctively  imperial  innovations  in  worship, 
should  be  scrupulously  investigated  and  carefully  pondered. 

From  Virgil,  and  also  from  Horace,  Ovid,  and  others,  we 
may  learn  of  the  new  ideals  of  this  wider  and  broader  day 
which  transformed  even  traditional  religion  in  the  Roman 
dominions.  The  hearts  of  the  subjects  of  imperial  Rome, 
the  hopes  of  the  Roman  proletariat,  were  centred  not  so 
much  in  the  old-time  Roman  religion  as  in  the  new-come 
reign  of  peace.  The  emperors  could  not,  if  they  would, 
escape  the  homage  of  their  subjects.  It  was  the  part  of 
wisdom  not  to  stifle  but  to  guide  this  spontaneous  zeal, 
this  uplifting  of  grateful  hearts  toward  the  ideal  of  a  bene 
ficent  and  omnipotent  imperial  fatherhood.  Those  ancient 
Caesars  could  as  little  escape  such  a  worship  as  can  the 
modern  Caesars  of  Russia.  Therefore  it  was  well  to  bind 
up  with  the  new  worship  the  religious,  social,  and  political 
life  of  various  orders  and  classes,  particularly  that  of  the 
lower  and  most  numerous  class,  which  was  more  or  less 
unprovided  for  by  traditionally  existing  religious  usage  and 
ceremonial.  Certainly  a  beginning  of  social  and  religious 
life  was  absolutely  needed  for  those  whom  Republican  Rome 
had  left  in  outer  darkness.  Without  the  part  assigned  in 
imperial  services  to  freedmen  and  small  tradesmen  the 
empire  would  never  have  been  in  so  advantageous  a  posi 
tion  for  reaping  the  benefits  of  Christianity  as  it  really  was 
when  the  critical  moment  arrived. 

Let  us  view  this  imperial  service  in  its  relation  (i)  to 


40  APPENDIX  I 

the  earlier  religion  of  Greece  and  Rome ;  (2)  to  the  new 
political  and  religious  needs  of  the  hour ;  and  (3)  to  the 
political,  social,  and  religious  needs  of  a  new  class  of  people, 
i.e.  of  a  class  of  people  who  had  hitherto  been  almost  com 
pletely  ignored. 

It  is  certain  that  the  notion  of  deifying  the  emperor  or 
any  human  being  must  have  been  rooted  in  previous  habits 
of  mind.  No  flattery,  however  base,  could  on  the  spur  of 
the  moment  have  invented  just  this  form  of  homage  with 
any  chance  of  securing  its  adoption.  The  fact  is  that  it 
was  not  the  work  of  clever  men.  They  had  to  set  their 
hands  to  it  in  obedience  to  a  popular  impulse  which  they 
were  too  clever  to  withstand.  No  one  supposes  that  the 
Senate  would  have  done  homage  to  Jupiter  Julius  of  their  own 
accord.  The  burst  of  popular  admiration  and  gratitude  which 
in  Greece  required  that  divine  honours  should  be  paid  to 
Flamininus,1  was  analogous  to  the  enthusiasm  felt  by  the 
Roman  populace  for  their  benefactor  Caesar.  Precedents 
therefore  must  be  sought  in  the  religion  of  the  people.  The 
time-honoured  worship  of  the  genius  of  the  Roman  people  or 
of  Rome  had  always  appealed  especially  to  the  people,  and 
this  was  naturally  and  promptly  associated  with  and  finally 
passed  into  a  worship  of  the  emperor.  Was  not  he  their 
good  genius?  the  people  asked.  Even  Augustus  allowed 
himself  to  be  worshipped  by  circumlocution  as  "  the 
clemency  of  Augustus,"  and  throughout  the  empire,  if  not 
in  Rome  itself,  were  erected,  with  his  officially  reluctant 
sanction,  altars  and  shrines  for  Rome  and  Augustus  like  the 
one  on  the  Athenian  Acropolis. 

This  Athenian  homage  may  serve  to  recall  the  history 
of  deification  in  Greece,  which  can  be  read  plainly  and  had 
run  a  long  course  before  the  days  of  the  Caesars  at  Rome. 

In  the  middle  of  the  Peloponnesian  war  a  gallant  Spartan 

1  See  his  life  by  Plutarch  (chap,  xvi.),  where  mention  of  a  survival  of 
this  worship  down  to  Plutarch's  time  is  made.  It  is  interesting  to  note 
that  this  deification  of  Flamininus  was  by  the  Greeks  of  Chalcidice,  near 
neighbours  of  the  Amphipolitans  who  long  before  paid  divine  honours  to 
Brasidas. 


THE  DEIFICATION  OF  ROMAN  EMPERORS          41 

soldier,  Brasidas,  died  in  Thrace  while  defending  Amphipolis 
from  the  attacks  of  Athens.  The  enthusiastic  Amphipolitans 
put  Brasidas  in  the  place  of  their  Athenian  founder  Hagnon, 
ordering  an  altar  to  be  dedicated  to  him,  and  bestowing 
upon  him  the  other  quasi-divine  honours  usually  given  by 
Greek  colonists  to  the  founder.  What  a  hold  was  gained  by 
this  manner  of  testifying  to  the  great  qualities  of  a  con 
temporary  is  shown  by  the  deification  of  Lysander,  which 
took  place  at  the  end  of  this  same  war.  The  novelty  here 
consists  in  the  fact  that  Lysander  received  sacrifices  and  all 
the  rest  of  it  while  he  was  yet  living.  Brasidas,  on  the 
other  hand,  had  died  before  Amphipolis  worshipped  him. 
Thus  long  before  the  Ptolemies  and  the  days  of  Roman 
imperialism  the  Greeks  in  Asia  had  capped  the  climax  of 
apotheosis  for  Lysander.  So  far  Rome  did  not  easily  go.  It 
was,  in  fact,  so  little  habitual  at  any  time  to  deify  a  living 
emperor  that  the  bare  proposal  was  treated  as  involving  his 
"  promotion  into  the  next  world."  There  was  a  moderating 
common-sense  at  Rome  which  kept  this  custom — half  Greek 
and  wholly  Oriental — within  certain  bounds,  and  associated 
it  with  the  reasonable  and  popular  worship  of  the  genius  of 
Rome. 

Such  were  the  Greek  and  Roman  possibilities  of  which 
imperial  apotheosis  was  the  enhancement  and  the  realisa 
tion.  Now  a  word  may  be  said  of  the  new  religious  needs 
to  which  this  apotheosis  gave  a  measure  of  satisfaction. 
These  new  needs  were  felt  alike  by  the  higher  and  lower 
orders  in  the  empire,  though  by  the  latter  most  keenly  and 
consistently.  Quintilian,  Tacitus,  and  Pliny  may  fairly 
represent  the  higher  orders.  They  stood  aloof  from  the 
popular  religious  point  of  view,  and,  like  many  who  took 
refuge  in  Stoicism  or  Epicureanism,  rejected  much  if  not 
most  of  the  mythology  in  which  the  popular  mind  still 
found  a  religious  satisfaction.  It  is  curious,  in  spite  of  all 
this,  to  note  the  way  in  which  Tacitus  reports  a  miracle 
performed  by  Vespasian.  He  really  seems  to  be  willing, 
for  a  moment  at  least,  to  recognise  a  supernatural  power  in 


42  APPENDIX  1 

the  emperor.1  Quintilian,  without  having  a  systematised 
philosophy  of  his  own,  talks  of  a  god  who  is  the  "  father 
and  contriver  of  the  world,"  one  who  "administers"  the 
universe.2  Plainly  the  emperor  ruling  the  Roman  world  is 
the  prototype  in  this  case.  Quintilian  does  not  think  of 
the  emperor  as  god,  but  he  thinks  of  god  as  the  emperor.3 
Pliny,  the  sceptical  naturalist,  was  especially  proud  of 
being  superior  to  popular  religion.  Few  things  awake  his 
enthusiasm,  and  his  usually  limping  prose  takes  sudden 
wings  only  for  a  moment  when  he  soars  toward  his  god 
made  manifest,  the  shining  sun,  "mind  of  the  universe."4 
But  Pliny  himself  has  not  wholly  escaped  the  religious 
contagion  of  his  time.  His  thoughts  constantly  hover 
around  the  person  of  the  emperor ;  his  illustrations  and 
explanations  are  always  bringing  the  emperor  in.  He  calls 
Nero  the  "  foe  of  mankind,"  and  mentions,  as  it  were  with 
bated  breath,  that  he  came  into  the  world  feet  foremost.5 
Again,  after  rejecting  various  superstitions,  he  exclaims — 
"  The  help  that  man  lends  to  man  is  god  ;  this  is  the  way  of 
glory  eternal.  This  is  the  way  taken  by  Roman  worthies 
of  old,  and  this  way  with  heavenly  step  now  goes  that 
'  maximus  aevi  rector '  (greatest  latter-day  guide),  Vespasian 
Augustus,  and  by  his  side  his  children  walk."  And  then 
he  adds — "Of  all  ways  for  paying  due  thanks  to  men 
of  great  desert,  to  enrol  them  as  gods  is  the  most  time- 
honoured."  6 

If  the  new  and  incalculable  power  of  the  Roman  em 
perors  had  such  a  dazzling  influence  over  minds  trained 

1  Igitur  Vespasianus  cuncta  fortunae  suae  patere  ratus  nee  quicquam 
ultra  incredibile,  laeto  ipse  vultu,  erecta  quae  adstabat  multitudine,  iussa 
exsequitur.      Statim   conversa   ad  usum  manus,    ac   caeco  reluxit    dies. 
Utrumque,  qui  interfuere,  mine  quoque  memorant,  postquam  nullum  men- 
dacio  pretium  (Hist.  iv.  81).      Cf.  Ann.  iv.  20,  where,  in  the  account  of 
Lepidus  and  Tiberius,  Tacitus  represents  the  favour  of  an  emperor  as  a 
sort  of  gift  of  grace,  not  to  be  won  but  allotted  by  fate. 

2  Inst.  Or.  ii.  16,  12.     Cf.  ibid.  xii.  2,  21. 

3  Cf.  one  of  Dante's  phrases  for  the  deity,  ' '  II  consiglio  che  il  mondo 
governa"  (Par.  xxi.  71).  4  Nat.  Hist.  II.  vi.  12. 

5  Ibid.  VII.  viii.  45.  6  Ibid.  II.  vii.  18  and  19. 


THE  DEIFICATION  OF  ROMAN  EMPERORS         43 

by  habits  of  philosophic  thought,  what  must  have  been  the 
popular  state  of  mind  ?  Certain  bursts  of  enthusiasm  which 
are  chronicled,  numerous  dedications  inscribed  on  stone, 
may  help  us  to  some  conception  of  this.  The  successful 
career  of  demagogic  and  unscrupulous  informers  gives 
further  light.  No  doubt  Eprius  Marcellus,  one  of  the 
ablest  and  most  unscrupulous  of  these  informers,  depended 
for  his  backing  upon  the  populace,  or  he  would  never  have 
taken  the  tone  which  he  did  in  the  Senate.  Eprius  and 
others  like  him  were  backed  by  popular  indignation  in  their 
fierce  attacks  upon  those  who  refused  to  take  the  prescribed 
oath,  "In  acta  divi  Augusti  et  divi  lulii."1  His  motto 
was  also  the  people's — Pray  for  good  emperors,  but  take 
any  you  can  get  •  and  this  represented  the  people's  state  of 
mind. 

Among  the  people  who  were  thus  blindly  loyal  to  the 
imperial  master,  were  large  numbers  whose  first  franchise 
connected  itself  with  this  new  order  of  things.  It  is  curious 
to  note — as  far  as  the  scanty  means  of  information  allow — 
what  a  seemingly  incongruous  compound  of  Asiatic  piety  and 
European  bureaucracy  gathered  around  the  institution  of  this 
new  imperial  rite. 

In  Italy,  Sicily,  Gaul,  Spain,  on  the  Danube,  and  in 
Africa  a  new  class  of  men  sprang  into  notice.  They  were 
freedmen  and  small  tradesmen,  and  formed  an  especial  class 
or  caste,  calling  themselves  Augustales.  They  had  to  do 
with  local  celebrations  analogous  to  the  Augustalia  at  Rome. 
Furthermore  there  were  provincial  meetings  of  notables. 

1  Eprius  Marcellus  was  identified  with  the  new  cult  since  he  was  one  of 
the  Sodales  Augustales  (see  Henzen's  Inscription  5425,  quoted  in  Nipper- 
dey's  note  on  Tacitus,  Ann.  xii.  4).  The  customary  oath,  "  In  acta  divi 
Augusti  et  divi  lulii"  may  be  insisted  upon  as  a  purely  secular  act,  since 
it  was  required  as  a  preliminary  to  the  performance  of  secular  functions 
(Ann.  iv.  42;  xiii.  n).  Still  the  use  of  the  word  divus  certainly  involves 
a  religious  attitude  toward  those  to  whom  it  applies.  Moreover  Tiberius 
plainly  regarded  this  sollemne  iusiurandum  as  promoting  him  to  a  condition 
beyond  mortal  mishaps,  if  Tacitus  speaks  truly  (Ann.  i.  72) — Neque  in  acta 
sua  iurari,  quamquam  censente  senatu,  permisit,  cuncta  mortalium  incerta, 
quantoque  plus  adeptus  foret,  tanto  se  magis  in  lubrico  dictitans. 


44  APPENDIX  I 

Convened  for  the  purposes  of  the  new  worship,  these 
assemblies  soon  became  centres  of  provincial  life,  and 
played  in  later  Roman  days  no  unimportant  political  part, 
although  Christianity  deprived  them  of  all  connection  with 
religion. 

Such  conventions  of  notables  existed  in  the  East  as 
well  as  in  the  West ;  but  not  so  the  new  order  of  Augustales.1 
In  Greece  and  Asia  Minor,  and  in  general  wherever  the 
empire  of  Alexander  had  planted  the  seeds  of  specifically 
Greek  political  organisation,  there  was  noticeable  here  and 
there  a  sort  of  church  organisation.  The  old  hieratic  term 
vewKo/305  (familiar  at  Eleusis,  for  example),  quite  removed  at 
last  from  its  original  meaning  of  temple-sweeper,  got  itself 
applied  rather  to  whole  communities  than  to  individual 
men,2  and  is  found  upon  many  coins  of  Asiatic  cities  where 
periodical  festivals  in  honour  of  deified  emperors  were  held. 
There  was  competition  for  this  privilege  of  holding  high 
imperial  festivals, — for  the  vewKo/na.3  Ephesus  stood  pre 
eminent  in  having  had  it  granted  four  times.4  In  conjunc 
tion  probably  with  these,  and  certainly  with  other  features 
of  the  imperial  worship,  there  came  into  existence  a  board 
of  ten  High  Priests  for  the  province  of  Asia.  To  take  one 
Eastern  province  as  an  example  of  many,  they  were  called 
'Ao-Lapxat,5  and  were  necessarily  men  of  substance  and  posi 
tion.  They  were  elected  by  representatives  from  various 
cities  who  assembled  yearly  at  Ephesus.  Of  these  ten 
'Aaridpxai  one  apparently  ranked  6  above  all  the  others,  and 

1  Although  no  evidence  of  the  fact  is  forthcoming,  the  Augustales  prob 
ably  existed  in  the  free  municipia  established  in  the  East.      But  these  as 
well  as  the  colonies  may  be  neglected  in  speaking  broadly,  since  they  were 
not  of  the  East  as  such. 

2  See,  in  Pauly,  Krause's  articles  "  Certamina  "  and  "  Neocoroi";  also 
his  more  detailed  monograph  on  the  veuKopia. 

3  See  Tacitus,  Ann.   iv.    55,  where  it  is  plain  that  something  like  the 
vewKopia  is  involved. 

4  Coins  of  Caracalla's  and  of  Elagabalus'  reign  bear  the  inscription — 
'E0e<n'u»'  fj.6i>b}v  ctTracrcDj/  rerpdm  veuKdpwv. 

3  For  other  provinces  there  were  other  titles — 
KctTTTraSoKapx?;?,  etc. 

G  See  Marquardt,  Rom.  Staatsvenvalt.  i.  p.  513. 


THE  DEIFICATION  OF  ROMAN  EMPERORS          45 


bore  either  the  unqualified  title  of  'J^viapx^s,  or  of 
-rijs'Ao-i'as.  He  performed  functions  analogous  to  those  of  a 
bishop.  Thus  it  came  to  pass  that  the  chief  rallying  point 
of  Paganism  in  its  last  battle  with  Christianity  was  one  of 
its  very  latest  phases  —  the  worship  of  Rome  and  Augustus 
elaborated  in  two  ways,  one  for  the  West  and  one  for  the 
East. 


II 

DEMETER  AT  ELEUSIS  AND  CNIDUS 

THE  worship  of  Demeter  does  not  agree  with  war,  since  she 
never  used  her  golden  sword l  to  slay.  Remembering,  when 
wronged  even,  that  she  was  the  giver  of  good  things,  she 
found  comfort  to  her  griefs  in  blessing  all  mankind.  Such 
a  goddess  had  no  place  in  Homer's  Iliad  \  the  whole  of  its 
heroism  is  alien  to  her.  Other  peaceful  gods  might  go  to 
war,  limping  Hephaestus  might  join  the  force  that  favoured 
Greece,  while  Aphrodite  smiling  fought  for  Troy ;  but 
neither  side  claimed  Demeter.  So  far  from  seeking  her  aid 
were  the  haughty  heroes  of  Homeric  song,  that  her  good 
gifts  were  sometimes  even  misprized.  Ajax  defied  all  who 
were  mortal  and  ate  of  the  fruits  of  Demeter,  all  whom  a 
spear-thrust  could  pierce  or  a  rock  could  crush  and  maim.2 
The  golden  grain  in  abundance,  for  which  a  farmer  will 
always  be  thankful,  often  seemed  to  those  valiant  men  of 
war  an  unwelcome  mark  of  mortality  and  weakness,  a  blot 
upon  the  brightness  of  undying  fame.  When  the  Achaean 
host  is  under  a  cloud  of  dust  and  its  burnished  helmets  and 
bristling  spears  are  tarnished,  then  the  poet  bethinks  him 
of  Demeter  the  yellow -haired,  where  she  so  often  stands 

1  Hymn  to  Demeter,  1.  4.  -  Iliad,  xxi.  76  ;   xiii.  322. 


ii  THE  HOME  GODDESS  OF  GRAIN  47 

among  the  winnowers  on  a  farm,  parting  the  wheat  from  the 
chaff  that  spreads  over  the  brightness  of  day  like  the  dust 
that  chokes  the  Achaeans.1  Plainly  it  is  rather  in  moments 
of  trial  and  humiliation  that  the  kindly  goddess  of  earth's 
fruits  is  remembered  in  the  Iliad.  Demeter  had  in  fact 
been  left  at  home  when  the  host  set  sail  for  Troy.  She 
remained  among  the  farms  and  flowery  fields  of  Thessalian 
Pyrasus  where  was  her  sanctuary,  and  its  name  of  Pyrasus  2 
came  from  the  abundance  of  wheat  which  was  her  gift. 

In  the  Odyssey,  on  the  other  hand,  farmers  and  farming 
are  looked  upon  with  more  interest;  and  naturally,  since 
the  intense  theme  of  the  Trojan  war  is  there  exchanged  for 
a  less  thrilling  but  more  charming  narrative  of  adventure. 
In  the  Iliad,  when  nothing  is  said  to  the  contrary,  we  may 
be  sure  of  fighting ;  in  the  Odyssey,  no  matter  what  else  is 
going  on,  there  is  continual  feasting.  The  whole  of  the  last 
half  of  the  Odyssey  has  its  scene  laid  at  home  among  the 
farmsteads  or  in  the  hall  of  Odysseus.  Unhappily  the 
domain  of  Odysseus  was  no  Pyrasus,  no  wheatland,  and 
therefore  though  we  become  familiar  with  the  domestic 
economy  of  Eumaeus,  whose  faithfulness  to  his  lord  Odysseus 
tempts  the  translator  to  be  absolutely  literal  and  to  call  him 
the  "divine  swineherd,"3  this  brings  no  mention  of  Demeter 
the  home  goddess  of  grain,  the  Kornmutter  or  Mother  of 
Corn.  Something  of  her  history  may  be  gathered  from 
Homer,  though  he  chiefly  knows  that  Zeus  was  her  husband, 
and  that  he  slew  her  beloved  lasion,  whom  she  met  upon 
a  thrice-ploughed  fallow  field  of  Crete.4 

Homer  either  did  not  imagine   that   Persephone   was 

1  Iliad,  v.  500.  '  Ibid.  ii.  695. 

3  Mr.  Gladstone  has  recently  and  truly  said  that  in  point  of  goodness 
Eumaeus  excels  every  one  of  the  Homeric  gods. 

4  Odyssey,  v.  125. 


48  DE METER  AT  ELEUSIS  AND  C NIDUS  n 

daughter  to  Demeter,  or  did  not  think  that  the  relation, 
so  ineffably  beautiful  according  to  the  story  which  finally 
prevailed,  had  any  very  great  significance.  He  speaks  of 
Persephone  simply  as  the  daughter  of  Zeus,1  whereas  the 
idea  that  finally  prevailed  made  Persephone  nothing  if  not 
her  mother's  daughter,2  and  sometimes  indeed  left  a  doubt 
whether  Zeus  or  Poseidon  were  her  father.  It  mattered 
little  to  her  later  worshippers  who  her  father  was,  since  she 
became  all  her  mother's, — the  eternal  type  of  a  daughter 
dearly  loved  and  lost,  sought  for  in  grief  and  found  at  last. 
But  to  Homer  Persephone  was  nothing  of  all  this  ;  she  was 
the  queen  of  the  dead ; 3  dread  Persephone.  Terror  was 
in  her  name,  and  in  spite  of  the  lovelier  phases  through 
which  she  passed,  a  word  of  slaughter  can  still  be  heard 
when  she  is  named.  For  Homer  she  was  always  to  be 
feared,4  a  divinity  only  then  to  be  called  glorious  when  by 
so  naming  her  you  might  forestall  some  dreadful  harm.  She 
could  send  forth,  from  where  she  ruled  among  the  dead, 
that  awful  Gorgon's  head  that  turned  to  stone  all  those 
whose  eyes  it  met;  her  anger  was  therefore  to  be  feared 
and  in  every  way  to  be  appeased.  This  gloomy  picture  of 
Persephone  is  drawn  in  the  Odyssey  >  where  she  dwells  and 
queens  it  in  a  dusky  realm  that  may  be  above  or  below 
ground ;  the  only  thing  which  is  certain  about  it  is  its  situa 
tion  with  reference  to  the  rest  of  the  world.  It  is  a  land 
far  off  in  the  darkness  of  the  west,  beyond  the  twilight  of 

1  Iliad,  xiv.  326  ;   Odyssey,  xi.  217. 

2  Eur.  Phoen.  687. 

3  Arcadian  legends  give  Persephone  the  name  Afrwoiva  (Paus.  VIII. 
xxxvii.)  and  her  mother  is  Demeter  Erinys.     At  Eleusis  under  this  aspect 
Persephone  was  named  Daeira.     See  Preller's  Greek  Mythology. 

4  Persephone  is  called  e-jrcuvfi  four  times  in  the  Iliad  and  four  times  in 
the  Odyssey,  where  she  is  also  (euphemistically)  four  times  called 

and  once  ayvr]. 


ii  TWO  HOMERIC  OTHERWORLDS  49 

Cimmeria.  It  is  an  outer  or  utter  world,  not  necessarily  an 
under  world. 

In  the  Iliad,  on  the  other  hand,  we  hear  of  Hades  or 
Aidoneus,  the  husband  of  dread  Persephone.  So  awful 
were  the  abodes  where  these  two  dwelt  and  ruled  supreme 
that  the  poet  speaks  of  the  disclosure  of  Hades'  dominions 
to  the  light  of  day  as  a  thing  too  awful  almost  to  mention. 
Unlike  the  far  off  country  of  the  dead  visited  by  Odysseus, 
the  otherworld  of  the  Iliad  is  under  men's  feet.  The 
righting  of  the  gods  before  Troy,  Poseidon  shaking  the 
earth,  and  Zeus  filling  the  air  with  his  thunders,  nearly 
broke  through  into  the  undiscovered  country  of  the  dead. 
Aidoneus  upon  his  nether  throne  was  filled  with  fear  and 
trembled. 

It  cannot  be  too  often  repeated  that  the  Demeter  known 
to  the  Homeric  poems  had  no  affinity  with  Persephone  in 
either  of  her  two  realms.  According  to  a  flickering  tradi 
tion  we  hear  that  Persephone  was  deemed  by  some  to  be 
not  Demeter's  daughter  but  a  child  borne  by  the  dreadful 
river  goddess  Styx.1  Perhaps  if  Persephone's  mother  had 
been  named  by  Homer,  he  would  have  said  she  was  the 
Styx.  Anything  rather  than  mother  to  the  queen  of  death 
was  Homer's  Demeter.  She  is  a  goddess  of  peace  and 
plenty.  For  another  presence  like  hers  we  may  look  to  a 
place  far  nearer  home  than  Greece,  to  English  Northumber 
land.  Hutchinson  says  in  his  history,2  "In  some  places 
I  have  seen  an  image  aparelled  in  great  finery,  a  sheaf  of 
corn  placed  under  her  arm  and  a  scycle  in  her  hand, 

1  Apollodorus,  Bib  I.  \.  3,  £. 

2  A    View  of  Northumberland,    with  an    Excursion  to   the  Abbey  of 
Mailross  in  Scotland,  by  W.  Hutchinson,  Anno  1776,  published  at  New 
castle  in   1778,   vol.   ii.   p.   17  of  the   Appendix  in  the   account  of  Mell 
Supper. 

E 


50  DE METER  AT  ELEUSIS  AND  CNIDUS 

carried  out  of  the  village  on  the  morning  of  the  con 
clusive  reaping  day,  with  music  and  much  clamour 
the  reapers  into  the  field,  where  it  stands  fixed  on  a  pole 
all  day,  and  when  the  reaping  is  done  is  brought  home  in 
like  manner.  This  they  call  the  Harvest  Queen,  and  it 
represents  the  Roman  Cerec."  Hutchinson  might  have 
added  that  it  corresponded  to  what  we  know  from  Homer 
of  Demeter,  who  resembles  nothing  so  much  in  those 
earliest  stories  as  this  Harvest  Queen  of  England  or  the 
Corn  Lady  whose  divinity  is  honoured  in  Scotland  by 
hanging  up  a  small  package  of  grain  when  the  reapers  have 
finished.1 

After  the  Homeric  poems  came  the  works  of  Hesiod, 
but  it  is  uncertain  whether  all  the  traditions  preserved  by 
Hesiod  and  not  recorded  by  Homer  are  of  an  origin  later 
than  Homer.  In  fact  they  both  give  us  glimpses  of 
customs  and  habits  of  mind  as  old  as  time.  It  is  con 
venient,  however,  and  not  seriously  misleading,  to  think  of 
the  cheerful  yellow -haired  Homeric  Demeter  as  of  one 
coming  to  woman's  estate  through  the  deeper  experiences 
with  which  Hesiod's  poems  invest  her.  Here  she  becomes 
acquainted  with  grief  through  her  dear  daughter  Persephone. 
Hesiod  knows  far  more  of  the  goddess's  kindred  than 
Homer.  Rhea  is  Demeter's  mother  and  Cronos  is  her 
father ;  Zeus  is  her  husband,  to  whom  she  bore  white-armed 
Persephone.  Hesiod  also  has  heard,  while  Homer  has  not, 
of  the  carrying  off  of  Persephone  by  Aidoneus.2  To  him 
Zeus  granted  his  daughter's  hand,  and  by  him  Persephone 
is  seized  and  carried  off  in  a  chariot.  Still,  Hesiod  is  not 
always  very  far  in  his  notion  of  Demeter  from  the  simple 

1  Parallels  from  various  country  customs  are  multiplied  by  Mannhardt, 
and  also  by  Frazer  in  the  Golden  Bough,  2  Theog.  912, 


ii         THE  GODDESS  OF  ENOUGH  AND  TO  SPARE       51 

and  uncomplicated  idea  of  Homer.  Demeter  the  goddess, 
crowned  with  those  very  fruits  which  she  alone  can  give,  is 
famine's  foe.  "  Work,  Perseus,  make  famine  your  foe  and 
fair-crowned  Demeter  your  friend."1  This  is  Hesiod's 
advice  to  his  kinsman. 

The  Greeks  in  Sicily  worshipped  Demeter  from  the 
earliest  days,  and  well  they  might,  since  the  island  of  their 
homes  was  so  especially  favoured  by  her  that  it  came  to 
deserve  the  name  of  the  granary  of  Rome.  One  Sicilian 
sanctuary  was  dedicated  to  Hadephagia,2 — a  strange  name 
indeed  until  by  translation  you  discover  that  this  object  of 
Sicilian  reverence  was  simply  the  Genius  of  a  square  meal, 
the  goddess  of  Enough-and-to-spare,  a  divinity  much  prayed 
to  even  now  by  cow-boys  and  many  other  people  who  have 
long  wildernesses  to  cross,  and  often  fast  perforce  for  many 
hours  together.  At  this  same  Sicilian  shrine  Demeter  herself 
was  worshipped  under  the  surname  of  Sito,  that  is,  of 
Mother  Rye.  This  Sicilian  service  paid  to  Demeter  Sito 
and  to  Hadephagia  is  but  the  logical  outcome  of  the  utili 
tarian  view  of  Demeter  as  famine's  foe  presented  by  Hesiod 
in  his  Works  and  Days^  that  oldest  of  farmers'  almanacs. 

But  beyond  the  simple  aspect  of  Demeter  as  the  giver 
of  food  there  lurked  in  the  earliest  adoration  of  her  some 
thing  most  solemn  and  secret.  How  early  Herodotus 3 
considered  this  worship  to  have  come  into  Greece  may  be 
judged  by  his  story  that  the  daughters  of  Egyptian  Danaus 
showed  unto  the  women  in  Pelasgian  days  what  were  the 
rites  to  be  celebrated  in  honour  of  Demeter.  Herodotus  is 
speaking  more  especially  of  the  special  festival  in  her 
honour  called  the  Thesmophoria,  where  she  was  worshipped 
as  giving  sanction  to  certain  Thesmoi  or  laws  upon  which 

1  Works  and  Days,  298.  2  Athenaeus,  x.  9.  3  ii.  171. 


52  DE METER  AT  ELEUSIS  AND  C NIDUS  \\ 

the  family  and  other  social  facts  were  based.  This  some 
what  vague  statement  of  Herodotus  is  at  least  sufficiently 
definite  to  show  that  he  regarded  Demeter's  worship  as 
among  the  most  ancient  forms  of  divine  service.  Before 
Homer  or  Hesiod  sang,  Demeter  was ;  and  the  sentiment  of 
awe  which  consecrated  her  goddess  and  mistress  of  what 
men  held  most  sacred  and  most  dear  had  existed  in  Greece 
from  the  very  first.  Before  the  poets  came,  a  whole  ritual 
must  have  grown  up,  the  significance  of  which  Demeter's 
peasant  worshippers  could  not  expound.  These  were  men 
whose  only  argument  was  the  observance  of  times  and 
ceremonies,  and  who  knew  no  higher  logic  than  the  telling 
of  a  tale  about  this  god  or  that.1  Hesiod,  who  first 
chronicles  in  full  their  stories  of  Demeter's  parentage,  and 
who  first  though  briefly  mentions  the  grief  that  came  when 
her  daughter  was  stolen  away  from  earth,  must  surely  some 
where  indicate  a  deep  and  solemn  view  of  the  fair-haired 
Demeter's  power  upon  the  lives  of  men.  Of  this  deeper 
view  there  are  traces,  though  it  may  be  necessary,  if  we 
would  clearly  understand,  to  read  between  his  lines. 

In  place  of  Homer's  phrase,  "  the  fruits  of  Demeter," 2 
Hesiod  prefers  to  speak  of  "the  holy  fruits  of  Demeter."3 
This  adjective  holy  contains  a  first  and  half  articulated 
expression  of  the  mystery  and  awe  which  overpowered  the 
pious  adorers  at  Eleusis  when  it  found  its  full  utterance  in 
ritual.  Another  hint  is  given  by  Hesiod  that  throws  light 
upon  this  homely  farm-religion  of  the  early  days  in  Greece. 

1  ' '  There  is  a  certain  illogical  logic  about  all  mythologies.    Where  philo 
sophy  leaps  at  once  to  abstract  terms  and  speaks  of  an  omniscient,  omni 
potent,  omnipresent  deity — mythology,  aiming  at  the  same  notions,  proceeds, 
agreeably  to  its  nature,  by  positive  imagery  in  place  of  negative  afo-tractions. " 
— C.  F.  Keary,  The  Vikings,  p.  61. 

2  Iliad,  xiii.  322.  3    Works  and  Days,  466. 


ii  THE  HALLOWED  FRUITS  OF  DEMETER  53 

I  mean  where  he  says,  in  words  not  always  rightly  under 
stood,  that  a  field  new  ploughed  and  newly  sown  has  power 
to  charm  a  babe  and  still  his  cries.  That  is,  if  the  babe  be 
only  laid  upon  it,  no  doubt.  This  receives  illustration  from 
various  customs  not  yet  extinct  in  Europe,  which  bring  the 
new  sowing  of  seed  and  the  tending  and  growth  of  young 
babes  into  one  and  the  same  scheme.  Hesiod,  just  after 
dwelling  upon  the  above  point  in  regard  to  stilling  young 
children,  proceeds,  "  Now  make  prayer  to  Zeus  and  to  holy 
Demeter,  that  they  may  make  perfect  and  heavy  of  growth 
the  hallowed  fruits  of  Demeter."1  Plainly  the  inscrutable 
power  which  gives  and  withholds  abundance  in  harvests  can 
somehow  hinder  or  help  the  health  and  growth  of  a  babe. 
Every  language  and  every  country  works  out  in  some  way 
the  old  story  and  utters  with  a  new  voice  the  time-worn 
truth  that  the  growth  of  babes  and  children  is  one  with  the 
growth  of  trees  and  flowers  and  grass.2  All  fruitfulness, 
every  species  of  multiplication  in  the  land,  is  linked  to  that 
of  every  other  kind  by  some  mystical  bond  which  makes 
one  of  them  all,  and  binds  the  growth  of  men  and  of  things 
into  a  single  and  continuous  scheme.  This  belief,  now 
argued  out  by  science  and  subjected  to  scrutiny  in  all  its 
parts  and  all  its  meanings,  was  represented  among  the 
peoples  of  all  times  in  a  thousand  quaint  customs  of  the 
countryside,  many  of  which  survive  among  the  innocent 
and  unlearned  to-day.  Akin  to  all  these  customs,  but 

1  The  whole  passage  runs  as  follows  :  — 

dpovpav' 


veios  aXe^tdpTj  Traiduv  €UKTj\r]T€ipa' 

^X^ffdai  5ev  Aii  xfloju'y,  ^r]fj.r]TepL  6'  ayvy, 
(KTe\ta  (3pi6eiv  A-rj/j.rjrepos  itpbv  aKrriv. 

Works  and  Days,  463-466. 
2  See  Mannhardt's  posthumously  printed  essay  "  Korn  und  Kind." 


54  DEMETER  AT  ELEUSIS  AND  CNIDUS  n 

fuller  of  the  perfect  truth  of  poetry  which  is  beauty,  and 
beauty,  and  once  again  beauty,  was  the  service  of  fair- 
crowned  Demeter  in  Greece  by  chosen  spirits  of  Greece, 
and  when  Rome  came,  in  Rome. 

Plainly  the  godhead  of  Demeter  and  her  kinship  with 
the  queen  and  king  of  darkness  were  bound  up  somehow 
with  the  deeper  and  more  mysterious  suggestions  made  ever 
and  anon  by  Hesiod.  Homer  knew  nothing  of  all  this,  and 
accordingly  Homer  knows  little  or  nothing  of  the  real  god 
head  of  Demeter.  The  first  complete  account  of  the  myth 
of  Demeter  is  contained  in  a  poem  of  later  date  than  the 
Iliad)  the  Odyssey,  or  the  writings  of  Hesiod.  This  is  the 
Homeric  Hymn  to  Demeter,1  called  Homeric  not  because  it 
was  written  by  Homer  but  as  the  work  of  some  poet  versed 
in  Homeric  lore,  and  its  probable  date  is  about  600  B.C. — five 
hundred  hexameter  lines  written  at  least  2500  years  ago, 
which  remained  absolutely  unknown  from  the  fourteenth  to 
the  end  of  the  eighteenth  century  of  our  era.  Mr.  Sidney 
Colvin 2  has  most  aptly  described  this  beautiful  poem,  and 
gives  the  following  account  of  its  substance  and  style : — 
"  There  is  nothing  liturgical  about  it ;  it  is  rather  in  the 
nature  of  a  ballad,  recited,  it  may  be,  by  a  patriotic  minstrel  of 
Eleusis  to  the  groups  of  strangers  who  thronged  to  the  city, 
or  in  competition  with  other  such  ballads  at  one  of  those 
poetical  tournaments  which  formed  part,  we  know,  of  many 
of  the  Greek  religious  festivals.  I  say  a  minstrel  of  Eleusis, 
because  of  his  special  tone  of  pride  in  the  town  and  locality, 
and  because  he  ignores  Athens,  while  his  Ionian  dialect 
would  be  quite  proper  to  an  Attic  rhapsodist.  It  is  their 

1  About  one  hundred  years  ago  Ruhnken  received  it  for  publication 
from  a  learned  friend  who  stumbled  upon  it  in  an  old  monkish  library  at 
Moscow. 

2  Cornhill Magazine,  vol.  xxxiii.  June  1876,  "A  Greek  Hymn." 


„  THE  HOMERIC  HYMN  TO  DEMETER  55 

ballad  character,  and  the  community  they  have  of  style  and 
diction  with  the  Iliad  and  Odyssey,  which  have  earned  the 
title  of  Homeric  for  a  certain  number  of  Greek  hymns  or 
narrative  poems  in  praise  of  particular  divinities  which  have 
come  down  to  us.     This  is  the  most  beautiful  of  them  all." 
This  beautiful  Hymn  contains,  according  to  the  view  set 
forth  below,  at  least  three  ballads  or  stories,  and  the  incon 
sistencies  and  roughnesses  in  its  composition  spring  from  a 
too  conscientious  effort  to  make  the  three  into  one.    At  the 
time  when  this  Hymn  was  composed,  or  perhaps  we  should 
say    compiled,  many   interesting  facts   and   fancies   about 
Demeter  and  her  worship  were  at  the  author's  command. 
The   most   conspicuous   of   these   are   two   stories   telling 
how  Aidoneus  carried  off  Proserpina,  and  of  Proserpina's 
final  restoration  to  Demeter.     One  of  these  may  be  called 
the  Iliad   story,   since   it   is   based   upon  the   Iliad  con 
ception  of  the  world  of  death  as  an  underworld.     The  other 
may  be  called  the  Odyssean  version,  since  according  to  it 
Persephone  was  spirited  away  into  the  land  beyond  Cim 
merian  darkness,  whither  Odysseus  went  to  talk  with  dead 
Tiresias.     Over   and  above  these  two  accounts  of  Perse 
phone's  disappearance  and  reappearance,  that  had  woven 
themselves  into  consistency  upon  the  lips  of  men,  there 
were  other  tales  inextricably  connected,  now  in  one  way  and 
now  in  another,  with  the  two  just  named,  which  gave  to  the 
bereavement  of  Demeter  and  the  robbery  of  Persephone  a 
local  habitation  and  a  name.     These  stories  may  be  con 
ceived  of  as  having  been  as  numerous  as  were  the  temples 
of  Demeter,  but  certainly  the  one  that  men  most  heeded 
was  the  one  which  localised  the  whole  myth  at  Eleusis  in 
Attica.     Eleusis  is  about  twelve  miles  from  Athens,  but  the 
fertile  and  extensive  Thriasian  and  Rarian  plains,  the  first 


56  DEMETER  AT  ELEUS1S  AND  CNIDUS  n 

one  towards  Athens,  and  the  second  one  towards  Megara, 
surround  it  and  make  it  a  natural  home  for  the  goddess  of 
grain.  It  is  evident  that  many  generations  of  simple  people 
on  the  farms  of  the  Thriasian  and  the  Rarian  plains  in 
Attica  had  been  absorbed  in  the  due  worship  of  the  goddess 
Demeter.  Gradually,  in  this  direction  and  in  that,  con 
nected  accounts  of  the  goddess  and  of  the  Eleusinian  rites 
in  her  honour  grew  up  and  found  credence.  The  most 
pleasing  and  popular  of  these  were  woven  into  one  narrative, 
which  forms  the  beautiful  and  yet  most  bewildering  Hymn  in 
question. 

The  perseverance  of  Dr.  Wegener  has  triumphed  over 
the  author  of  this  Hymn  to  Demeter.  What  the  poet  joined 
together  Wegener  has  triumphantly  put  asunder.  Too 
anxious,  like  many  compilers  of  religious  articles  and  creeds, 
that  no  one  concerned  should  find  cause  of  offence  in  his 
work,  the  author,  in  spite  of  the  poetic  exquisiteness  of  his 
touch,  left  such  inconsistencies  and  patent  incongruities  that 
each  tradition  can  with  more  or  less  certainty  be  disentangled 
from  the  others ;  and  this  is  what  Dr.  Wegener  has  done.1 

The  Iliad  story  of  the  carrying  off  of  Proserpina  is 
briefly  as  follows  :— Zeus  conspired  with  Gaia,  the  earth,  to 
get  Persephone,  his  child  by  Demeter,  for  his  brother 
Aidoneus  to  wife.  Earth  snared  the  smiling  maid  by  a 
most  fatal  blossom  called  the  narcissus.  Persephone 
reached  forth  to  pluck  the  wondrous  flower,  and  lo !  the 
ground  opened,  and  Aidoneus  dragged  the  shrieking  girl 
down  to  his  underworld  home.  Hecate  meanwhile  was 
sitting  in  her  cave  thinking  delicate  thoughts.  She  and 
she  alone  could  see  the  robber  on  his  downward  way,  and 
she  it  was  who  made  haste  with  the  news  to  Demeter.  The 

1  Philologus,  xxxv.  (1876)  pp.  227-254. 


ii  THE   WEDLOCK  OF  PERSEPHONE  57 

bereaved  mother  stands  at  Zeus's  throne  and  asks  for  resti 
tution.  Zeus  urges  that  Aidoneus  is  a  worthy  husband  for 
Persephone.  Then  Demeter  shuns  Olympus,  and  resorts 
to  the  fields  and  towns  of  men.  She  retires  into  her  temple 
(at  Eleusis),  where  she  passes  a  whole  year.  The  world  is 
stricken  in  all  the  produce  and  increase  of  earth.  Zeus, 
forced  by  the  cutting  off  of  all  fruits,  sends  Iris  first,  and 
then  the  other  gods.  All  others  fail,  and  Rhea,  Demeter's 
mother,  last  of  all  goes  to  her  and  she  makes  peace.  Zeus 
grants  a  compromise.  Two-thirds  of  the  year  Persephone 
is  to  stay  with  her  mother  and  see  the  glad  light  of  day,  but 
for  one-third  of  each  twelve  months,  during  the  sad  season 
of  winter  and  darkness,  the  daughter  and  mother  are  to  be 
parted,  Persephone  is  to  be  with  her  husband  Aidoneus. 
Appeased  at  last,  the  mother  welcomes  back  her  child, 
and  earth  once  more  covers  itself  with  the  holy  fruits  of 
Demeter. 

The  second  version,  told  in  the  same  breath  by  the 
Homeric  author  of  Demeter's  Hymn,  is  the  Odyssean 
version,  and  it  must  be  admitted  that  the  landscape  of 
Eleusis  does  not  suit  the  demands  of  this  form  of  the  story 
so  well  as  it  does  the  Iliad  tale  just  given,  but  on  the  other 
hand  this  account  of  the  story  brings  in  the  pomegranate 
seed — a  mystical  emblem  often  seen  in  the  hands  of  Demeter 
and  Persephone  at  Eleusis  and  elsewhere.  There  is  much 
to  tempt  an  unwary  person  here ;  certainly  the  rash  would 
incline  to  pronounce  this  Odyssean  tale  the  older  of  the 
two.  However  this  question  of  age  be  decided,  there  is  no 
doubt  as  to  a  strong  affinity  with  the  Odyssey,  not  only  in 
the  Odyssean  aspect  of  the  tale  which  has  yet  to  be  given, 
but  in  the  whole  of  the  poem.  Mr.  Colvin  has  descried 
and  inimitably  described  the  likeness  as  follows :  "  It  (the 


58  DEMETER  AT  ELEUSIS  AND  CNIDUS  n 

Hymn)  moves  with  much  of  the  same  easy  grandeur  as  the 
Odyssey,  it  has  the  same  romantic  charm,  and  delights  us 
with  similar  pictures  of  heroic  manners,  of  chiefs  trusted  by 
their  people,  of  beautiful  unabashed  virgins,  of  noble  hos 
pitality  to  strangers.  Like  the  Odyssey,  it  tells  us  of  gods 
going  to  and  fro  among  mortals,  unrecognised  till  they 
choose;  of  disguises  and  feigning  answers  and  sudden 
revelations." 

And  now, — to  turn  from  the  Odyssean  touch  in  the  whole 
poem  to  what  has  been  talked  of  as  the  Odyssean  version  of 
the  carrying  off  of  Proserpina, — that  runs  as  follows  : — On  a 
flowery  mead  close  by  Oceanus,  Persephone  is  gathering 
flowers  with  the  daughters  of  Oceanus.  But  a  sudden  fear 
arrives — Hades  dashes  across  the  flowery  field  with  his 
chariot  and  spirits  the  maiden  away.  Zeus  knows  nothing 
of  the  deed,  but  is  busy  in  a  far  off  temple  accepting 
sacrifice  from  men.  The  robber  king  of  death  meanwhile 
drives  ever  onward  toward  the  darkling  West.  Persephone 
cries  ever  and  anon,  but  most  of  all  when  at  the  very  last  of 
the  weary  journey  she  sees  that  she  must  lose  the  sight  of 
day.  Hades  comforts  her  by  telling  her  of  the  honours  she 
shall  have  as  queen  among  the  dead ;  and  furthermore,  that 
fate  may  never  take  her  from  him  quite,  he  secretly  thrusts 
a  pomegranate  seed  into  her  mouth.  Demeter  hears  her 
daughter  scream,  and  rends  her  garments,  and  wraps  her 
shoulders  in  the  garb  of  mourning.  Thus,  seeking  and  ask 
ing,  for  nine  days  long  does  she  pass  over  land  and  sea. 
None  of  Persephone's  playmates,  not  one  of  all  the  gods 
and  men,  can  tell  her  who  the  robber  is,  or  where  her 
daughter  tarries.  So  therefore  she  goes  to  the  all-seeing 
sun,  and  he  shows  to  her  the  utmost  regard  and  kindness. 
Filled  with  pity,  he  tells  her  who  the  robber  is,  and  whither 


ii  THE  ILIAD  AND  THE  ODYSSEAN  STORIES         59 

Persephone  has  been  carried.  Then  Demeter  in  wrath 
betakes  her  to  her  Eleusinian  temple.  Zeus  meanwhile 
hears  her  wrongs,  and  sends  his  messenger  Hermes  with 
instant  reprimand,  and  with  command  that  Hades  make 
restitution.  Hades,  the  Zeus  of  the  netherworld,  made  no 
retort  in  anger,  but  smiled  serenely  and  bade  Persephone 
do  as  the  word  of  Hermes  commanded.  He  was  sure, 
through  the  fatal  seed  of  the  pomegranate  which  she  had 
taken,  that  Persephone  would  not  forsake  him  utterly. 
Hermes  accordingly  leads  her  back  to  her  mother,  who  is 
glad  of  her  return,  but  grieves  to  find  what  an  unbreakable 
spell  from  Hades  is  on  her. 

These  two  legends  are  curiously  but  not  at  all  indis- 
tinguishably  interwoven  through  the  whole  of  the  first  and 
the  last  portions  of  the  Homeric  Hymn  to  Demeter.  If 
in  no  other  material  detail,  they  coincide  in  taking  for 
granted  the  existence  of  a  temple  of  Demeter.  This  temple 
is  not  necessarily  at  Eleusis.  Both  of  these  versions  would 
suit  any  other  centre  for  the  worship  of  Demeter  just  as  well 
or  ill  as  they  suit  Eleusis,  for  it  is  possible  to  substitute 
another  name,  and  thus,  so  to  speak,  the  venue  of  the  myth 
can  be  changed  with  the  alteration  of  one  word.  The 
second  or  Odyssean  story  lays  the  scene  of  the  robbery  in  a 
far  off  land,  which  is  certainly  neither  Eleusis  nor  any  other 
centre  of  Greek  life,  and  as  for  the  place  whither  the  stricken 
Demeter  retires,  it  is  simply  spoken  of  as  her  temple,  and 
no  local  details  are  given,  but  only  the  name  Eleusis.  The 
Iliad  tale,  on  the  contrary,  might  be  supposed  to  consecrate 
Eleusis,  not  only  as  the  place  where  Demeter's  temple  was, 
but  as  the  actual  theatre  of  the  robbery  of  Proserpina.  But 
Cnidus  (whither  the  story  wandered  by  sea  from  Thessaly 
no  later  perhaps  than  it  went  to  Eleusis)  would  answer  just 


60  DE METER  AT  ELEUSIS  AND  CNIDUS  n 

as  well.  The  same  may  be  said  of  any  other  place  where 
Demeter  was  worshipped,  and  where  there  was  a  rock-form 
ation  suggestive  of  the  rending  and  yawning  of  the  earth, 
and  a  smiling  and  fertile  plain  near  by  for  the  flower  gather 
ing  of  Persephone.  Accordingly  neither  of  these  stories  of 
the  carrying  off  of  Proserpina  necessarily  localises  the  myth 
and  worship  of  the  two  goddesses,  mother  and  daughter,  at 
Eleusis.  But  towards  the  hundredth  line  of  the  Homeric 
Hymn  a  strange  thing  suddenly  happens.  Just  as  the  earth 
opened  when  Persephone  reached  out  her  hand  to  gather 
the  fatal  narcissus,  so  when  the  reader  seeks  to  follow  onward 
the  narrative  thread  of  Persephone's  robbery,  woven  together 
out  of  two  strands,  he  finds  that  it  becomes  tangled  suddenly 
from  one  line  to  the  next,  and,  before  he  knows  it,  he  is 
dragged  down  to  where  sorrowing  Demeter  sits  on  the 
Laughless  Stone  by  the  Eleusinian  well  called  Maidenswell. 
The  way  of  Persephone's  story  comes  to  a  sudden  chasm, 
and  a  legend  of  different  quality,  though  not  less  beautiful, 
is  disclosed. 

This  is  the  purely  and  most  sweetly  Attic  tale  of  Deme- 
ter's  stay  at  Eleusis.  Here  we  have  united  together  certain 
local  traditions  that  grew  up  at  and  near  Eleusis  in  the  early 
days  when  Athens  and  Eleusis  were  on  so  nearly  equal  a 
footing  as  independent  states,  each  exercising  a  local  leader 
ship,  that  they  with  ten  others  could  eventually  become  the 
twelve  members  of  the  Attic  confederation — if  it  were  sure 
that  this  word  suited  the  politics  of  those  earliest  days. 
This  interjected  Eleusinian  tale  of  Demeter's  stay  at  Eleusis 
gives  the  needed  consecration — the  only  one  respected  in 
those  early  days— to  the  temple  and  observances  of  Demeter 
at  Eleusis.  By  this  tradition  is  founded  the  Eleusinian 
claim  to  be  the  greatest  centre  for  Demeter -worship  in 


ii  THE  ELEUSINIAN  LEGEND  OF  DE METER          61 

Greece.  This  story,  however,  is  not  told  in  one  way  only 
any  more  than  that  of  Persephone's  taking  away.  Here 
again  the  scrutiny  of  minute  perseverance  discovers  various 
inconsistencies  and  a  double  version.  As  the  points  of 
difference  seem  less  vital  here,  they  shall  for  the  most  part 
be  given  in  one  narrative,  while  a  few  matters  can  be  kept 
till  the  end  for  consideration. 

The  Eleusinian  legend  of  how  Demeter  came  to  dwell  at 
Eleusis  is  then  substantially  as  follows  : — Near  a  well  close 
to  the  Acropolis  of  Eleusis  the  sorrowing  goddess  Demeter 
rested  from  exhaustion,  for  she  had  been  long  in  search  of 
her  lost  Persephone.  What  was  the  name  of  the  well,  Par- 
thenos  (Maidenswell)  or  Callichoros  (Dancewell)  ?  However 
that  may  be  decided,  the  goddess  rested  there  on  the  Laugh- 
less  Stone,  the  Agelastos  Petra.  Was  Demeter  disguised  as 
an  old  woman,  or  was  she  there  in  the  undisguised  majesty 
of  her  divine  beauty  ?  The  stories  varied ;  but  in  this  they 
agree,  that  the  four  daughters  of  the  king  of  Eleusis, 
Celeus  son  of  Eleusin  by  name,  came  thither  with  pitchers 
of  bronze  that  they  might  draw  water  from  the  well.  They 
question  the  goddess,  who  tells  them  a  tale  neither  plain 
nor  unvarnished.  She  has  been  enslaved  by  men  who 
kidnapped  her  in  Crete.1  When  her  captor  landed  at  Thori- 

1  The  use  of  Crete  in  this  feigned  narrative  is  by  some  supposed  to  amount 
to  a  recognition  of  Crete  as  an  early  cradle  of  the  worship  of  Demeter.  It 
should,  however,  be  remembered  that  Crete  was  the  most  obvious  of  places 
to  mention  in  any  invented  tale  of  seafarers  and  seafaring.  This  is 
proved  by  its  constant  occurrence  in  feigned  adventures  in  the  Odyssey 
(xiii.  256  and  ff. ;  xix.  171  and  ff. ),  and  by  the  way  in  which  it  often  creeps 
into  the  Homeric  MSS. ,  either  instead  of  places  of  less  frequent  resort  or  in 
addition  to  them  (cf.  Odyssey,  i.  93,  where  two  verses  about  Crete  are  added ; 
ibid.  285,  where  Zenodotus  substituted  Crete  for  Sparta  ;  the  same  thing 
occurs  in  Odyssey,  ii.  214  and  359.  See  La  Roche's  critical  edition, 
Teubner,  1867.  Something  of  the  kind  is  reported  at  Odyssey,  iv.  702). 
The  real  proof  of  a  widespread  and  very  early  belief  in  Crete  as  one 
of  the  starting-points  of  the  Demeter  myth  is  in  Hesiod's  localising  in 


62  DE  METER  AT  RLE  US  IS  AND  C NIDUS  n 

cus  in  Attica  she  made  her  escape  under  cover  of  night. 
Now  she  wishes  to  be  taken  into  service.  The  girls  go  to 
their  mother,  Metanira,  and  return  with  a  message  that 
she  will  be  welcomed  as  their  brother  Demophoon's  nurse. 
According  to  one  story  Demeter,  in  the  undisguised  splen 
dour  of  her  divinity,  dazzled  Metanira,  the  babe's  mother, 
who  arose  as  if  to  give  the  place  of  honour  to  the  entering 
guest,  whose  more  than  human  skill  was  required  to  deliver 
the  infant  Demophoon  from  the  evil  spells  cast  over  his  life 
by  a  wicked  nurse.  The  other  way  of  telling  the  tale  makes 
Demeter  none  the  less  a  good  fairy,  only  her  gentle  offices 
are  given  in  the  disguise  of  a  grief-stricken  woman  over 
burdened  with  years  and  misfortunes.  Thus  disguised  she 
takes  the  boy — the  child  of  Metanira  and  Celeus  latest  born 
— and  gives  him  the  care  without  which  he  could  never  have 
been  brought  to  man's  estate.  Of  whatever  nature  the  god 
dess's  service  was,  all  tales  agree  in  saying  that  the  child  grew 
apace,  without  the  ordinary  food  of  mortal  babes,  fondly 
cherished  upon  her  immortal  bosom  and  lulled  to  rest.  At 
this  point  a  curious  turn  is  taken  by  the  myth,  which  relates 
that  Demeter  sought  to  make  the  boy  Demophoon  immortal, 
and  to  that  end,  when  all  the  house  was  asleep,  set  him  in 
the  flames.  One -night  she  was  watched  either  by  one  of 
the  sisters  or  by  the  mother  Metanira.  Catching  sight  of 
Demophoon  in  the  flames,  his  indiscreet  and  misguided 

Crete  ( Theogony,  969  and  ff. )  the  commerce  between  Demeter  and  lasion, 
reported  but  not  localised  in  the  fifth  Odyssey,  vv.  125  and  if.  This  is 
confirmed  by  a  reference  to  Bacchylides  (Bergk,  fr.  64),  where  the  rape  of 
Proserpina  is  localised  in  Crete.  Common  report,  however,  had  it  with 
equal  certainty  that  Proserpina  was  carried  off  from  the  fertile  fields  of 
Sicily.  I  think  it  therefore  unjustifiable  to  appeal  to  the  early  poets  as 
giving  an  undisputed  pre-eminence  in  Demeter-worship  to  Cretan  tradi 
tions.  Ariadne — a  sort  of  Persephone — came  from  Crete,  it  will  be 
remembered,  and  she  has  little  or  no  direct  connection  with  Demeter. 
See  Appendix  X. 


ii  INTERLOPERS  IN  THE  STORY  63 

kinswoman  screamed  aloud,  whereupon  the  goddess,  having 
laid  hold  upon  him,  was  moved  to  sudden  anger  and  let  him 
fall.  The  family  is  awakened,  and  the  women  minister  to 
the  affrighted  child.  Then  Demeter  takes  her  departure, 
but  not,  as  the  previous  episode  would  seem  to  suggest,  in 
anger.  No ;  she  waits  to  give  full  commands  concerning 
the  building  of  her  temple  at  Eleusis,  and  she  enters  into  all 
the  rites,  the  orgies,  as  they  were  called,  which  were  to  be 
celebrated  in  her  service  there.  These  commands,  accord 
ing  to  one  story,  were  laid  by  the  goddess  upon  Metanira 
and  her  daughters,  who  did  not  call  upon  Celeus  and  his  sons, 
— among  whom  was  Triptolemus, — until  morning  dawned. 
The  intervening  hours  through  all  the  night  were  spent  by 
the  women  in  propitiating  the  goddess.  The  alternative 
version  is  that  Demeter  on  the  eve  of  departure  spoke  to  the 
women  of  her  worship  and  its  orgies,  and  then  summoned 
King  Celeus  and  his  sons  Triptolemus,  Diocles,  Eumolpus 
and  Polyxenus,  and  gave  to  them  all  needful  commands  for 
the  building  of  her  temple  and  the  institution  of  her 
service. 

As  points  of  divergence  arose,  they  have  been  indicated 
in  the  above  summary  of  the  Eleusinian  story  of  Demeter. 
Two  main  versions  there  plainly  were,  but  even  after 
making  allowance  for  such  a  variation,  there  remain  diffi 
culties  to  be  cleared  up.  First  of  all  the  whole  story  of 
Demeter's  seeking  to  make  Demophoon  immortal  by  im 
mersing  him  in  fire  seems  incongruous  and  incomprehens 
ible.  This  fact,  taken  together  with  the  identity  not  only 
in  substance  of  the  account  of  the  fire-baptism  of  Demophoon 
by  Demeter,  and  one  preserved  elsewhere1  of  the  fire-baptism 

1  Apollodorus,  Dibliotheca,  iii.  13,  6.  Compare  the  account  of  Demo- 
phoon's  fire-baptism  given  also  by  Apollodorus,  i.  5,  4.  The  story  suits 


64  DE METER  AT  ELEUSIS  AND  C NIDUS  n 

of  Achilles  attempted  by  his  mother  Thetis  and  foiled  by  his 
father  Peleus,  removes  one  difficulty.  The  whole  fire  episode 
was  probably  imposed  upon  this  story ;  it  has  no  place  there, 
at  least  not  in  the  form  in  which  it  has  been  transmitted. 
And  this  smooths  the  way  for  clearing  the  second  difficulty. 
It  is  plain  that  there  is  a  surplus  of  proper  names  here. 
The  king  of  Eleusis,  in  whose  house  Demeter  tarried,  is 
not  in  all  accounts  of  the  myth  called  Celeus.  Panyasis l 
names  the  Eleusinian  king  Eleusis  or  Eleusin,2  whereas  in 
this  Homeric  Hymn  the  king  is  Celeus,  and  his  father's  name 
is  Eleusin.  Moreover  the  youthful  hero  worshipped  at 
Eleusis,  and  especially  in  the  Rarian  plain  near  by,  as 
Demeter's  favoured  child,  whom  she  had  instructed  in 
the  arts  of  farm  labour,  is  Triptolemus,  not  Demophoon. 
This  circumstance  would  lead  us  to  expect  Triptolemus  to 
take  Demophoon's  place  in  the  story  of  Eleusis  given  in  the 
Homeric  Hymn,  and  such  is  the  case  in  what  are  considered 
later,  but  may  represent  earlier  versions  of  it.3  Now  if  from 
the  Homeric  Hymn  be  subtracted  the  fire-baptism  of  Demo 
phoon,  there  is  nothing  left  for  Demophoon  in  all  the  story. 
It  looks  as  if  Demophoon  and  his  father  Celeus  were 
interlopers  in  this  Eleusinian  tale,  and  it  is  not  im 
possible  that  their  presence  here  may  be  a  chapter  of  early 

Achilles,  and  does  not  suit  Demophoon.  Thetis  wished  to  make  him 
immortal  by  burning  out  the  mortal  part  which  he  had  from  his  father. 
We  are  not  told  how  the  thrusting  of  Demophoon  "like  a  torch  into  fire" 
was  supposed  to  make  him — of  mortal  father  and  mortal  mother — superior 
to  mortality.  Furthermore  Achilles  never  tasted  mother's  milk,  and 
hence  his  first  name  was  Ligyon.  A  point  is  made  of  Demophoon's  not 
taking  the  breast,  but  nothing  remarkable  comes  of  it.  The  whole 
Achillean  fire-legend  loses  reality  in  the  alien  story  of  Eleusis. 

1  Apollodorus,  Bibliotheca,  i.  5,  2  (where  Pherecydes  is  quoted  as  saying 
Eleusis  was  a  son  of  Oceanus  and  Ge.     In  these  tales  Eleusis  figures  as 
Cecrops  and  Cychreus  do  at  Athens  and  Salamis) ;   Hyginus,  Fab.  147. 

2  Cf.  Pausanias,  I.  xxxviii.  7,  end. 

3  Hyginus,  Fab.  147,  who  was  followed  bv  Ovid. 


ii  CELEUS  AND  DEMOPHOON  65 

religious  history  in  disguise.  Supposing  the  religious  im 
portance  of  Phliasian  Celeae l  to  have  been  overshadowed 
and  all  but  clean  forgot  in  very  early  days,  we  should 
then  have  a  survival  of  it  if  the  local  hero  of  Celeae  was 
Celeus.  Accordingly  the  supposition  would  be  that,  before 
Eleusis  and  its  legends  completely  won  the  day,  there  was 
an  interregnum,  a  period  when  neither  Eleusis  nor  Celeae 
nor  Andania2  had  appropriated  exclusively  the  story  of 
Demeter's  sorrows  upon  earth.  What  variety  of  names  and 
episodes  there  may  have  been  in  all  these  rival  tales  cannot 
be  known.  But  the  uncertainty  of  many  of  the  important 
proper  names  in  the  Attic  story  as  it  has  reached  us  is 
most  significant.  For  Metanira  some  give  Cothonea ; 3  the 
name  Demophoon  crowds  Triptolemus — Demeter's  real 
favourite — into  the  position  of  an  elder  brother;  and 
Celeus  is  invited  into  the  Eleusinian  story  of  Demeter, 
taking  the  place  of  Eleusin,  who  becomes  his  father.  Celeus 
could  easily  (in  a  compromise-version)  fill  the  unimportant 
place  of  the  child's  father  Eleusin  in  the  narrative,  but  it 
was  not  so  easy  to  supplant  Triptolemus,  a  local  demi-god 
whose  worship  was  almost  on  a  par  with  that  of  Demeter 
herself.  This  is  the  reason  why  Demophoon  appears  in 
this  story  only  to  disappear,  and  indeed  there  is  very  little 
beyond  the  record  of  a  Demophoon,  son  of  Theseus,  to 
show  where  Demophoon  came  from.4  Of  him  we  have  but 
the  name,  though  it  is  certain  that  in  some  early  story  he 
played  a  leading  part,  for  the  name  reappears  in  Euripides' 

1  Pausanias,  II.  xiv.  2  Ibid.  IV.  iii.  10. 

3  Pausanias  (I.  xxxviii.  3)  says  that  Pamphos  "  Kara  ravra  Kal  "O^pos  " 
calls  the  daughters  of  Celeus,  Diogenia,  Pammerope,  and  Saisara.     These 
names  are  unknown  in  our  Homeric  poems. 

4  Hyginus  tells    of  the  nine   journeys  to  the  shore  near  Amphipolis 
in  Thrace  of  Phyllis,    betrayed  by  Demophoon,    Fab.    59.       Cf.    Ovid, 
Her.  ii. 


66  DEMETER  AT  RLE  US  IS  AND  C NIDUS  n 

Heradidae^  where  a  Demophoon  figures  as  king  of  Athens, 
and  indeed  elsewhere  frequently  but  with  no  defined  asso 
ciations.  The  original  Demophoon,  unlike  Celeus,  could 
hardly  have  belonged  to  Celeae. 

The  composer  of  the  Homeric  Hymn  to  Demeter 
escaped  some  embarrassment  by  leaving  out  entirely  the 
great  Eleusinian  myth  of  Triptolemus.  The  record  of 
this,  which  has  been  preserved,  is  chiefly  in  the  shape 
of  pictures  and  a  few  fragmentary  lines  of  poetry  from 
a  lost  play  of  Sophocles.  Plainly  Triptolemus,  the  hero- 
prince  of  Eleusis,  was  adopted  by  Demeter;  he  was  her 
son  in  whom  she  was  well  pleased,  and  through  whom 
she  granted  to  men  all  manner  of  good  things  above  and 
beyond  what  it  was  his  especial  province  and  privilege 
to  give,  the  boon  of  plenteous  grain,  and  the  knowledge 
needed  for  its  planting  and  due  preservation.1  Temples 
were  built  and  altars  established  for  the  grateful  worship 
of  Triptolemus,  the  especial  favourite  not  of  Demeter 
only  but  also  of  Persephone.  Especially  sacred  to  him  was 
the  plain  where  first  he  showed  men  how  to  plant  and 
plough,  the  Rarian  plain,  which  was  set  apart  as  holy 
ground  for  ever,  and  from  which  was  derived  the  grain  for 
making  the  cakes  offered  up  in  the  Eleusinian  temples. 
Its  produce  came  as  a  part  of  the  revenue  of  the  Eleusinian 
temple  of  Demeter,  and  one  of  the  peculiar  duties  of  the 
priests  in  charge  at  Eleusis  was  to  keep  this  plain  of  Tripto 
lemus  free  from  all  pollution. 

Thus   by  examining  closely  the  Homeric  Hymn  more 

1  Whether,  as  I  have  perhaps  too  positively  suggested  above,  Triptolemus 
should  play  Demophoon's  vacated  part  in  the  Demeter  myth  is  another 
question.  On  this  whole  point  M.  Lenormant's  article  "  Ceres,"  in  Darem- 
berg  and  Saglio's  Dictionnaire  des  antiques  grecques  et  romaines,  may 
be  profitably  consulted. 


ii  WOMAN'S  LOVE  FOR   WOMAN  67 

even  than  a  phase  of  the  religious  activity  of  the  early  Attic 
mind  has  come  to  light.  Stories  grouped  themselves  about 
Demeter  at  Eleusis  which  first  revealed  the  greatness  of 
the  goddess  herself,  next  the  bond  between  her  and  her 
child  Persephone.  With  this  was  involved  the  worship  of 
Aidoneus — more  or  less  identified  with  the  local  hero 
Eubouleus — and  Persephone,  rulers  in  the  undiscovered 
country  of  the  dead.  Quite  unexpectedly  at  the  end  of  the 
story,  where  the  immemorial  observances  in  Eleusinian  wor 
ship  are  receiving  sanction  and  institution  from  Demeter,  it 
is  borne  in  upon  the  attentive  reader  that  Demophoon  and 
Triptolemus  do  not  belong  to  the  same  group  of  local 
traditions,  and  thus  a  glimpse  at  the  local  history  of  early 
Eleusis  and  of  some  neighbouring  shrine,  say  its  Pelopon- 
nesian  neighbour  Celeae,  is  given.  Furthermore  Triptolemus, 
and  perhaps,  in  his  own  forgotten  story  at  home,  Demo 
phoon  also,  represent  the  beneficent  influence  of  Demeter 
the  mother  of  corn  and  the  goddess  of  beautiful  abundance. 
This  beneficence  of  hers,  this  overflowing  generosity  in 
her  nature,  provides  for  more  than  creature  comforts, — it 
makes  for  what  is  highest  and  best  in  home  existence  and 
civilised  life. 

One  noticeable  touch  of  poetic  truth  in  the  story  of 
Demeter  at  Eleusis  is  the  way  in  which  woman's  love  and 
care  and  need  for  woman  are  portrayed.  When  Demeter  is 
sitting  all  forlorn  the  daughters  of  Celeus  come  upon  her, 
cheerful  and  careless  maidens  sent  forth  to  fetch  water. 
The  spectacle  of  self-forgetful  sorrow  which  the  goddess 
presents  seems  to  transform  them ;  they  ask  her  why  she 
tarries  in  so  lonely  a  place,  quite  aloof  from  the  town.  She 
ought  to  be  in  some  home,  they  urge,  for  there  in  the 
shadowing  halls  dwell  women  of  her  age  and  older  too. 


68  DEMETER  AT  ELEUSIS  AND  CNIDUS  n 

They  will  be  kind  in  word  and  in  deed.  Such  is  the  tender 
promise  of  consolation  which  the  maidens  give, — and  the 
promise  is  fulfilled ;  Demeter  is  as  much  loved  as  she  her 
self  is  loving  in  the  house  of  Eleusinian  Celeus,  her  home 
on  earth.  Through  the  whole  story  men  are  kept  in  the 
background.  lambe,  the  wayward  daughter  of  the  house, 
cheers  Demeter  with  her  gibes,  and  Metanira  refreshes  her 
not  with  wine,  but  water  perfumed  with  herbs  and  made 
more  strong  and  sweeter  for  the  tired  taste  with  barley. 
Demeter  is  thus  made  whole  by  her  own  bounteous  gift  of 
grain.  Silent  and  eloquently  sad  was  Demeter,  as  she 
moved  with  the  gentle  maidens  towards  their  home.  Not 
a  sound  was  heard  as  they  went,  nothing  save  her  footfall 
and  the  dulled  rustle  of  her  heavy  raiment,  dark  with  the 
colour  of  mourning.  Ministered  to  at  last  by  these  kindly 
womenfolk  she  smiled,  she  laughed,  and  her  spirit  was  glad 
within  her. 

This  pathetic  picture  lends  a  divine  sanction,  as  it  were, 
to  the  need  which  woman  in  trial  has  for  kindly  women, 
and  throws  light  upon  one  whole  side  of  the  worship  of 
Demeter.  For  Demeter,  as  the  upholder  of  the  ties  of 
marriage,  was  called  Thesmophoros,  and  a  festival  in  her 
honour  called  the  Thesmophoria  was  celebrated  by  women 
and  women  only.  To  this  worship  some  of  the  very 
noblest  aspects  of  the  Eleusinian  service  would  seem  to  be 
allied.  In  his  little -known  picture  of  the  Women  at  the 
Thesmophoria,  Aristophanes  has  made  abundantly  merry  at 
the  expense  of  Demeter's  Thesmophorian  woman's  festival, 
but  for  all  that  it  remains  more  than  ever  sacred. 

How  is  it  possible  to  translate  into  modern  words  the 
pious  aspirations  of  the  old-time  farmers  who  worshipped 
Demeter  at  Eleusis?  How  can  the  divinity  of  Demeter 


ii  DEMETER'S  LOVE  FOR  PERSEPHONE  69 

be  made  comprehensible  or  even  plausible  to  us  ?  Perhaps 
not  at  all,  but  yet  there  is  a  charm  in  the  goddess's  simple 
story  of  trial  and  triumph  through  sorrow  that  seems  to 
claim  the  hearts  of  men,  no  matter  how  alien  to  Greece 
their  birth  and  breeding  may  chance  to  be.  The  central, 
the  efficacious  and  communicable  grace  of  Demeter's  story 
is  the  love  she  bears  Persephone.  This  is  a  home  tie,  and 
through  this  Demeter  becomes  the  home  goddess.  It 
sometimes  seems  that  the  whole  range  of  ideas  dwelt  upon 
in  Demeter's  service  by  Greeks  is  covered  by  that  beautiful 
and  nobly,  broadly  English  word  harvest -home.  Under 
the  mastery  of  the  home  impulse,  of  love  for  her  own,  the 
great  goddess's  whole  beneficent  nature  gradually  unfolded 
itself.  If  you  should  say  that  Aphrodite l  loved  to  be  loved, 
I  might  by  way  of  contrast  maintain  that  Demeter  asked 
only  and  chiefly  to  love,  to  lavish  her  care  and  minute  pains 
upon  some  one  who  needed  protection. 

The  daughter  thus  beloved  of  Demeter  was  a  wondrous 
creature,  in  no  way  resembling  that  dread  Persephone  of 
Homeric  song.  A  child  of  Demeter  and  not  of  the  awful 
Styx,  her  face  bears  the  look  of  a  flower  freshly  opened. 
The  gentle  and  shyly  smiling  curves  of  her  lips  show  the 
lines  sometimes  seen  in  blossoms,  delicately  closed  because 
the  day  is  done.  The  maiden's  only  care  is  for  flowers,  and 
the  unmeasured  love  of  her  mother  is  her  shield  against  all 
harms,  until  the  fatal  hour  when  Hades  comes  and  robs 
her  of  the  pleasant  light  of  day,  snatching  her  away  from 
joy  in  flowery  things.  There  is  an  almost  adequate  repre 
sentation  of  Persephone  the  flower  maiden,  the  dear  and 
delicate  child  in  whom  dwelt  the  graces,  the  perfumes  and 

1  For  a  further  presentation  of  the  relation  between  these  two  divinities 
see  chapter  vii.  below,  on  Aphrodite  at  Paphos,  near  the  end. 


70  DEMETER  AT  ELEUSIS  AND   C NIDUS  n 

the  colours,  all  that  earth  shows  forth  in  all  the  lilies  of 
all  her  fields.  This  representation  is  a  statue  of  whitest 
Parian  marble,  so  small  that  were  it  less  perfect  it  would 
be  what  Mr.  Pater  so  prettily  calls  it,  the  merest  toy. 
This  wondrous  figure  was  found  by  Sir  Charles  Newton 
within  the  sanctuary  of  Demeter  and  Persephone — one  of 
their  most  ancient  sanctuaries,  that  of  Cnidus  on  the 
Triopian  promontory  in  south-west  Asia  Minor.  The 
first  and  untried  loveliness  of  a  maiden  unacquainted  as  yet 
with  grief  and  untested  by  the  world  has  passed  into  this 
most  delicate  Praxitelean  work. 

No  greater  contrast  can  be  imagined  than  that  between 
this  statue  of  Persephone  and  another  found  near  it,  and 
like  it  to  be  seen  now  in  the  British  Museum.  This  second 
statue  is  possibly  that  of  an  aged  and  careworn  priestess  of 
Demeter.  But  at  the  same  time  in  it  we  have  before  us 
the  embodiment  of  Demeter  herself,  as  she  was  in  act  of 
going  sad  and  despairing  to  the  house  of  Celeus.  Upon 
this  speaking  marble  the  unwitting  artist,  under  the  uncon 
scious  inspiration  of  the  sad  sweet  story,  has  set  the 
impress  of  sorrow,  and  with  it  a  touch  of  that  remembrance 
of  happier  things,  which  is  "sorrow's  crown  of  sorrow." 
Here  truly  is  the  goddess  Demeter,1  in  outward  semblance 
like  her  priestess,  a  stricken  woman  well  advanced  in  years, 

1  See  M.  Lenormant  in  his  article  "  Ce"res"  above  referred  to  : — "  On 
parle  d'une  De'me'ter  Fpcua  (Hesych.  s.v.}  ou  '  vieille  femme'  ;  ce  surnom 
fait  allusion  a  la  forme  que  la  de'esse  avait  prise  en  arrivant  a  Eleusis  et 
pendant  son  sejour  dans  la  maison  de  Celeos.  II  semblerait  en  re"sulter 
que  Ton  a  quelquefois  adore1,  et  par  suite  repre'sente'  De'me'ter  sous  ce 
de"guisement  emprunte".  M.  Newton  (Discov.  at  Halic.,  etc.,  p.  399), 
M.  R.  Foerster  (Raub  der  Kara,  p.  248),  et  M.  Heuzey  (Monum.  de 
I'Assoc.  des  Etudes  Grecques,  p.  10),  ont  me'me  cru  reconnaltre  la  De'me'ter 
Graia  dans  une  statue  de  vieille  femme  en  pied,  d'un  travail  fort  remar- 
quable  et  d'un  accent  tres  e'leve',  qui  provient  des  ruines  du  sanctuaire  des 
grandes  dresses  a  Cnide  (Newton,  op.  cit.  planche  Ivi.)  " 


ii  DE METER'S  RULES  FOR  RIGHT  LIVING  71 

but  noble  in  her  mien.  The  folds  of  her  apparel,  the  eager 
forward  leaning  of  her  head,  tell  of  vain  seeking  and  un 
availing  grief.  Here  stands  the  mater  dolorosa  mourning 
for  her  child.  But  somehow  hers  is  not  a  passive  woe,  for 
her  there  is  still  room  for  hope.  There  is  therefore  a 
strenuousness  in  Demeter's  sorrow  unlike  the  total  self- 
surrender  to  grief  of  many  sweet  portrayals  of  the  fainting 
Mother  Mary  at  the  Cross. 

For  Demeter  there  was  still  hope,  and  while  she  waited 
all  her  sorrow  and  the  fruitlessness  of  her  search  only 
served  to  bring  into  active  life  and  motion  her  impulse  to 
do  good.  Many  a  home  has  been  blessed  and  cheered  by 
some  such  selfless  presence  as  was  sorrowing  Demeter's  at 
Eleusis.  Deprived  of  the  home  love,  and  of  the  light  of 
her  sweet  daughter,  Demeter  became  the  good  fairy  and 
the  friend  of  the  Eleusinian  home  of  Celeus,  the  faithful 
and  all-wise  nurse  and  instructress  of  the  son  of  the  house, 
and  through  them  the  devoted  friend  and  helper  of  all  the 
homes  of  men  on  earth.1  For  Triptolemus,  with  the  know 
ledge  of  agriculture,  gave  the  laws  of  Demeter  to  men. 
These,  the  goddess's  rules  for  right  living,  were  no  doubt 
preserved,  with  momentary  glimpses  at  one  of  the  most 
elevating  of  the  many  beautiful  myths  of  early  Attica,  by 
Sophocles  in  his  lost  play  called  Triptolemus.  We  almost 
see  the  kindly  goddess  appearing  on  the  scene  and  giving 
her  beneficent  injunctions  to  young  Triptolemus,  for  a 
learned  expounder  of  the  eleventh  Olympian  Ode  of  Pindar 
quotes  from  the  play,  which  lay  open  before  him,  these 
solemn  words,  "  Set  my  commandments  on  the  tablets  of 

1  Ovid  brings  out  the  human  side  of  the  story  by  an  artifice  used  in 
Euripides'  Electro..  He  discrowns  Celeus,  and  brings  Demeter  to  a  poor 
man's  home. 


72  DEMETER  AT  ELEUSIS  AND  CNIDUS  n 

thy  heart."  But  we  really  know  nothing  except  that  these 
words  occurred  in  the  play.  The  sacred  words  may  have 
been  given  by  Demeter  to  Triptolemus,  or  quoted  from  her 
by  Triptolemus  to  some  favoured  man.  It  would  be  best 
of  all  to  know  what  the  commandments  were.  Perhaps 
some  notion  of  their  import  is  contained  in  the  Pythagorean 
rules  of  life  which  Porphyry1  puts  into  the  mouth  of 
Triptolemus. 

"Thou  shalt  honour  thy  father  and  thy  mother,  thou 
shalt  make  glad  the  gods  with  offerings,  and  do  no  wanton 
harm  to  beasts."  Upon  some  such  commandments  as 
these  Demeter  based  her  laws,  and  the  penalty  for  disobey 
ing  them  was  a  withdrawal  of  her  favour  and  a  denial  of 
all  her  good  gifts.  Our  own  Jewish  fifth  commandment  is 
not  very  different  from  Demeter's,  which  required  men  to 
honour  their  parents  in  order  that  the  earth  might  yield 
her  increase.  Before  Demeter  gave  her  gifts,  wretched 
men,  so  say  the  poets,  were  forced  to  live  upon  acorns. 
The  Demeter  who  preserved  the  homes  and  hearths  of 
men  from  want,  and  sanctified  the  bonds  of  family  life,  was 
a  noble  type  of  divine  womanhood,  above  and  beyond  all 
other  types  that  Greek  men  worshipped,  and  the  noblest  of 
the  three  great  Cnidian  statues  found  by  Sir  Charles  New 
ton  is  undoubtedly  a  representation  of  this  Demeter. 
Mother  of  peace  and  giver  of  plenty,  there  she  sits,  the 
Lady  Bountiful  and  Beautiful  of  Greece.  Her  gaze  is  now 
at  last  more  nearly  serene,  but  in  it  there  is  sadness  as 
a  memory  of  past  sorrows.2  The  goddess  has  made  her 

1  De  Abstin.  iv.  22. 

-  As  I  see  this  statue,  the  sadness  of  its  look  is  not  overpowering. 
Hence  I  venture  to  differ  from  those  who  see  in  it  Demeter  Achaia,  the 
mother  of  lamentations,  so  to  say.  If  I  understand  Mr.  Pater  aright,  I 
see  it  as  he  does. 


ii  DE METER  OF  CNIDUS  73 

peace  with  evil  and  the  power  of  death,  and  takes  joy  in 
such  sweet  communion  with  her  child  as  the  fates  allowed. 
It  would  be  too  much  to  say  that  this  Demeter  smiles,  but 
cheerfulness  lurks  half  suppressed  about  her  mouth,  just  as 
in  her  attitude  there  is  relief  and  great  repose  in  spite  of 
something  that  seems  almost  to  be  constraint.  A  curious 
mingling  of  opposites  there  is  both  in  her  posture  and  her 
face.  A  cheerful  look  that  tells  of  mystery  and  wherein  lurks 
the  memory  of  woe,  a  contradiction,  as  it  were,  between 
her  eyes  that  are  not  glad,  and  the  lower  lines  of  mouth 
and  chin  that  are  not  sorrowful.  A  posture  of  evident  rest 
and  yet  an  impression  of  bashfulness  and  almost  of  hesit 
ancy.  These  contrasting  expressions  existing  side  by  side, 
hard  to  seize  and  harder  still  to  describe,  together  with  the 
manner  of  holding  the  head,  and  the  .uneasy  grace  with 
which  the  limbs  are  disposed,  are  seen  alike  in  the  Demeter 
and  the  Persephone  of  Cnidus  which  are  attributed  to  the 
Praxitelean  School.  There  both  mother  and  daughter  are 
marked  by  these  same  family  traits,  a  shyness  which  goes 
with  all  natures  delicately  noble  and  free  from  self-seeking, 
that  shyness  which  men  learn  by  wandering  much  alone, 
and  musing  oft  when  only  the  trees  and  the  streams,  only 
the  green  earth  and  her  fruitful  fields,  are  there  to  sym 
pathise  and  understand. 

On  many  vases  and  in  some  bas-reliefs  it  is  hardly 
possible  to  distinguish  Demeter  from  Persephone.  This  is 
as  it  should  be  according  to  the  worship  rendered  them  at 
Eleusis.  Excepting  in  her  days  of  thoughtless  youth,  before 
her  trial  came,  Demeter's  Persephone  is  Demeter's  self 
twice  told.  During  the  third  of  every  year,  the  wintry 
season  when  Persephone  was  the  unwilling  bride  of  Hades 
and  abode  with  him  in  sadness,  Demeter  was  forlorn.  Joy 


74  DE METER  AT  ELEUSIS  AND  CNIDUS  .    n 

came  back  to  her  with  spring  when  Persephone  was  freed 
again  to  stay  with  her.  Their  sorrows  and  their  joys,  their 
life,  their  love,  their  happiness,  are  always  one.  If  under 
Demeter's  name  be  symbolised  power  to  grow  and  bear 
full  fruit  inherent  in  each  living  thing,  then  Persephone 
may  be  called  the  outward  blossoming  into  leaf  and  flower 
and  fruit.  But  who  shall  surely  say  which  of  these  two 
processes  or  powers  is  Demeter  and  which  Persephone? 
Where  does  the  domain  of  either  begin  or  end  ?  Demeter 
and  Persephone  each  represent  the  power  to  grow  and  the 
process  of  growth.  Of  these  two  elements  commingled  is 
their  soul,  which  is  one  though  it  dwells  in  two  bodies. 
Both  are  two  aspects  of  one  and  the  same  fact  in  nature, 
and  each  is  the  incarnation  in  her  joy  of  the  yearly  burst 
of  springtide  life  on  earth,  and  of  the  glad  abundance  of 
the  riper  year,  while  in  the  sorrows  suffered  alike  by  each 
is  shown  the  yearly  march  of  living  things  towards  death. 
Each  of  these  goddesses,  linking  her  happiness  to  sorrow 
and  rising  out  of  grief  to  gladness,  bears  the  testimony  of 
her  being  to  the  indissoluble  link  that  joins  life  to  death 
and  death  to  life ;  while  the  unfathomable  love  that  joins 
them  both,  and  makes  them  live  one  life  when  they  are 
sundered  just  as  when  they  are  together, — this  mirrors  for  us 
that  unity  which  pervades  the  world  and  makes  all  growth 
and  all  life  a  blossoming  from  the  unknown  depths  of  ever- 
fruitful  love — tokens  of  the  "nevei- dying  flowers  of  joy 
eternal," l  given  for  a  space  and  for  a  space  withdrawn. 

i  Perpetui  fiori  dell'  eterna  letizia.     Dante,  Par.  xix.  22. 


Ill 

DIONYSUS  IN  THRACE  AND  OLD  ATTICA 

THE  goddess  Persephone,  like  many  tragic  heroines  of 
lore  mortal  mould,  whose  mischances  moved  Athenian 
icarts  in  the  theatre  of  Dionysus,  loved  light  and  life.  The 
queen  of  the  netherworld  tarried  in  her  realm  of  darkness, 
longing  always  for  the  upper  earth  and  its  bright  ray. 
Nevertheless  she  was  the  wife  of  Hades,  and  stayed  in  his 
underworld  for  one  third  of  every  year.  This  bestowal  of 
the  loveliest  life  divine,  even  for  a  brief  season,  on  the 
fellowship  of  the  dead — or,  if  you  will,  this  transfiguration 
of  Homer's  death-dealing  goddess  of  the  dead  into  a  creature 
so  lovely  and  so  loving  that  she  charms  alike  and  com 
forts  the  realms  of  life  and  death — indicates  a  progress. 
The  wondrous  flower -change  suffered  by  Demeter's 
daughter  images  a  widened  and  deepened  view  of  the  life 
beyond. 

This  progress  began  even  in  those  minds  from  which 
the  Homeric  poems  sprang,  but  here  was  only  its  beginning. 
Homer's  Elysium  was  but  a  shadowy  and  merely  painless 
place  of  abode  when  compared  with  the  islands  of  the 
blest  of  the  latter-day  Greeks.  Such  satisfactions  as  Homer 
granted  in  that  neutral-tinted  place  to  a  favoured  few  were 


76          DIONYSUS  IN  THRACE  AND  OLD  ATTICA          in 

not  a  well-earned  meed  of  righteousness,  nor  were  the 
punishments  of  Tantalus  or  Sisyphus  conceived  of  by 
Homer  as  more  than  shadows  of  their  life  on  earth  ;  they 
were  dim  semblances  of  what  those  men  of  unworth  suffered 
ere  they  died. 

The  new  Persephone,  flower-changed  from  Persephone 
the  dread,  went  down  and  lighted  up  the  silent  home  where 
hitherto  the  spirits  of  men,  of  just  alike  and  unjust,  had  led 
a  shadowy  life  where  joy  laughed  not  but  only  smiled, 
where  sorrow  brought  no  pain.  Men's  ideas  underwent  a 
corresponding  change,  and  we  can  read  between  the  lines 
of  the  new  legend  of  Eleusis l  that  a  great  revolution  came 
to  pass  in  the  belief  concerning  immortality.  Hand  in 
hand  with  this  there  was,  partly  its  cause  and  partly  its 
result,  an  alteration  in  men's  ideals  of  duty  and  perfection 
in  the  present  life. 

The  clearest  and  most  musically  devout  expression  of 
these  new  feelings  and  thoughts  is  found  in  Pindar,  a  poet 
of  Boeotian  Thebes,  who  flourished  in  the  first  half  of  the 
fifth  century  B.C.  It  is  not  surprising  that  a  Theban  should 
have  spoken  as  one  having  authority  about  the  life  here 
after,  since  the  transformation  of  religious  belief  in  question 
was  especially  associated  with  Thebes  through  Dionysus. 
The  more  definite  and  substantial  expectation  of  future 
rewards  and  punishments,  to  which  the  Greeks  finally 
accustomed  their  meditations,  was  connected  everywhere 
with  the  worship  of  Dionysus,  a  late -born  god,  whose 
Theban  mother  died  at  Thebes  in  Boeotia,  that  he  might 
come  to  being.  In  Attica  this  changed  point  of  view 
which  Dionysus  everywhere  brought  with  him  was  associ- 

1  An  interesting  connection  by  way  of  derivation  has  been  suggested 
between  the  words  Eleusis  and  Elysium. 


in  THE  CRAZED  GOD  OF  HOMER  77 

ated  not  alone  with  him,  but  with  a  holy  alliance  sealed  at 
Eleusis  in  secrecy  and  mystery  between  Demeter  and 
Persephone,  with  Hades  hovering  near,  on  the  one  hand, 
and  on  the  other  this  new  godhead  of  Dionysus  freshly 
come l  to  Greece  from  the  north  and  east. 

From  Thrace  in  the  north,  and  from  Phrygia,  where  his 
first  worshippers  called  him  by  many  names,  but  chiefly 
Sabazius,  Dionysus  brought  much  that  was  barbarous. 
And  the  barbarous  and  non-Hellenic  quality  of  the  new 
god  made  him  a  sad  puzzle  to  the  Homeric  public.  In 
the  only  extended  and  discriminating  Homeric  account  of 
Dionysus,2  his  behaviour  is  represented  as  the  reverse  of 
courageous,  and  he  is  surnamed  " mainomenos,"  "beside 
himself,"  or  "crazed."  The  Greeks  before  Troy  knew  as 
little  of  Dionysus  as  of  Demeter,  and  the  ideal  heroic 
quality  was  inconsistent  with  the  worship  of  either  divinity. 
The  most  that  Homer's  heroes  did  was  to  admit  that 
Bacchus  was  a  god,  and  to  own  a  wholesome  fear  of  scorn 
ing  him. 

Lycurgus,  the  fierce  Thracian,  so  runs  the  short  and 
simple  story  of  Homer,  warred  against  the  new  divinity. 
He  pursued  the  crazed  young  god,  and  drove  him  to  fling 
himself  into  the  ocean.  In  the  depths  of  the  sea  Thetis 
showed  him  kindness,  and  kept  him  safe  from  the  dread 
hatchet  of  Lycurgus,  whose  ferocity  was  finally  punished  by 
total  blindness.  In  the  eyes  of  a  typically  vigorous  hero, 

1  Herodot.  ii.  52  ;  see  also  iv.  79. 

3  /Had,  vi.  135  ;  it  is  not  uncommon  to  regard  this,  and  that  other 
Homeric  place  where  Dionysus  appears,  as  untergeschoben  or  suppositi 
tious.  Until  some  knowledge  is  positively  gained  of  the  circumstances 
under  which  they  made  their  way  into  the  text,  the  whole  question  may 
be  neglected.  Whoever  formulated  these  accounts  had  behind  him  or 
them  a  really  established  conception  of  the  god,  and  this  is  what  concerns 
the  present  inquiry. 


78          DIONYSUS  IN  THRACE  AND  OLD  ATTICA          in 

Ajax,  let  us  say,  Dionysus  was  disgraced  by  his  incompetence 
for  war  if  not  by  his  flight,  and  no  punishment  miraculously 
overtaking  the  enemy  could  make  him  other  than  a  crazed 
and  cowering  being.1  Therefore  it  is  not  wonderful  that 
this  late -born  god  was  not  a  favourite  in  the  days  of 
Homeric  chivalry,  so  far  as  he  was  then  known. 

The  truth  is  that  Dionysus  was,  from  the  outset,  a  god 
of  contradictions.  He  represented  death  as  well  as  life. 
He  was  a  god  of  fiery  manifestations,  though  born  in  the 
lowland  plain  of  mountain-watered  Nysa,  and  though  he 
is  constantly  worshipped  as  the  representative  of  abundant 
vegetation.  He  was  attended  by  the  seasons,  by  the 
nymphs  of  flowing  waters  and  of  growing  trees,2  by  the 
Muses  and  by  old  Silenus — the  type  of  all  things  that  flow 
upon  the  earth — by  the  Satyrs,  always  half  beasts  and  half 
men.  A  prophet  divine,  Dionysus  was  sometimes  over 
come  by  his  own  gift  of  wine.  The  leader  and  inspirer  of 
holy  choral  song,  the  god  whose  worship  awakened  and 

1  Whatever  may  be  said  of  the  Homeric  conception  of  courage  and 
cowardice,  it  cannot  be  successfully  denied  that  a  certain  grotesqueness 
as  of  cowardice  attached  to  some  aspects  of  Dionysus  as  popularly  con 
ceived.  The  jokes  at  his  expense  in  the  Frogs  of  Aristophanes  are  always 
harping  on  this  string,  and  they  certainly  did  not  shock  but  pleased  the 
people  assembled  to  do  him  honour. 

2  Of  course  at  the  time  when  these  personifications  of  the  various 
movements  and  growths  in  nature  sprang  into  being,  there  was  nowhere 
any  consciousness  of  the  relation  they  bore  to  what  we  should  distinguish 
from  them  as  "  the  real  things"  or  "the  things  themselves"  ;  they  were 
the  real  things  for  those  in  whose  imaginations  they  first  sprang  into 
being,  and  their  confusing  multiplicity  and  elusive  nature  reproduce  the 
confusion  which  lies  upon  the  shifting  face  of  woodlands,  streams,  and 
meadows.  After  generations  had  dreamed  and  talked  of  these  baffling 
wildwood  creatures,  at  a  time  when  there  was  a  conscious  analysis  of 
popular  stories,  and  a  systematic  attempt  to  revive  ancient  belief,  we  find 
the  poet  Callimachus  giving  the  true  account  of  what  nymphs  were.  In 
the  Hymn  to  Delos,  w.  82  and  ff. ,  he  exclaims  :  "  O  Muses,  tell  me  truly, 
goddesses  mine,  did  oak-trees  then  come  to  be  when  the  nymphs  were 
born  ?  Nymphs  are  glad  when  showers  bring  increase  to  the  oaks,  nymphs 
are  sad  when  the  oak-trees  have  lost  their  leaves." 


in  DIONYSUS  A  GOD  OF  CONTRADICTIONS  79 

sustained  the  loftiest  strains  of  sacred  tragedy,  was  himself 
amid  the  brawls  of  leering  drunkards,  and  his  unreproving 
presence  sanctioned  all  the  worst  excesses  bred  of  unmixed 
wine.  His  Maenads  and  his  Bassarids,  when  they  wandered 
off  to  honour  him  by  penance  in  the  wilderness,  were  often 
seized  by  frenzy  fits,  that  made  his  name  a  signal  for  most 
murderous  deeds  of  harm,  and  yet  he  was  a  saviour  god 
who  suffered  death  and  insult  every  year  to  redeem  man 
kind.  Such  was  the  conflict  of  elements  in  this  new  divinity. 
With  wandering  tribes  Dionysus  came  from  Thrace ;  and 
Daulis  on  Mount  Parnassus  with  Boeotian  Thebes  received 
him  in  the  earliest  days. 

The  gradual  adoption  of  this  strange  worship  through 
out  Greece  may  be  called  a  first  Macedonian  conquest  or 
supremacy,  which  had  its  day  in  the  world  of  the  spirit, 
long  before  that  of  Philip  the  crafty  and  his  son  Alexander, 
the  "  great  Emathian  conqueror  "  of  Milton's  song.  Indeed 
Emathia,  the  cradle  of  Philip's  power,  was  that  district 
north  of  Mount  Olympus,  where  upon  the  spurs  of  Mount 
Bermius  were  those  fabled  rose-gardens  of  Midas  that 
early  harboured  the  myth  of  Dionysus.  These  prehistoric 
associations  gave  to  Philip's  intrigues  and  Alexander's 
masterful  ambitions  a  sort  of  home  sanction  from  the  god 
of  their  home,  and  hence  perhaps  came  the  great  con 
queror's  fondness  for  appearing  with  the  attributes  of 
Dionysus.1  Dionysus  was  the  first  Thracian  conqueror  of 
the  spirit  of  Hellas,  and  the  later  Greeks  so  conceived  him 
when  they  created  the  type  called  the  Indian  Dionysus, 
who  is  the  arch-conqueror, — a  deification,  as  it  were,  of  the 
Eastern  exploits  of  Alexander. 

1  Compare  the  masquerade  of  Antony  as  Dionysus  at  Ephesus. 
Hut.  Antony,  24. 


8o          DIONYSUS  IN  THRACE  AND  OLD  ATTICA         in 

The  undeniable  touch  of  his  original  Thracian  ferocity 
which  Dionysus  has,  even  in  his  most  highly  developed  and 
sweetly  civilised  aspects,  is  startling  at  first,  and  never  easy 
either  to  understand  or  to  combine  with  his  other  aspects. 
Nevertheless  a  contemplation  of  the  god  throughout  his 
whole  career,  and  especially  as  he  was  worshipped  at  Athens, 
will  outweigh  whatever  disgust  might  be  felt  at  the  lower 
phases  of  his  ritual,  and  leads  us  to  wonder  at  the  high 
purposes  and  great  truths  which  finally  associated  them 
selves  with  him. 

Now  it  is  important  to  define  terms  and  explain  the 
meaning  here  attached  to  the  word  Thracians.  Those 
Thracians  from  whom  Greece  learned  to  worship  Dionysus 
were,  of  course,  not  the  Thracians  personally  known  to 
Herodotus.  Before  his  day  the  earlier  Thracians  had 
migrated  southward  from  Thrace,  and  had  established 
themselves,  first  of  all  in  Phocian  Daulis,  and  then  in 
various  parts  of  Boeotia.  The  mountains  of  Attica  near 
Marathon  appear  to  have  been  visited  by  these  early  in 
vaders  from  Thrace,  and  a  record  of  this  survives  in  the 
legends  of  the  mountain-deme  Icaria.  Cadmus,  the  maternal 
grandparent  of  Dionysus  in  the  Theban  story,  is  said  to  have 
sojourned  in  Thrace  on  his  way  from  Phoenicia  to  Greece. 
The  fabled  visit  of  Dionysus  to  Icaria  and  King  Icarius  is 
perhaps  best  explained  by  connecting  it  with  the  migration 
headed  by  Butes.  Thracians  are  known  to  have  wandered 
over  the  islands  of  the  Aegean  under  his  leadership,  and 
not  far  from  their  track  was  Marathon,  whence  they  might 
easily  penetrate  into  Attic  Icaria.  With  the  name  of  Butes 
associates  itself  the  so-called  Thracian  sea  supremacy,  a 
time  in  prehistoric  days  when  Thracians  are  said  to  have 
controlled  the  Archipelago.  The  seat  of  their  power  was 


1 1 1  PROBA  TIONS  OF  A  THENIAN  DIONYSUS  8 1 

Naxos.  The  early  presence  of  Thracians  on  Naxos  ac 
counts  for  the  plentiful  growth  of  stories  connecting 
Dionysus  with  that  island,  called  in  the  end  especially  his 
own,  and  described  as  having  the  shape  of  his  vine-leaf. 
Naxos,  we  hear,  was  the  place  where  Dionysus  was  born 
and  bred.  This  tale  and  that  of  the  god's  visit  to  Icarius 
have  plainly  no  close  affinity  to  the  Theban  story.  In  Crete 
also  there  was  during  the  two  hundred  years  traditionally 
allotted  to  the  Thracian  sea  supremacy  abundant  chance  for 
the  creation  of  a  vigorous  legend  of  Dionysus.1 

Now  the  local  tales  of  Dionysus  in  vogue  upon  Naxos, 
the  other  Cyclades  and  Crete,2  would  be  sure  to  play  no 
inconsiderable  part  at  Athens,  which  was  in  especially  close 
communion  with  the  islands  of  the  Aegean.  From  the 
Archipelago,  therefore,  as  well  as  from  Boeotian  Thebes 
and  the  favouring  oracle  of  Apollo  at  Delphi,  can  be 
traced  influences  that  combined  at  Athens  with  the 
aboriginal  and  old  Attic  tale  of  Icarius  and  Icaria.  A  late 
comer  in  Athens,  the  Thracian  god  was  the  gainer  through 
long  waiting;  for  an  unconscious  selection  performed  by 
his  Athenian  votaries  neglected  the  wildest  and  basest 
features  of  his  story,  taking  from  Icaria,  Thebes  and  Naxos 
only  the  higher  traits.  Thus  Dionysus  at  Athens  became 
the  godhead  and  the  centre  of  the  widest  and  best  worship 
known  to  the  best  spirits  in  the  best  days  of  the  best  com 
munity  of  Hellas. 

His  ritual  underwent  a  triple  probation  before  Athens 
fully  adopted  him  and  he  so  shone  before  men  that  he 
became  the  tutelary  god  and  great  inspirer  of  Aeschylus, 
Sophocles,  Euripides,  and  Aristophanes.  There  was  first 

1  For  a  different  view  of  these  Thracians,  see  the  second  note  on  chapter 
vi.  below.  '-  See  an  account  of  them  by  Hoeck  in  his  Creta. 

G 


82          DIONYSUS  IN  THRACE  AND  OLD  ATTICA         in 

the  Icarian  or  old  Attic  probation,  then  the  probation  in 
Naxos  and  Crete,  and  thirdly,  the  probation  in  Thebes  and 
Boeotia.  All  these  were  preceded  by  the  god's  first  estate 
in  Thrace,  and  succeeded  by  his  public  adoption  and 
glorification  in  Athens.  We  ought,  therefore,  to  consider 
five  stages  of  Dionysus:  ist,  Dionysus  in  Thrace;  2d, 
Dionysus  in  old  Attica,  i.e.  Icaria;  3d,  Dionysus  in  the 
Archipelago  at  Naxos  and  Crete ;  4th,  Dionysus  at  Thebes ; 
and  5th,  Dionysus  in  Athens.  But  it  will  be  neither  con 
venient  nor  possible  to  consider  these  five  stages  with  equal 
fulness. 

After  an  account  of  Dionysus  in  Thrace,  the  considera 
tion  of  certain  debatable  points  belonging  to  Dionysus 
everywhere  will  lead  to  such  treatment  of  him  in  Thebes 
and  in  the  Archipelago  as  is  required  for  understanding 
him  first  in  old  Attica  or  Icaria,  and  finally  in  Athens. 
In  Athens  he  reached  his  final  stage  of  perfection,  and  a 
consideration  of  his  worship  there  will  form  the  climax  and 
be  our  abundant  reward  for  the  present  rather  perplexing 
study  of  the  details  of  earlier  phases.  And  yet  the  most 
painstaking  scrutiny,  the  minutest  examination  of  such 
evidence  as  may  be  had,  will  never  disentangle  completely, 
never  make  perfectly  plain,  just  what  elements  constituted 
the  Dionysus  first  worshipped  in  early  Greece.  His  charac 
ter  was  composite  from  the  moment  Greeks  worshipped 
him ;  for  in  Boeotia,1  as  in  Attica  2  and  on  Naxos,3  some 
part  of  him  was  native  to  the  soil,  and  he  was  nowhere 
wholly  Thracian.  There  are  dimly  visible  traces  of  the 
merging  of  an  early  Greek  worship  of  trees  into  the  more 
soul-stirring  rites  of  the  Thracian  newcomer.  The  confusing 

1  Dionysus,  surnamed  ZvdevSpos,  was  a  Boeotian  god. — Hesychius. 

2  Pausanias,  I.  xxxi.  4.  3  Athenaeus,  iii.  p.  78. 


in  THRACIAN  BIRTH-MARKS  OF  DIONYSUS  83 

thing  about  him  is  that  in  Thrace  as  well  as  in  Greece  he 
appears  as  in  part  a  tree -god,  attaching  to  himself  the 
attributes  of  a  primitive  and  barbarous  Jack-in-the-Green. 
But  it  is  convenient  to  make  abstraction  for  the  moment  of 
his  vegetable  antecedents  in  Thrace.  We  may  safely  con 
sider  that  their  chief  effect  upon  him  in  his  new  Hellenic 
dwelling-places  was  to  give  him  instincts  which  were  the 
remote  ties  of  a  half -forgotten  kinship  allying  him 
with  indigenous  tree -spirits  and  tree -worships.  The 
probably  milder  and  less  clearly  marked  observances  which 
he  found  in  Greece  were  soon  merged  into,  and  were 
obliterated  by,  his  intenser  and  more  brilliant  strain.1  This 
belonged  to  him  by  Thracian  birthright,  and  here  we  have 
his  birth-mark,  the  one  constant  element  in. early  Dionysus 
worship, — he  was  and  is  and  always  will  be  a  god  of 
Thracian  quality.  The  great  modern  historian  of  Rome 
has  thus  indicated  what  this  Thracian  quality  was.2 
"Maidens  dashing  at  midnight  down  the  mountainside 
with  brandished  torches,  the  boom  of  deafening  instruments, 
the  rush  of  streaming  wine  and  streaming  blood,  a  religious 
holiday-making  that  lashed  all  the  senses  to  a  furious  pitch 
of  frenzy  and  hurled  men  headlong  on  to  madness, — 
Dionysus,  in  all  the  glory  and  the  terror  of  his  name,  was 
a  Thracian  god."  That  portion  of  the  god's  character 
which  came  from  Thrace  in  early  times  may  therefore  claim 
examination  first. 

In  such  an  examination  it  must  be  taken  for  granted  that 
the  latter-day  Thracians  (and  so  far  as  religion  is  concerned, 
the  Macedonians)  reproduce  the  leading  qualities  of  the 


1  I  have  not  been  able  to  look  at  Rapp's  Beziehungen  des  Dionysos- 
kultes  zu  Thracien,  but  he  is  reported  as  taking  this  view. 
-  Mommsen  in  his  5th  vol.  ch.  vi.  p.  189. 


84  DIONYSUS  IN  THRACE  AND  OLD  ATTICA         in 

earlier  tribes,  who  finally  left  Thrace  (whether  of  Asia  Minor  or 
of  Europe  is  not  certain)  and  wandered  westward  and  south 
ward  with  their  native  god  to  Greece.    Making  this  proviso, 
we  may  say  that  the  Thracians  were  of  stubborn  spirit, 
uncompromising  like   the    rocky   lands   which    they   have 
always  defended  as  their  home.     A  wild  race  of  mountain 
robbers,  skilled  in   some   things   beyond   the   measure   of 
barbarians,  they  would   not   be  brought  under  the  yoke. 
"The    Satrae,"   says    Herodotus,   "have   never   been   the 
subjects  of  any,"  and  Thucydides,  himself  of  semi-Thracian 
parentage,  tells  of  mountain-dwelling  Thracians,  men  whom 
nothing  could  force  into  military  service.     In  later   days  ' 
they  were  ferocious  in  rejection  of  Christianity,  and  then, 
when    Christianised    at    last,    they    proved    most    faithful 
defenders  of  the  Church.     To  their  reckless  defiance  of  all  ' 
invaders  these  Thracians  joined  certain  views  about  religion, 
death,  and  life  hereafter,  which  bear  directly  upon  the  early 
type  of  Dionysus.     There  was  a  great  contempt  for  this 
present  life,  a  vivid  faith  in  a  better,  and  to  them  a  more 
real   and   important   life   hereafter.     The   saying  that  the 
body  is  the  grave  of  the  soul  was  originally  Thracian,  and : 
the  Thracians  used  to  gather  in  bitter   mourning  around  j 
each  new-born  child.     They  wept  for  sorrows  sure  to  come,  j 
But  if  a  tribesman  died  they  rejoiced  and  spoke  of  his. 
happy  deliverance.     A  dying  chieftain  left  many  wives,  andi 
after  his  death  high  court  was  held  to  know  which  wife  he: 
loved  most  dearly.     The  chosen  widow  was  rewarded  by| 
death  upon  her  husband's  tomb,  and  all  the  others  enviedi 
her  good  fortune.1 

Without  some  personal  god  to  lead  the  tribes  of  the 
dead,  such  an  intense  realisation  of  life  hereafter  would 

1  Herodot.  v.  4,  5. 


in  ASIATIC  THRACIANS  85 

hardly  have  thriven.  Indeed  the  real  beginning  of  it  all 
was  an  intensely  real  person  whose  dwelling-place  and 
whose  power  interested  the  tribes  in  Thrace  more  than 
even  their  native  hills,  whose  favour  they  prized  more  highly 
than  liberty  itself.  Such  a  person  possessed  their  pious 
souls,  and  was  the  god  of  Thracians  everywhere.  Hero 
dotus  was  astonished  at  the  intensity  of  their  devotion,  and 
remarks  especially  that  they  believed  that  there  was  no  other 
god  save  only  their  own  god.  This  enthusiastic  intensity 
and  almost  Mohammedan  intolerance  imposed  the  Thracian 
worship  even  upon  communities  otherwise  far  in  advance 
of  them.  Here  then  is  an  Asiatic  touch l  in  the  beginnings 
of  Dionysus ;  indeed  his  Thracian  origin  was  partly  Asiatic.2 
Thracians  in  Thrace,  Thracians  in  Asia  Minor  where  were 
settled  their  Phrygian  cousins,3 — all  these  tribes  worshipped 
Dionysus  under  a  name  of  their  own  choosing,  and  cele 
brated  in  his  honour  most  strange  and  violent  festivals, 

1  For  admirable  suggestions  about  the  eastern  aspects  of  Dionysus 
worship,  see  the  quotation  from  Sir  George  Birdwood,  K.C.I.E.  (who 
suggests  that  the  name  Dionysus  is  of  Phoenician  origin),  given  in 
Appendix  III.,  p.  164. 

-  Aristophanes,  Birds,  874,  with  scholiast's  note,  and  Wasps,  9. 

3  Servius  on  Aeneid,  iii.  i$adfin.\  and  especially  Herodot.  i.  28  ;  iii.  90; 
vii.  75.  See  also  Strabo  passim.  He  is  constantly  harping  on  the  affinity 
between  Thracians  and  Mysians,  Bithynians,  and  the  like.  But  Mommsen 
himself  could  not  be  more  in  despair  about  confusions  and  uncertainties 
regarding  the  peoples  of  Thrace  and  the  interior  of  Asia  Minor.  See  xii.  p. 
564,  where  he  gives  for  this  state  of  things  the  same  reason  given  recently 
by  Mommsen  :  diopiaai  8£  xa^€7rov-  ainov  5e  TO  rot>s  e-rrrjXvdas  fiap- 
ftdpovs  Kal  crrpartwras  &VTO.S  fj.ij  /3e/3a£ws  Kar^xeiV  rV  Kpa.rr)del<rav,  dXXd 
TrXai/TjTas  elj/ai  rb  ir\tov  e/c/SaXXoiTas  Kal  e/c/SaXXo/^i/ous.  airavTO.  5e  ra 
tdvT)  TO.VTO.  QpqiKid  rts  et/fd^ot  av  §td  r6  TTJV  Trcpaiai>  vt/j.e<rdai  TOVTOVS 
Kal  5id  TO  /XT?  7roXi>  e^aXXdrreu/  dXXiJXwj/  e/carfyous.  We  know  as  little 
(perhaps  less)  of  this  region  of  the  Balkan  peninsula  and  of  Asia  Minor  as 
Strabo  did,  simply  because  the  confusing  cause  has  continued  to  work. 
Here  has  been  and  is  still  a  confused  maelstrom  of  tribal  and  national 
antagonisms  in  constant  motion,  occasional  waves  of  more  or  less  temporary 
invasion  have  always  broken  in  upon  any  permanent  and  clearly  defined 
shaping  of  political  life  in  the  ancient  realm  of  the  Thracian  tribes. 


86          DIONYSUS  IN  THRACE  AND  OLD  ATTICA         in 

both  by  night  and  by  day.  The  barbarian  violence  of  these 
led,  no  doubt,  to  the  epithet  of  the  "crazed  god  "  for  Diony 
sus,  who  had  not  been  very  long  or  very  far  away  from  Thrace 
in  Homer's  day.  So  close  a  love  bound  worshippers  to  this 
god  that  they  sent  solemn  messages  to  him,  informing  him 
of  their  needs,  once  in  every  five  years.  A  messenger  once 
appointed  by  lot,  the  faithful  first  gave  their  messages,  then 
three  tribesmen  stood  forth  holding  with  points  stretched 
firmly  heavenwards  three  upright  spears.  Others  then  laid 
hold  upon  the  favoured  emissary's  hands  and  feet,  and 
tossed  him  upward.  He  was  greatly  blamed  and  another 
messenger  was  chosen  if  he  did  not  light  upon  the  spear- 
points  and  die  ;  if  he  died  all  was  well.1 

It  is  wonderful  to  see  how  indestructible  was  this  worship 
in  Thrace  of  the  leader  of  Elysian  joys,  of  the  marshaller 
of  the  blessed  dead,  the  real  king  of  the  real  world,  call 
him  Zamolxis,  Sabazius,  or  Gebeleizis,  what  you  will,  for  he 
has  later  names  in  Rome  and  Greece. 

The  Roman  festivals  of  mid-winter,  called  Brumalia  be 
cause  they  fell  upon  the  shortest  day  (bruma  or  breuissima) 
of  the  year,  and  also  those  called  Rosalia 2  for  the  mid 
summer-night  of  perfect  blooming  roses,  all  these  maintained  j 
themselves  with  astonishing  persistence  on  Thracian  soil.   ' 
Their  centres  lay  just  where  the  ancient  cradle  of  Thraco-  ; 
Macedonian3  Dionysus -worship  was   to   be  found.      The  I 

1  Herodot.  iv.  94. 

2  Tomaschek,  "  Brumalia und  Rosalia,"  Reports  of  Vienna  Acad.  Phil.    ! 
hist.  Class.,  1868. 

3  The   Macedonians   had  no   distinctive   religion.     As   soon   as   they   I 
appear   in   history   they  are  in   most  respects    Greek,  but  imbued  with   | 
Thracian   religious   ideas,   as   were  also   other  tribes   of  Illyrian  origin.    l 
These  ideals  from  Thrace  they  never  abandoned,  and  modified  only  by 
degrees  as  Macedonia  allied  itself  with  the  glories  and  greatness  of  Greece. 
Accordingly  the  distinction  between   Macedonia   and   Thrace,   Thracian 
and  Macedonian,  may  be  ignored  in  treating  of  the  history  of  Thracian 


in  BRUMALIA  AND  ROSALIA  87 

districts  are  two,  the  first  of  which  lies  among  the  snowy 
mountains  and  the  mountain  spurs  of  Olympus.  This 
district  is  extensive  if  it  be  understood  also  to  include 
Emathia,  the  heart  of  early  Macedonia,  and  to  take  in 
Mount  Bermius  and  the  fabled  roses  of  the  gardens  of 
Midas.  This  district,  contiguous  to  Greece,  may  be  con 
veniently  called  Pieria.  Distinct  from  this,  and  farther  to 
the  north,  lies,  near  the  river  Strymon  of  Orphic  fame,  the 
second  centre  of  this  worship,  which  bears  the  name  of 
Pieris. 

Not  far  from  Philippi,  which  lies  in  this  district  of  Pieris, 
was  found  an  inscription  belonging  to  a  Roman  epoch,  but 
in  the  spirit  of  its  piety  towards  Bromius  can  be  detected 
the  ancient  and  lingering  worship  of  the  Thracian  Dionysus : l 
"Hercules  shed  tears"  the  mourner  says,  " tfien  why  not  I, 
for  Venus  marks  thee  all  her  own  by  beauty  less  than  by  thy 
loving  heart  of  excellence  ?  Now  whether  the  mystic  maids  for 
Bromius*  service  scaled  chose  thee  on  flowery  meads  their 

Dionysus.  Only  we  may  have  reason  to  think  that  the  constant  and  close 
communion  between  Greece  and  Macedonia  reinforced  all  along  the  line 
certain  cruel  and  crude  Thracian  aspects  of  the  god  which,  without  Mace 
donia,  might  have  been  more  completely  softened  by  native  Greek  ideas 
and  observances.  As  to  a  fusion  between  Illyrians  and  Thracians  as  far 
as  matters  religious  are  concerned,  this  is  made  more  than  probable  by 
the  fact  that,  apart  from  their  share  in  Thracian  rites,  the  Illyrians  can 
not  be  found  to  have  had  a  traditional  religion.  Their  rudimentary 
observances  were  early  absorbed  in  the  wild  Thracian  cult,  just  as  were 
certain  local  cults  of  early  Greece.  See  in  the  Fragments  of  Olympiodorus 
(Dindorf,  §  27),  an  account  of  Valerius  in  Thrace  during  the  reign 
of  Constantine.  Hearing  of  treasure -trove,  he  got  orders  from  the 
emperor  to  take  possession.  He  found  the  ground  was  sacred,  dug  there, 
found  three  silver  statues,  upon  whose  removal  by  him  Thrace  and  Illyria 
were  overrun  by  Goths,  Huns,  and  Sarmatians.  Connected  with  these 
statues  and  their  holy  ground  were  mystical  observances  which  protected 
both  districts  :  iv  fj.£<ri$  yap  avrrjs  re  QP^KTJS  Kol  rou  'IXXiynKou  /car^/ceiTO 
TO,  rrjs  reXer^s.  One  of  the  statues  was  to  keep  Goths  out,  the  second  kept 
out  the  Huns,  and  the  third  was  a  bar  against  the  Sarmatians. 
1  C.  I.  L.  iii.  i,  686. 


88  DIONYSUS  IN  THRACE  AND  OLD  ATTICA         in 

mate  and  Satyr-friend  to  be,  or  whether  the  Naiads  require 
thee  to  join  their  torch-led  bands  and  hold  with  them  high 
festival,  wheresoever  thou  art,  dear  boy,  and  whatsoever  .  .  ." 
and  here  the  marble  record  ends,  yet  not  before  bearing  its 
testimony  to  the  persistence  through  Roman  days  of  Diony- 
siac  customs  inherited  from  those  unremembered  Thracian 
tribes  who  lent  their  god  to  Greece.  Dionysus  in  Thrace, 
accordingly,  must  be  looked  upon  as  the  head  of  a  world 
hereafter,  but  not  of  such  an  Elysian  realm  as  that  com 
monly  thought  to  have  satisfied  Greek  religious  belief.  The 
hereafter  presided  over  by  Thracian  Dionysus  was  the  world 
of  worlds,  the  real  life,  far  better  and  brighter  than  this. 
The  reality  of  this  Dionysus  world  and  Dionysus  worship  is 
witnessed  to  by  the  many  and  vain  struggles  made  by 
Christian  bishops  to  eradicate  from  Christian  merrymaking 
certain  heathen  practices  derived  from  ancient  Dionysiac 
festivals.  We  hear  repeatedly  of  these  practices  in  Christian 
documents,  especially  in  the  decrees  of  councils.1  The 

1  Ralli  and  Potli,  Hvvray^a,  etc.,  Athens,  1852-59,  ii.  p.  450.  Hav 
ing  failed  to  find  any  trace  of  this  monumental  work  in  the  catalogues  of 
the  British  Museum,  the  Bodleian,  or  the  Taylorian  Libraries,  I  applied 
to  my  distinguished  and  learned  friend  Mr.  Panagiotes  D.  Kalogeropoulos, 
Librarian  of  the  Greek  Parliament  Library  in  Athens.  In  his  answer  he 
gives  me  the  full  title  of  the  six  volumes  ;  I  quote  from  the  second.  The 
general  title  is  :  "Zvvray/Jia  r&v  deiuv  Kal  iep&v  Kavbvwv  rwv  re  dyiuv 
/cat  7ravev(pr)iJ,uv  diro<rrb\uv,  /cat  r&v  iep&v  olKOVfj.eviK&i'  Kal  TOTTLK&V 
vvvbduv,  Kal  r&v  Kara  ^fj.4pos  ayluv  Trartpw,  €Kdod£v  o~vv  TrXe/crrats 
&\\a.is  rr\v  6/c/cX?7(ria<7Tt/cV  Kardcrrao-LV  5te7roi5<rats  5taTa£e<rt,  yuera  TUV 
dpxaiuv  e^yrjrCov,  Kal  dia(f>6pui>  dvayvw<T[ji,dr(i)i>  virb  T.  A.  'PdXX?;  /cai 
M.  riorX??,  67/cpto-et  TTJS  (rytas  /cat  fj.eyd\-rjs  rov  Xpla-rov  €KK\T)<rias,  'A6ijt>ai, 
1852.  The  title  of  the  second  volume,  from  which  I  quote,  is  :  ot  0etot 
/cat  iepoi  Kav6ves  r&v  ayiuv  /cat  ira.veuty'fjij.wv  airoffrbKuv  rwv  h  Nt/cata, 
iv  ~K.ovffTavTivovTr6\ei,  fi>  'E0^<ry,  h  Xa\K7]d6vi,  ev  r(p  TpovXXy  rov 
/3a<riAt/coO  TraXart'ou,  ev  Nt/cata  rb  B.'  olKov/j.eviKuv  <rvi>65wj>,  /cat  rwv  cv 
K.ovffTai>TLVOVTr6\ei,  rfjs  re  ev  r^>  vd^>  rdov  aytw  a.iroar6\(j}v  Trpdorrjs  /cat 
Seur^pas,  Kal  rfjs  £v  r$  TTJS  aylas  2o0tas,  yevo^vwv  iep&v  avvbSuv, 
fiera  rrjs  e&yrnreus  'Iwdvvov  rov  Zuvapa,  Qeodupov  rov  BaXa-a/xtDi/oj,  /cat 
A\e%iov  rov  Apiaryvov,  Kal  HtvaKos  dvdXvriKov  airdvrwv  r&v  fv  rQ  devrfyy 
Kavovuv  fj-erd  rfjs  <rvfj.<puvtas  avruv,  'A&7Jvai,  1852. 


1 1 1  MODERN  MA  Y-DA  Y  FES  77  VA  LS  89 

Rosalia  is  described  as  a  "  wicked  and  reprehensible  holi 
day-making  " — Trawqyvpis  aXXo/eoTo? — celebrated  at  Easter 
in  remote  country  districts  through  the  persistence  of  an  evil 
traditionary  custom.  This  occurs  in  a  note  on  an  order  of 
the  sixth  council  at  Trullo,1  commanding  the  suppression 
of  various  heathenish  festivals,  including  also  the  Brumalia. 
Again,  in  the  tenth  century,  there  was  a  decree  against 
these  festivals ;  but  they  apparently  kept  their  hold  upon 
the  peoples  inhabiting  the  Balkan  peninsula  even  unto 
modern  times.2  There  is  a  curious  record  of  what 
seems  very  much  like  an  adaptation  to  Albanian  peasant 
life  of  the  Athenian  festival  of  Dionysus,  and  is  probably  a 
survival  of  the  Thracian  festival  called  Rosalia.3  During 
the  first  week  of  May  a  festival  is  held,  when  the  people 

1  January  15,  A.D.  706. 

2  The  earlier  custom  fixed  the  Rosalia  at  or  about  Whitsuntide. 

3  Arabantinos,    xp°v°ypa(t>'ia    T??S     'Hirelpov,    Athens,    1857,    vol.     ii. 
p.    191  :  T6re   (palverai  rots  (Hapytot.?)  Trapexwpirjd'r]  KCU  TO  5i/ca£w/za  rod 
KpoTflv  TT]V  Ka\ov/j^vrjv  eoprr^v  PoffdXiav  T)  Povad\ia  diapKovcav  airo  TT)S  A 
fj-eXP*1-  T^s  H'  Mafov,  6're  6  Xa6s  iKMytav  TroXir^v  Tivd  cbs  dpxnyov  evdv/u.ei. 
did  5ia.(j>6ptt}v  KUfJUK&v  (ncrjv&v'  /uera£i)  5£  TQVTUV  e/cp6rei  /cat  TrXaffTrjv  TLVO, 
/jLaxirjv,  (TXTj/Aartfo/u.eVwj'  5vo  (TTpaTiuTiKuv  au/mdrw^,  TOV  ^v  xPL<J"riaviKOVi 
TOV  5£  6dofj.aviKou  dpxnyov/J^vov  virb  ir\a.ffTov  TLaffffd  6'crris  <rwe\a.[Ji.[3dveTO 
aix/^aXwros,  /aerd  r^v  yevofjifrijv  ev  rr\  reXeurat'a  TjfJ.tpg.  TTJS  copras  \f/evdo- 
/j-dxTlv.     See  in  Folk-Lore  for  December  1890,  p.  518,  some  interesting 
notes  on  May- Day  observances  in  North-Western  Greece,  especially  the 
Ionian  Islands.      If  Mr.  J.  G.  Frazer's  informant  had  known  more  of  the 
festal  rites  of  antiquity,  he  would  no  doubt  have  carried  the  origin  of  actual 
customs  far  beyond  the  days  of  Venetian  supremacy.     As  it  is,  he  has 
enabled  Mr.  Frazer  to  give  a  most  graphic  description  of  the  flower  festival 
as  celebrated  in  medieval  times  at  Corfu  and  elsewhere.      After  all,  the 
best  authorities  on  such  a  point  are  the  Greeks  themselves.      I  have  my 
kind  friend  Mr.  Kalogeropoulos  to  thank  for  the  following  references.      He 
writes  :   "As  for  Roussalia,  you  will  find  in  the  fifth  number  of  'AvaroXiKT] 
'E7ri0ec6picns   (January   1873)  a   dissertation    of  Politis   irepi    Pov<ra\iov. 
This  periodical  was  published  in  Athens.       Kampouroglous  wrote   also 
about  Roussalia  on  the  24151  page  of  his  History  of  Athens.      Kam 
pouroglous  has  also  written  something  about  Roussalia  in  the  'E/SSo^ads 
(a  weekly  periodical  published  ^in  Athens).      Pandora,  another  periodical, 
contained  another  dissertation  of  Politis  rather  shorter  than  the  article  of 
the  Epitheorisis." 


90          DIONYSUS  IN  THRACE  AND  OLD  ATTICA         in 

choose  them  a  leader  and  give  themselves  over  to  pleasure 
in  various  comic  performances,  and  among  these  they 
especially  applaud  a  sham  fight  between  two  champions, 
one  a  Christian  soldier  and  the  other  a  Turkish  pasha.  It 
is  needless  to  say  that  the  pasha  is  worsted  and  carried  off 
prisoner  in  this  patriotic  Punch-and-Judy  show. 

But  now  the  main  features  of  Dionysus  in  Thrace  must 
be  brought  into  comparison  with  Dionysus  as  he  was  wor 
shipped  in  Greece.  From  being  the  god  of  the  only  real 
world,  he  comes  further  to  underlie  all  that  is  most  real,  all 
that  in  nature  arrests  the  eye,  startles  the  ear,  or  awes  the 
mind.1  The  two  views  of  the  god's  nature  lay  confused  in 
the  childlike  stories  and  rites  of  Thrace  and  the  Thracians. 
To  gather  a  complete,  an  early,  and  a  plain  record  of  the 
second  and  more  obviously  poetical  view  of  the  god,  not 
Thrace,  but  Phrygia  and  Thrace  together, — the  larger  Thrace, 
— must  be  applied  to.  Out  of  Phrygia,  as  has  been  intimated, 
came  in  part  the  ancient  Thracians,  and  in  Phrygia  dwelt 
of  old  cousins  of  theirs  who  had  fundamental  beliefs  prac 
tically  the  same  with  theirs.  From  the  dim  traces  which 
are  still  preserved  in  Thrace  and  Phrygia  may  still  be  read 
a  conception  of  Dionysus,  which  is  that  of  later  Greece 
reduced  to  simpler  terms. 

In  this  disentangling  process  it  will  be  convenient  to 
forget  the  so-called  infernal  character  of  the  god,  to  forget, 
that  is  to  say,  the  otherness  of  the  world  where  he  was 
thought  to  rule,  and  to  remember  alone  its  reality.  To  the 
tribes  of  Thrace  these  two  qualities  were  no  doubt  dimly 
identical.  Dionysus  the  god  of  reality  soon  becomes  an 
incarnation  of  the  elements.  Wine  was,  in  those  early  days 

1  Also  as  a  god  of  the  underworld  he  would  be  conceived  of  popularly 
as  sending  up  trees  and  plants  and  as  the  author  of  springs. 


in  THE  ANGEL  OF  THE  DARKER  DRINK  91 

of  story-making,  quite  as  much  an  element  as  water.1  Wine 
was  in  fact  regarded  as  a  perpetual  source  of  miracles,  and 
came  to  be  looked  upon  as  a  tertium  quid  in  whose  essence 
the  natural  and  the  supernatural  met  together,  sometimes 
for  good  and  sometimes  for  evil.  It  was  at  the  same  time 
an  elixir  of  life  and  a  draught  by  which  men  lost  their  senses 
and  their  lives ;  it  represented  and  incarnated  as  it  were  the 
sterner  as  well  as  the  more  charming  aspects  of  Bacchic 
power,  for  Dionysus  was  not  only,  as  Homer  2  calls  him,  a 
"spring  of  joy  for  mortal  men,"  but  he  was  also  the  "angel 
of  the  darker  drink." 3  He  came  offering  his  cup  and  invit 
ing  the  souls  of  men  "  forth  to  their  lips  to  quaff,"  and  thus, 
beguiled  by  wine,  they  accomplished  his  will,  following  after 
him  through  madness  and  the  gates  of  death.  On  the 
other  hand,  one  of  the  streams  with  which  Odysseus  filled 
that  trench,  out  of  which  the  flitting  ghosts  had  to  drink 
before  he  could  get  speech  of  them,  was  a  stream  of  sweet 
wine.  And  so  it  seems  that  wine  had  some  power  to  lead 
back  for  an  instant  to  the  gates  of  life  the  very  spirits  swept 
forth  by  its  spell  into  darkness  and  death.  Elsewhere  in  the 
Homeric  poems  we  hear  again  of  the  power  of  wine  to 
awaken  and  make  glad  the  anguished  spirits  of  the  beloved 
dead,4  and  a  modern  voice  has  uttered  for  the  Persian 
Omar 5  the  same  belief  that  wine  makes  glad  the  dead — 

And  not  a  drop  that  from  our  cups  we  throw 
For  earth  to  drink  of,  but  may  steal  below 

To  quench  the  fire  of  anguish  in  some  eye 
There  hidden, — far  beneath,  and  long  ago. 


1  See,  for  instance,   the  way  in  which  the  Pramnian  wine  given  by 
Maron  to  Odysseus  is  praised  in  the  ninth  Odyssey,  vv.  196-213. 

2  Iliad,  xiv.  325.  3  Fitzgerald's  Rubaiyat,  quatrain  xliii. 

4  Iliad,  xxiii.  220.  6  Fitzgerald's  Rubaiyat,  quatrain  xxxix. 


92          DIONYSUS  IN  THRACE  AND  OLD  ATTICA         in 

The  worship  of  unusual  brightness,  of  motion  and  flash, 
attached  itself  to  the  four  elements  of  wine,  fire,  water,  and 
gold.  Dionysus  was,  accordingly,  not  the  god  of  any  one 
of  these  only.  He  was  a  god  of  flush  and  flame,  made 
manifest  in  all  flashing  and  flowing.  It  is  not  by  making 
distinctions  between  the  various  elements,  but  rather  by 
translating  each  into  terms  of  the  others  that  he  is  best 
understood.  Gold  was  his  especial  element,  hidden  in  the 
bowels  of  the  earth  or  flowing  in  the  fabled  floods  of 
Phrygian  Pactolus.  Chrysopator  was  his  traditional  epithet 
used  by  a  Christian  poet  of  Egyptian  birth,  Nonnus,  who 
endeavoured  to  sum  up  the  legends  of  Dionysus  in  forty- 
seven  books,  each  consisting  of  a  large  array  of  Greek 
hexameters.  Father  of  gold  Dionysus  really  was  from  the 
first  in  Thracian  Pieris.  There  is  a  hill  near  Philippi  where 
the  mountain  tribes  of  Thrace  used  to  get  gold.  They 
called  it  Dionysus'  own.  Then  there  is  the  story  of  King 
Midas,1  which  belongs  to  Thrace  as  well  as  to  Asia  Minor. 
Midas  turned  all  he  touched  to  gold ;  and  the  story  is  in 
reality  a  blurred  record  of  Dionysus  as  Father  of  Gold 
where  Midas  stands  for  Dionysus.  From  Dionysus,  as  a 
mark  of  gratitude  for  hospitality  received,  was  lent  to 
Midas,  by  his  own  choice,  the  power  of  transmuting  all  he 
touched  to  gold,  and  when,  because  of  it,  he  was  brought 
near  to  starvation,  his  prayer  for  deliverance  was  to  Diony 
sus.  Dionysus  bade  him  wash  in  the  floods  of  Pactolus,  and 
from  this  bath  of  Midas  that  river  derived  its  fabulous  rich 
ness  in  gold.  In  Thrace  Midas  had  miraculous  rose-gardens 
on  the  flanks  of  Mount  Bermius.  There  he  sought  to  take 


1  See  Herodot.  viii.  138,  and  also  i.  14  and  35  ;  Pausanias,  I.  iv.  5  ; 
Xen.  Anab.  i.  2,  13  ;  and  Hyginus,  whose  igist  fable  gathers  nearly 
all  the  threads  of  the  story  together. 


in         ELEMENTS  OF  WINE,  FIRE,  WATER,  GOLD         93 

the  elusive  Silenus.  Long  his  efforts  were  in  vain,  but 
Silenus  at  last  drank  of  a  spring  with  which  the  wily  Midas 
had  mixed  wine.  Heavy  with  the  unknown  fumes,  Silenus 
was  seized,  and  Midas  never  loosed  his  hold  until  he  had 
heard  prophecies  about  things  to  come.  This  legend,  in 
which  Dionysus  is  both  Midas  and  Silenus,  both  captive 
and  captor,  couples  itself  with  the  abundant  record  of 
Dionysus  as  a  god  of  prophecy.  But  this  power  of  prophecy 
came  to  him  chiefly  if  not  solely  as  the  god  of  wine.  In 
wine  lurks  truth,  the  adage  says,  and  this  is  why  no  one 
was  answered  by  the  Thracian  oracle  on  Mount  Zilmissus,1 
before  he  had  taken  much  pure  wine.  Wine  belonged  to 
Dionysus  as  the  good  gift  that  freed  man's  soul  from  man's 
self  and  made  way  for  the  power  of  the  god  to  speak  his 
will. 

Fire  belonged  to  Dionysus,  partly  no  doubt  from  causes 
which  made  other  divinities  of  the  hereafter  who  were  also 
nature  gods  most  easily  appeased  by  torch -bearing  wor 
shippers,  and  which  gave  rise  to  various  fire  -  festivals. 
Furthermore,  traces  of  sun-worship  may  also  be  detected 
in  this  aspect  of  the  cult  of  Dionysus,  but  beyond  this 
the  violent  and  all  -  possessing  power  of  flooding  fire 
marked  it  as  his  own.  However  this  may  be,  Dionysus 
was  looked  upon  as  leader  of  the  band  of  fire -breathing 
planets  in  the  sky.  The  wielder  of  fire,  the  fire -faced, 
the  sower  of  fire  seed,  the  fire-begotten,  the  fire-thunderer, 
or  the  spirit  that  roars  in  high  flames, — all  these  epithets 
bestowed  on  Dionysus  mark  him  as  the  mover  and  maker 
of  fire.  Aristotle  tells  an  anecdote  that  attaches  this 
aspect  of  the  god  also  to  Thrace.  There  was,  he  says,2  a 

1  Macrobius,  Saturnalia,  i.  18,  T. 

-  Trepi  Oavfj-aatiov  cucoucr/xoTUH/,  cxxii.  133. 


94          DIONYSUS  IN  THRACE  AND  OLD  ATTICA          in 

well- known  place  of  Thracian  assembly  where  Dionysus 
promised  good  crops  for  the  coming  year  by  a  miraculous 
manifestation  of  flame  from  the  top  of  his  holy  hill  hear  by. 

Water  again,  which  like  wine  was  one  of  the  streams  of 
the  draught  poured  out  by  Odysseus  for  the  dead,  is  the 
element  of  the  Thracian  Dionysus,  as  is  shown  by  his 
ancient  Thracian  name  of  Dyalos,  god  of  springing  water. 
In  countless  stories  traceable  to  Pieria  and  Pieris,  the  god 
had  for  his  nurses  the  spirits  of  flowing  waters,  his  child 
hood's  companions  and  woman-helpers  were  nymphs  and 
naiads  of  the  mountain  sides.  These  were  but  one  com 
pany  of  all  the  elusive  troops  of  water-folk  that  flood  the 
whole  career  of  Dionysus.  To  water-nymphs  must  be 
added  innumerable  Satyrs  and  Sileni.  Nothing  is  plainer 
than  the  meaning  of  those  curious  representations  where 
Satyrs  are  pictured  in  the  act  of  smiting  the  ground,  whence 
obedient  to  their  stroke  a  nymph  arises.  Here  is  water 
calling  forth  water,  and  the  bubbling  up  of  a  mountain 
spring  is  the  gist  of  these  beautiful  picture-poems.1 

Silenus,  companion  of  Dionysus'  revels,  sharer  of  his 
adventures  in  early  and  in  later  days,  was  an  incarnation  of 
fluid,  a  water-man  who  might  at  any  time  change  again  to 
the  fluid  from  which  he  sprang.  This  being  true  of  Silenus 
and  all  Sileni,  it  is  in  a  measure  true  of  Satyrs  also,  who  are 
a  more  youthful  repetition  of  the  type  of  Silenus.  A  Silenus 
is  an  Asiatic  Satyr,  just  as  the  Curetes  are  the  Satyrs,  the 
Sileni,  the  Tityi,  and  the  Corybantes  of  Crete.2  Without 
at  all  pressing  this  statement,  we  may  learn  from  it  that 
the  attendants  of  Dionysus  are  as  elusive  as  the  god 


1  This  interpretation  was  first  given  to  them  by  Carl  Robert. 

2  Strabo,  X.  ch.  iii.  pp.  463-474.     This  whole  chapter  is  of  the  utmost 
importance  for  understanding  Dionysus  and  allied  divinities. 


in  DIONYSUS-SILENUS  THE  WATER-MAN  95 

himself,  and  as  each  of  the  four  elements  in  which  his 
power  was  chiefly  manifested.  You  no  sooner  begin  to 
see  what  Dionysus  and  his  creatures  are  than  they  are 
instantly  something  new.  Each  melts  into  another  when 
you  try  to  single  him  out  in  the  whirling  dance  of  Dionysus. 
Nonnus  tells  at  length  the  following  tale  of  Silenus.1  Hav 
ing  danced  his  best  in  eagerness  to  win  a  prize,  Silenus 
overreached  himself.  So  swift  became  his  motions,  so 
numberless  the  undulating  curves  and  swerves  of  his  limbs, 
that  all  at  once  he  was  himself  no  more,  but  swiftly  flowed  as 
a  river  onward  to  the  sea.  His  paunch  became  the  river-bed, 
his  hair  showed  upon  the  stream  in  guise  of  bulrushes  in 
the  shallows  near  the  shores,  and  the  pipes  he  played  on 
resumed  their  ancient  stand  and  grew  once  more  as  reeds. 
Through  his  attendants  the  Sileni,  the  god  has  been  abund 
antly  identified  with  water.  But  the  Thracians  and  the 
Phrygians  did  more  than  this,  they  frequently  identified  the 
god  himself  with  the  watery  element.  As  the  representative 
of  resistless  water's  flow,  Dionysus  was  in  their  conception 
bull-shaped.  The  usual  art-type  of  a  river  is  the  bull,  often 
times  a  man-headed  bull,  as  may  be  seen  on  many  ancient 
coins.  Horace  was  not  unmindful  of  this  when  he  wrote  : 

Sic  tauriformis  volvitur  Aufidus 
Qui  regna  Dauni  praefluit  Apuli.2 

There  was  a  notion  that  in  the  bull  resided  exhaustless 
vigour,  and  thus  the  bull -form  represented  flowing  and 
falling  waters  as  the  cause  of  growth  and  abundance.  The 
bull  serves  in  many  mythologies  along  with  the  cow  to  re 
present  any  sort  of  a  river  and  water  in  general,  and  even 
the  later  Greek  artists  remembered  this,  so  that  the  beautiful 

1  Dionysiaca,  xix.  261  ff.  2  Odes,  iv.  14,  25. 


96          DIONYSUS  IN  THRACE  AND  OLD  ATTICA          in 

Dionysus  of  later  days  is  sometimes  represented  with  the 
horns  of  a  bull. 

Thus  Dionysus  in  Thrace  at  last  stands  forth  well-nigh 
complete,  moving  in  all  that  makes  real  the  world  that  is — 
water,  fire,  wine,  and  flashing  gold.  Moreover  as  dwelling  in 
the  land  of  the  real  he  knew  the  truth,  and  would  declare  it 
when  rightly  approached  by  the  use  of  his  element,  whether 
of  wine  or  of  water.1  He  was  the  giver  of  oracles  in 
Thrace.2 

Other  features  still  remain  for  a  necessarily  discursive 
consideration.  The  Thracian  Dionysus  was  a  fierce  and  a 
pitiless  hunter,  a  man-slaying  power  that  would  rend  all 
creatures  in  sunder,  an  eater  of  raw  flesh.  This  feature  is 
undoubtedly  Thracian  and  only  survived  in  Greece.  The 
Thracian  Dionysus  appears  also  to  have  been  the  god  of 
lovely  song  and  the  leader  of  rhythmic  dancing ;  but  per 
haps  this  was  really  added  to  him  in  Greece.  The  under 
standing  of  both  these  aspects  requires  first  of  all  the 
observation  of  those  minor  and  less  constantly  heeded 
persons  who  form  his  countless  following. 

Muses,  Hours,  Graces,  Seasons,  mountain  nymphs,  Oreads 
or  hill  spirits,  Dryads  or  forest  maidens,  and  Hamadryads, 
— these  beautiful  emanations  from  the  central  divinity  of 
Dionysus  dance  around  the  triumphant  god,  and  mourn  him 
when  he  departs  from  them.  But  there  are  figures  more 
intimately  belonging  to  him,  a  numerous  band  of  so-called 
Bacchae,  Bacchants,  or  Bacchanals.  Just  so  in  early  Thrace 
and  Phrygia  the  wildly  roaming  woman  votaries  of  Sabazius 
were  called  Sabae.3  It  is  the  god  in  them,  not  they  them 
selves,  that  prompts  their  cries, — even  when  with  loud  lament 

1  Macrobius,  Saturn.  i.  18,  i.  2  Euripides,  Hecuba,  1267. 

3  See  schol.  on  Aristoph.  Birds,  874. 


in     BASS  A  KIDS,  SATYRS,  CORYBANTES,  CURETES     97 

they  mourn  Dionysus  dead  they  are  still  possessed  by  him. 
Through  them  he  cries  aloud  and  seeks  with  shouts  and 
wizard  motions  to  break  the  spell  of  death  and  call  to  life 
the  spring  —  his  quickened  self.  The  great  Thracian 
originals  for  these  Bacchanals  or  Bassarids,  as  they  are 
also  named,  were  called  Mimallones  and  Clodones.1  These 
were  women  nerved  to  more  than  woman's  work,  who  were 
much  feared,  and  who  followed  the  god  through  the  valleys 
of  Thrace.  They  all  did  nothing  of  themselves,  but  the  god 
in  them  cried  aloud,  as  they  darted  through  the  wilderness, 
the  well-known  Bacchant  cries  "  evoe  "  (eu  hoi  or  eu  sot)  and 
"  saboi."  Here,  in  the  myriad  women  possessed  of  the  god, 
we  have  a  personification  of  the  passive  side  of  nature,2  that 
into  which  the  god  as  motion,  as  moisture,  enters  to  make  it 
wholly  his. 

Turning  to  the  male  figures  that  swarm  continually  about 
him,  there  is  such  confusion  that  no  discrimination  can  at 
first  be  made.  Satyrs  and  Sileni,  already  spoken  of,  men  of 
the  water  and  the  wood ;  Telchines,  those  workers  of  metal 
from  Rhodes ;  Corybantes,  attendants  given  to  Dionysus  by 
his  Phrygian  mother  the  great  nature  goddess  Cybele; 
Curetes  from  Crete, — all  these  and  others  have  their  func 
tion.  Some  of  them,  as  the  Satyrs,  represent  ever  and  anon 
the  coarser  aspects  of  wine-drinking ;  some,  like  the  Telchines, 
have  to  do  with  Dionysus  as  father  of  gold,  and  naturally 
associate  themselves  with  one  who  at  Eleusis  is  almost 
identified  with  Plutus,  or  rather  Pluto,3  the  god  of  nether 

1  Plutarch,  Alexander,  ch.  ii. 

2  See  T.    A.    Voigt's   article   "Dionysus"   in   Roscher's  Mythological 
Lexicon,  where  this  is  made  very  plain  in  an  admirable  presentation  with 
which  Mannhardt  himself  would  not  have  found  any  fault. 

3  According  to  Hesiod  Plutus  was  a  son  of  Demeter,  and  therefore  not 
her  son-in-law.      His  father  was  lasion,   and  he  was  begotten   in  Crete 
(Theogony,  969  and   fif. )      This  Plutus  was  an  errant  and  elusive  god, 

H 


98         DIONYSUS  IN  THRACE  AND  OLD  ATTICA  in 

gold  and  nether  realms.  In  these  figures  who  can  rush  into 
excesses  unworthy  of  the  god  himself  his  majesty  is  so  far 
saved.  Another  group  of  Dionysus'  male  followers  must 
now  be  sought  to  represent  his  most  darkly  cruel  aspect. 
Those  curious  and  elusive  beings,  called  Pans  or  Aegipans, 
swarm  in  every  Bacchanalian  rout,  and  though  they  make 
less  noise  perhaps  than  the  Corybantes  with  their  drums, 
not  the  shouting  Bacchanal  women  themselves  do  such 
savage  deeds  as  the  Pans  when  they  are  roused.  Aegipans 
and  Bacchanals,  therefore,  are  often  possessed  with  the  native 
savagery  of  the  barbarous  man-eating  Dionysus. 

Pan  or  Aegipan  was  originally  capable  of  better  things,1  and 
in  fact  a  pure  and  sweetly  simple  worship  of  Pan  was  cherished 
at  Athens.  He  was  originally  a  shepherds'  god,  and  could 
not  withstand  successfully  the  superior  claims  of  greater  gods 
not  confined,  like  himself,  to  the  hamlets  and  haunts  of 
lonely  shepherds — rocky  places  on  the  very  summits  of 
mountains.  Hence  he  surrendered  his  independence  and 
is  found  in  many  shapes,  both  large  and  small,  swarm 
ing  in  Dionysus'  train.  All  Pans  have  goat's  legs  and 
horns,  but  not  all  are  made  utterly  savage  by  contact  with 
the  Thracian  god.  Some  Pans  are  young  and  graceful, 

a  wanderer  whom  the  lucky  would  fall  in  with  and  straightway  become 
rich  and  glorious.  The  Hesiodic  conception  of  him  bears  a  striking 
resemblance  to  the  poetic  notion  of  Dionysus,  to  meet  whom  and  feel  whose 
power  was  to  be  forever  blessed,  only  the  typical  element  of  Plutus  was 
only  gold  and  never  wine.  Hesiod's  Plutus  is  certainly  not  yet  identical 
with  the  brother  of  Zeus  and  Demeter,  Hades  (ibid.  455),  who  was  also 
Persephone's  husband  (ibid.  769).  The  Aidoneus  who  robbed  Perse 
phone  from  Demeter  (ibid.  913)  was  apparently  thought  of  as  dimly 
identical  with  Hades.  Pluto  was  a  daughter  of  Oceanus  and  Tethys 
(ibid.  355),  of  those  who  with  lord  Apollo  and  the  rivers  take  in  hand  the 
bringing  up  of  men  from  their  boyhood,  according  to  the  allotment  of 
Zeus,  dvdpas  Kovpt£ov<n  crw  'A7r6XXawi  &VO.KTL  \  /ecu  TJora/iots, 
AIOS  Trdpa  fj-dipav  Covert  (ibid.  347  and  f. ) 
1  See  Preller's  Greek  Mythology, 


in  THE  SAVAGE  PAN  OR  A  EG  IP  AN  99 

spending  the  time  piping  on  reeds,  and  in  the  end  this  type 
of  dear  little  Pan  loses  his  goat's  legs  and  becomes  a  civil 
ised  and  harmonious  young  Satyr. 

The  fiercer  type  of  Pan  or  Aegipan  is  distinguished  from 
the  Satyr  by  courage.  For  the  Satyrs  personify  among  other 
things  the  aspect  of  Dionysus  which  made  Aristophanes  lam 
poon  him  as  a  weakling  and  a  coward.  "  Always  drunk  with 
wine,  the  Satyr  kind  is  insolent  through  and  through.  Their 
brawling  threats  are  loud,  but  war  drives  them  in  headlong 
flight.  Of  plentiful  readiness  they  in  the  dance,  and  skilled 
beyond  others  in  draining  the  widest  cups  and  the  deepest  to 
their  very  dregs."1 

A  savage  god  had  need  of  courage  in  his  deeds  of  grim 
and  reckless  cruelty,  and  this  quality  was  neither  in  Satyr 
nor  in  Silenus,  but  only  in  Pan.  All  three  had  horns  and 
tails  and  pointed  ears,  and  all  seem  at  times  to  be  nearly 
the  same,  since  all  represent  the  frenzy  of  the  god  superfi 
cially  called  drunkenness,  yet  in  a  crisis  your  Satyr  is  a  tear 
ful  drunkard,  your  Silenus — in  spite  of  all  his  wisdom — is  a 
maudlin  drunkard,  while  your  Pan  is  always  fighting  drunk. 
The  Aegipans  of  Dionysus  did  not  wear  horns  and  hoofs  for 
naught,  since  they  appear  to  have  bequeathed  these  append 
ages  to  the  devil  of  many  a  modern  legend.  With  these 
Pans  were  associated  all  panic  terrors  inspired  through 
them  by  Dionysus.2  These  savage  and  sudden  inroads  of 
terror  form  a  counterpart  to  those  equally  mysterious  and 
equally  sudden  ecstasies  and  bursts  of  reckless  joy  sent  most 
frequently  by  Dionysus  to  his  women  votaries  the  passive 
Bacchanals.  There  is  just  this  difference :  the  Bacchanals 

1  Nonnus,  Dionysiaca  t  xiv.  120  ff. 

2  Eurip.  Rhesus,   36  ;   Bacchae,    303  ff. ;    Pausanias,   II.  xxiv.  6  ;    see 
also  Nonnus,  Dionysiaca,  x.  at  beginning. 


ioo        DIONYSUS  IN  THRACE  AND  OLD  ATTICA          in 

feel  the  Bacchic  bliss,  whereas  the  Pans  inspire  the  panic 
fears.     And   here   begins  the  second   stage   of  this   short 
inquiry   into  the   savage  Dionysus,   the   rending  god,  the 
power  from  the  world  underground  that  directs  the  earth 
quake  and  its  various  attendant  catastrophes  of  fire,  of  water, 
and   of  endless   panic  fears.      The  Bacchanals   and  Pans    ! 
associate  themselves  even  in  Grecian  story  with  frenzied  rend-    : 
ings  of  men  and  animals.     The  frenzy  prompting  these  acts 
is  so  plainly  from  the  god  that  such  rendings  may  be  called   ; 
acts  of  worship — features  of  his  ritual.     From  this  we  may   j 
argue  backwards  to  a  considerable  degree  of  cruelty  and   ' 
savagery  in  the  worship  of  the  Thracian  Dionysus.     Human   i 
sacrifice    was   assuredly    not   uncommon   in    the    earliest 
worship  of  Thracian  tribes,  and  it  is  likely  to  have  been 
begun  with  an  effort,  like  that  described  by  Herodotus,  to  i 
send  a  messenger  to  the  god  in  his  world  beyond.     A  thirst 
for  blood  of  some  kind  is  very  universally  attributed  to  the 
dead  by  early  legends.     This  thirst  for  blood  is  no  doubt 
often  a  thirst  for  substantial  life,  as,  for  instance,  was  that  of 
the  shades  who  flocked  around  Odysseus,  but  it  allies  itself 
to  cruelty,  and  its  wildest  fullest  realisation  is  in  the  Bacchic 
Thiasos.    This  is  but  a  much-needed  collective  name  .cover 
ing  all  followers  of  the  god,  all  who  are  so  full  of  him  that 
they  know  not  what  they  do — the  Bacchanals  or  Maenads 
on  the  one  hand,  and  the  Aegipans  on  the  other. 

When  this  mysterious  frenzy  seized  his  Thiasos,  woe 
betide  man  or  beast  whom  they  found  on  their  way ;  the 
god  possessed  them  utterly,  and  they  wrought  his  miracles 
unarmed.  The  warlike  excesses  of  horns  and  hoofs  into 
which  Dionysus  hurried  his  outrageous  Pans  beggar  de 
scription.  In  these  tales  the  god  is  revealed  in  his  most 
awful  aspect.  He  was  named  Anthroporraistes  or  man- 


in  DIONYSUS  OMESTES  101 

wrecker  on  the  island  of  Tenedos,  while  the  Chiotes  spoke 
of  Dionysus  Omadios,  glad  of  raw  flesh.  For  this  last 
savage  trait  another  epithet  elsewhere  used  was  Omestes, 
favourer  of  raw  flesh ;  and  these  names  may  serve  to  indicate 
the  dark  background  of  the  Thracian  legends  concerning 
the  god.  "  Dionysus  Omophagus,  the  eater  of  raw  flesh, 
must  be  added,"  says  Mr.  Pater,1  "to  the  golden  image  of 
Dionysus  Meilichius,  the  honey-sweet,  if  the  old  tradition  in 
its  completeness  is  to  be  ...  our  closing  impression ;  if 
we  are  to  catch  in  its  fulness  that  deep  undercurrent  of 
horror  which  runs  below  this  masque  of  spring,  and  realise 
the  spectacle  of  that  wild  chase  in  which  Dionysus  is  ulti 
mately  both  the  hunter  and  the  spoil."  Indeed,  what  the 
same  gifted  writer  says  of  the  Bacchanals  of  Euripides  may 
be  applied  to  the  legend  of  Dionysus  as  a  whole  :  "  It  is  it 
self  excited,  troubled,  disturbing,  a  spotted  or  dappled  thing 
like  the  oddly  shaped  fawn-skins  of  its  own  masquerade,  so 
aptly  expressive  of  the  shifty,  twofold,  rapidly  doubling 
creature  himself."  Truly  "  the  darker  stain  "  of  the  gloomier 
Thracian  legend  is  always  "shining  through";  no  matter 
what  cheerful  aspect  of  the  Hellenised  Dionysus  you  may 
choose,  he  is  always  a  god  of  tragedies  more  than  in  name. 
This  is  exemplified  in  the  old  Attic  legend  of  Icaria  as  well 
as  elsewhere. 

Before  taking  up  that  legend,  however,  a  further  considera 
tion   is  desirable  of  what   the   god  whose  power  it  exalts 
came  to   represent  for  religious-minded  Greeks.     Already 
•   in  proving  that  Dionysus  stood  for  wine,  water,  fire,  and 
!   gold,   and   in   telling    of  his   savage   aspect,   it   has   been 
impossible  entirely  to   exclude   points   that   are  surely  of 
later  growth ;  so  now  in  speaking  of  Dionysus  as  the  com- 

1  "The  Bacchanals  of  Euripides,"  Macmillan's  Magazine,  May  1889. 


102        DIONYSUS  IN  THRACE  AND  OLD  ATTICA          in 

peer  of  Apollo,  leader  of  the  Muses,  and  himself  the  god 
of  song,  it  will  not  be  possible  to  exclude  the  earlier  germs 
and  signs  of  this  later  transformation,  this  translation 
of  Dionysus  from  the  depths  of  the  Thracian  wilderness 
and  the  world  of  the  dead  to  the  peaks  of  Grecian 
Parnassus. 

All  the  elements  of  Dionysus  associate  with  themselves 
a  notion  of  swift  brightness,  of  inevitable  sparkle.  The 
ecstasy  that  words  cannot  utter  finds  a  near  escape,  its 
native  utterance  in  song.  Hence  the  pious  Pindar  sings  in 
a  famous  prelude  that  "  water  is  best,  but  gold  is  like  a 
beacon  blazing  through  the  night,  while  songs  that  celebrate 
Olympian  glories  shine  pre-eminent  even  like  the  flaming 
noonday  sun."1  In  another  prelude2  the  same  poet  sings 
of  three  things  most  useful  to  man :  "  Winds  that  blow  and 
waters  that  fall  in  fertilising  showers, — showers  that  are  the 
children  of  the  clouds;"  and  then  as  a  climax,  song,  in 
which  no  doubt  he  would  have  us  feel  the  swiftness  of  fresh 
winds  and  the  richness  of  glad  rain :  "  But  if  any  show 
bravery  in  deeds,  honey-sweet  song  shall  spring  forth  and 
fly  from  tongue  to  tongue  a  pledge  assured  of  glorious 
achievements  to  come." 3 

The  many  familiar  phrases  connecting  poetic  inspiration 
with  springing  waters  and  pure  flowing  streams,  or  with 
wine, — as  when  we  hear  that  Alcaeus,  Xnacreon,  Sophocles, 
and  others  could  write  and  sing  their  best  only  when  under 
its  influence, — all  these  fancies  group  themselves  around 
Dionysus  as  an  incarnation  of  the  swift  flashing  power  and 
resistless  beauty  that  attaches  both  to  wine  and  water,  but 
finds  its  fullest  utterance  in  the  changeful  cadences  of 

1  So  begins  his  first  Olympian  Ode. 
2  Twelfth  Olympian  Ode,  beginning.  3  Ibid. 


in  DIONYSUS  AND  THE  MUSES  103 

perfect  song,  the  graceful  undulations  and  fitful  variations 
of  an  ordered  and  yet  wayward  Bacchic  dance. 

The  worship  of  song  and  dance  implied  in  their  associ 
ation  with  Dionysus  came  as  an  afterthought,  or  rather  as  a 
climax,  for  in  this  worship  his  diviner  essence  was  most 
made  manifest.  In  these,  at  last,  were  fused  and  expressed 
all  the  elements  in  which  the  power  of  Dionysus  moved. 
The  elemental  force  in  wines  and  waters,  in  gold  and  fire, 
had  been  rudely  associated  and  yoked  together  in  the 
Thracian  and  Phrygian  notion  of  Sabazius.  Whence  came 
the  further  step  which  made  Dionysus-Sabazius  the  god  of 
harmonious  songs  and  rhythmic  dances  ?  This  may  be  left 
in  doubt,  though  tradition  and  the  story  of  Thracian  Orpheus 
indicate  that  this  transformation  was  thought  of  as  beginning 
far  back  in  Thrace. 

Thracians,  we  are  told,  established  on  Mount  Helicon 
the  worship  of  the  Muses,1  and  one  of  the  sayings  at  a 
Boeotian  festival,  which  had  other  features  of  Thracian 
origin,  shows  how  close  a  bond  united  Dionysus  and  the 
Muses.  At  this  wild  and  Thracian -seeming  festival,  ap 
propriately  named  the  Agrionia,2  the  Boeotian  women 
searched  long  and  anxiously  for  the  god  with  many  lament 
ations  ;  then,  as  at  a  sudden  flash  of  light,  they  said  each 
to  her  neighbour, — "  He  is  not  here  but  hath  fled  away  to 
hide  him  with  the  Muses."  The  Muses,  as  known  to  their 
earliest  adorers,  were  emanations,  so  to  speak,  from  Dionysus 
the  god  of  song.  The  higher  and  least  earth-born  of  his 
qualities  required  the  same  separate  incarnation  and  im 
personation  which  was  given  in  Satyrs  and  Sileni  to  his 
coarser  strain.  The  history  of  the  worship  of  the  Muses, 
how  they  came  to  be  nine  instead  of  three,  their  original 

1  Strabo,  X.  Hi.  17,  p.  471.  2  Plutarch,  Sympos.  viii.     Proem. 


104        DIONYSUS  IN  THRACE  AND  OLD  ATTICA          m 

number,1  would  lead  too  far  afield.  It  appears  that  both 
the  Muses  and  the  Graces  were  adjuncts  to  Dionysus  and 
Apollo  when  these  divinities  appeared  as  representatives  of 
idealised  song  and  dance.  Dionysus  was  called  Melpomenos 
in  this  capacity,  and  under  the  same  aspect2  Apollo  was 
surnamed  Musagetes.  As  before 3  in  speaking  of  the  more 
catholic  and  benign  aspects  of  Apollo,  so  now,  in  penetrat 
ing  into  the  higher  regions  and  more  inspiring  features  of 
Dionysiac  worship,  in  treating  of  the  perfected  Dionysus, 
you  come  face  to  face  with  the  perfect  unison,  the  flawless 
concord  of  the  two  great  gods  of  poetry,  dancing  and  song. 
Not  only  did  Apollo  share  with  Dionysus  his  mountain  of 
Parnassus  and  his  Delphian  temple,  but  Dionysus  freely 
gave  room  for  a  temple  of  Delphian  Apollo,  "  the  Pythion 
of  the  Icarians,"  4  in  his  own  first  Attic  home,  close  to  the 
flanks  of  high  Pentelicus. 

Thither  we  now  must  go.  Having  examined  closely  the 
aboriginal  Dionysus  in  Thrace,  and  having  considered  the 
prime  factors  in  the  Bacchic  godhead  from  various  points 
of  view,  we  turn  to  that  stage  in  the  history  of  Bacchic 
worship  which  our  own  countrymen  have  done  so  much  to 
illuminate — the  first  worship  of  Dionysus  in  the  highlands 
of  Attica  at  Icaria. 

In  this  legend 5  traces  of  old  Thracian  savagery  survive 

1  See  Oscar  Bie,  Die  Musen  in  der  Antiken  Kunst. 

2  The  Muse  Melpomene  may  be  regarded  as  an  emanation  from  this 
Dionysus,  who  is  the  Dionysus  of  Eleutherae.     See  Pausanias,  I.  ii.  5. 

3  Introductory  Lecture,  at  the  end. 

4  See  Mr.  Carl  Buck  on  this  and  other  discoveries  (p.  174,  Am.  Journal 
of  Archaeology ,  June  1889).    The  worship  apparently  came  up  to  Icaria  from 
the  Marathonian  tetrapolis,  where  there  was  a  Delion  whose  rites  were  con 
nected  later  with  the  Athenian  Delia.     See  chap.  viii.  below. 

5  See  Otto  Ribbeck  on  the  whole  subject,  Anfaenge  des  Dionysoskults 
in  Attica,  Schriften  der  Univ.  zu  Kiel,  1869.     Also  F.  Osann  in  the  sixth 
meeting  at  Cassel,  October  1843,  of  the  Verein  deutscher  Philologen  und 
Schulmaenner. 


in  THE  ICARIAN  LEGEND  105 

in  spite  of  transformations  wrought  by  the  Attic  instinct, 
which  always  seeks  to  observe  measure.  The  Thracian 
legend  thus  moderated  to  suit  Attic  taste,  and  brought  into 
parallelism  with  the  Eleusinian  Demeter-legend,  runs  as 
follows 1 :  "  Under  King  Pandion — the  fifth  since  Cecrops— 
Demeter  and  Dionysus  came  to  Attica.  Dionysus  was 
entertained  by  Icarius,  in  Epacrian  Icaria,  while  Demeter 
was  the  guest  of  King  Celeus."  Icaria  comprised  an  upland 
valley  hemmed  in  on  one  side  by  Mount  Pentelicus 2  and 
separated  from  Marathon3  by  a  huge  mountain  wall,  which  is 
cleft  by  the  stream  that  flows  from  Rapendosa.  Two  other 
forest  cantons,  Plothea  and  Semachidae,  formed  the  triple 
confederation  of  mountaineers  to  which  Icaria  belonged. 
The  three  bore  a  collective  name,  Epacria.  An  especial 
bond  between  Semachidae  and  Icaria — like  that  between 
Eleusis  and  Celeae — is  suggested  by  the  existence  of  a 
parallel  legend  to  the  effect  that  Semachus  at  Semachidae 
first  entertained  the  god.4  Icarius,  who  has  been  truly 
called  "the  heroic  type  of  the  Athenian  farmer,  devoted  to 
his  trees,  his  crops,  and  his  only  daughter  Erigone,"  was  so 
irresistibly  hospitable  that  to  the  latest  days  the  wor 
shippers  of  Dionysus  were  fond  of  seeing  him  sculptured 
in  the  act  of  entertaining  their  god,  a  bearded  and  portly 
presence,  who  arrives  noisily  and  numerously  attended.5 
He  is  pictured  in  the  act  of  having  his  sandals  removed. 
This  office  is  deftly  performed  by  an  obsequious  dwarf  of  a 

1  Apollodorus,  Bibl.  iii.  14,  7. 

2  For  a  view  taken  from   Icaria  and    looking  toward   Pentelicus,  see 
Appendix  XI.  i.  49. 

3  For  the  view  toward  Marathon,  see  ibid.  48. 

4  See  Stephanus  Byzantinus,  s.v.  ST/^CIX^CU.     See  Appendix  II. 

5  A  doubt  has  been  raised  whether  this  might  not  be  anybody  enter 
taining,  rather  than  Icarius  in  particular.     See  Professor  Gardner,  Journal 
of  Hellenic  Studies,  v.  p.  137. 


io6       DIONYSUS  IN  THRACE  AND  OLD  ATTICA          in 

Satyr.  So  overcome  is  father  Dionysus  with  the  journey 
upward  from  Marathon, — where  no  doubt  his  Thracians 
landed  him, — and  by  copious  draughts  of  retzinato l  on  the 
way,  that  he  requires  a  second  Satyr  to  lean  upon.  In  one 
bas-relief  a  palm  shows  upon  the  right,  and  a  fig-tree  on  the 
left,  symbolising,  both  of  them,  that  epithet  of  Dionysus 
which  is  least  certainly  Thracian,  Dendrites  2  or  the  spirit  of 
growing  trees.  Here  perhaps  is  a  something  added  to  the 
incoming  god,  which  came  to  him  from  a  primitive  worship 
of  trees,3  inherited  by  the  Icarian  shepherds  from  remote 
and  fetish-worshipping  ancestors. 

Dionysus  proved  no  ungrateful  guest,  but  rewarded  Icarius 

1  Plut.  Quaest.  conviv.  v.  3.     My  attention  was  called  to  this  passage 
by  the  much  lamented  Dr.   Schliemann.      Plutarch  (or  whoever  speaks 
under  Plutarch's  name)  discusses  the  dedication  to  Poseidon  and  Dionysus 
of  the  pine  tree,  accounting  for  it  by  their  common  element  of  moisture 
and    productivity  :    Kal   Noaeiduvl    ye    <pvTa\fj.i^,    Atofivay  de    devdpiry, 
wdvres  (ws  eiros  eiTretv)  "E\\-rjves  dtiovo-iv.    Then  he  accounts  for  Poseidon's 
especial    claim   on  the    pine  by  its  use  in  shipbuilding,   adding  r$  de 
Aiovvcrct)  rrjv   irirvv  dvLepwffav,   ws   e<f)T]dijvovo~av  rbv  olvov'  Kara,  yap  ra 
TTiTvudr]  xwP'tt  \tyovo~iv  TjSiV  olvov  ryv  a^ireXov  (frepeiv'  Kal  rty  dep^brrira 
rrjs  yijs  Geo^pacrros  amcmu  .   .   .   ou  pty  d\\d  /ecu  TTJS  Trirvos  avTrjs  et'/cos 
a.7ro\aveu>    rrjv    a^ireXov,     exotiffrjs     eTriTrjdei6TrjTa     TroXXTji/     TT/SOS 
o-UTrjpiav    olvov  Kal  5iafj,ovr]i>'  ry  re  yap  TT/TTT;  Trdvres  e%a\et- 
(frowi  TO,  ayyela,  xai  TTJS  pfTivrjs  virofjt,iyvvov(ri  TTO\\O!  ry  oiVy, 
Kadairep  Ei;/3oe?s  TUV  "E\\rjviK&v,    Kal  TUV  'IraXiK&v  oi  trepl  rbv  Udoov 
oiKovvres.      Thus  the  Greek  peasant  of  to-day  need   not  be  too  much 
abashed  when  the  vials  of  Occidental  scorn  are ; poured  upon  him  because 
he  likes  still  the  resinated  wine  of  antiquity — a  high-bred  taste,  hard  for 
some  to  acquire. 

2  Ibid,  and  also  iv.  6,  where  a  curious  attempt  is  made  by  the  Athenian 
Moeragenes  to  prove  that  the  god  of  the  Jews  is  none  other  than  Dionysus. 
The  season  and  also  the  manner  of  their  chiefest  feast  is  appropriate  to 
Dionysus  :    rty  yap  Xeyofj^vrjv  vrjo-Teiav  aK/j,dfoi>TL  rpvyrjTCp  rpair^as  re 
TTporldevTCU   Travrodair^  oTrcipas,   VTTO  (T/c^z/ats    Kal   KaXidalv  eKK\-r)fj.dro}v 
MaXicrra  Kal  KLTTOV  8iaireir\eyiJLevais.      Little  as  Moeragenes  convinces  by 
his  argument,  he  yet  supplies  interesting  touches  in  a  picture  of  country 
side  and  greenwood  festivals  in  honour  of  Dionysus. 

3  With  this  same  primitive  worship  may  also  be  connected  the  cere 
mony  of  "  Aiorai,"  or  the  hanging  of  effigies  on  trees  which  characterised 
the  Icarian  festival,  and  was  accompanied  by  the  song  Aletis.     See  for  the 
facts  Miss  Harrison's  Mythology  and  Monuments,  p.  xl.  and  ff. 


in  ICARIUS  ENTERTAINS  THE  GOD  107 

by  showing  him  how  to  plant  and  tend  the  vine,  and  how 
to  make  wine.  Till  then,  shepherd-like,  Icarius  drank  water 
chiefly  and  milk  sometimes.  Dionysus  held  forth  to  him 
a  goblet  crowned  with  foaming  wine,  and  said,  according  to 
the  gist  of  Nonnus' l  report :  "  Lo,  thou  art  blessed,  for  men 
shall  sing  in  future  days  thy  praises  thus :  Icarius  rather 
than  Celeus  himself  be  praised,  and  Erigone,  his  daughter, 
beyond  the  praises  of  Metanira,  Celeus'  spouse.  Tripto- 
lemus  gave  the  wheaten  ear,  but  from  Icarius  we  have  the 
wine-flashing  clusters  of  summertide.  Compared  to  these, 
what  are  the  gifts  of  Demeter  ?  Corn  brings  not,  as  wine,  a 
sweet  release  from  grief."  This  comparison  is  taken  from 
Nonnus,^  but  it  meets  us  in  every  version  of  the  story,  and 
doubtless  represents  the  typically  Attic  way  of  regarding  the 
boons  of  corn  and  wine,3  and  it  foreshadows  the  ultimate 
union  at  Eleusis  and  Athens  of  Demeter  and  Dionysus. 

The  story  of  Icarius  may  now  be  continued  in  borrowed 
words :    "  The  vine  is  carefully  tended  and  reared ;  but  a 

1  See  the  forty-seventh  book  of  his  Dionysiaca,  from  which  details  have 
been  borrowed  in  the  following  account. 

2  See  Dionysiaca,  xlvii.  47  and  99. 

3  There  are  no  traces  of  a  local  Icarian  attribution  of  Demeter's  gift  to 
Dionysus,  but  Pliny  (Nat.  Hist.  vii.   59)  says  that  Eumolpus  introduced 
the  cultivation  of  the  vine  and  trees.      Here  then  is  the  trace  of  a  local 
Eleusinian  legend  attributing  the  Icarian   gift   of  Dionysus   to  another 
Thracian  figure  in  early  legends.     But  this  variation  had  no  hold  upon  the 
imagination  of  religious  men.     In  fact,  as  Dr.  Merriam  has  abundantly 
shown,  ' '  the  legends  of  Eleusis  and  Icaria  were  so  closely  connected  in 
the  minds  of  the  mythologists  that  the  one  naturally  suggested  the  other. 
Not  only  has  Statius  linked  Icaria  with  Eleusis,  but  Apollodorus  (iii.  14,  7) 
has  done  the  same  ;    as    also    Schol.  Aristophanes,   Knights,  697  (here 
Icarius  welcomes  Dionysus  who  is  a  fugitive  from  outrageous  Pentheus) ; 
Philostratus,L£/z'.tf.  39  ;  Gregory  Nazianzen,  Orat.  iii.  p.  iooc;  and  Lucian, 
De  Saltatione,  39,  40,  where  he  speaks  of  both  stories  being  represented 
in  full  by  the  dancers  of  the  day.     In  some  writers  they  were  even  confused, 
as  in  the  Etymologicum  Magnum,  62,  n,  where  Erigone  is  interchanged 
with  Persephone.      In  Servius  ad  Virgilii  Georgica,  i.  19  Triptolemus  is 
called  the  son  of  Icarius.      Nonnus  links  the  two  stories,  Dionys.  xxvii. 
283-307." 


io8        DIONYSUS  IN  THRACE  AND  OLD  ATTICA          in 

he-goat  breaks  into  the  enclosure  and  injures  it  with  char 
acteristic  voracity.  Icarius  in  anger  slays  the  goat,  offers 
him  in  sacrifice  to  the  god,  blows  up  the  skin,  oils  it,  and 
gives  it  to  his  companions  to  dance  about,  thus  originating 
the  sport  of  askoliasmos,  a  usual  accompaniment  of  the 
Dionysiac  festival.  The  divine  gift  is  not  destroyed  by  the 
goat J ;  but  Icarius  is  soon  enabled  to  follow  the  injunctions 
of  the  god,  to  travel  about  the  country  with  a  waggon  loaded 
with  wine  skins,  proclaiming  the  joys  of  the  vine,  with 
practical  applications,  and  without  water.  "  2  The  Epacrian 
shepherds  marvelled  at  the  glorious  gift.  "  Whence  comes 
it  ?  "  they  cried,  "  for  it  is  not  from  the  Naiads,  their  water- 
streams  are  not  sweet."  Furthermore,  as  Nonnus,  with 
rather  frigid  elaboration,  makes  them  proceed,  "it  cannot 
be  oil  from  olives,  that  is  not  for  man  to  drink.  It  is  not 
honey,  for  that  begets  a  most  swift  and  strong  surfeit." 

But  the  shepherds  abused  the  gift  and  were  made  drunk 
with  too  much  wine.  Just  here  the  more  awful  significance 
of  Dionysus  and  his  worship  shows  through  the  transparent 
innocence  of  these  shepherd-simpletons.  Ignorant  of  what 
drunkenness  was,  the  legend  goes  on  to  say,  they  thought 
themselves  undone.  In  fact  it  was  the  savage  god  who 
entered  in  and  possessed  them  wholly.  Dionysus  required 

1  K-fjv  pe  <f>dyr]s  tiri  j>i£av,  fytws  gri  Kapiro^op^ffu 
&<rffov  eTTiffTre'tffa.l  <roi,  rpdye,  Ovoftfry. 

— Evenus  of  Ascalon,  Anthol.  ix.  75. 

2  I  quote  from  p.  65  of  Dr.  Merriam's  "  Report  to  the  Committee 
of  the  School  of  Classical  Studies  at  Athens,"  a  monument  of  brilliant  and 
accurate  scholarship  which  does  honour  to  the  American  School.  To  this 
Report  I  refer  for  an  exhaustive  account  of  the  Icarian  legend  in  all  its 
bearings,  with  full  enumeration  of  all  the  sources.  It  is  fortunate  that  the 
singularly  interesting  relics  of  the  country  deme  Icaria  have  found  such  an 
able  interpreter.  This  great  contribution  to  our  knowledge  of  old  Attica 
is  contained  in  the  same  publication  of  the  American  Archaeological  In 
stitute  with  the  seventh  Annual  Report  of  the  Managing  Committee  of  the 
American  School,  Cambridge,  1889. 


in  THE  SHEPHERDS  SLAY  ICARIUS  109 

the  sacrifice  of  what  was  most  prized  and  best  beloved,  and 
so,  crazed  by  him,  these  shepherds  slew  Icarius,  their  bene 
factor  and  their  friend,  and  then  they  swooned  away,  wholly 
overcome  by  the  power  of  the  god.  When  sense  returned 
they  woke  and  saw  what  they  had  done.  .  The  repentant 
murderers  buried  Icarius  after  washing  him  in  a  mountain 
stream  on  the  edge  of  the  forest,  the  very  stream  perhaps 
which  all  visitors  of  Icaria  see  to-day  flowing  through  a  glen 
near  by.  It  is  shaded  by  mighty  plane  trees  so  gnarled  and 
hoar  that  it  would  seem  as  if  no  antiquity  could  outstrip 
theirs.  Not  far  below  these  trees  the  stream  plunges  down 
a  steep  and  reaches  the  valley  of  Rapendosa,  whence  it  flows 
past  Marathon  into  the  bay  towards  Euboea.  Near  the 
plunge  made  by  this  stream  is  a  cave  still  inhabited  during 
the  heats  of  noon  by  shepherds  as  simple  as  those  in  our 
old  Icarian  story.1 

Icarius  having  been  slain  and  buried,  a  terrific  vision  of 
him — so  says  Nonnus,  and  it  is  by  no  means  sure  that  he 
invented  this  episode  of  the  ghost 2 — clad  in  a  blood-flecked 
garment,  the  "  dappled  herald  telling  of  a  murder  to  which 
none  living  bare  testimony,"  appears  to  Erigone.  The 
daughter  wildly  seeks  his  grave.  Search  is  long  in  vain. 
Tired  of  her  own  way,  at  last  Erigone  follows  her  faithful  dog 
Maera,  whose  instinct  leads  her  to  the  place.  Then,  fordone 
with  horror,  she  hangs  herself  beside  her  father's  grave. 

At  this  point  Nonnus  is  truly  pathetic  in  his  account  of 
the  lament  over  Icarius  and  Erigone.  "  Wine,  gift  by  my 
own  Bromius,  given  to  make  men  cease  from  care,  sweet 
wine  hath  brought  but  bitterness  to  Icarius.  Gladness  it 

1  For  photographs  see  Appendix  XL  i.  46,  47. 

2  The  dog  Maera  is  made  into  the  messenger  in  some  versions,  e.g. 
Apollodorus,  above  cited. 


i io          DIONYSUS  IN  THRACE  AND  OLD  ATTICA        in 

gave  to  all  mankind,  but  death  to  him.  Sweet  wine  was  a 
foe  to  Erigone;  for  truly  Dionysus,  who  comes  to  chase 
dull  care  away,  hath  pursued  to  her  death  and  slain  with 
grief  our  own  Erigone." 

As  if  to  leave  no  doubt  that  Dionysus  wielded  a  power 
of  possession  which  drove  men  to  madness  and  despair,  and 
was  not  solely  a  god  of  wine  and  jollity,  the  last  episode  of 
the  Icarian  legend  gives  a  woful  account  of  how  Erigone's 
death  was  atoned  for.  A  mania  laid  hold  upon  all  the 
maids  of  Icaria.  With  one  accord,  stung  by  a  conscious 
ness  of  guilt  for  what  their  fathers  and  brothers  had  done, 
they  flew  to  the  mountain-side  and  hanged  themselves  upon 
the  forest  trees.1  Apollo's  oracle,  questioned  in  extremity, 
could  only  urge  the  punishment  of  the  guilty  murderers  of 
Icarius.  These  slayers  slain,  a  respite  came  at  last.  Ever 
afterwards  the  shepherds  of  Epacria  worshipped  Icarius, 
Erigone,  and  Maera  (the  faithful  dog  whom  she  had 
tenderly  reared),  and  kept  their  memory  green  at  a  yearly 
festival.  On  these  occasions  small  effigies  were  suspended 
from  the  branches  of  forest-trees,2  to  commemorate — so  at 
least  the  story  ran — Erigone's  manner  of  death.  Meanwhile 
the  father,  his  daughter,  and  the  sagacious  Maera  were 
translated  to  the  firmament.3  Icarius  with  his  waggon 
becomes  Bootes  with  his  Wain ;  Erigone,  the  Virgin ; 

1  A  similar  mania  for  hanging  is  recounted  as  overtaking  the  maidens 
of  Miletus  ;  see  Gellius,  xv.  io  ;  and  Plutarch,  De  Anima.     For  an  extra 
ordinary  array  of  similar  epidemics  of  suicide,  see  E.  Bachut,  Histoire  de 
la  Mddecine  et  des  Doctrines  Mddicales,   Paris,  1873,   vol.   i.   p.    56  and 
ff.,  where  several  curious  modern  parallels  to  this  feature  in  the  legend  are 
given. 

2  For  an  admirable  discussion  of  this  practice,  see  Miss  Harrison  on 
the   "Mythology   of  Athenian  Local   Cults,"  pp.    xxxix.  and  ff.    of  the 
Mythology  and  Monuments  of  Ancient  Athens. 

3  Hyginus,  Fab.  130,  end  :  ' '  Erigone  signum  virginis,  quam  nos  Justi- 
tiam  appellamus  ;    Icarius  Arcturus  in  sideribus  est  dictus,   canis  autem 
Maera  Canicula." 


in          LEGENDS  PARALLEL  TO  ICARIAN  STORY       in 

Maera,    the   Dog-star;    and   the   Cantharus   of  Dionysus 
appears  close  at  hand  as  the  Crater.1 

As  to  the  legendary  epoch  to  which  this  visit  of  Icarius 
and  this  old  Attic  legend  should  be  assigned,  the  whole 
question  hangs  together  with  similarly  insoluble  ones, — with 
the  date,  for  instance,  of  Demeter's  arrival  at  Eleusis. 
King  Pandion — himself  of  most  elusive  date — under 
whose  rule  Demeter  and  Dionysus  came,  is  said  to  have 
had  dealings  with  Thracians.  These,  however,  were 
Thracians  already  established  south  of  Thrace  at  Daulis, 
on  a  spur  of  the  Parnassus,2  toward  the  confines  of  Boeotia. 
All  this  favours  at  least  the  supposition  that  Thracian  in 
fluence  made  itself  felt  in  very  early  times  on  Attic  ground. 
Accordingly  the  Icarian  legend  of  Dionysus  is  likely  to  be 
as  old  as  other  tales  told  in  Boeotia  or  on  the  islands  of  the 
Archipelago,  such  as  Naxos  and  Icaros,  both  of  which 
claim  the  glory  of  having  given  birth  to  the  god  of  wine. 
In  fact  the  mountains  of  old  Attica  gave  to  the  Thracian 
god  a  home  no  less  his  own  than  the  islands  conquered  by 
Thracians.  Icaria  became  to  him  in  Attica  what  Pieria  was  in 
Thrace ;  and  the  shape  which  tragedy  took  in  Attica,  and 
with  it  the  course  of  the  history  of  poetry  even  to  the  present 
day,  was  determined  by  the  way  in  which  Icarian  shepherds 
understood  the  worship  and  the  power  of  Dionysus.  No 
doubt  their  manner  of  taking  the  whole  story  was  a  far  simpler 
one  than  that  familiar  to  Sophocles,  yet  it  was  sufficiently 
complex  to  make  Icaria  the  cradle  of  tragedy  and  comedy.3 

1  To  this  Dr.   Merriam  adds  (p.  67)  :   "The  bright  star  c,  near  the 
right  wrist  of  Virgo,  was  called  provindemiator  (TrpuroTpvyTjTrip,  Aratus, 
Phaenomena,  138),  as  rising  shortly  before  the  vintage.     Icarius  is  Bootes, 
as  vindemiator  (Tpvy-rjT-rjp,  Schol.  Arat.  Phaen.  91). 

2  See  Appendix  II.  at  the  end  of  this  chapter,  on  Dionysus  of  Eleutherae. 

3  Athenaeus  ii.  40  A  :  airb  /JLtdrjs  ical  TJ  TTJS  Tpayydtas  evpe<ris  ev 

JS  'Arn/c^s  evptdr),  Kal  KO.T  avrbv  rbv  rrjs  rpvyrjs  Kaip6v  '  cuf>' 


ii2         DIONYSUS  IN  THRACE  AND  OLD  ATTICA         in 

This  momentous  beginning  associated  itself  with  the  two 
mountaineer  festivals  celebrated  by  the  Icarians  —  with  their 
thanksgiving  and  fast  -day  observances,  so  to  speak.  Little 
or  nothing  definite  is  known  of  the  former  of  these.  It 
must  have  been  a  May-day  festival  like  the  Thracian  festival 
afterwards  called  the  Rosalia;  the  other  corresponded  to 
the  Brumalia,  in  so  far  at  least  as  it  was  celebrated  in  the 
bitterest  cold  of  the  bitter  month  Lenaeon.1 

Their  joy  was  in  Dionysus  revealed  in  streams  flowing 
free  and  fast,  no  longer  bound  in  wintry  fetters  of  ice, 
revealed  also  in  the  plenteous  foliage  and  brilliant  blossoms 
of  their  native  woodland  home.  The  masses  of  yellow 
narcissus,  found  at  their  ancient  home  to-day,  painting 
whole  mountain  -  sides  with  their  bright  yellow,  gleaming 
in  the  sunshine,  and  spreading  fragrance  near  and  far, 
may  fitly  give  ocular  demonstration  of  their  old  time 
gladness. 

Their  sorrowing  was  lament  for  Dionysus,  loved  and  lost. 
They  mourned  when  nature's  flash  and  flow  and  all  her 
lively  colours  seemed  to  vanish  from  their  eyes.  Partly  to 
lament  the  loss  of  Dionysus'  presence,  partly  to  recall  his  life, 
they  sought  in  winter  time  bleak  and  storm-swept  summits. 
There  tarried  those  upon  whom  came  compulsion  from  the 
god.  There  Maenads  and  Bacchanals  vied  in  wild  dances 
and  loud  cries,  which  they  thought  would  have  power  to 
bring  back  growth  and  life  to  trees  and  plants  and  streams. 
Dionysus  slept  or  was  among  the  dead.  It  was  as  though 


08  5?j  Kal  rpvyydla  TO  irpurov  eK\rj6-rj  77  KW/jupdia.  See  the  last  volume  of 
Bergk's  Griechische  Litteraturgeschichte  (Aus  dent  Nachlass,  pp.  7 
and  8). 

1    Dread  are  the  days  of  the  month  Lenaeon,  the  flayer  of  oxen, 
Under  the  blasts  of  the  north  wind,  of  ice  that  Boreas  sharpens, 
Scouring  o'er  Thrace  and  her  pastures  of  horses  to  breadths  of  the 
ocean.  —  Hesiod,  Works  and  Days,  504  ff. 


in  WINTER  FESTIVALS  ON  PARNASSUS  113 

they  thought  each  winter  would  last  for  ever  if  they  did  not 
beat  the  ground  and  summon  spring. 

The  sincerity  with  which  they  made  these  desperate  forays 
into  winter's  fastnesses  is  best  shown  by  hardships  actually 
endured.  The  incident  in  point  which  has  been  preserved 
relates,  not  to  the  festival  at  Icaria,  but  to  that  on  Mount 
Parnassus.  This  matters  little,  since  both  were  centres  of 
Dionysus  worship,  strongly  influenced  from  Thrace  in  early 
days.  On  Mount  Parnassus  we  hear  that  a  band  of  frenzied 
votaries  were  blocked  upon  high  levels  above  Delphi.  In  the 
midst  of  their  incantations  to  revive  the  life  of  spring,1  snows 
imprisoned  them.  Those  who  climbed  the  steep  from 
Delphi  for  their  rescue  suffered  from  the  utmost  rigours  of 
the  cold.  Their  raiment  grew  stiff,  and — so  the  unknown 
narrator  declares — became  absolutely  brittle  and  friable. 

Though  it  may  safely  be  said  that  of  the  two  peasant 
festivals  celebrated  in  conjunction  with  Dionysus  worship 
at  Icaria  and  elsewhere  one  was  sad  and  the  other  glad,  no 
more  definite  account  can  be  safely  given.  With  the  sad 
festival  glad  features  were  associated,  and  the  glad  festival 
was  not  without  its  mourning.  Thus  are  the  inherent  con 
tradictions  of  Dionysus  in  his  very  nature  mirrored  out 
wardly  by  his  festivals. 

It  is  reasonably  certain  that  the  whole  observance  which 
spread  all  over  Attica  and  called  itself  the  Rural  Dionysia 
took  its  characteristic  shape  in  Icaria.  Wonderful  to  relate, 
the  Attic  salt  of  moderation  and  due  measure  came  from 


1  See  the  Plutarchian  De  primo  frigido,  xviii. :  h  d£  AeX0ots  avrbs 
6'n  r&v  els  rbv  Hapvaabv  avaftavruv  /SoTj^crcu  rcus  Qvdaiv, 
a.Trei\r)/j.fj.tvais  virb  Trvev/n-aTos  xaAeTroC  Kal  xt^os,  OUTOJS  tytvovro  Sia  rbv 
Trdyov  ffK\r)pal  Kal  £uXu>5ets  ai  x\a/xi55es,  w?  Kal  dpafcadai  8iaTeivo/JLei>as 
Kal  priyvvffdai.  In  the  light  of  recent  events  this  passage  gains  new 
interest  as  the  earliest  circumstantial  account  of  a  blizzard. 


ii4        DIONYSUS  IN  THRACE  AND  OLD  ATTICA         in 

the  unaided  taste  of  a  community  wonderfully  aloof  from 
all  the  rest  of  Attica.     Icaria  is  deserted  now,  and  lay  far 
from  the  course  of  travellers  until  the  American  excavations 
unearthed  fragments1  which  claim  a  visit  from  all  interested 
in  Greek  antiquity.     And  truly  the  visitor  is  well  repaid, 
finding  how  beautiful  is  the  spot  where  highest  art  had  its 
first   outset,  and   earliest   took   shape.2     Near   its   ruined  ; 
church3  begin  the  forests,  the  very  "wilderness  of  Marathon," 
through    which,    the    poet    Statius    says,    Erigone    once  j 
wandered  seeking  wood  to  place  upon  Icarius'  funeral  pyre.  | 
Here,  in   recent   days,    but   fortunately   past,  were   secret 
haunts  of  brigands  now  unknown  to  Greece.     The  lovely 
vale  of  Rapendosa  is  not  far,  for  it  belonged  to  Icaria's 
demesne. 

This  is  the  setting  of  the  earliest  legend  of  Dionysus  in 
Attica;  here  were  celebrated  his  Icarian  festivals.  Their 
importance  is  shown,  and  also  their  unusually  noble 
character,  by  the  fact  that  Susarion  was  first  invited  there, 
and  that  there  was  performed  the  first  comedy,  unless  it  be 
premature  to  give  the  name  of  comedy  to  Susarion's  great  j 
invention.  The  requirements  of  the  Icarian  holiday-makers 
must  have  been  raised  amazingly  by  a  gradually  purified 
and  elevated  taste  to  lead  them  to  invite  Susarion  from 

1  See  Appendix  XI.  i.  52-55. 

2  This  was  tacitly  recognised  when  a  sculptured  scene,    representing 
Icarius,   Erigone,  and  the  dog  Maera,  was  used  to  adorn  the  stage  of  the 
Athenian  theatre  of  Dionysus.     This  scene,  together  with  Hermes  carrying ; 
the  infant  Dionysus  to  Zeus,  is  still  admired  at  Athens  by  those  who  visit! 
the  ruined  theatre.     An  examination  of  these  sculptures  has  satisfied  many 
that  they  were  not  originally  made  for  the  comparatively  inconspicuous! 
place  into  which  they  are  now  crowded,  for  they  are  really  too  high  suitably; 
to  adorn  what  may  be  called  the  parapet  of  the  raised  stage  built  byi 
Phaedrus  where  they  now  are.     They  originally  decorated  a  stately  fafade 
upon  the  stage  itself,  which  was  probably  built  by  the  munificence  of  Nero,  ;.i 
mindful  of  his  duty  as  Apollo  in  the  flesh. 

3  See  Appendix  XI.  i.  50,  51. 


in  SUSARION  AND  THESPIS  AT  1C  ARIA  115 

Megara.  Report  had  reached  them  of  his  new  scheme  of 
methodised  revelry  jocosely  acted  before  the  festive  wor 
shippers,  and  he  had  perhaps  heard  of  the  observances  at 
Icaria,  which  were  certainly  his  great  opportunity.  This 
happened,  it  appears,  while  the  great  Icarian  Thespis  was 
still  a  youth.  Perhaps  from  seeing  Susarion's  first  perform 
ance  in  his  native  place  Thespis  received  the  fruitful 
thought  which  prompted  him  later  when  he  became  the 
father  of  tragedy.  And  yet  the  invention  was  his  very  own, 
springing  from  the  legends  which  had  surrounded  his  child 
hood. 

It  was  an  Icarian  custom  to  sing  a  mournful  song  named 
"Aletis,  the  Wanderer"  in  honour  of  Erigone,  and  relating 
her  sad  fate.  To  these  songs  Thespis  had  listened  no 
doubt,  and  also  to  the  strangely  sad  and  wildly  joyous  Dithy- 
rambus  wherein  were  mirrored  the  contradictions  involved  in 
the  nature  of  Dionysus.  These  recounted  the  poetic  vicis- 
situdes  of  the  great  divinity — his  sufferings  and  his  triumphs. 
Filled  with  a  higher  and  a  new  apprehension  of  the  scope 
of  all  these  sad  stories,  Thespis  transformed  the  Aletis  and 
the  Dithyrambus.  '  He  had  heard,  perhaps,  of  the  wonder 
ful  performances  of  tragic  choruses  at  Sicyon,1  which  so  exer 
cised  the  mind  of  royal  Clisthenes  that  he  interfered  with  the 
subjects  represented,  for  fear  of  their  strong  hold  upon  the 
people  who  heard  them.2  Of  Thespis  Dr.  Merriam 3  has 
truly  said  :  "  The  sad  story  of  the  father  of  his  gens  (Icarius), 
the  rites  attendant  upon  the  festival,  the  dithyrambic 
choruses  in  vogue  predisposed  him  to  this  end  (progress  in 
tragedy),  and  gave  him  a  nucleus  to  which  he  added  the 

1  As  Bernhardy  says  (Litteralurgeschichte,  i.  p.  417),  the  connection 
between  early  tragedy  in  Attica  and  these  events  at  Sicyon  has  been  over 
stated.     Perhaps  Bentley  was  right  in  saying  there  was  none. 

2  Hdt.  v.  67,  end.  3  Report,  pp.  71  and  72. 


ii6        DIONYSUS  IN  THRACE  AND  OLD  ATTICA         in 

actor,  the  prologue,  and  speeches  between  the  choral  songs, 
and  he  employed  different  masks  to  enable  him  to  take  the 
parts  of  several  persons  consecutively  in  the  same  play. 
This  proved  him  the  Columbus  of  a  new  world, — a  mimic 
world,  but  one  calculated  to  excite  the  interest,  as  it  is  said 
to  have  engaged  the  hostility,  of  the  great  law-giver.  It 
must  have  been  a  few  years  only  after  Susarion's  advent  in 
Icaria  that,  as  Plutarch  tells  us,1  the  novelty  of  the  inven 
tion  was  attracting  many,  and  Solon  in  his  old  age,  being 
fond  of  amusement  and  music,  also  went  to  see  Thespis  acting 
in  his  own  play.  It  is  a  legitimate  inference  from  the 
language  of  Plutarch  that  the  play  was  produced  at  some 
distance  from  Athens — in  other  words,  in  Icaria;  for  we 
can  hardly  imagine  Solon,  a  true  Greek,  to  have  remained 
away  from  a  festival  of  importance,  with  novel  features, 
celebrated  at  his  own  door.  Later  than  this  event  fell  his 
censure  of  Pisistratus,  for  the  latter's  bad  acting  in  the 
game  which  he  played  in  winning  his  first  tyranny." 

The  Dithyrambus,  we  may  suppose,  was  so  modified  by  i 
Thespis  that  its  calmer  course  allowed  interruption;  nay,'i 
even  required  it.2     Out  of  the  most  lawless  and  wayward  of 
lyric  strains  Thespis  made  tragedy,  no  doubt,  by  requiring  i 
both  singers   and  holiday-makers  to  leave  their  uncouth 
ways  of  wildness  and  listen  while  the  glorious  sufferings  of 
Dionysus  came  before  them,  recited  by  a  single  man  (for  ] 
it  is  not  right,  as  yet,  to  call  him  actor)  standing  for  and; 
speaking  for  the  bacchanalian  concourse.     He  spoke,  no 
doubt,  in  much  the  same  spirit  in  which  a  speaker  addresses 

1  Solon,  29. 

2  Bernhardy  (Litteraturgeschichte,  i.  p.  417)  quotes  Themistius'  dis 
sentient  citation  of  Aristotle,  who  plainly  thought  tragedy  was  built  up  on 
the  Dithyrambus  :  -rb  ^v  irpurov  6  xfy>os  etVtwi/  T-fSej/  ets  TOI)S  0eofc,  Qt<riri< 

re  KO.I  pijaiv 


in  FIRST  TRAGEDIES  AT  ICARIA  117 

a  Quaker  meeting.  It  was  he  whom  Dionysus  chose  for 
speaking.  While  he  spoke  he  only  was  the  living  god  made 
flesh.  This  advance  towards  full-fledged  tragedy  was  made 
as  early  as  the  second x  quarter  of  the  sixth  century  B.C. 

Upon  this  one  step  all  others  depended,  and  they  were 
soon  taken  ;  for  the  perfection  of  Attic  tragedy  came  early 
in  the  fifth  century  B.C.  Before  long,  in  fact,  Icaria  was  no 
more  the  centre  of  the  Attic  worship  of  Dionysus,2  and 
Dionysus  was  brought  in  triumph  to  take  his  place — last 
come  but  not  the  least  of  all  the  gods  in  Athens. 

1  The  Parian    Marble  fixes  Susarion's  advent,  with  the  first  comedy 
performance,  about  the  beginning  of  this  second  quarter.    Icaria  no  sooner 
gave  Attica  its  comedy  than  Icarian  Thespis  commenced  tragedian  and 
tragedy  as  well.     For  the  various  authorities,  see  Dr.  Merriam's  Report. 

2  But  Icarians  played  an  important  part  in  Athenian  history,   contri 
buting  a  great  comedian,   Magnes,  who  died  not  long  before  424  B.C., 
as  well  as  many  pious  and  generous  men  of  note.     See  Dr.  Merriam's 
Report,  pp.  80-93,  where  they  are  all  enumerated  and  characterised.     This 
account  closes  as  follows  :   ' '  We  still  seem  to  have  enough  to  draw  some 
conclusions   as   to  the  characteristics  of  the   people  who  dwelt  in  that 
picturesque  mountain-hemmed  spot,   and  traced  their  ancestry  back   to 
Icarius,  who  entertained  the  god.     We  find  no  generals  of  renown,  no 
statesmen  active  in  moulding  for  good  or  ill  the  affairs  of  Athens,  no 
orators  of  power,  no  one  especially  active  in  proposing  and  pushing  laws 
in  the  public  assembly  for  public  weal  or  private  gain,  no  historians,  no 
philosophers,   no  artists.      They  are  distinguished  by  two  traits,   which 
claim  our  respect  and  admiration.     These  are  a  deep  devotion  to  religion 
and  a  sound  and  sturdy  integrity." 


APPENDIX  II 
DIONYSUS    ELEUTHEREUS 

A  WHOLE  cycle  of  stories  touching  the  first  arrival  of 
Dionysus  in  Athens  ignores  Icaria,  and  centres  itself 
around  Eleutherae,  a  Boeotian  town  on  the  Athenian 
frontier,  much  claimed  by  Athens.1  Eleutherae  may  be 
visited  with  immense  profit  to-day  by  those  desirous  of 
gaining  the  unforgettable  impression  which  a  strong  Greek 
fortification  built  in  the  days  of  superb  workmanship  can 
make.  On  the  high  road  from  Athens  to  Thebes  it  lies 
not  far  from  the  point  where  the  road  dips  down  from  the 
mountain  spurs  of  Cithaeron  into  the  level  Boeotian  plain. 
Its  wonderfully  preserved  battlements  are  not  seen  well 
except  from  the  Boeotian  side ;  but  he  who  turns  back 
from  the  right  point  in  the  road  will  understand  what  is 
meant  by  saying  that  the  Greeks  could  build  nothing  that 
was  not  subject  to  the  laws  of  harmony  and  proportion. 
Go  closer,  and  the  symmetry  with  which  square  stones  are 
here  grouped  together  into  a  massive  wall  makes  you 
wonder  why  so  many  strong  walls  in  the  world  have  been 
made  so  uninteresting. 

But   to  return  to   Dionysus.      That  this   Dionysus  of 
Eleutherae  played  an  important  part  is  sure,  since  the  beauti- 

1  Strabo,  IX.  ii.  (p.  412).  The  Dionysus  of  Eleutherae  was  considered 
to  be  a  later  arrival  than  the  Lenaean  god  at  Athens,  and  the  old  Icarian 
Dionysus  of  course  was  thought  of  as  having  preceded  him.  See  the  article 
Dionysia,  in  Pauly's  Real- Encyclopaedic,  by  Preller. 


DIONYSUS  ELEUTHEREUS  119 

ful  seat  of  honour  still  standing  in  place  at  the  theatre  of 
Dionysus  in  Athens  is  by  inscription  marked  as  belonging  to 
the  priest  of  Dionysus  Eleuthereus.  Moreover,  in  the  older 
of  the  two  temples,  close  to  this  Athenian  theatre,  was  an 
image  of  Dionysus,  said  to  have  come  from  his  temple  at 
Eleutherae ;  and  custom  required  this  image  to  be  borne 
once  a  year  to  a  small  outlying  temple  upon  the  road  which 
led  to  Eleutherae.1 

Furthermore,  the  coming  of  Dionysus-worship  to  Athens 
from  Eleutherae  is  associated  with  the  reign  of  King  Amphic- 
tyon.  Now  the  common  version  of  the  coming  of  Diony 
sus  into  old  Attic  Icaria  goes  back  only  to  King  Pandion, 
when  also  Demeter  came  to  Eleusis.  Without  pretending 
to  assign  such  a  thing  as  a  date,  it  is  roughly  true  that 
Amphictyon  belongs,  according  to  Attic  tradition,  to  an 
earlier  time  than  Pandion.  Too  much  importance  need 
not,  however,  be  attached  to  this  matter  of  precedence, 
since  the  coming  of  Dionysus  to  Semachus  and  Semachidae 
is  assigned2  to  the  reign  of  Amphictyon.  This  story  of 
Dionysus  coming  first  to  Semachidae — one  of  the  three 
Epacrian  denies,  of  which  Icaria  and  Plotheia  were  the 
other  two — is  probably  an  attempt  to  mediate  between  and 
combine  the  Icarian  and  the  Eleutheraean  legends.  The 
chief  differences  between  Dionysus  of  Eleutherae  and  of 
Icaria  are  accounted  for  if  we  consider  that  he  came  more 
immediately  from  savage  Thrace  to  Icaria  than  to  Eleutherae. 
At  Eleutherae,  Eleuther  was  the  one  who  taught  his  right 
worship,  and  fashioned  his  first  image.3  Now  Eleuther  was 
no  less  a  personage  than  the  son  of  Apollo  by  Aithusa,4  who 
was  herself  a  daughter  of  Poseidon  by  Alcyone,  one  of  the 
seven  Pleiades,  daughters  of  Atlas.5  Eleuther  stands  for  a 

1  See  Miss  Harrison's  account  at  pp.  254  and  571  of  Mythology  and 
Monuments  of  Ancient  Athens. 

2  Syncell.  p.  157  (125):  KO.TCI,  'AfJuftiKTiJova  rbv  Aeu/caXWos  vlbv  rivts 
(fxiffL    kibvvffov    ei's    Ti)i>    'ATTiKrjv    e\6bvra    ^evwdTJvat    ZTj/udxy    Kal    rrf 
6vyarpl  avrov  vefipida  ScopTjcraaflcu.     £repos  5'  fy  oCros  e/c  Se^cA^s.     The 
next  king  was  Erichthonius.  3  Hyginus,  Fab.  225. 

4  Pausanias,  IX.  xx.  i.  6  Apollod.  III.  x.  i. 


I20  APPENDIX  II 


softening  or  Hellenising  influence,  which  Dionysus  had  sub 
mitted  to  before  he  reached  Athens  from  this  quarter — an 
influence  emanating  from  Eleuther's  father,  Delphian  Apollo. 
Voigt,  in  Roscher's  Ausfuhrliches  Lexicon,  thus  sketches  the 
legend  of  Eleutherae :  "The  daughters  of  Eleuther  saw 
Dionysus  clad  in  a  black  goat-skin.  For  scoffing  at  this 
vision  they  were  visited  with  madness.  To  cure  them 
Eleuther  was  commanded  by  an  oracle  to  worship  Dionysus 
Melanaigis,  of  the  black  goafs  fell"  Accordingly,  Hesychius 
describes  Dionysus  Eleuthereus  as  a  god  who  gives  release 
from  the  madness  that  comes  upon  Dionysiac  revellers,  and 
Pegasus,  the  Eleutheraean  priest,  stands  for  a  moderate  use 
of  wine  not  unmixed  with  water. 


IV 
DIONYSUS  AT  ATHENS 

IN  the  last  chapter  Dionysus  was  brought  from  Thrace, 
and  found  an  Attic  home  in  Icaria.  Now  he  must  be  brought 
to  Athens,  and  accompanied  thence  to  Eleusis,  where  his 
power  has  already  been  recognised  while  we  tarried  in 
Icaria.  The  last  and  fullest  presentation  of  the  perfected 
god,  at  the  moment  when  the  widest  reach  of  religious 
thought  and  richest  depth  of  religious  fervour  attached  to  his 
worship  in  Athens  and  Greece,  has  been  given  us  in  the 
Bacchanals  of  Euripides.  It  only  remains,  therefore,  to 
connect  what  we  know  of  Dionysus  in  Thrace  and  Dionysus 
in  Icaria  with  this  worship  of  Dionysus  at  Athens.  The 
culminating  truth  about  the  god  will  be  revealed  to  us 
after  attentive  consideration  of  the  most  perfect  play  of 
Euripides,  and  then  we  shall  in  the  next  chapter  close  our 
consideration  of  the  Eleusinian  divinities  with  a  concluding 
if  not  a  final  word  about  the  Eleusinian  ritual,  and  some 
account  of  the  monuments  of  Eleusis. 

The  cult  of  Dionysus  was  not  adopted  at  Athens  as  a 
matter  of  course.  Patrician  traditions  firmly  rooted  at 
the  capital  long  resisted  popular  pressure  in  favour  of  the 
peasant  divinity  from  Icaria.  There  was  evidently  far  less 


122  DIONYSUS  AT  ATHENS  iv 

disposition  in  early  Athens  to  meddle  with  Icarian  merry 
making  than  there  was  to  take  part  in  Eleusinian  mysteries, 
with  which  at  a  very  early  date  Athenian  family  traditions 
connect  themselves.1  Still  the  political  fusion  of  Attica, 
which  was  associated  with  the  glorious  name  of  Theseus, 
could  not  in  the  long  run  be  maintained  without  such 
broadening  of  religious  observance  in  Athens,  and  in 
Attica  at  large,  as  should  more  completely  make  one  the 
heart  of  Attic  and  Athenian  religion.  The  exclusiveness  of 
local  or  township  ritual  had  to  disappear,  and  some  sort 
of  religious  fusion  had  to  be  brought  about. 

Fortunately  this  needed  fusion  tended  to  accomplish 
itself  in  spite  of  official  discouragement.  The  spontaneous 
impulses  of  a  people,  religious  without  being  superstitious, 
accomplished  many  large-hearted  alterations,  and  among 
them  the  triumph  at  Athens  of  Eleusinian  Demeter  and 


1  Pausanias,  I.  xxxviii.  3,  gives  the  terms  of  peace  concluded  between 
Athens  and  Eleusis  after  the  war  in  which  were  killed  King  Erechtheus 
of  Athens  and  Immarados  of  Eleusis,  a  son  of  the  Thraco-Eleusinian 
Eumolpus.  The  Eleusinians  were  to  submit  to  Athenian  supremacy, 
saving  only  that  they  were  to  regulate  the  mysteries  in  their  own  way. 
Eumolpus  and  the  daughters  of  Celeus  were  in  immediate  charge  of 
T&  lepa  TOLV  deoiv,  the  sacred  observances  in  honour  of  the  two  goddesses. 
But  Pausanias  names  as  successor  to  Eumolpus,  Ceryx,  a  son,  not  of 
Eumolpus,  but  of  Hermes  and  the  daughter  of  Athenian  Cecrops,  Aglaurus. 
The  same  belief  in  an  early  intervention  of  purely  Athenian  families  at  the 
rites  of  Eleusis  is  shown  by  Strabo's  explanation,  XIV.  (p.  633),  of  certain 
existing  privileges  attaching  to  the  so-called  /3a<riAets  at  Ephesus.  They 
represented  a  remote  Athenian  ancestor,  Androclus,  the  son  of  Codrus, 
who  founded  Ephesus  and  assumed  control  of  the  Eleusinian  rites  trans 
planted  thither  at  that  early  date.  The  most  convincing  authority  for  an 
early  fusion  in  some  sort  of  Athenian  and  Eleusinian  observances  is  perhaps 
Herodotus  in  his  account  of  Solon's  answer  to  King  Croesus,  who  wished 
him  to  tell  of  the  happiest  man  he  knew  (i.  30).  Tellus  the  Athenian  was 
his  man,  for  after  a  life  otherwise  completely  happy  Tellus  died  most 
gloriously  fighting  for  his  country  against  near  neighbours  (the  Megarians). 
He  fell  at  Eleusis,  and  was  buried  where  he  fell  with  all  possible  honours. 
Such  burial  of  an  Athenian  on  Eleusinian  soil  suggests  more  or  less  com 
plete  fusion  of  Athenian  and  Eleusinian  rites  and  customs. 


iv  EPIMENIDES  OF  CRETE  123 

Icarian  Dionysus.  Here  was  indeed  one  of  the  earliest  of 
many  brilliant  victories  gained  by  Athenian  democracy. 
Solon's  actually  visiting,  or  even  a  commonly  credited  report 
that  he  visited,  the  rustic  play  at  Icaria  or  elsewhere  in 
Attica,  marks  a  turning-point  in  the  history  of  Athenian 
state  religion. 

It  is  abundantly  evident  that  in  the  earlier  ages  of  their 
worship  Demeter  and  Dionysus  were  alike  divinities  of  the 
common  people.  Consequently  we  are  not  surprised  to  find 
either  that  the  earliest  recorded  enlargement  of  the  official 
religion  at  Athens  was  a  recognition  of  Eleusinian  Demeter, 
or  that  there  is  a  still  more  complete  record  of  the  later 
official  adoption  of  Dionysus  forced  by  popular  discontent. 
The  people  failed  to  win  power  under  Cylon's  leadership, 
but  they  succeeded  in  altering  the  state  religion.  Epimen- 
ides,  a  wise  man  of  Crete,  was  called  in,  after  Cylon's 
attempt  had  been  suppressed,  to  devise  means  for  allaying 
popular  disaffection.1  The  gratitude  felt  for  Epimenides 

1  The  date  of  this  purification  of  Athens  by  Epimenides  was  about 
596  B.C.  That  a  humanising  change  did  come  over  Attic  religion  and  its 
officially  constituted  observances  is  beyond  the  possibility  of  a  doubt. 
What  in  general  terms  this  change  was,  and  that  it  was  definitely  associ 
ated  with  Epimenides,  appears  from  Strabo,  who  says  (p.  479):  £K  5£  TTJS 
QaiffTov  rov  TOI)S  Ka.6apfj.oijs  TroirjcravTa.  5ta  TUV  eir&v  "Etiri^vL^v  (paalv 
elvai, — and  Plutarch's  Life  of  Solon  (12,  §  4  and  ff.),  where  we  read  of 
Epimenides  substantially  what  follows  : — He  was  reputed  a  friend  of 
the  gods,  with  especial  skill  and  knowledge  touching  mysteries  and 
enthusiastic  rites  (TTJV  tvdovcriacrTiKTiv  /ecu  TeXecmxV  <ro<t>lav).  He 
did  much  to  prepare  the  way  for  Solon's  legislation  by  reducing  to 
simplicity  (ev&TaXeis  eiroirjo'f)  the  official  sacrifices,  and  by  softening  down 
in  them  the  observances  'of  mourning.  He  introduced  certain  sacrifices 
into  funeral  rites,  and  thus  banished  harsh  and  barbaric  usages  to  which 
most  women  had  previously  clung.  But  above  all,  by  certain  propitiatory 
and  purifying  rites,  and  by  instituting  new  observances  and  sanctuaries,  he 
made  the  city  ready  for  sacred  orgies  and  hallowed  it  for  the  service  of 
justice,  bringing  it  to  a  readier  obedience  of  the  promptings  of  concord. 
See  Bernhardy,  Griechische  Litteraturgeschichte,  i.  p.  409.  Since  the 
above  was  written  the  newly-discovered  Aristotelian  Constitution  of  Athens 
has  gone  far  to  justify  the  importance  here  attached  to  the  intervention 


124  DIONYSUS  AT  ATHENS  iv 

at  Athens  was  commemorated  by  a  statue,1  and  we  may 
conclude  that  his  reforms  were  made  in  the  interest,  not  of 
concord  only,  but  of  the  maintenance  of  the  state  religion 
on  a  broader  and  therefore  a  more  universally  acceptable 
basis. 

In  the  month  which  was    afterwards    selected  for  the 
flower-festival  of  Dionysus,  when   that  also  was   officially 


of  Epimenides,  by  briefly  mentioning  it  as  follows  :  'ETrt/zei/^s  5'  6 
eirl  Totfrois  tKadype  TTJV  Tr6\iv.  This  comes  at  the  very  beginning  of  the 
newly-discovered  MS.  ,  and  is  preceded  by  an  account  of  Myron's  arraign 
ment  of  the  Alcmaeonidae  and  of  their  perpetual  banishment  carried  out 
even  upon  the  buried  remains  of  their  dead.  This  was  all  done  in 
expiation  of  their  outrageous  suppression  of  Cylon  and  his  faction.  When 
the  grateful  Athenians,  marvelling  at  his  work,  pressed  riches  and  honours 
upon  him,  Epimenides  took  for  his  guerdon  a  branch  of  the  sacred  olive, 
and  went  his  way. 

1  Pausanias  says  (I.  xiv.  3  and  4)  :  —  "  I  intended  ...  to  give  such 
account  as  is  possible  of  the  sanctuary  at  Athens  called  the  Eleusinion, 
but  was  prevented  by  a  vision  in  a  dream.  I  will  turn  to  what  may  law 
fully  be  told  to  every  one.  In  front  of  this  temple,  where  is  the  image  of 
Triptolemus  "  ["We  are  undoubtedly  justified,"  says  Miss  Harrison  in 
her  admirable  commentary  (p.  93  of  Mythology  and  Monuments  of  Ancient 
Athens],  "in  supposing  that  the  two  temples"  (one  of  Demeter  and 
Kore,  the  other  of  Triptolemus)  '  '  went  by  the  name  of  Eleusinion  "],  "  is 
a  bronze  bull,  apparently  being  led  to  sacrifice,  and  a  seated  figure  of 
Epimenides  of  Cnossus."  [Strabo  (as  well  as  Plutarch)  tells  us  more 
accurately,  X.  iv.  14  (p.  479),  that  Epimenides  came  from  Phaestus 
in  the  Cnossian  district.  ]  '  '  Epimenides  is  said  to  have  gone  into 
a  field  and  to  have  fallen  asleep  there  in  a  cave,  and  the  sleep  did 
not  depart  from  him  for  forty  years  ;  and  after  his  awakening  he 
wrote  poems  and  purified  various  cities,  among  them  Athens."  I  have 
been  quoting  from  Mrs.  Verrall's  excellent  translation,  p.  86  of  Mythology 
and  Monuments  of  Ancient  Athens.  The  statue  of  Epimenides  would 
naturally  be  placed  in  the  Eleusinion,  if  Osann  is  right,  as  I  have  thought 
him,  in  attributing  to  him  the  official  recognition  of  the  Lesser  Mysteries 
and  their  whole  definite  organisation.  Whether  the  actual  place  of  its 
erection  should  be  in  front  of  one  temple  or  the  other  was  a  matter  of  chance 
or  momentary  convenience.  I  venture,  therefore,  to  believe  in  the  literal 
accuracy  of  Pausanias,  both  as  regards  the  place  of  this  statue  and  the 
person  whom  it  represented.  Indeed  the  witness  of  Pausanias  confirms 
Osann's  views  adopted  below.  The  name  of  Epimenides  was  indissolubly 
associated  in  the  minds  of  religious  Athenians  with  their  Eleusinion  and 
its  Lesser  Mysteries.  Not  to  have  raised  his  statue  in  just  that  precinct 
would  have  been  like  denying  to  Browning  his  place  in  Westminster 
Abbey. 


iv  THE  PISISTRATIDAE  AND  DIONYSUS  125 

recognised  at  Athens,  Epimenides  caused  a  new  festival  to 
be  celebrated.  This  was  a  specifically  Athenian  observance 
in  honour  of  Demeter  and  Persephone,  but  especially  of 
Persephone,  called  the  Lesser  Mysteries,  as  distinguished 
from  the  older  -  established  Greater  or  Eleusinian  Mys 
teries  to  which  it  served  as  a  prelude.1  In  the  Lesser 
Mysteries  Dionysus  became  associated  with  Demeter,  the 
mystery  of  his  birth  under  the  name  of  lacchos  being  duly 
commemorated.  This  was  all  the  more  natural  because 
Dionysus  had  already  found  his  way  to  Eleusis.2  Not  until 
nearly  a  century  later  is  there  record  of  a  second  step 
taken  in  the  occupation  of  Athens  by  Attic  Dionysus.  This 
time  the  changes  were  conclusive  and  effectual ;  they  were 
the  result  not  of  the  people's  disaster  under  a  Cylon,  but  of 
the  triumphs  of  the  enlightened  friend  of  the  people,  a 
native  of  the  Attic  highlands.  The  famous  tyrant  Pisi- 
stratus  and  his  family  appear  to  have  been  the  providential 
defenders  of  the  faith  in  Dionysus.3  Before  passing  to  the 

1  Diogenes  Laertius,  lib.  I.  cap.  x.  sec.  6  :   Idpfoaro  (sc.   'E7riyuei>/577s) 
5e   /cat   Trap'    'Adrjvalois  T&    iepbv   TWJ/    <refj.vui>   6euv,    ws  $770-1    Kbfiwv  6 
'A/ryetos  kv  rip  Trepl  irot.'rjTwv. 

2  See  the  first  note  on  chapter  v.  below. 

3  The  evidence  connecting  Pisistratus  with  the  revised  and  enlarged 
Bacchic  worship  at  Athens  is  sufficient,  due  regard  being  had  to  the  sort 
of  evidence  which  is  at  all  possible  in  a  matter  of  the  sort.     What  the 
evidence  is  Ribbeck  has  not  stated  adequately.     I  will  here  try  to  give  a 
suggestion  of  it : 

(a)  The  only  positive  evidence  connecting  Pisistratus  with  Dionysus  is 
a  somewhat  inconsequent  utterance  of  Athenaeus  (p.  533  C):  6  5£  Heiffto'- 
rparos  /cat  ev  TroXXots  /3api)s  tytvero,  STTOU  Kal  TO  ^  h.Q"t]vr\ai  rov  Aiovvcrov 
TrpbauTTov  fKeivov  TII^S  <j>a<riv  eiK6va.  Some  one  will  be  sure  to  see  in  this 
a  mere  bit  of  invention  springing  from  the  well-known  latter-day  habit  of 
making  statues  of  living  potentates  with  the  attributes  of  various  divinities. 
But  this  objection,  if  well  taken,  leaves  still  more  assured  the  certainty  of 
an  especially  close  relation  between  Pisistratus  and  Dionysus.  The  story 
could  not  otherwise  have  got  itself  invented. 

(t>)  Pisistratus  is  indirectly  but  very  really  connected  through  Ono- 
macritus  with  the  whole  reshaping  by  the  Orphic  school  of  the  religion  and 
the  mysteries  of  Dionysus  and  Demeter. 


126  DIONYSUS  AT  ATHENS  iv 

new  festivals  that  were  then  instituted,  the  new  features 
now  discernible  in  the  myth  of  Dionysus  must  be 
given.  These  are  said  to  have  been  shaped  and  codified, 

(c)  Plutarch,  Theopompus,  and  Athenaeus  never  tire  in  relating  anec 
dotes  to  show  how  Pisistratus  befriended  the  tillers  of  the  fields.     The 
newly-recovered  Constitution  of  Athens  (see.  chapter  16)  re-enforces  this  point. 
Pisistratus  did  everything  in  his  power  to  make  the  country  people  in 
dustrious  and  keep  them  in  the  fields  rather  than  allow  them  to  congregate 
anywhere  and  agitate.      He  instituted   TOI)J  Karen,  STOOL'S  St/cacrrds,  made 
frequent  country  visits  of  inspection,  and  settled  disputes.      Then  comes 
the  brief  story — well  told,  as  Aristotle's  pointed  stories  always  are — of  the 
labourer  in  the  Hymettus  district  accosted  at  Pisistratus'  command  while 
digging  with  a  (?)  "spike."    Asked  what  crops  he  grew,  the  countryman 
promptly  answered,  ' '  Curses  in  plenty  and  abundant  distress,  and  Pisistratus 
is  sure  of  his  tithes."     In  spite  of  the  burning  question  of  tithes,  Aristotle 
goes  on  to  say  of  Pisistratus :  ovdev  S£  TO  Tr\rj6os  ovd'  ev  rots  d'XXois  Trctpc&xXei 
Kara  TTJV  ap^v,   dXX'  cu'ei  Trapeo~Keva£ei>  elprjvrjv  /cat  erripec    5t'  i](rw)(iav ' 
5to  Kctl  7roXXd/as  [7ra/3y/u.tdf]eTo  u>s  [77]  Uto'LO'TpdTov  rvpavvls  6  e?rt  Kp6voi» 
/Si'os  e'ifj.     In  spite  of  oppressive  taxation  the  countryman  of  Attica  be 
lieved  in  Pisistratus,  who  conciliated  all  his  prejudices.      Among  these 
was  a  childlike  belief  in  the  bodily  intervention  of  the  gods  skilfully  flattered 
and  practised  upon  by  the  return  of  Pisistratus  with  Athene  Sotera  to  guide 
him, — some  say  she  was  a  Thracian  girl  named  Phye  (cf.  Athen.  p.  609). 
Since  there  was  a  reshaping  of  the  worship  of  Dionysus  in  his  day  and  by 
his  friend  Onomacritus,  can  we  suppose  Pisistratus  to  have  stood  aloof 
from  such  an  incomparable  means  of  currying  favour  with  his  agricultural 
constituents,  and  satisfying  his  own  religious  impulses  ? 

(d)  That  there  was  this  new  departure  under  Pisistratus  is  abundantly 
shown  by  the  facts  and  dates  in  the  career  of  Thespis.     Furthermore, 
Plutarch's  anecdote  (Solon,  29)  of  Solon  railing  at  Thespis  for  his  play 
actor's  trick  of  manifold  lying  goes  with  his  account  of  how  the  same  law 
giver  jeered    at  Pisistratus'  bad  acting  in  the  role  of  Odysseus  (Solon, 
chap.  30),   and  manifestly  associates  Pisistratus  with  the  new-fashioned 
play-acting. 

(e)  One  striking  fact  is  added  to  the  above  from  Aristotle's  Athenian 
Constitution,  15.      Pisistratus  spent  his  second  exile  in  the  district  where 
Dionysus  was  earliest  worshipped  :  irepl  rbv  Qep/j.a?ov  K6\Trov  .   .   .   exeWev 
be    irapijXOev   ets   TOI)S    irepl   TLdyyatov  TOTTOVS   86ev    •%pr)fj.a,Ti(rdiJ,€vos  KCU 
o-rpartwras  /Ma-Ouadfj.ei'os    .    .    .    ai>a.(r6<ra.<r6ai  /3ia  TTJV   dp^V  e7rexefy>ei. 
See  Mr.  Kenyan' s  note. 

(/)  Aristotle  mentions  the  tyranny  of  Pisistratus  and  the  Pisistratidae 
as  among  the  longest  of  duration  known  to  him  (Politics,  v.  12).  General 
ising  from  his  facts  in  the  previous  chapter  (n)  with  an  especial  eye  to 
the  career  of  the  Pisistratidae  (e.g.,  ZTI  d£  ^  pbvov  avrbv  0cuW0cu 
/j.r}5tva  T&V  dpxofj^vwi'  vfipifrisTa,  /mrjTe  viov  ^-fjre  v£av,  dXXd  /mrjd^  &\\ov 
(M-ridtva  ruv  Trepl  avr6v),  he  says  that  a  tyrant  "should  appear  to  be 
particularly  earnest  in  the  service  of  the  gods  ;  for  if  men  think  that  a 


iv        ONOMACRITUS  AND  ORPHIC  MYTH-MAKING     127 

as  it  were,  by  Onomacritus.1  However  this  may  be,  the 
connected  story  that  got  itself  together  at  this  time  in 
Athens  and  Attica  explains  the  aspect  of  Dionysus  upon 
which  the  name  lacchos  was  bestowed.  lacchos  is  the 
Dionysus  whose  mystic  birth  came  into  the  Lesser  Mys 
teries  long  since  instituted  by  Epimenides,  whose  Cretan 
birth  had  much  to  do  with  shaping  the  popular  conception 
of  this  new  aspect  of  the  god. 

This  story  had  been  in  the  air,  and  was  only  recorded, 
it  were,  by  Pisistratus  and  Onomacritus,  who  were 
encouraged  by  the  Delphian  oracle  to  do  this.  It  is 
evidently  made  up  of  a  motley  and  legendary  material, 
ultimately  Thracian  perhaps,  but  immediately  contributed 
from  the  islands  of  the  Archipelago,  and  most  especially 
from  Crete. 

ruler  is  religious  and  has  a  reverence  for  the  gods,  they  are  less  afraid  of 
suffering  injustice  at  his  hands,  and  they  are  less  disposed  to  conspire 
against  him,  because  they  believe  him  to  have  the  very  gods  fighting  on 
his  side.  At  the  same  time  his  religion  must  not  be  thought  foolish  " 
(Dr.  Jowett's  translation,  pp.  181  and  ff. ) 

1  Onomacritus  (see  Herodotus,  vii.  6)  was  on  confidential  terms  with 
Hipparchus,  but  had  to  leave  Athens  because  he  introduced  into  the  oracles 
of  Musaeus  one  of  his  own,  wherein  the  imminent  destruction  of  an  island 
near  Lemnos  was  predicted.  Lasus  of  Hermione,  Pindar's  teacher,  detected 
his  fraud.  When  the  Pisistratids  were  in  exile  he  appears  to  have  been 
reconciled  with  them,  and  to  have  joined  them  in  their  visit  to  the  court  of 
Xerxes,  whom  he  incited  to  war  against  Athens  by  various  prophecies. 
Herodotus  calls  him  •x.p-rja  [j.oKb'yov  re  K.o.1  dLad^rrjv  yjp-qviL&v  TUV  Mof<ra£ou. 
He  was  one  of  the  codifiers  of  the  Homeric  text,  into  which  he  introduced 
various  interpolations.  What  is  thus  known  of  his  treatment  of  Musaeus 
and  Homer  makes  it  regrettable  that  we  do  not  know  the  Orphic  materials 
out  of  which  he  wrought  what  was  possibly  an  officially  sanctioned  account 
of  the  Zagreus-Dionysus  myth.  But  so  far  are  we  from  knowing  his  sources 
that  our  means  of  knowing  what  was  made  of  them  are  very  scanty.  It  is 
only  from  comparatively  late  authorities  that  we  hear  of  his  dealings  with 
Orphic  materials.  The  Orphic  brotherhood  first  showed  itself  in  his  day 
and  at  Athens.  See  for  a  sufficient  account  the  articles  "Onomacritus" 
and  "Orpheus"  in  Pauly's  Real- Encyclopaedic,  both  by  Dr.  G.  F.  Bahr. 
See  for  a  yet  fuller  account  Bernhardy,  Litteraturgeschichte  (i.  419-421, 
and  ii.  425-440),  and  Lobeck's  Aglaophamus. 


1 28  DIONYSUS  AT  A THENS  I v 

Onomacritus,  let  us  say,  striving  to  weave  conflicting 
accounts  of  Dionysus  into  one,  hit  upon  the  idea  (or  else 
finding  it  ready  to  his  hand  made  skilful  use  of  it)  of  a 
succession  of  births  each  one  of  which  was  a  reincarnation 
of  the  one  god. 

Zagreus,  or  Dionysus,  under  his  more  savage  and  uncom- 
forting  aspect,  Zagreus  the  wild  Huntsman,  an  incarnation 
of  the  pitiless  harms  and  blasts  of  winter-time,  was  the  son 
of  Zeus  and  Persephone.  This  Persephone  was  not  the 
flower-faced  maiden  fair  and  sweet,  but  the  threatening 
queen  of  death.  Her  son,  this  Dionysus  -  Zagreus,  was 
mightily  favoured  by  his  father  Zeus.  So  true  was  this  that 
Zeus  was  about  to  give  the  child  his  throne,  and  to  sur 
render  with  it  his  thunders  and  lightnings.  But  this  plan, 
like  so  many  others  of  Zeus,  was  defeated  by  Hera's 
jealousy.  Hera  set  the  Titans  upon  him,  and  they  were 
his  undoing,  although  he  shifted  into  many  shapes  to  get 
free.  The  Titans,  fourteen  in  number,  took  Zagreus  while 
he  was  under  the  shape  of  a  bull.  They  tore  him  into 
fourteen  pieces,  which  Apollo  buried  at  Delphi.  Only  his 
heart  was  not  buried  there,  for  Athena  took  it  and  gave  it 
to  Zeus,  who  swallowed  it  and  then  brought  forth  to  new 
birth  the  babe  Dionysus,  specifically  named  lacchos.  A 
favourite  subject  for  sculptured  bas-reliefs  was  this  mystically- 
born  babe  lacchos,  wildly  swung  by  a  Maenad  and  a  Satyr 
in  the  mystical  sieve  to  which  he  owed  his  Orphic  epithet 
of  Liknites — lacchos  or  Dionysus  of  the  mystic  sieve. 

And  now  the  necessity  of  understanding  the  definite 
official  form  into  which  this  newly-recognised  worship  fitted 
itself  brings  up  the  more  or  less  chaotic  mass  of  Athenian 
festivals  in  honour  of  Dionysus.  The  following  attempt  to 
deal  with  it  shall  be  at  least  characterised  by  a  certain  neglect 


iv        EPIMENIDES,  PISISTRATUS,  ONOMACRITUS     129 

of  complicated  and  subordinate  questions.1  It  goes  without 
saying  that  Pisistratus  and  his  advisers  did  not  invent  anything 
new  when  they  instituted  the  Athenian  festivals  in  honour  of 
Dionysus.  This  was  equally  true,  no  doubt,  of  Epimenides 
and  the  Lesser  Mysteries.  They  merely  reorganised  exist 
ing  popular  usage,  and  by  official  recognition  gave  it  per 
manence,  and  secured  its  orderly  observance.  Pisistratus 
would  naturally  feel  that  he  would  tighten  his  hold  upon  his 
enthusiastic  highlanders,2  and  that  his  power  could  thereby 
be  more  secure,  if  only  he  gave  legal  sanction  to  their 
favourite  worship  of  Dionysus.  He  accordingly  organised 
in  the  god's  honour  the  most  brilliant  national  ceremonies, 
and  by  instituting,  as  it  were,  a  yearly  triumph  of  Dionysus 
at  Athens,  he  made  his  partisans  sure  of  his  and  their 
supremacy. 

The  accomplishment  of  all  this  led  him  to  make  a  series 
of  religious  innovations,  which  completed  the  work  begun  a 
century  before  by  Epimenides,  and  by  this  great  addition  to 
previously  recognised  religious  observances  the  religion  of 
the  Attic  people  in  the  broadest  sense  was  finally  estab 
lished  as  the  Attic  state  religion.  Pisistratus,  to  put  the 
gist  of  the  matter  shortly,  introduced  into  the  official 
calendar  the  two  peasant  festivals  long  observed  in 
Dionysus'  honour  by  the  people  in  Icaria  and  elsewhere.3 

1  In  attacking  this  much  debated  theme,  I  shall  not  attempt  to  arm 
myself  in  all  cases  with  my  sources.      I  have  followed  Ribbeck  in  many, 
though  not  in  all,  material  respects. 

2  Aristotle's    Constitution   of  Athens,    13  :    ^trav   8'   at    ordaets    T/sets 
.   .   .   TplrTj  5'  77  T&V  diaKpiwv  e0'  r/  TeTay/j.tvos  fy  IIct0i0T/Mroff,  STJ^UOTIKW- 
raros  elvai  doK&v.     Cf.  Herod,  i.  59,  and  Aristotle's  Politics,  v.  9. 

3  The  whole  problem  of  these  festivals  is  complicated  by  a  third  set  of 
autumn  festivals  celebrated  specifically  for  the  vintage.     These  assumed  an 
enormous  importance  at  all  centres  of  Dionysus  worship,  and  probably 
shifted  the  dates   of  the    chief  celebrations    in  his  honour.      It  is  even 
rash  to  say  that  these  were  not  as  old  as  any  observances  which  I  call 

K. 


1 30  DIONYS US  AT  A THENS  i v 

In  doing  this,  however,  he  was  careful  to  make  such  splen 
did  additions  and  gorgeous  modifications  as  gave  the 
incoming  god  a  great  pre-eminence,  and  caused  his  newness 
to  be  forgotten.  And  now  a  brief  account  of  these  two 
festivals,  the  Anthesteria  or  flower-festival  and  the  Lenaean 
festival,  is  requisite. 

As  Pisistratus  found  the  flower -festival  of  Dionysus,  it 
was  apparently  an  occasion  for  greeting  gladly  the  return  of 
spring.  There  were  children  garlanded,  and  garlanded 
worshippers  young  and  old.  There  was  tasting  of  wine 
newly  opened,  and  there  was  competitive  potation,  ending 
no  doubt  in  some  sort  of  wordy  row  like  that  which  assails 
the  ear  to-day  near  frequented  pot-houses  of  the  cheaper 
sort  in  Greece.  This  flower -festival,  as  Pisistratus  left  it, 
was  all  that  it  had  been,  with  the  addition  of  a  triumphal 
entry  of  Dionysus  into  Athens.  Furthermore  there  was 
instituted  a  symbolical  marriage  of  Dionysus,  the  idea  of 
which  was  cleverly  borrowed  from  the  yearly  marriage  rite 
of  Dionysus  and  Ariadne,  celebrated  on  the  island  of  Naxos 
in  the  Archipelago.1  Cleverly  borrowed,  I  say,  because  the 
Attic  rite  of  marriage  was  only  dimly  connected  with  nature 
worship.  Ariadne  represents  the  spring,  and  her  annual 
wedlock  with  Dionysus  symbolises  the  yearly  renewal  of 

Dionysus-festivals.  It  has,  however,  seemed  suitable  to  leave  them  out 
here.  For  a  full  but  rather  confusing  account  of  all  festivals  of  the 
one  kind  and  the  other,  see  Preller's  article  "Dionysia"  in  Pauly's  Real- 
Encyclopddie.  The  confusion  is  certainly  not  in  Preller's  admirable  pre 
sentation,  but  rather  in  conflicting  and  insufficient  information  which  alone 
is  available. 

1  The  merit  of  first  seeing  this  connection  belongs  to  Dr.  Thiersch, 
who  speaks  of  it  in  his  introduction  to  Pindar,  p.  156.  Speaking  of  the 
archons  Aristotle  says,  Constitution  of  Athens,  3  :  $Kr)crav  5'  ou%  a/na 
irdvres  01  evvta  dpxofres,  dXX'  6  /*&>  /3a<riXei)s  e[l]xe  TO  vvv  Ka\ovfj.€vov 
JZovK&Xiov,  TrXtja'lov  TOV  Upvravetov  (a"r)/j.eiov  d£'  £TI  Kal  vvv  yap  rrjs 
TOV  /3a<nX^ws  yvvaiKos  17  (ri;/i/xi£is  evravda  ylverai  T$  Atovvay  /cat  6 


iv  REFORMED  ATHENIAN  FESTIVALS  131 

nature's  vivid  powers  of  growth.  But  the  wife  of  the  King- 
archon  at  Athens  did  not  necessarily  suggest  the  spring. 
Dionysus'  yearly  marriage  with  her,  celebrated  in  the  Bu- 
colion,  was  certainly  witnessed  at  Athens  with  feelings  much 
the  same  as  those  entertained  by  Venetians  who  witnessed 
the  annual  espousals  of  their  doge  with  the  sea.  In  the  one 
case  the  bridegroom  and  in  the  other  the  bride  represented 
the  body  politic,  and  the  espousals  in  both  cases  proclaimed 
most  loudly  the  existence  of  a  tie  between  each  member  of 
the  body  politic  and  a  power  which  all  regarded  as  assuring 
safety  to  the  state.  Only  for  Venice  the  power  was  of  this 
world  and  material,  while  for  Athens  it  was  the  glorious 
divinity,  the  mysterious  grandeur  and  intensity,1  that  trans 
figured  the  Attic  Dionysus. 

Such  were  the  important  and  significant  ceremonies 
added  in  Pisistratus'  day  to  the  flower -feast  of  Bromius, 
and  they  made  of  it  a  festival  which  later  on  Thucydides 2 
calls  "the  older  Dionysia,"  and  which  coincided  with  the 
summer  feast  of  joy  observed  in  the  country,  particularly  at 
Icaria.  Another  phrase  applied  to  it,  also  in  Thucydides' 
day,  is  the  "  Lesser,  or  the  Rural  Dionysia."  For  this  Mr. 
Browning  has  a  shorthand  phrase  taken  directly  from  the 
Greek  :  he  has  called  this  holiday  the  "  Little-in-the-fields." 
These  various  names  of  Older  and  Lesser  served  evidently 
to  distinguish  the  feast  of  Pisistratus  from  the  Greater 


1  It  should  be  mentioned  (simply  by  way  of  showing  what  a  puzzling 
mixture  of  life  and  death,  gladness  and  grief,  all  Dionysus  worship  was) 
that  the  last  day  of  Pisistratus'  reformed  flower-festival  was  a  commemor 
ation  of  funereal  kind.  Fourteen  women  took  a  solemn  oath  of  purity  and 
obedience  to  tradition.  Then  they  made  due  sacrifice  at  fourteen  altars, 
one  for  each  of  the  fragments  into  which  the  fourteen  Titans  rent  Dionysus- 
Zagreus.  Possibly  these  were  of  the  nature  of  those  humanised  funereal 
rites  whose  institution  Plutarch  attributes  to  Epimenides  of  Crete. 


132  DIONYSUS  AT  ATHENS  iv 

Dionysia  which  came  into  existence  long  after  his  day  and 
generation. 

And  now  for  the  second  festival  of  Dionysus  as  instituted 
by  Pisistratus.  Remarking,  no  doubt,  the  great  and  growing 
importance  of  the  Icarian  mid-winter  celebration,  he  legal 
ised  the  popular  winter  holiday  of  Dionysus  at  Athens,  and 
made  various  innovations  in  the  direction  of  the  new  Icarian 
fashion  of  representing  plays.  And  here  we  come  to  an 
important  and  more  or  less  certain  date.  In  535  B.C.,  eight 
years  before  Pisistratus  died,  Thespis  of  Icaria  brought 
out  his  first  play,  at  the  winter  or  Lenaean1  festival  in 
Athens.2  Thus  the  Icarian  Satyr-play  and  tragedy  were 
brought  to  Athens  just  at  the  time  when  Dionysus  came 
into  power,  so  to  speak.  In  fact  their  introduction  was  his 
triumph. 

This  renovation  and  new  consecration  of  their  imme 
morial  merry-makings,  and  of  the  time-honoured  rites 
resorted  to  for  ensuring  a  fruitful  year  on  the  Attic  country 
side,  gave  immense  satisfaction  to  the  people  at  large.  It 
lifted  a  load  of  apprehension  and  discontent  from  their 
hearts,  and  the  echo  of  their  longing,  made  less  mournful  by 

1  Ribbeck  maintains  with  good  arguments  that  the  Icarians  held    to 
the  old  name  for  Gamelion,  which  was  Lenaeon  (see  Plutarch's  fragments 
on  Hesiod,  No.  29).     That  Athens  once  used  this  name  for  the  month  of 
the  winter  -  festival  of  Dionysus  -  Lenaeus  (for  this  name  see  Hesychius) 
and  of  the  biennial  processions  of  the  Lenae  (Strabo,  x.  468)  or  Lenides 
(Eustath.   on  Iliad,  vi.    132)  to  the  mountains,  especially  to  Parnassus 
(Pausan.  X.  iv.  3),  is  proved  by  its  survival  in  Asia  Minor,  whither  Athenians 
early  transplanted  it  (see  the  inscriptions  cited  by  Pape  in  his  Diet,  of 
Proper  Names}.     The  Boeotians  possibly  had  the  same  name  for  the  same 
season  (Hes.  Works  and  Days,  504),  though  Plutarch  (fragments  on  Hesiod, 
No.  29)  denies  it. 

2  The  precise  date  depends  upon  the  Parian  Marble,  but  not  the  fact, 
which  is  very  widely  vouched  for.    It  is  noteworthy  that  the  trustworthiness 
of  dates  given  on  the  Parian  Marble  has  received  confirmation,  here  and 
there,  by  the  newly-recovered  Athenian  Constitution  of  Aristotle.     See  Mr. 
Kenyon's  note  on  Damasias,  p.  33. 


iv  THE  PEOPLE'S  FAITH  IN  DIONYSUS  133 

trust  in  the  coming  of  the  god,  is  still  heard  in  a  prayer 
written  by  Sophocles,  and  uttered  for  luckless  Antigone  at 
a  time  of  breathless  crisis  in  her  fate.1 

"  Come,  for  all  the  people  tremble  at  the  threatened  harm. 
Pass  thou  with  purifying  footsteps  down  Parnassus'  slope, 
ay  !  or  cross  the  booming  gulf  of  waters.  Help !  Leader 
thou  of  fire-flaming  planets  in  the  dance.  Help  !  Overseer 
thou  of  cries  men  make  through  sleepless  watches  of  the 
night.  Show  thee  now,  son  of  Zeus  begotten  !  Come  and 
bring  from  Naxos  in  thy  train  the  frenzied  Thyiades  thy 
handmaidens,  who  all  the  livelong  night  dance  thee, 
thee  lacchos,  O  dispenser  thou  and  steward  for  mankind." 
The  peculiar  and  startling  locution  here  may  be  supposed 
to  represent  the  acts  of  the  dancing  Thyiades,  as  a  sort  of 
materialisation  of  his  power  in  them,  their  dancing  is  the 
god  in  them  made  manifest.  The  invocation  which  intro 
duces  this  same  prayer  well  shows  the  wide  tolerance  which 
gathered  into  Athenian  worship  epithets  and  rites  from  every 
home  of  the  god  in  Greece ;  for  many  are  his  homes,  and 
numerous  indeed  were  the  places  of  his  birth.  "  Thou  who 
bearest  many  names,  Semele's  delight,  who  watchest  over 
far-famed  Icaria 2  and  rulest  where  all  are  welcome,  in  the 
sheltered  lowlands  of  Eleusinian  Deo.  O  dweller  in 
Thebes  of  the  softly  gliding  Ismenus,  in  Thebes  the  mother 
of  Bacchanals.  Thou  hast  shown  thee  amid  the  smoky 
glare  of  flaming  torches,  arriving  on  Parnassus — mountain  of 
twin  peaks.  Near  these  two  peaks  live  nymphs  close  to 
the  cave  Corycian,  and  there  flow  Castalia's  fountain  springs. 

1  Antigone,  1115-1152. 

2  That  this  is  the  right  reading  can  hardly  be  denied  by  any  one  who  has 
read  pp.  96  and  97  of  Dr.  Merriam's  Report.     The  conditions  upon  which 
Professor  Jebb  said  he  would  read  Icaria  (see,  in  his  ist  edition,  note  on 
line  1119)  are  now  fulfilled,  and  he  is  half  convinced  (cf.  his  2d  edition). 


1 34  DIONYSUS  AT  A THENS  i v 

Yea,  the  ivy  tangled  in  the  folds  of  Nysa's 1  hills,  the  tender 
green  of  lofty  promontories  covered  with  luxuriant  vines 
send  down  to  Thebes  the  Saviour  God,  and  her  streets  are 
filled  with  the  heavenly  clangour  of  his  echoing  name. 
Thebes,  dearly  of  him  and  of  his  thunder-smitten  mother 
dearly  loved  ! "  Truly  the  ring  of  a  most  genuine  piety 
sounds  in  many  a  passage  of  the  Attic  tragedians,  but  here 
Sophocles  has  certainly  surpassed  himself.  The  anxiety 
felt  by  Dionysus'  peasant  worshippers  was  that  their  god 
should  be  duly  propitiated.  They  wished  to  conciliate  his 
favour  for  Athens  and  Attica.  Nothing  could  accomplish 
this  unutterably  desirable  end  but  the  official  celebration  of 
his  festivals. 

In  satisfying  this  demand  by  his  two  feasts,  the  Anthe- 
steria  and  the  Lenaea,  Pisistratus  called  Thespis  and  Tragedy 
from  Icaria.  But  this  performance  of  plays,  once  trans 
ferred  to  the  broader  horizon  of  the  capital,  soon  grew  to 
such  proportions  that  it  threatened  to  crowd  out  the  indis 
pensable  and  immemorial  religious  acts  required  to  be 
done  by  all  worshippers  for  the  health  of  the  state.  Thus 
the  deep  religious  purpose  for  which  Dionysus  and  his  wor 
ship  were  honoured  at  Athens  would  have  been  unfulfilled. 
To  remedy  this  came  the  later  institution  of  the  Greater 
Dionysia,  by  which  the  most  important  representations  of 
tragedies  and  comedies  were  relegated  to  a  third  and  al 
most  exclusive  theatrical  occasion  made  for  the  worship  of 
Dionysus.  The  month  of  March  was  fixed  upon  for  this 
festival,  which  seems  to  have  been  wonderfully  free  from 
the  trammels  of  mystic  nature-worship.  It  was  indeed  a 
new  institution  made  in  the  spirit  of  the  democratic  reforms 

1  On  Nysa,  see  Appendix  III.  at  the  end  of  this  lecture,  upon  the  second 
birth  and  eastern  affinities  of  Dionysus. 


iv  THE  GREATER  DIONYSIA  135 

of  Clisthenes  to  honour  the  people's  most  beloved  god. 
The  name  by  which  it  was  sometimes  called  was  the 
City  Dionysia^  but  its  commoner  name  was  the  Greater 
Dionysia. 

The  gorgeous  pageants  of  Venice  in  all  her  glory  seem 
unspontaneous  and  almost  insincere  when  compared  with 
this  great  Athenian  glorification  of  Dionysus.  Patriotism 
intensified,  exultant  freedom,  delight  in  beauty,  delicate 
skill  in  all  graceful  arts,  animated  and  adorned  this  World- 
Exhibition  of  high  thoughts  and  melodious  speech.  There 
Dionysus  shone,  a  leader  of  the  Muses,  and  the  Graces 
moved  in  his  train.  So  transfigured  was  the  Thracian  god 
that  all  the  savagery  of  his  ancient  worship  now  became  a 
unison  of  speech  and  song  and  dance,  a  dazzling  mani 
festation  of  all  concordant  arts,  wherein  there  shone  the 
blithest  and  the  best  that  sculpture  ever  shaped  or  poetry 
conceived. 

And  now  it  is  time  to  contemplate  the  god  himself  in 
all  his  Attic  and  comprehensive  majesty.  The  enlightened 
Attic  worshipper  of  Bacchus  sought  at  the  Athenian  Diony- 
siac  theatre,  in  the  presence  of  the  older  gods,  solemnly 
represented  by  their  priests,  a  relief  from  that  sense  of 
spiritual  oppression  from  which  the  human  conscience  has 
never  been  entirely  free.  As  an  analogy  to  the  idea  that 
"  in  Adam's  fall  we  sinned  all,"  may  be  found  underlying 
this  Attic  ceremonial  the  idea  of  a  vicarious  complicity  in 
the  old-time  murder  of  the  Icarian  king,1  and  of  a  predes 
tined  responsibility  for  the  sad  fate  of  Erigone. 

All  the  great  Athenian  tragedies  were  acts  of  worship 

1  To  speak  of  Icarius  as  a  king  is  to  use  the  language  of  the  later  and 
Athenian  version  of  the  legend  which  makes  him  king  of  Athens.  This 
seems  to  be  the  version  alluded  to  by  Hyginus,  Fab.  130,  and  by  Pausanias, 
I.  ii.  7,  end. 


136  DIONYSUS  AT  ATHENS  iv 

dedicated  to  Dionysus,1  but  one  of  them  is  called  the 
Bacchanals,  the  last  but  one  of  all  plays  written  by  the 
latest  born  of  the  three  great  tragedians,  Euripides.  The 
Bacchanals  enacts  and  explains  as  its  sole  plot  and  plan, 
utters  as  the  burden  of  all  its  choral  songs,  the  fulness  of  the 
power  of  Dionysus.  Here  no  aspect  of  the  Bacchic  god 
head  is  forgotten.  Nowhere  in  all  literature  is  the  strange 
baffling  quality  of  Dionysus  presented  with  such  complete 
ness  and  consistency  as  in  this  play  of  Euripides.  Here  is  a 
tragedy  written  with  the  sincerely  pious  intent  of  revealing  the 
spirit  and  the  will  of  the  god — a  veritable  Gospel  according  to 
Euripides.  A  gospel  truly,  and  of  Dionysus ;  but  of  what 
Dionysus  ?  Is  the  god  of  this  marvellous  play — the  vision 
of  him,  that  is,  which  was  granted  to  Euripides  in  the  ful 
ness  of  his  powers — a  revelation  from  the  purely  Attic 
worship  of  Dionysus,  or  a  reminiscence  of  the  fiercer 
Thracian  god,  or  is  he  a  philosopher's  fiction  argued  about 
and  reduced  to  consistency  until  he  has  lost  the  wild-wood 
tang  attached  to  his  native  self?  The  answer  must  be  that 
he  is  not  one  but  all  of  these  in  one.  Euripides  no  doubt 
would  be  the  first  to  feel  that  a  poet  could  only  gain  by  a 
seeming  inconsistency  in  an  attempt  to  interpret  the  most 
inconsistent  of  all  divinities. 

Hence  the  many  aspects  of  the  Bacchanals.     It  may  be 

1  This  fact  needs  frequent  reassertion  in  answer  to  those  who  are 
inclined  to  put  a  poet  like  Euripides  in  the  position  of  a  modern  assailant 
of  religion.  The  circumstance  that  he  wrote  tragedies  at  all  ought  to  clear 
him  of  any  such  charge.  As  Bernhardy  has  most  truly  said,  ' '  People  are 
sometimes  oblivious  of  the  fact  that  the  whole  structure  of  Greek  religious 
service  and  their  whole  scheme  of  nature -worship  remained  wholly  un 
shaken  and  intact  until  the  Peloponnesian  war  or  thereabouts.  Accord 
ingly  such  fault-finding  or  doubt  as  poets  and  thinkers  express  relates  to 
morals  and  to  certain  misrepresentations  of  the  divine  nature,  and  did  not 
come  near  the  heart  of  their  national  religion."  Litter aturgeschichte,  i. 
p.  420. 


iv  THE  BACCHANALS  OF  EURIPIDES  137 

called  the  Passion-play  of  Attica,  and  it  has  been  compared  to 
the  Medieval  Morality.  And  yet  a  certain  loftiness  of  religious 
tone  does  not  here  exclude  the  most  unmistakable  reminis 
cences  of  the  fierce  and  awful  god  of  ancient  Thrace.  The 
philosophic  and  discerning  reader  may  see  in  the  mother 
who  rends  her  own  son  —  in  Agave  and  all  her  Maenad 
train  —  a  spirited  personification  of  the  power  in  a  roaring 
mountain  torrent,  and  of  the  fury  of  lapping  flames  ;  but  he 
will  not  press  the  point  too  far  if  he  remembers  that  Euri 
pides  wrote  his  play  while  he  was  staying  at  the  court  of 
Archelaus  in  Macedonian  Pieria,  that  very  portion  of  Thrace 
whence  Dionysus  issued.  It  is  as  if  a  pilgrimage  to  the  land 
of  his  birth  were  required  by  the  god  himself  of  a  poet 
whose  presentation  of  the  Bacchic  godhead  was  to  perpetuate 
its  undimmed  memory.1 

Something  wilder  than  religious  Athens  knew  surrounded 
Euripides  and  every  Athenian  who  visited  Macedonia  in 
those  days.  Such  visitors  "would  hear,  and  from  time 
to  time  actually  see,  something  of  a  religious  custom 
in  which  the  habit  of  an  earlier  world  might  seem  to 
survive.  As  they  saw  the  lights  flitting  over  the  moun 
tains,  and  heard  the  wild  sharp  cries  of  the  women,  there 
was  presented  a  singular  fact  in  the  more  prosaic  actual  life 
of  a  later  time,  an  enthusiasm  otherwise  relegated  to  the 
wonderland  of  a  distant  past,  in  which  a  supposed  primitive 
harmony  between  man  and  nature  renewed  itself."  2  It  is 

1  It  has  already  been  noted  above  that  Pisistratus,  during  his  second 
exile,  spent  most  of  his  time  in  Thrace.  On  returning  —  this  was  his 
second  restoration  —  he  was  enabled  to  encourage  the  innovations  of 
Thespis,  and  give  official  sanction  to  the  peasant  worship  of  the  Thracian 
god.  See  Aristotle's  Constitution  of  Athens,  15,  and  Mr.  Kenyon's  refer 
ence  there  to  Herodotus'  allusion  to  supplies  drawn  by  Pisistratus  airo 


2  Here,  as  below,  I  quote  from  Mr.  Pater's  essay. 


138  DIONYSUS  AT  ATHENS  iv 

well  known  that  the  women  of  the  house  of  Philip  and 
Alexander  were  carried  into  measureless  excesses  by  the 
possession  of  the  god.  They  were  Bacchanals  with  a 
vengeance,  and  all  the  dreadful  deeds  attributed  by  Euri 
pides  to  his  Bacchanals  would  therefore  be  looked  upon 
by  them  as  a  poetic  amplification  only  of  what  lay  within 
their  own  experience.  "Later  sisters  of  Centaur  and 
Amazon,  the  Maenads,  as  they  beat  the  earth  in  strange 
sympathy  with  its  waking  up  from  sleep,  or  as  in  the 
description  of  the  messenger,  in  the  play  of  Euripides, 
they  lie  sleeping  in  the  glen  revealed  among  the  morning 
mists,  were  themselves  indeed  as  remnants  —  flecks  left 
here  and  there,  and  not  quite  evaporated  under  the  hard 
light  of  a  later  and  common  day — of  a  certain  cloud-world 
which  had  once  covered  all  things  in  a  veil  of  mystery." 

It  is  indeed  marvellous  that  our  poet  in  this  very  play,  so 
well  fitted  to  please  a  semi -barbarous  Macedonian  taste, 
so  full  of  the  proto-Thracian  spirit  of  Dionysus,  has  been 
able  to  remain  true  to  the  loftier  teachings  of  Anaxagoras 
and  Socrates.  The  fact  nevertheless  remains  that  a  leaven, 
half  of  philosophy  and  wholly  religious,  so  pervades  this 
play  that  it  not  only  sums  up  the  past  but  prefigures  the 
future.  It  contains,  revealed  here  and  there  in  brief  flashes, 
what  may  be  called  a  Messianic  vision.  Less  manifestly 
perhaps  than  Virgil,  and  yet  perhaps  more  deeply,  Euripides 
is  moved  by  a  vision  long  beforehand  of  religious  truth  to 
come.  In  the  sorrows  and  the  joys  of  Dionysus  and  his 
train,  touches  come  here  and  there  which  are,  it  would 
seem,  the  outcome  of  a  Dionysus-granted  power  of  prophecy. 
Euripides  had  vision  long  beforehand  of  the  mysteries 
of  faithful  sorrowing,  the  ecstasies  of  Christian  joy.  This 
is  no  new  discovery,  for  in  the  days  of  the  early  Church 


iv  MESSIANIC  VISION  OF  EURIPIDES  139 

the  Christian  poet  Nonnus  devoted  his  energies  to  a 
long  and  most  loving  work,  chronicling  with  minute  and 
pious  care,  often  in  most  sweetly  flowing  verse,  all  that  has 
ever  been  sung  or  said  of  Dionysus.  Still  more  striking 
was  the  appreciation  of  Dionysus  shown  by  the  pious  com 
piler  of  a  curious  work  called  Christus  Pattens.  To  this 
devout  Christian's  uncramped  and  unpremeditated  piety  was 
given  the  vision  of  a  real  analogy  between  the  passion  of 
Christ  and  the  passion  of  Dionysus.  Moreover  we  owe, 
strangely  enough,  the  preservation  of  some  important  lines 
in  the  play  of  Euripides  to  his  curious  cento. 

And  now  before  analysing  the  Bacchanals  one  caution 
must  be  given.  None  must  suppose  that  the  personage 
named  Dionysus,  who  proclaims  himself  to  be  the  god  dis 
guised  in  mortal  form,  is  the  only  presentation  of  the  god 
head.  This  disguised  Dionysus  is  in  some  ways  still  what 
we  may  suppose  the  one  speaker  used  by  Thespis  in  his 
earliest  Icarian  ventures  to  have  been.  He  is  not  Dionysus 
but  only  a  focal  point  around  which  gather,  with  endless  and 
flickering  play  of  change,  the  constantly  shifting  figures  in 
the  plot.  The  Maenads  who  followed  the  Dionysus-man 
from  Asia,  and  who  form  the  chorus  of  the  play,  are  them 
selves  the  god.  The  Dionysus-driven  women  of  Thebes 
who  against  their  will  are  dashed  from  home  to  revel  on  the 
mountain  side,  these  are  but  passive  receivers  of  the  Bacchic 
godhead — "Impotent  pieces  of  the  game  he  plays." 

King  Pentheus,  though  he  loudly  proclaims  himself  the 
foe  of  Dionysus,  is  only  possessed  by  the  mad  frenzy  sent 
by  Zagreus-Dionysus,  the  wild  Thracian  huntsman  who  has 
found  his  furious  way  through  Cretan  legends.  Dionysus, 
in  some  of  his  many  phases,  is  manifested  by  any  and  every 
personage  in  the  play.  Even  the  scene  of  its  enactment  is 


1  40  DIONYS  US  AT  A  THENS  i  v 

full  of  the  god,  for  it  is  the  holy  spot  in  Thebes  l  where 
Semele  died  in  bringing  Dionysus  prematurely  to  his  birth. 
We  see  the  thunder-smitten  ruins  where  Semele  was  slain. 
Their  smoking  embers  are  hallowed  by  a  fitful  flame.  This 
is  the  god  of  fire  actually  present,  and  to  him  the  spot  had 
long  been  made  consecrate  by  order  of  Cadmus,  when  Cad 
mus  still  was  king.  To  mark  them  twice  his  own,  Dionysus 
has  covered  these  smoking  ruins  with  his  cherished  vine. 

The  play  opens  with  the  arrival  at  this  place  in  Thebes 
of  the  youthful  god  who  soliloquises  :  Behold  Zeus's  son 
arrived  at  Thebes,  where  fire  consumed  his  mother  Semele. 
He  surveys  the  scene,  and  proclaims  his  own  presence 
marked  by  flame  and  by  the  vine.  Then  our  Dionysus,  in 
mortal  disguise,  looks  out  upon  the  world  and  sees  the  same 
godhead  made  manifest  on  every  side,  and  recognised  every 
where  as  pre-eminent.  Nature  with  her  floods  and  flames 
and  all  her  luxury  of  green  is  his,  for  his  power  has  made  it  ; 
and  man  also,  up  to  the  present  moment,  has  been  prompt 
to  pay  him  homage.  The  Lydians  on  their  fields  flooded 
with  gold,  the  Phrygians  and  Persia's  burning  plains,  the 
forts  of  Bactria  and  the  snow-swept  reaches  of  Media,  all 
Araby  the  blest  with  the  swarming  cities  of  Asia  Minor, 
have  acknowledged  the  god.  Now  Greece  must  bow  the 
knee  before  him.  His  night-long  revels  must  be  loved  ;  and 
the  reveller,  in  order  no  doubt  to  keep  before  him  a  sign  of 
the  star-flecked  sky  at  eve,  at  midnight,  and  at  dawn,  must 
adopt  the  garb  beloved  of  Dionysus  —  a  dappled  fawn  skin. 

The  god  made  man  proceeds  to  tell  what  brought  him  to 
Thebes.  The  madding  impulses  that  Dionysus  sends  invited 

1  Cf.  Pausanias,  IX.  xii.  3  :  0a<rt  S£  ol  G^/Moi,  Kadon  r??s  d/f/307r6X'ea?y 
dyopd  afaaiv  e0'  TJ/J.UV  ireiroiTjTat.,  KaS/xou  TO  dpxcuov  oiK.lav  elvac 


TOV 
TOVTOV  5£  Kal  es 


iv  ANALYSIS  OF  THE  INTRODUCTION  141 

by  the  insults  heaped  upon  his  dearest  mother  Semele  at 
Thebes,  and  by  an  impious  resistance  there  offered  to 'his  wor 
ship,  have  seized  and  carried  away  the  womankind  of  all  the 
town.  The  men  are  slower,  more  stubborn  and  prone  to  fight, 
though  the  battle  be  a  losing  one.  The  women  of  Cadmus' 
house  have  gone  stark  mad  and  have  left  their  homes  to 
revel  in  the  wilderness  of  Cithaeron,  and,  lest  they  should 
not  be  filled  enough  with  his  power,  Dionysus  in  the  flesh 
now  goes  to  them,  and  leaves  those  other  Dionysus-driven 
creatures, — his  Maenad  followers  who  came  to  Thebes  with 
him  from  far-off  Asia.  With  his  exit  toward  Mount 
Cithaeron  ends  the  opening  and  introductory  scene  of  the 
play — its  prologue. 

In  the  act  of  going  Dionysus  makes  a  sign  at  which  the 
Asiatic  Maenads,  the  Bacchanals  from  whom  the  play  is 
named,  troop  wildly  into  the  theatre ;  and  the  street  around 
Pentheus'  royal  palace  resounds  with  the  beating  of  drums 
and  Bacchic  cries.  The  disguised  Bacchus,  the  man- 
Dionysus,  their  youthful  captain,  is  gone,  but  the  god 
Dionysus  still  is  there,  and  his  voice  is  heard  in  the  strains 
now  sung  by  this  Bacchanalian  throng. 

They  tell  of  their  weary  journeyings  made  sweet  by 
Dionysus'  love.  They  warn  the  polluted  and  the  profane 
to  beware  and  give  way,  for  the  god  is  to  speak  in  them. 
Then  comes  a  divine  song  which  tells  of  blessed  mysteries. 
Blest  is  he  who  hallows  his  manner  of  life,  and  cleaves  with 
his  soul  most  straitly  to  the  Thiasos  fulfilled  with  the  god, 
the  Thiasos  madly  scouring  the  mountains ;  for  thus  shall 
his  soul  be  purged  and  made  most  clean.  Yea,  and  with 
the  worship  of  the  god  must  be  joined  reverence  for  the 
great  mother  goddess  Cybele.  "  Then  on,  Bacchanals,  on  ! 
ye  Bacchants,  lead  ye  Dionysus  home  to  Thebes." 


1 42  DIONYSUS  AT  ATHENS  iv 

The  next  strain  sings  of  thunder  -  smitten  Semele,  of 
Zeus,  who  snatched  her  babe  to  his  thigh,  whence  it  came 
in  due  season  to  full  and  fated  birth.1 

"  Thebes,"  so  shouts  the  Maenad  throng,  "  Thebes,  the 
nurse  of  Semele,  deck  thee  now  and  yield  to  Bacchic 
promptings — 

Branches  of  ivy  or  of  oak 

Take  thou,  a  very  Bacchanal ; 
Nor  let  the  Bacchant's  dappled  cloak 

Of  fawn-skin  from  thy  shoulders  fall, 
White-fringe  it  all  with  wool-tufts  small ; 
The  ferule  wield  with  reverent  care, 
And  of  its  wantonness  beware." 

And  now  the  risen  surge  of  song  beats  higher  still,  and 
higher  rises  the  quickening  pulse  of  the  inflowing  god,  for 
they  cry — 

Soon  shall  the  country  rejoice  in  the  dance, 
Soon  with  his  revellers  Bacchus  advance, 
Into  the  hills,  the  hills  shall  he  fare. 

Then  for  a  time  not  Dionysus,  but  a  peculiar  aspect  of 
Zeus  is  the  theme  of  the  sacred  song.  They  sing  of  the 
Zeus  of  Crete,  who  is  after  all  not  the  father  of  Dionysus  but 
Dionysus  himself.  Euripides  certainly  was  conscious  of 
this,  and  he  means  at  least  to  suggest  that  Crete  was  a 
debatable  ground,  where  the  legends  of  Dionysus  and 
Zeus  met  and  overlapped.  The  Curetes  in  their  Cretan 
haunts,  the  wildly  dancing  Corybantes  so  picturesquely  sung 
of  here,  surrounded  the  birth  and  protected  the  rearing  of 

1  Dr.  Sandys,  in  his  excellent  note  on  this  passage,  refers  to  the  epithet 
rjfUT&effTos,  half-matured,  used  of  Dionysus  by  Nonnus,  Dionysiaca,  xlv. 
99  ;  see  also  i.  5,  and  he  also  cites  Ovid,  F.  iii.  717,  puer  ut posses  maturo 
tempore  nasci,  expletum  patrio  corpore  matris  onus.  See  upon  this  curious 
feature  of  the  myths  the  Appendix  at  the  end  of  this  lecture. 


iv  ENTRANCE-SONG  OF  THE  BACCHANALS         143 

this  Cretan  Zeus-Dionysus.  The  mother  goddess  Rhea  is 
associated  with  him  in  Crete,  just  as  Cybele  is  united  with 
Dionysus  in  Phrygian  worship.  Moreover  the  Maenads 
of  our  chorus  immediately  pass  to  a  song  of  the  invention 
of  various  Bacchic  instruments — the  drum,  the  flute,  and 
the  cymbal.  We  hear  that  all  these  were  instruments  for 
praising  Cybele  and  Dionysus,  or,  if  so  you  choose  to  say  it, 
of  Rhea  and  of  Cretan  Zeus. 

That  this  whole  song  was  profoundly  religious  is  no  doubt 
sufficiently  evident,  but  the  religious  intent  is  nowhere  more 
undisguisedly  present  than  in  its  closing  strain,  where  the 
ecstasies  of  pious  revellings  are  wildly  sung  with  the  cry, 
"Evoe,  Bacchus  leads  on  and  hearts  are  thrilled,"  which 
comes  from  promptings  of  the  god  himself.  After  this  the 
miracles  worked  by  Dionysus  are  touched  upon.  Here  and 
elsewhere  in  the  play  the  poet  tells  of  the  miraculous  flow  of 
milk  and  honey1  that  springs  from  the  ground  at  the  bidding 
of  his  Bacchant  revellers.  On  they  go,  beating  drums,  singing 

1  In  this  and  other  passages  of  the  play  where  Dionysus'  followers 
show  miraculous  command  over  honey,  Euripides  indicates  his  familiarity 
with  an  out-of-the-way  legend  of  Dionysus  at  home  in  the  island  of  Euboea, 
which  lay  near  by,  under  the  jealous  governance  of  Athens,  and  which  was 
largely  occupied  in  the  poet's  day  by  Athenian  colonists.  According  to 
these  legends  Dionysus  was  reared  in  Euboea  (anciently  called  Macris, 
or  Long-island)  by  Aristaeus,  the  giver  of  honey,  who  was  his  constant 
instructor.  His  nurse  in  Euboea  was  a  nymph,  who  is  sometimes  said  to 
have  been  named  Macris.  Nysa,  a  name  familiar  as  applied  in  many 
Dionysus-stories  to  the  moist  and  wooded  place  where  the  fiery  god  came 
to  birth,  was  a  second  name  given  to  this  Euboean  nurse  of  Dionysus. 
Whether  her  name  was  Nysa  or  Macris,  this  Euboean  maiden  was  Aris 
taeus'  daughter.  The  remarkable  point  to  remember  from  these  Euboean 
legends  is  the  prominence  in  them  of  milk  and  of  honey,  two  good  gifts 
from  which  Dionysus  is  dissociated  in  the  earliest  Attic  story.  Further 
more,  it  is  noticeable  that  Euripides,  in  weaving  these  bright  Euboean 
strands  into  his  play,  made  it  plain  that  he  regarded  himself  as  a  religious 
interpreter  for  the  whole  of  Greece  and  not  for  Attica  alone.  For  an 
account  of  the  wider  scope  of  the  legends  of  Dionysus'  birth  at  Nysa,  see 
Appendix  III.  at  the  end  of  this  chapter. 


144  DIONYSUS  AT  ATHENS  iv 

Evoe  to  the  Evian  god.  Phrygian  shouts  they  shout,  while 
flutes  trill  in  their  revels  to  thrill  them  with  rippling  joys. 
But  while  we  look,  as  with  a  flash  from  many  white  limbs 
darting  forward,  they  have  passed  ere  has  died  on  the  ear 
their  shout,  "  On,  Bacchants,  on  !  " 

Now  the  plot  begins  to  thicken,  and  the  three  central 
acts,  courses,  or  periods — what  you  will — now  begin.  The 
whole  play  has  five  parts,  of  which  the  introductory  one  is 
already  over.  At  the  end  comes  the  fifth  and  concluding 
part,  a  winding-up  of  the  play. 

In  the  three  central  acts  now  beginning  is  portrayed  the 
Passion,  as  it  were,  of  Dionysus.  In  the  first  act,  Reason 
fails  to  turn  the  enemy,  King  Pentheus,  from  his  impious 
purposes  against  the  god.  Here,  at  the  very  outset,  the 
flutter  of  frenzy  to  come  hovers  over  Pentheus  the  arch- 
sinner,  and  he  already  belongs  to  Dionysus.  In  the  second 
act  comes  the  consummation  of  blinded  Pentheus'  sin. 
The  man-Dionysus,  the  vicar  of  the  god  in  Thebes,  is  seized 
and  thrown  into  prison.  The  Maenads  from  Asia  are 
threatened  with  violence,  and  the  Theban  revellers  on 
Mount  Cithaeron  are  hunted,  and  some  of  them  taken 
and  thrown  into  prison.  But  close  upon  the  heels  of  sin 
treads  punishment.  The  third  act  sets  forth  the  nature 
and  the  manner  of  an  awful  chastisement  inflicted  upon 
King  Pentheus,  on  all  his  house  and  on  the  land  of 
Thebes.  An  earthquake  comes  first  to  reveal  the  wrath 
and  majesty  of  outraged  Dionysus.  Out  of  the  midst  of 
the  earthquake  the  man-Dionysus  emerges  from  his  dark 
prison,  reminding  Eleusinian  hearers  of  Persephone  restored 
from  the  realms  of  Hades  to  Demeter  and  the  day.  He 
comes  to  foil  and  flout  his  half -crazed  persecutor,  as 
appears  in  the  next  event  on  the  stage.  Madness  seizes 


iv  CENTRAL  ACTS  AND  CONCLUSION  145 

Pentheus,  madness  and  the  judgment  of  flood  and  fire, 
those  attendant  ministers  of  the  earthquake  which  are 
personified  in  the  Theban  Maenads.  These  finally  rend 
Pentheus  and  reduce  him  to  a  shapeless  and  dismembered 
heap  of  fragments.  The  wages  of  sin  is  madness  first, 
and  finally  death.  So  ends  the  Passion  of  Dionysus  in 
the  destruction  of  his  persecutor  in  whose  dismember 
ment  we  see  rehearsed  the  tragic  fate  of  Dionysus-Zagreus. 
The  conclusion  or  Exodus  of  the  play  sets  forth  a  moral 
that  recalls  the  purifying  ritual  introduced  with  Delphian 
Apollo's  sanction  from  the  border  town  of  Eleutherae.1 
The  moral  in  point  is,  that  there  is  need  of  Dionysus'  help 
in  recovering  from  harms  brought  about  by  his  own  power, 
and  with  it  is  coupled  an  urgent  representation  of  the  folly 
of  all  resistance  to  the  god. 

Such  is  a  brief  outline  of  the  central  acts  and  conclusion 
of  the  play,  whose  noble  introduction  has  been  examined 
already  at  length.  It  now  remains  to  take  an  equally  care 
ful  view  of  these  four  main  parts.  To  begin  with  the  first 
scene,  where  the  intending  sinner  is  still  on  his  probation. 
The  transgressor  in  question  is  Pentheus  the  son  of  Agave, 
Semele's  sister,  into  whose  hands  his  grandfather  Cadmus 
has  resigned  the  royal  power  at  Thebes.  Tiresias  the  seer 
is  first  seen  upon  the  stage.  His  name  is  Tiresias,  but  he 
has  suffered  a  Bacchic  change  into  something  not  himself; 
and  his  gospel  is  the  method  of  Bacchic  madness.  This 
Tiresias  is  not  the  dread  shade  that  defies  in  Homeric 
song  the  power  of  darkness  and  seems  to  live  in  death. 
Nor  is  he  the  Tiresias  of  Sophocles,  that  majestic  incarna 
tion  of  wisdom  whose  mighty  wrath  and  burning  scorn 
cowed  even  the  spirit  of  Oedipus  the  Great.  Tiresias  in 

1  See  above,  Appendix  II.,  on  Dionysus  of  Eleutherae. 
L 


146  DIONYSUS  AT  ATHENS  iv 

the  Bacchanals  is  grotesque,  if  we  forget  that  Dionysus 
has  entered  into  him  and  possessed  him,  when  he  comes 
upon  the  stage  attired  in  a  Bacchic  garb,  ill-suited  alike  to 
his  years  and  his  priestly  office.  He  is  bent  upon  taking 
his  part  in  Bacchic  revellings,  and  is  in  the  act  of  seeking 
another — a  companion  old  like  himself,  and  like  himself  ill- 
suited  for  the  dance.  This  companion  appears ;  he  is  the 
royal  Cadmus,  and  shows  at  the  outset  eagerness  even 
greater  than  that  of  Tiresias  for  gambolling  in  the  wilder 
ness  of  Cithaeron,  saying  : 

Where  leads  the  dance,  where  must  we  take  our  turn 

And  toss  our  gray-haired  heads  ?     Interpret  thou, 

Aged  Tiresias  ;  lead  my  old  age, 

For  thou  art  wise.      The  livelong  day  and  night 

Untiring  with  my  thyrsus  I'll  smite  earth. 

'Tis  sweet  for  us  when  we  our  age  forget.1 

Tiresias  ends  by  seeming  the  less  grotesque  of  the  two ; 
it  is  he  who  turns  apologist  for  Dionysus,  and  very  skilfully 
his  argument  begins : 

"  We  reason  not  o'er  nicely  of  the  gods, 
They  are  the  heirlooms  by  our  fathers  left, 
As  old  as  time  ;  no  logic  shall  destroy  them, 
Not  though  the  keenest  wit  should  prompt  the  thought. 

Scoff  not  at  old  men  dancing,  mock  not  at  these  ivy  crowns 
on  our  silvered  heads,"  he  says, — 

1  The  merit  of  having  established,  by  changing  one  letter  in  the  MS., 
the  undoubted  reading  here  belongs  to  Milton.  See  Dr.  Sandys  on  this 
line.  He  says  of  Milton's  emendations,  "They  were  written  in  the 
margin  of  his  copy  of  the  edition  of  Euripides  printed  by  Paul  Stephens 
at  Geneva  in  1602,  2  vols.  410,  now  in  possession  of  Henry  Halford 
Vaughan,  Esq.,  of  Upton  Castle,  Pembroke.  Milton  bought  it  in  1634, 
the  very  year  in  which  he  wrote  Comus,  which  was  acted  at  Michaelmas  of 
that  year,  and  shows  in  several  points  special  familiarity  with  this  and 
other  plays  of  Euripides.  Cf.  especially  Comus,  297-301  with  Iph.  T., 
264-274. 


iv  FIRST  GREAT  ACT  147 

The  god  hath  not  distinguished  if  the  young 
Or  if  the  older  man  should  join  the  dance, 
Claiming  from  all  alike  service  and  honour. 

But  look,  Tiresias  and  Cadmus  cease  their  talk  and  retire 
up  the  stage.  Hurrying  footsteps  interrupt  them,  and  they 
see  from  afar  Pentheus ;  he  comes  breathless,  and  quivers 
from  head  to  foot  when  he  pauses.  This  sort  of  tremor 
is  a  well-known  sign  of  approaching  madness.  The  tie  is 
close  that  binds  the  aged  Cadmus  of  Sidon  to  this  new-comer. 
Pentheus  is  twice  over  Cadmus'  grandson ;  through  Echion 
his  father  earth-born,  sprung  from  the  dragon's  teeth  which 
Cadmus  sowed  near  the  spring  of  Dirce  at  Thebes ;  and 
through  his  mother  Agave,  Cadmus'  own  daughter.  Accord 
ingly  in  this  scene  where  Pentheus  is  most  nearly  his  native 
self  he  shows  a  certain  affection  for  Cadmus,  the  only  glad 
ness  in  him,  for  he  was  otherwise  all  grief,  even  as  his  name 
implied.1  Tiresias  and  Cadmus  moved  our  laughter  when 
they  first  entered ;  contrasted  now  with  Pentheus  they  take 
on  the  semblance  of  calm  and  almost  of  dignity. 

Words  chase  each  other  out  of  Pentheus'  mouth. 
He  was  abroad.  News  came  of  Theban  women  revelling 
on  Cithaeron,  wild  with  that  strange  impostor  Dionysus. 
"  It  is  a  shame !  these  women  cloak  impure  desires  with 
professed  piety.  But,"  he  screams,  "  I  have  caught  some 
of  them  "  ;  and  then  adds,  with  a  cruel  sneer,  "  I  am  after  the 
others.  Agave  my  mother,  and  her  sisters  Ino  and 
Autonoe,  are  of  the  band,  and  all  shall  be  prisoned." 
Thinking  of  the  Dionysus-man,  he  adds  : 

To  us  a  being  strange  is  come,  they  say, 
From  Lydian  lands,  a  wizard  and  a  cheat, 

1  His  name  is  a  Greek  equivalent  for  Tristan,  the  man  of  sorrow. 


148  DIONYSUS  AT  ATHENS  iv 

With  golden  curls  and  fragrant  flowing  hair. 
His  wine-flushed  glance  hath  Aphrodite's  charm,1 
And  day  and  night  he  wanders  with  them  there. 

Here  Pentheus  refers  to  the  revellers  on  Cithaeron.  Soon, 
losing  all  self-control,  he  says  blasphemously  : 

I'll  end  the  thumping  of  his  thyrsus-wand 
And  all  the  tossing  of  his  locks :  his  head 
Shall  fall,  by  this  hand  from  his  body  sundered. 

Sneers  at  Dionysus'  fire-birth  and  Semele's  fame  and  fate ; 
scoffs  at  Dionysus'  second  birth  from  Zeus's  thigh,2  coupled 
with  insults  heaped  upon  the  memory  of  Dionysus'  well- 
beloved  mother,  now  follow  quickly,  and  the  blasphemer  is 
so  wholly  engrossed  in  his  blaspheming  that  he  fails  to  see 
the  two  old  men  who  have  been  hovering  in  the  background 
awaiting  opportunity  to  address  him.  Now,  with  a  wild 
start  at  their  Bacchic  trappings,  Pentheus  whirls  a  torrent 
of  angry  words  upon  them.  He  fairly  goes  mad  with  rage. 
Tiresias  is  responsible ;  Tiresias  must  be  gaoled  with  the 
women  captured  from  Cithaeron.  A  gleam  of  moderation 
revisits  him  just  here ;  after  all,  Tiresias  is  too  old  for  such 
treatment,  he  says.  Those  women,  though,  must  and  shall 
be  kept  from  wine.  "  Wine-bibbing  is  no  meat  for  woman 
kind  ! "  exclaims  the  tumultuous-minded  king.  The  quality 
of  Tiresias,  as  Sophocles  portrayed  him,  now  shows  in  his 
wrathful  answer  to  the  king. 

1  Dr.  Sandys  is  very  happy  in  quoting  here  two  lines  of  the  Comus, 
752,  753,  which  very  probably  were  inspired  by  this  passage  : 

What  need  a  vermeil-tinctured  lip  for  that, 
Love-darting  eyes,  or  tresses  like  the  morn  ? 

-  Here  as  in  other  blasphemies  of  Pentheus  we  have  a  picturesque  state 
ment  of  the  doubts  and  difficulties  felt  by  reasoning  men  concerning  certain 
grotesque  features  in  the  myth  of  Dionysus.  See  Appendix  at  the  end  of 
this  chapter. 


iv  THE  PROBATION  OF  PENTHEUS  149 

Sham  wisdom  oils  and  glibly  wags  thy  tongue, 
But  all  thy  argument  is  foolishness. 
A  bold  man  skilled  in  overmastering  speech, 
If  sense  abides  not  in  him,  harms  the  state. 

And  here  Tiresias,  inspired  less  by  the  god  perhaps  than 
by  the  over-subtle  reasonings  about  gods  and  men  which  he 
so  eloquently  scorned  earlier  in  the  play,  subjects  the  tale 
of  Dionysus'  second  birth  to  a  treatment  half  meant  in 
earnest  and  half  intended  as  an  answer  to  the  fool  according 
to  his  folly.1  To  understand  the  subtlety  of  his  argument 
here,  it  must  be  understood  that  meros  is  the  Greek  word 
meaning  piece  or  part,  while  meros  means  thigh. 

"Scorn  not,"  says  the  subtle  seer  Tiresias,  "but  rever 
ently  repeat  the  tale  how  Zeus  plucked  to  Olympus  the  babe 
unborn.  Thence  Hera  strove  to  fling  him  down,  but  she 
was  foiled.  To  defeat  her  Zeus  took  a  piece," — a  meros — 
"from  earth -encircling  ether.  This  phantom  babe  was 
abandoned  to  Hera.  The  real  Dionysus  babe,  meanwhile, 
was  firmly  sewed  with  golden  needles  into  Zeus's  thigh  " — 
his  meros.  After  this  sophism  Tiresias  ends  his  justification 
of  the  ways  of  Dionysus  by  telling  Pentheus  of  the  god's 
miraculous  power.  "  Bacchus,"  he  says,  "  is  a  prophet  and  a 
warrior.  The  radiant  peaks  of  lofty  Parnassus  are  redolent 
of  the  god  by  day  and  night — 

Thou  shalt  descry  him  still :  on  Delphi's  rocks 

He  bounds  torch-dancing  o'er  their  twin-peak'd  alps, 

Flinging  and  whirling  the  leafy  thyrsus-wand." 

Another  phase  of  Tiresias'  defence  of  the  god  is  an  answer 
to  the  king's  wild  accusation  taxing  the  Bacchanals  with 
wantonness.  This  passage  is  worth  remembering,  because, 

1  For  further  discussion  of  this  curious  defence  of  the  faith,  see  Appendix 
III.  at  the  end  of  this  chapter. 


1 50  DIONYS US  AT  A THENS  i v 

taken  with  a  passage  from  the  messenger's  speech,1  it  con 
tains  our  poet's  pious  understanding  of  the  revels  of  Bacchan 
alian  women,  and  because  Milton  admired  it,  and  expanded  it 
in  a  well-known  passage  of  his  Comus?  It  runs  as  follows:3 

1  The  other  passage  is  one  of  the  most  wonderful  in  the  play,  vv.  677- 
688  : 

Late  as  to  pasture  forth  I  led  my  kine.   .   .    . 
While  gleaming  sunrise  sped  its  warmth  to  earth, 
I  saw  three  bands  of  women-revellers  : 
The  first  Autonoe  ruled,  the  second  band 
Thy  mother  Agave.  Ino  led  the  third. 
Lapped  all  in  slumber  lay  their  limbs  relaxed, 
Some  couched  on  heaped-up  twigs  of  silver  fir, 
Some  pillowed  on  oak-leaves,  their  heads  low  laid 
Reclining  where  they  might,  yet  as  they  should  ; 
Not  right  thy  word,  that,  overcome  with  wine 
And  with  the  sounding  flute,  they  left  their  lords 
To  hunt  for  Cypris  through  the  wilderness. 

2  Directly  inspired  by  vv.  314-318  which  follow  in  the  text,  indirectly 
by  the  ones  quoted  in  the  last  note,   and  by  the  wildwood  spirit  of  the 
Bacchanals  which  he  has  woven  in  a  wonderfully  original  fashion  into  the 
whole  of  his  masque,   are  Milton's  justly  celebrated   lines  in  praise  of 
chastity,  Comus,  418-475.     The  process  of  picturesque  expansion  to  which 
the  most  classical  of  English  poets  has  thus  subjected  the  most  romantic  of 
the  Greek  classics  has  its  parallel  in  Goethe's  expansion  of  Iph.  Taur. 
1401,  1402  : 

A  sister's  love  thou  feelest,  goddess,  too  ; 
I  yielded  but  to  that,  I  love  my  kin, 

into  the  following,  where  the  terseness  of  the  original  is  sadly  lacking  : 
Du  liebst,  Diane,  deinen  holden  Bruder 
Vor  allem  was  dir  Erd'  und  Himmel  bietet, 
Und  wendest  dein  jungfraulich  Angesicht 
Nach  seinem  ew'gen  Lichte  sehnend  still. 
O  lass  den  einz'gen  Spatgefundnen  mir 
Nicht  in  der  Finsterniss  des  Wahnsinns  rasen. 

It  is  curious  to  find  Milton  and  Goethe  playing  so  decidedly  the  part  of 
romanticists  as  compared  with  Euripides  ;  it  is  equally  curious  to  find 
in  these  instances  so  complete  an  exemplification  of  Mr.  Sidney  Colvin's 
definition  of  classical  and  romantic  writing  :  "in  classical  writing  every 
idea  is  called  up  to  the  mind  as  nakedly  as  possible,  and  at  the  same  time 
as  distinctly  ;  it  is  exhibited  in  white  light  and  left  to  produce  its  effect  by 
its  own  unaided  power.  In  romantic  writing,  on  the  other  hand,  all 
objects  are  exhibited  as  it  were  through  a  coloured  and  iridescent  atmo 
sphere."  Preface  to  Selections  from  Landor.  3  Vv.  314-318. 


iv  PENTHEUS  IS  DEAF  TO  REASON  151 

Not  Dionysus'  strength,  when  Cypris  calls, 

Shall  make  a  woman  chaste.      Inborn  and  bred, 

Bone  of  her  bone,  is  thorough  chastity 

Where  she  is  chaste.      'Tis  worth  our  weighing  well : 

She  that  is  chaste  may  not  corrupted  be 

For  all  her  Bacchanalian  revellings. 

Finally,  despairing  utterly  of  converting  so  blatant  a 
sinner  as  Pentheus,  the  prophet  shows  a  sad  foreknowledge 
in  his  closing  words  : 

Thou  art  crazed  to  death,  nor  hast  thou  drugs, 
Nor  findest  none  to  cure  thee,  drugged  with  folly ! 

Now  Cadmus  seconds  his  companion's  urgent  reasonings 
and  beseeches  Pentheus  not  to  persecute  Dionysus,  not 
to  neglect  the  mountain  revels,  but,  by  leaving  Thebes 
for  Cithaeron,  to  stay  at  home  with  righteousness.  After 
the  plea  of  wisdom  has  failed,  the  voice  of  pleading  love 
still  sounds :  "  My  son,"  says  Cadmus,  "  stay  at  home 
with  us;  cross  not  the  threshold  into  outer  lawlessness." 
Here  Cadmus  strives  by  an  ingenious  way  of  putting  his 
thought  to  humour  in  words  the  dangerous  frenzy  of 
Pentheus  while  he  really  contradicts  him.  Pentheus  is  all 
for  staying  at  home,  and  therefore  Cadmus  talks  of  going 
out  to  the  wilderness  as  the  only  real  way  left  open  for 
staying  at  home,  a  novel  way  of  presenting  the  gist  of  the 
adage  ubi  bene  ibi  patria^  But  Pentheus  is  obdurate,  and 
Cadmus  humours  him  still  more ;  granting  that  he  may  be 
right  in  scorning  Dionysus'  godhead,  there  are  considera 
tions  of  family  policy  which  ought  to  make  Pentheus  wink 
at  the  divine  pretensions  of  his  cousin — the  son  of  Semele, 
the  sister  of  his  mother  Agave. 

But  all  arguments  and  all  management  are  vain.     Pen- 

1  Or,  as  Menander  puts  it,  T£  yap  /caAws  irpavcrovTi  ira<ra.  yr]  Trarpts. 


1 52  DIONYSUS  AT  A THENS  i  v 

theus  can  no  longer  contain  himself.  He  fairly  foams 
with  rage  at  the  end  of  Cadmus'  expostulation.  Fiercely 
he  turns  away,  and  despatches  men  to  the  holy  places 
where  Tiresias  practises  augury.  These  must  be  entirely 
destroyed.  Against  the  cheating  stranger,  the  Dionysus- 
man,  he  sends  guards  saying  : 

That  girl-shaped  vagrant,  bringer  of  this  pest 
We  know  not  of,  the  man-shaped  Dionysus, 
The  worker  of  abominations — Stone  him  ! 

Thus  cries  frantic  Pentheus,  and  Tiresias  bodingly  mur 
murs  to  Cadmus  as  they  go, 

May  Pentheus  never  bring 
His  namesake  Grief,  O  Cadmus,  to  thy  home. 

This  first  act  of  the  play  now  closes  with  a  lyric  cry 
from  the  Maenad  worshippers  of  Dionysus. 

Holiness  with  her  pinions  of  gold  is  summoned  to 
earth  that  she  may  record  the  blasphemies  of  Pentheus. 
Dionysus  is  praised  as  the  god  of  garlands  and  feasts,  of 
dancing,  thyrsus  in  hand,  and  of  sweet  shrillings  from  flutes. 
His  gifts  are  wine  and  riddance  of  lingering  sadness,  with 
sleep  that  closes  great  joy.  "  Lawless  folly  ends  in  harm. 
Peace  and  soundness  of  mind  under  the  watchful  gods  bring 
concord  and  happiness.  But,"  the  song  and  its  singers  main 
tain,  "there  is  wisdom  and  wisdom.  Man's  wisdom  can 
bring  him  to  folly  : 

No  true  wisdom  comes  from  being  wise 

In  dizzy  thought  that  past  man's  level  flies." 

But  enough  of  calm  reasonings,  the  lyric  song  now 
breaks  away  from  contemplation,  and  revelry  is  its  theme. 
Revelry  and  some  place  not  curst  like  Thebes  with  Pentheus' 


iv  THE  SECOND  GREAT  ACT  153 

sin.  "  Oh  for  Cyprus,  Aphrodite's  isle  !  Lead  on  to  Pieria 
and  high  Olympus'  steep,  great  Bromius  !  "  The  course  of 
song  finally  grows  more  calm.  Through  jollity  of  feasts, 
through  prosperity  and  peace  that  breeds  stout  men,  the 
lyric  ode  goes  on  its  way,  showing  the  mercy  of  Dionysus 
and  his  loving-kindness. 


To  him  whose  fortunes  rise, 

To  him  whose  hopes  decline, 

He  gives  glad  gifts  alike  ;  to  none  denies 

The  painless  joys  of  wine. 


At  the  end  comes  a  prayer  which  is  partly  an  argument : 

Through  every  night  and  day 

To  live  through  life  the  happy  way 

From  froward  men  withdrawn  apart ! 

For  me  the  throngs  of  lowlier  men  ;  their  creed, 

Their  way  of  life,  be  graven  on  my  heart ! 

Thus  closes  the  first  great  act,  which  we  may  call  the 
Probation  of  Pentheus.  The  second  great  act  which  now 
begins  gives  an  account  of  the  Sin  of  Pentheus^  which  is  the 
Passion  of  Dionysus^  and  the  third  and  last  great  act  depicts 
the  Perdition  of  Pentheus. 

At  the  opening  of  the  second  act  a  guard  leads  in  the 
Dionysus-man,  the  Asian  reveller  whose  unresisting  ways  had 
won  his  captors'  hearts,  and  awed  them  into  recognition  of 
his  godhead.  This  prisoner  is  welcomed  by  Pentheus  with 
blasphemous  exultation,  although  a  warning  comes  with 
him.  The  guard  who  leads  Dionysus  prisoner  reports  the 
first  of  the  miracles  that  foreshadow  the  awful  judgment  of 
Dionysus.  The  women  from  Cithaeron,  in  whose  capture 
Pentheus  so  exulted,  have  been  freed  as  by  enchant 
ment  from  their  bonds.  Self- loosened,  their  shackles  fell 


1 54  DIONYSUS  AT  A THENS  iv 

away  and  invisible  hands  have  burst  their  prison  bars. 
Pentheus  hears  all  this  unmoved,  and  scorns  the  pressing 
appeal  of  the  rough  and  ready  guard  to  change  from  his 
wilful  impiety.  Turning  to  the  prisoner,  the  king  pays  an 
unwilling  tribute  to  his  loveliness  in  words  that  well  describe 
the  latter-day  Dionysus,  a  type  with  which  we  are  most 
familiar — 

Thy  frame  is  not  unshapely,  stranger, 
Not  wrestling  made  this  hair  of  thine  so  long. 
Its  gracious  flow  half  hides  thy  very  cheek  ; 
Thy  skin  is  white  to  help  thy  scoundrel  schemes. 
Not  sunburnt  thou,  but  pampered  in  the  shade. 

Then  begins  a  strange  duel  of  words  between  Pentheus  and 
the  god.  "What  is  thy  name?"  King  Pentheus  harshly 
asks.  "Not  hard  to  know,  for  I  was  born  in  flowery 
Tmolus,"  is  the  answer.  "What  are  these  new  rites  of 
thine  ?  "  the  king  then  asks.  At  the  answer,  "  They  are  of 
Dionysus,"  Pentheus  loses  all  self-control,  and  pours  out 
abuse  upon  Zeus  and  Semele  and  the  night  orgies  in  honour 
of  Dionysus.  "What  shape,"  the  king  again  asks,  "do 
these  precious  orgies  take?"  "That  may  not  be  told  to 
men  unholy ;  the  revellers  have  gifts  well  worth  the  know 
ing,  though  thou  shalt  not  hear."  Flurried  by  the  god's 
unwavering  tone  of  reprimand,  Pentheus  nevertheless  puts 
a  bold  face  upon  the  matter,  and,  after  sneering,  invites 
still  sterner  reproof  by  asking  how  the  god  looked  when 
he  showed  himself  to  the  faithful.  "  Even  as  he  willed," 
the  answer  comes,  "  not  shaped  by  my  command." 

After  this  the  king  crazily  dashes  out  with  wild  attacks 
upon  the  Bacchic  ritual,  but  at  each  onset  he  is  checked 
and  checkmated  by  stern  reprisals  from  the  inflexible  god 
whose  human  representative  stands  before  him.  Gradually 


iv  THE  SIN  OF  PENTHEUS  155 

Pentheus  loses  his  head  so  completely  that  he  has  to  be 
guided,  so  to  speak,  towards  his  own  iniquitous  purpose. 
His  sin  approaches  consummation  when,  in  answer  to 
Dionysus'  words  "  Tell  me  my  fate,  thy  threatened  terrors 
name,"  he  declares  that  he  will  shear  off  his  prisoner's  soft 
and  silken  locks. 

"My  hair  is  consecrate,  I  wear  it  for  the  god,"  the 
beauteous  stranger  answers.  In  spite  of  this  and  repeated 
warnings,  Pentheus  snatches  away  the  thyrsus-wand,  and  is 
for  putting  "  the  insolent  fellow  "  in  prison. 

At  this  point  the  possession  by  Dionysus  of  the  Dionysus- 
man  culminates  and  gradually  becomes  complete.  Till  now 
he  distinguished  between  himself  and  the  god,  but  now 
he  declares  confidently  that  the  god  will  free  him,  and  to 
silence  Pentheus'  sneers,  he  says  of  Dionysus — 

Now  present,  he  now  sees  what  I  endure  ! 

Soon   after  this  the  culmination  com^s,   and  he  is  com 
pletely  the  god  when  he  says — 

He  is  in  me  :  wicked  and  blind  thou  art. 

Pentheus,  worsted  in  argument,  is  about  to  carry  the  day 
by  an  appeal  to  brute  force.  He  has  Dionysus  bound  and 
prepared  for  imprisonment,  each  step  being  in  spite  of 
solemn  warning.  The  most  solemn  of  these  is  where  the 
god  says — 

Thy  life,  thy  name,  thy  sin  thou  knowest  not. 

At  this  Pentheus'  spirit  cowers,  and  all  in  a  tremor  he  cries 
in  a  dazed  way — 

Pentheus,  Echion's  son  and  Agave's  I  am. 


1 56  DIONYSUS  AT  A THENS  i v 

And  then  he  hears  with  terror  from  the  prisoner  whom  his 
men  are  leading  to  confinement— 

Thou  and  thy  name  are  meet  for  deep  disaster. 

"  Coop  the  fellow  in  the  stables,"  he  cries  in  fear  and  anger, 
"let  him  dance  there  with  dumb  beasts."  Speaking  of  sure 
requital  to  come  speedily,  the  disguised  Dionysus  disappears 
at  last  with  the  threatening  words  addressed  to  Pentheus — 

Though  thou  declares!  Dionysus  is  not, 
In  binding  me  thou  art  confining  him. 

Now  the  Bacchanals  sing  a  song  of  fear  and  woe.  This 
is  the  winter  of  their  discontent,  and  truly  this  darkest  point 
in  the  play  mirrors  the  sadness  and  the  longing  of  those 
mysterious  winter  festivals  on  Mount  Parnassus  and  in 
Attica  which  were  always  attached  to  the  worship  of 
Dionysus.  The  analogy  of  this  festival,  which  included 
rejoicing  for  the  new  and  mystic  birth  of  the  god  lamented 
so  lately,  accounts  for  one  theme  of  this  song,  a  glad  wel 
coming  of  the  birth  of  Dionysus. 

"Achelous'  daughter  Dirce  makes  Theban  lands  yield 
abundantly.  Dirce,  whose  waters  welcomed  the  new-born 
Dionysus  and  bathed  him  that  the  flames  from  his  father's 
bolt  might  leave  no  scar," — Dirce  is  now  unfriendly  to  the 
revellers  in  whom  dwells  the  fulness  of  Dionysus. 

Pentheus  and  his  sin  soon  engross  their  song.  He  is  a 
fierce-glaring  monster  fitly  spawned  by  earth  from  dragon's 
teeth.  They  close  with  a  prayer  to  Dionysus  for  help.  As 
it  proceeds  this  prayer  becomes  an  incantation  in  the  spirit 
of  rude  magic  charms  used  by  peasants  to  bring  forth  nature's 
power  and  ensure  full  crops. 

"  Dionysus,  dost  thou  leave  thy  prophets  here  to  strive 


iv          CAPTIVE  GOOD  ATTENDING  CAPTAIN  ILL       157 

in  vain?  With  brandishings  of  thy  most  golden  thyrsus 
come  down  Olympus.  Where  in  Nysa's  wilds  or  on  the 
heights  Corycian  art  thou,  Dionysus  ?  Art  thou  near  the 
Thracian  realm  of  singing,  whose  forests  followed  Orpheus, 
marshalled  with  all  wild  beasts  in  his  wake  ?  Lo  !  he  comes 
over  Axius.  He  comes  with  whirling  Maenad  train  across 
the  Lydias,  father  of  plenty  in  the  Thracian  land  of  good 
horses." 

This  is  the  frantic  prayer  for  help  of  "captive  good 
attending  captain  ill,"  transmuted  in  the  Bacchic  fires  of  faith 
so  as  to  become  an  invocation  which  reshapes  itself  at  the 
close  into  a  song  of  thanksgiving  and  praise.  It  ends  with 
the  strains  of  Bacchus'  triumphal  march  in  order  to  usher 
in  the  Lord  of  Vengeance  whose  coming  with  requital  is  at 
hand.1 

"  Make  way,"  so  runs  the  burden  of  this  song  of  the 
judgment  of  Dionysus, 

Let  justice  be  shown  and  be  dread, 
For  justice  make  way  and  her  sword  ; 


1  I  cannot  do  better  than  quote  from  Professor  Tyrrell' s  Introduction  to 
the  Bacchanals,  where  I  have  found,  just  at  the  moment  of  going  to  prf  3,  a 
presentation  of  the  deep  religious  significance  of  the  whole  play  from  which 
my  too  belated  knowledge  of  his  admirable  work  has  prevented  me  from 
profiting  sufficiently.  Of  the  various  choral  odes  Professor  Tyrrell  most 
truly  says  :  ' '  The  parados  and  the  four  stasima  not  only  are  suitable  in  a 
degree  rare  in  Euripides  to  the  parts  of  the  action  at  which  they  are 
respectively  introduced,  but  form  a  whole  in  themselves  and  an  elaborate 
picture  of  the  Bacchic  cult.  The  parodos  (vv.  64-169)  describes  the  out 
ward  form  and  ritual  of  the  Bacchic  worship  ;  the  first  stasimon  (vv. 
370-431)  describes  its  sacred  joys,  the  second  stasimon  (vv.  519-575) 
refers  to  the  birth  of  the  god,  the  third  (vv.  862-911)  breaks  into  tumultuous 
enthusiasm  and  anticipations  of  triumph,  and  the  fourth  (vv.  977-1024) 
urges  on  the  '  hounds  of  frenzy '  against  the  violator  of  the  rites  of  the 
Maenads."  Professor  Tyrrell  refers  to  Pfander's  Die  Tragik  des  Euri 
pides,  Bern,  1869  ;  and  also  to  Scheme's  similarly  striking  account  of  the 
choral  odes  of  the  Iphigenia  at  Aulis,  the  very  last  play  written  by 
Euripides. 


158  DIONYSUS  AT  A THENS  iv 

To  his  throat  shall  she  set  it  and  smite  off  the  head 
Of  Echion's  earth-spawned  offspring  untoward, 
The  godless,  the  lawless,  the  froward.1 

With  this  song  ends  the  Passion  of  Dionysus.  Now  comes 
the  third  and  last  great  act,  a  veritable  Vision  of  Judgment ', 
which  treats  of  the  Perdition  of  Pentheus.  It  begins  with 
the  first  revelation  which  this  play  contains  of  Dionysus  in 
his  terrific  might.  The  god  comes  to  the  rescue  of  his 
suppliants,  and  to  give  judgment  against  the  evildoer. 
Forth  from  the  earthquake,  which  is  Dionysus'  might,2  steps 
smiling  and  unharmed  the  prisoner  of  a  moment  since. 
The  veil  of  the  palace  of  Pentheus  has  been  rent,  but  Pen 
theus,  more  and  more  dazed  and  crazed,  is  still  unabashed. 
He  is  for  further  harm  to  Dionysus,  but  ere  he  attempts  it 
he  listens  to  a  messenger  from  Cithaeron.  There  the  might 
of  the  god  has  shown  itself  in  a  judgment  as  it  were  of  fire. 
Crashing  down  the  hills  and  spreading  terror  and  ruin  far 
and  wide  the  lava-stream  of  Maenad  women  has  proclaimed 
their  lord's  resistless  might.  This  appears  next  in  the  com 
ing  of  madness  which  enters  the  guilty  soul  of  Pentheus. 
Crazed  by  his  own  rising  frenzy  and  mocked  by  the  dis 
guised  Dionysus,  who  leads  him  towards  death,  Pentheus 
goes  to  spy  out  the  Maenads  at  their  revels  on  Cithaeron. 
He  is  himself  madly  accoutred  as  a  Bacchanal. 

Off  there  on  Cithaeron  comes  the  final  execution  of  the 
will  of  Dionysus  upon  the  luckless  king.  Perched  high  on 
a  pine  tree,  by  a  mad  freak  of  his  own  which  the  disguised 

1  Vv.  1010-1013.     This  is  the  last  song  of  the  chorus. 

2  The  chief  authority  for  this  statement  is  in  this  and  other  passages  of 
the  Bacchanals,  and  in  the  identification  of  Dionysus  with  fire.     Cf.  also 
a  fragment— relating  to  Bacchic  orgies— from  the   Edani  of   Aeschylus, 
Dind.    55  :     \f/a\/j.6s    §'    dAaXdfci   ravp6(f)doyyoL    5'    virofj.VKu>i>Tai    iroBlv 

fopepoi    [unoi,    rviravov  5'    ei/cajv    &<r6'     viroyaiov 
/Sapurap/S^s. 


iv  THE  PERDITION  OF  PENTHEUS  159 

Dionysus  made  haste  to  gratify  before  disappearing  in  a 
pillar  of  fire,  Pentheus  is  spied  by  the  Maenads.  They 
whirl  the  tree  to  the  ground  and,  mad  themselves  with 
Dionysus,  they  look  on  the  king  as  a  mountain -ranging 
lion.1  Thus  these  frenzied  women  surround  him,  now 
darting  and  dancing  light  as  flickering  flames,  and  now  in 
mass  resistless  whirling  like  a  torrent  down  the  mountain  side. 
Here  is  the  devastating  flood  that  comes  with  an  earthquake 
springing  upward  from  unseen  sources  terrific  in  its  might. 

1  In  this  wild  scene  Euripides  glorifies  and  does  a  sort  of  poetical 
justice  to  country  customs  which  still  subsist  and  have  often  been  fraught 
with  the  shedding  of  human  blood.  See  Mr.  Frazer's  interesting  chapter 
on  "  Killing  the  Tree  Spirit"  (pp.  240-253  of  vol.  i.  in  his  Golden  Bough] 
where  he  gives  —  following  Mannhardt  and  others  —  an  account  of  the 
Lower  Bavarian  custom  of  various  mock  executions  at  Whitsuntide.  The 
Pfingstl  thus  executed  is  like  Pentheus  here  a  king  of  the  wood,  and 
' '  his  defeat  and  death  at  the  hands  of  another  proved  that  his  strength 
was  beginning  to  fail,  and  that  it  was  time  his  divine  life  should  be  lodged 
in  a  less  dilapidated  tabernacle. "  See  also  F.  A.  Voigt  (article  ' '  Dionysus" 
in  Roscher's  Mythological  Lexicon,  p.  1061).  There  was  a  legend 
(Pausanias,  II.  ii.  5  and  6)  concerning  two  most  sacred  Bacchic  images  at 
Corinth,  one  of  Ai/crios  and  the  other  of  Bd/c%etos.  They  were  made,  not 
of  the  wood  of  the  true  Cross,  but  of  the  wood  of  the  very  fir  tree  upon 
which  Pentheus  was  placed  by  Dionysus,  as  we  read  in  this  play.  More 
over  a  command  had  come  from  Delphi  to  worship  the  tree  as  the  very 
god  himself — TO  dtvdpov  foa  T$>  0e<£  crtfieLV.  From  this  Voigt  rightly  con 
cludes  that  the  Maenads  must  have  worshipped  the  tree  before  felling  it. 
There  was  a  still  more  primitive  and  Asiatic  custom  in  the  Thraco- Phrygian 
home  of  Dionysus.  There  they  felled  a  fir  tree  once  every  year,  and 
carried  it  in  solemn  procession  to  its  home — the  god's  temple.  Thus  in 
the  Corinthian  tale  preserved  by  Pausanias  we  have  record  of  the  ancient 
transfer  of  worship  and  allegiance  from  the  tree,  which  was  the  older  incar 
nation  of  Dionysus  Dendrites,  to  the  graven  image  which  eventually  attached 
to  itself  all  worship.  Strabo,  a  wonderfully  acute  observer  of  the  broadest 
aspects  of  Greek  religion,  groups  together  (X.  p.  468)  Dionysus,  Apollo, 
Hecate  (Proserpina?),  the  Muses,  and  Demeter,  and  ascribes  to  their 
worship  TO  bpyiavTLKbv  TTO.V  KO!  TO  ^aK^KOv  /cat  TO  %opiAc6»',  Kal  TO  irepi 
ras  TeXeTas  /j.v<rTLK6v.  Then  he  adds  what  is  of  especial  interest  in  con 
junction  with  Pentheus  and  the  fir  tree  :  devdpofopiai  TC  Kal  \opeiai  Kal 
reXeTctl  Koival  TUV  deuv  efoi  TOITTWV.  After  this  he  seems  to  grow  con 
fused  and  to  take  Apollo  and  the  Muses  out  of  this  group,  where  of  course, 
viewed  under  certain  aspects,  they  are  not  at  home,  though  by  right  of 
descent  and  through  ties  of  early  ritual  they  are  indissolubly  bound  to  it. 


160  DIONYSUS  AT  ATHENS  iv 

The  tall  fir  tree  of  Pentheus  sways  and  yields,  it  crashes 
to  the  ground  overborne  by  these  flames  and  floods  of 
Dionysus.  Pentheus  himself,  when  once  he  touches  earth, 
is  seized  by  the  women.  Flames  they  are  no  longer, 
and  they  are  not  floods,  but  of  a  sudden  they  become  the 
many-handed  earthquake  which  has  shaken  Thebes,  and  so 
they  rend  and  hideously  mangle  Pentheus'  limbs.  His 
head  is  plucked  from  his  body,  his  feet  are  wrenched  from 
his  legs,  his  thighs  are  forced  from  their  sockets,  and  his 
sides  are  flayed  and  lacerated  foully.  Tossed  into  the  air, 
his  limbs  deface  the  leafage  of  the  trees,  and  his  head  is 
spiked  on  a  spear  to  be  carried  off  in  triumph  by  his  mother 
Agave.  The  earth  was  his  father's  mother,  and  Agave  his 
own  mother  with  her  three  Maenad  bands  impersonates  the 
mysterious  and  wrathful  powers  of  nether  earth. 

Here,  perhaps,  if  the  line  is  to  be  drawn  at  all,  comes  the 
division  at  which  the  Perdition  of  Pentheus  ends,  and  the 
fifth  part  of  the  play  begins. 

Filled  with  the  spirit  of  fierce  Dionysus,  the  wild  hunts 
man,  Agave  cries  aloud,  still  madly  thinking  that  she  bears 
in  triumph  a  lion's  head,  "Bacchus  led  on  in  the  chase 
wisely,  for  wise  he  is.  He  made  the  Maenads  dart  and 
hunt  this  quarry  to  its  lair."  With  mystical  significance  the 
chorus  of  Bacchanals  from  Asia  make  answer,  "Yea,  for 
our  king  is  a  huntsman."  Dionysus,  plainly,  is  a  jealous 
god,  visiting  the  iniquities  of  Pentheus  on  his  mother  Agave, 
and  his  power  is  so  strong  that  those  whom  it  has  once 
possessed  cannot  lightly  find  returning  sense.  So  it  is 
that  Agave,  glorying  in  the  slaughter  of  a  lion— the  unwitting 
murder  by  her  devoted  hands  of  her  own  and  only  son,  grows 
impatient  under  her  father  Cadmus'  vain  efforts  to  restore 
her  mind,  and  harps  upon  grievances  against  her  son. 


iv  CADMUS  CURES  AGAVE 'S  MADNESS  161 

"How  age  turns  men  to  crabbedness,"  she  cries.  "Would 
that,  like  his  mother,  my  son  were  a  lucky  hunter;  but  heaven- 
fighting  he  is  fit  for,  and  good  for  nothing  else."  Then  she 
turns  again  to  Cadmus,  saying,  "Father!  rebuke  him  roundly. 
Bring  him  here  to  me."  Her  mind  is  bent  on  having  the 
head,  her  glorious  hunting-prize,  fastened  trophy-fashion  on 
the  palace  front.  She  waits  for  Pentheus  to  do  it. 

Wondrously  true,  wondrously  sad  is  the  moment  when 
Agave  ceases  to  be  the  god,  and  comes  back  to  herself  at 
last. 

Cadmus  has  waited  for  a  pause  in  his  daughter's  ravings, 
and  when  it  comes  he  suddenly  says,  a  propos  of  nothing, 
"  Look  up  and  scan  the  sky." l  Surprise  seems  to  still  her 
frenzy,  and  she  asks,  "Why  bid  me  look  at  the  sky?"  Dis 
regarding  this  question,  he  asks  if  the  sky  seems  altered. 
Now  Agave  finds  that  she  sees  it  more  clearly.  "  Its  light 
is  brighter,  things  seem  to  stand  more  firmly  in  the  world." 
"Art  thou  restored  to  sense?"  finally  asks  Cadmus,  and 
his  daughter  answers — 

I  know  thy  meaning  not,  and  yet  somehow 
Sense  comes,  and  from  my  former  mind  I  change. 

Skilfully  Cadmus  pursues  his  advantage,  and  awakens  the 
slumbering  memories  of  calmer  days  in  Agave's  mind. 
Finally  she  turns  questioner,  and  presses  him  with  inquiries 
about  her  own  mad  doings.  With  a  shriek  of  despair  she 
finally  recognises  the  head  of  her  own  son  in  her  own 
hands,  and  sees  at  last  that  she  has  murdered  him  in  Bacchic 
frenzy,  and  cries,  "We're  Dionysus -slain,  I  see  it  now." 

1  Those  who  have  experience  in  cases  of  mental  aberration  must 
admire  the  truth  to  fact  in  this  representation  of  a  recovery  of  sanity  under 
wise  guidance. 

M 


1  62  DIONYSUS  AT  ATHENS  iv 

Cadmus,  speaking  for  the  god,1  makes  answer,  "Outrage 
breeds  outrage,  you  denied  his  godhead."2 

The  winding-up  of  the  Bacchanals  in  the  last  one  hundred 
lines  has  little  further  bearing  on  the  divinity  of  Dionysus 
and  needs  no  comment  here.  It  is  sufficient  to  have  had  a 
glimpse  in  this  sublime  play3  of  the  god  as  he  was  conceived 
by  the  Athenians,  who  worshipped  him  in  the  fulness  of  his 
Thracian  and  Old  Attic  godhead.  From  this  ruder  and 
earlier  conception  much  that  was  not  divine  but  cruel  and 
barbarous  had  been  separated,  but  enough  of  proto-Thracian 
harshness  and  pitilessness,  as  of  the  untamed  powers  of 
nature,  still  attaches  to  him  even  in  the  Bacchanals^  to  make 
it  once  more  plain  that  not  he,  but  rather  Apollo  his  brother, 
must  always  represent  the  most  purely  Hellenic  ideal  of  a 
righteous  and  beneficent  god. 

1  In  this  wonderful  scene  Cadmus  represents  the  god  —  he  incarnates 
Dionysus  the  saviour  from  Dionysiac  madness.  It  is  significant  that  this 
most  merciful  aspect  of  Dionysus  is  the  last  one  presented  in  the  play. 
After  this  Dionysus  appears  as  the  deus  ex  machina,  and  formally  justifies 
his  dealings  with  Thebes  and  the  house  of  Cadmus  by  appeal  to  Zeus.  See 
the  Appendix  (II.)  to  the  foregoing  lecture  for  an  account  of  this  Dionysus 
Eleuthereus.  Cf.  also  for  a  very  complete  presentation  of  the  cheerful  and 
beneficent  aspects  of  the  Bacchic  godhead,  the  Orphic  Hymn  to  Dionysus 
Lysios,  No.  50,  in  Hermann's  Collection  (Leipzig,  1883).  The  invoca 
tion  is  : 

K\vdi,  fj.aKap,  Aibs  vV,  ewiX^vie  Ed/t^e,   di/JLrjTup, 

\v<rte  daifj.ov. 


2  V.  1298.     After  the  speech  of  Cadmus  immediately  following,  at  v. 
1325,  Professor  Tyrrell  says  a  modern  play  would  have  ended. 

3  For  the  presence  of  the  sublime  in  Euripides,  denied  by  some,  we  have 
the  authority  of  Goethe,  who  knew  well  and  well  appreciated  the  Bacchanals. 
See  his  translation  of  the  great  scene  between  Agave  and  Cadmus  written 
in  1826  (vol.  xxix.  of  Cotta's  1868  edition,  pp.  34  and  ff.)     In  his  con 
versation  with  Eckermann  of  the  i8th  February  1831,  he  said:   "  Alle, 
die  dem   Euripides  das  Erhabene  abgesprochen,   waren  arme  Haringe, 
und  einer  solchen  Erhebung  nicht  fahig  ;    oder  sie  waren  unverschamte 
Charlatane,  die  durch  Anmasslichkeit  in  den  Augen  einer  schwachen  Welt 
mehr  aus  sich  machen  wollten  und  wirklich  machten  als  sie  waren." 


APPENDIX   III 

SECOND    BIRTH    OF    DIONYSUS— HIS    EASTERN 
AFFINITIES 

IN  Euboea,  as  in  other  Aegean  Islands  such  as  Naxos 
and  Icaria,1  the  legends  of  Dionysus  became  entangled 
with  a  mass  of  tradition  which  belongs  to  the  far  Eastern 
world.  With  this  is  closely  connected  a  record  of  pre 
historic  changes  in  the  tribe  and  family  which  survives 
in  the  curious  story  of  the  second  and  only  real  birth 
of  the  god  from  Zeus's  thigh.  This  complex  snarl 
of  variegated  tradition  is  perhaps  most  plainly  recorded 
in  the  first  of  two  fragments  of  a  hymn  to  Dionysus.2 
"Some  there  are  who  say  'twas  on  Draconus,3  'twas 
in  Icarus,  some  say ;  and  some  say  in  Naxos,  son  of 
Zeus  who  wert  sewed  in  with  needles,  some  say  'twas 
on  the  banks  of  Alpheius,  the  deep  eddying  river,  that 
Semele  went  with  thee  and  brought  thee  to  birth  for 
Zeus  who  rejoices  in  thunders ;  others  there  are,  my 
king,  who  relate  that  at  Thebes  thou  earnest  to  birth, 
— all  of  them  speaking  falsehoods.  For  verily  the  father 
of  men  and  of  gods  brought  thee  to  birth  where  men  were 
far  away,  and  in  secret  from  white-armed  Hera.  A  certain 
spot  there  is  called  Nysa,  a  lofty  mountain  covered  with 

1  Not  the  Attic  deme  Icaria.  2  Homeric  Hymns,  xxxiv. 

3  A  promontory  on  the  Aegean  island  of  Icaria,  or  Nicaria. 


1 64  APPENDIX  III 

blossoming  forests,  in  the  uttermost  parts  of  Phoenicia  it 
lies  close  to  the  streams  of  Aegyptus."  l 

Two  things  are  here  attested,  for  only  one  of  which — 
his  second  birth  —  we  are  prepared  by  the  ordinarily 
accepted  accounts  of  Dionysus.  For  the  birth  of  Dionysus 
in  the  far  East  nothing  in  the  Icarian  legend,  and  little  in 
the  Theban  legend,  save  the  importance  of  Cadmus  of 
Sidon,2  and  certain  Thraco- Phrygian  features  of  the  tale, 
have  prepared  us.  Moreover,  the  second  birth  of  Dionysus 
as  it  stands  in  the  purely  Grecian  legends  is  not  only  a 
most  mysterious  but  a  seemingly  grotesque  episode.  The 
idea  of  taking  this  episode  out  of  the  more  or  less  purely 
Greek  story  of  the  god,  and  of  connecting  it  with  his  fabled 
birth  in  the  far  East,  is  certainly  suggested  by  the  Homeric 
fragment  above  quoted,  but  it  had  never  occurred  to  me 
until  I  received  some  very  valuable  information  in  answer 
to  a  request  which  I  addressed  to  my  friend  Mr.  Clinton 
Dawkins.  I  had  asked  him  to  make  inquiry  about  the 
habitat  of  the  cinnamon  tree,  wishing,  if  possible,  to  deter 
mine  by  that  means  what  sort  of  place  Nysa,  Dionysus' 
birth-place,  was  thought  to  be,  when  it  was  identified  by 
Herodotus  with  a  place  where  the  cinnamon  grew.  I 
wished  to  know  whether  cinnamon  trees  grew  in  dark  low- 
lying  meadow-lands  or  on  rugged  mountain  sides.  The 
information  so  kindly  provided  by  my  friend  came  from  no 
less  eminent  and  learned  a  source  than  Sir  George  Bird- 
wood,  K.C.I.E.  With  his  kind  permission  I  reproduce  it 
here,  since  it  gave  the  right  clue  and  has  helped  me  towards 
a  very  fair  solution  of  the  difficulties  concerned. 

"Herodotus  (iii.  in)  says — 'Some  relate  that  it 
[/avi/a/Mo/Aov]  comes  from  the  country  in  which  Dionysus 
was  brought  up';  and  (iii.  97) — 'The  Aethiopians  border 
ing  upon  Egypt  .  .  .  and  who  dwelt  about  the  sacred  city 
of  Nysa,  have  festivals  in  honour  of  Dionysus';  and  again 

1  This  fragment  was  found  in  the  same  Moscow  MS.  where  the  Hymn 
to  Demeter  first  came  to  light.      It  is  also  known  through  Diodorus. 

2  Herodotus,  ii.  48  and  49. 


SECOND  BIRTH  OF  DIONYSUS  165 

(ii.  146)  he  says — 'But  Dionysus  was  no  sooner  born  than 
he  was  sewn  up  in  the  thigh  of  Zeus,  and  carried  off  to 
Nysa,  above  Egypt,  in  Aethiopia.'  Now  there  are  several 
Nysas.  Herodotus  meant  Nysa  in  Aethiopia,  that  is  the 
Troglodytic  country  beyond  the  Soudan ;  for  the  Soumali 
country  is  the  cinnamon  country.  On  the  other  hand  the 
story  of  Dionysus,  'the  Assyrian  stranger/  is,  inter  alia,  a 
myth  of  the  development  of  Phoenician  commerce,  of  which 
wine  was  everywhere  throughout  the  Eastern  Mediterranean 
(Levant)  the  staple ;  and  the  Greek  myths  associating  the 
wine  god  with  Mount  Meroe  l  in  Aethiopia  probably  arose 
from  the  fact  that  in  the  original  Phoenician  myth  he  was 
not  a  '  child  of  the  womb '  but  '  of  the  thigh  '  (/xr/po's).  That 
is  to  say,  these  myths  probably  arose  at  the  time  when 
kinship  among  men  had  ceased  to  be  traced  through  their 
mothers  and  had  already  begun  to  be  traced  through  their 
fathers.  Similarly  the  association  of  the  wine  god  with 
'  Nysa  above  Egypt '  was  presumably  due  to  there  having 
been  a  Nysa  near  Meroe,  and  to  his  Greek  name  being 
Atdwcros ;  this  Greek  form  of  his  name  being  probably  a 
folk  corruption  of  his  Phoenician  name,  which  would  almost 
certainly  end  in  nisi  '  man.' 

"  Of  course  the  cult  of  the  vine  and  the  manufacture  of 
wine  did  not  arise  in  Aethiopia  but  on  the  slopes  of  the 
Indo-Caucasus,  and  hence  Mount  Meroe  [Meru]  and  the 
Indo-Caucasian  Nysa  have  been  identified  as  the  seats  of 
the  education  of  the  young  Aiovwos." 

It  is  evident  most  abundantly  from  the  Homeric  Hymn 
and  from  Herodotus  that  the  notion  of  Dionysus'  second 
birth  was  often  connected  with  thigh  mountain,  Mount  Meroe> 
and  it  is  equally  plain  that  this  connection  might  involve 
rejecting  more  or  less  consciously — according  as  the  matter 
was  more  or  less  reasoned  out — the  current  reports  of  his 
birth  at  Thebes,  or  Naxos,  or  elsewhere  in  Greece,  or  Thrace, 

1  Cf.  Eustathius  (fol.  p.  310,  1.  6)  on  Iliad  ii.  637  :  8pos  de  TL  ' 
M7?p6s  €K\r)dr),  kiovvay  ava.Keifj.evov,  80ev  MT/por/xi^rjs  /j.€/u.udei>Tai, 
o 


1 66  APPENDIX  III 

or  Phrygia.  Perhaps  the  whole  story  identifying  Nysa — 
that  elusive  place,  which  never  stays  quite  where  you  put  it, 
but  has  a  trick  of  moving  far  East  if  you  seek  it  in  Greece, 
and  of  lurking  in  Thrace  if  you  seek  it  in  Egypt  or  Arabia — 
with  Mount  Meroe  and  the  far  East  may  have  been  called 
into  being  by  the  epithet  of  Dionysus  ^porpa^r^,  nursling 
of  the  thigh,  which  goes  hand  in  hand  with  that  other  one 
€ipa<f)ia>TT]<;,  sewed  in  with  needles.  Perhaps  some  mute 
inglorious  Euhemerus  could  settle  the  difficulty  quite  com 
fortably  by  saying  that  the  epithet  should  be  translated 
nursling  of  Mount  Meroe,  and  then  he  could  say  that  the 
other  epithet  was  a  mistake  produced  by  a  stupid  tale 
regarding  the  thigh  of  Zeus.  This  is,  however,  a  too  con 
venient  way  of  meeting  the  difficulty,  nor  is  that  adopted 
by  Euripides  in  the  Bacchanals  in  the  least  more  satisfactory, 
although  it  was  made  with  a  certain  Jesuitical  sincerity,  and 
in  its  day  probably  satisfied  many  religious  minds  in  difficulty 
about  the  patent  incongruity  of  the  tale.  For  when  Euri 
pides  wrote  the  Bacchanals  the  best  intellects  of  the  time, 
and  he  was  among  them,  still  clung  to  a  belief  in  the  efficacy 
of  a  subtle  analysis  of  words.1 

Tiresias,  a  holy  man,  utters  the  apology,  explanation,  or — 
if  you  chose  to  call  it  so — the  sophism2  by  which  Euripides 

1  Cf.  Mr.  Tyrrell's  admirable  note  (p.  xxx.  of  the  Introduction  to  his 
Bacchae] :  "The  reason  of  this  etymologising  " — he  speaks  of  that  at  v.  520 
of  the  play — "  is  to  be  found,  as  Schwalbe  well  observes,  in  the  deep  con 
viction  with  which  Greek  antiquity  was  imbued,  that  between  the  word  and 
the  thing  denoted  by  it  there  was  some  secret  bond  or  hidden  affinity." 

2  For  an  equally  curious  sophism  which  Sophocles  puts  into  the  mouth 
of  Antigone,  see  his  Antigone,  904-915.       Both  of  these   passages  are 
alien  to  modern  taste,  and  are  prompted  by  the  rhetorical  training  enjoyed 
by  Sophocles  and  by  Euripides.      Goethe,  Conversation  with  Eckermann  of 
28th   March   1827,   says   he   would   give   a   great  deal   if  a    "  tiichtiger 
Philologe"  would  prove  that  the  passage  from  the  Antigone  was  spuri 
ous.     The  chief  reason  why  this  desire  of  his  has  never  (pace  Jacob)  been 
gratified    is    found   in  Aristotle's  citation   of  lines  911   and   912,   and  in 
Herodotus,   who   has   put   the   same   rhetorical   commonplaces    into   an 
episode  of  Persian  history  (iii.  119). — Since  writing  the  above,  I  have  read 
Professor  Jebb's  Appendix,  where  he  rejects  lines  904-920  as  interpolated 
by  lophon  or  as  due  to  the  actors.     I  am  not,  however,  inclined  to  take 
this  view. 


SECOND  BIRTH  OF  DIONYSUS  167 

shames  the  blasphemies  of  Pentheus  and  other  scoffers  at 
the  second  birth  of  Dionysus.1 

Him  dost  thou  scorn,  and  mock  to  hear  the  tale, 

How  in  Zeus'  thigh  he  was  sewn  up.      Give  ear 

And  learn  of  me  that  this  is  as  it  should  be. 

When  Zeus  from  flames  and  lightnings  plucked  him  out, 

And  bore  Olympus-ward  a  god  unborn, 

Then  Hera  sought  to  fling  him  down  from  Heaven, 

Zeus  foiled  her  plot  with  counter-plots  divine. 

1  Bacch.  286-297.  Dindorf  rejects  these  lines  because  of  their  "  dictio 
inepta  confusa  omninoque  non  Euripidea,"  which  amounts  to  saying,  Euri 
pides  did  not  write  them  because  they  are  not  by  Euripides.  This  seems  to  be 
Wecklein's  view.  Professor  Tyrrell  makes  out  a  better  case  :  "It  seems 
hardly  too  much  to  say  that  vv.  286-297  must  be  interpolated,  because 
they  explain  away  a  story  taken  as  literally  true  by  the  chorus,  vv.  520- 
530,  and  also  in  the  second  strophe  of  their  entrance  song."  Theirs, 
he  maintains,  was  the  orthodox  version  opposed  by  Euripides  to  the  sceptical 
one  given  by  Pentheus.  ' '  It  can  hardly  be  maintained,  therefore,  that  Euri 
pides  would  have  assigned  to  Tiresias  (who,  as  well  as  the  chorus,  is  all 
along  the  exponent  of  the  views  of  the  believers)  a  theory  explaining  away 
the  myth  in  which  the  chorus  express  their  belief."  Here  Professor  Tyrrell 
seems  to  me  to  apply  essentially  modern  standards  of  faith  and  orthodoxy  to 
the  side  of  Greek  religion  which  is  most  absolutely  turned  away  from  them. 
To  me,  and  I  suppose  to  many,  such  a  divergence  is  far  from  inconceivable 
between  Tiresias  and  the  chorus, — both  of  them  equally  authoritative,  toth 
of  them  equally  orthodox,  if  such  an  alien  word  may  be  used  where  it  has  no 
real  application.  It  would  be  indeed  marvellous  if  the  god  of  transforma 
tions,  illusions,  and  contradictions  did  not  often  inspire  his  votaries  to 
contradict  each  other.  No  one  phase  of  the  elusive  manifestations  of 
Dionysus,  and  no  one's  account  of  any  feature  in  his  story,  must  be  treated 
as  final.  It  must,  furthermore,  be  remembered  that  these  offending  verses 
can  be  taken  as  a  very  clever  answering  of  the  fool  according  to  his  folly, 
an  attempt  to  mediate  between  the  blasphemous  scepticism  of  Pentheus 
and  a  story  which  he  was  incapable  of  accepting  as  the  true  believers  did. 
Regarded  in  this  light  the  sophism  of  Tiresias  is  a  Jesuitical  concession 
made  for  the  salvation  of  Pentheus'  soul  as  a  last  and  desperate  move. 
Cadmus  follows  with  the  last  appeal  of  all,  which  is  characterised  by  the 
same  spirit.  He  allows  that  Dionysus  is  a  man.  These  concessions  form 
part  of  the  plan  which  shows  in  Pentheus  the  self-deluded  and  self-devoted 
victim  of  wanton  wickedness.  ' '  No  one  can  ever  convince  every  one  that 
this  passage  is  spurious,"  says  Professor  J ebb  of  Ant.  904-915.  Change 
spurious  to  genuine,  and  the  remark  applies  to  Bacch.  286-297.  Every  one 
can,  however,  be  convinced  that  both  passages,  if  spurious,  were  the 
earliest  of  interpolations.  Thus,  in  any  case,  Bacch.  286-297  retains  its 
religious  significance. 


168  APPENDIX  III 

A  piece  l  torn  off  from  earth's  encircling  ether 
He  framed  to  be  a  pledge  of  peace  2  with  Hera, 
With  Dionysus'  semblance  cheating  her. 
But  men  report  that  for  a  time  the  god 
Grew  in  Zeus'  loin,3  contriving  all  the  tale, 
Exchanging  terms  because  a  changeling  pledge 
To  Goddess  Hera  was  conveyed  4  by  Zeus. 

The  temper  in  which  all  these  difficulties,  so  far  as  it  may 
be  said  that  they  are  still  difficulties,  in  the  legend  of  Dionysus 
are  now  met  is  a  very  different  one  from  that  in  which  Euri 
pides  wrote  the  above.  As  for  the  Nysa  placed  in  the  far  East, 
and  Dionysus'  eastern  birth,  that  goes  to  prove  the  probable 
infusion  of  a  strong  Phoenician,  Egyptian,  or  Arabian  strain 
into  the  habit  of  Dionysus  as  known  among  the  Aegean 
islands  in  early  days.  Add  to  this  the  apparently  Phoeni 
cian  character  and  derivation  of  his  name,  and  the  whole 
setting  of  the  beautiful  Homeric  Hymn  wherein  we  read 
how  sea-robbers  tried  to  carry  off  the  god,  and  how  they 
were  punished  for  it.  Then  the  outlines  for  understanding 
Dionysus  as  "the  Assyrian  Stranger,"  and  for  interpreting 
certain  touches  in  his  story  as  "  a  myth  of  the  develop 
ment  of  Phoenician  commerce,"  are  complete. 

The  mystery  of  his  second  birth  remains  to  be  cleared 
up.  In  the  fragment  of  a  Homeric  Hymn  quoted  at  the 
outset  it  is  noticeable  that  the  writer  rejects  all  maternity 
in  the  case  of  Dionysus,  puts  poor  Semele  entirely  out  of 
court,  and  maintains  that  Zeus  only,  and  Zeus  alone, 
brought  the  babe  to  birth.  Backofen,5  in  his  Mutterrecht, 
first  had  a  glimpse  of  the  fact  that  here  was  a  Greek  parallel 
to  the  more  primitively  grotesque  assertion  made  by  impli 
cation  in  the  curious  practice  known  as  the  couvade^  that  a 
child's  father  is  both  parents  in  one,  and  that  he  is  most 


2  Literally  a  hostage  (ofj.rjpov),  and  this  is  part  of  the  play  upon 
and  ^pos.  3  /u,77p6s.  4  w^pevtre. 

5  See  F.  A.  Voigt  on  "  Dionysus  "  in  Roscher's  AusfuhrUches  Lexicon, 
p.  1046. 


SECOND  BIRTH  OF  DIONYSUS  169 

particularly  and  especially  its  mother.  Curiously  enough 
not  Dionysus  only,  who  proves  it  in  his  own  person,  but  also 
Apollo,  here  again  his  ally,  maintains  this  strange  doctrine. 
Aeschylus,  with  a  deep  insight  into  the  mysterious  background 
of  his  own  faith,  makes  Apollo  say,  in  the  Eumenides,  OVK 
ICTTI  fJMJTrjp  .  .  .  To/<eu?,  .  .  .  TtKTet  8'  o  6 pwcT KM .  The  great 
principle  exemplified  in  the  second  birth  of  Dionysus  is  a 
triumphant  justification  for  Orestes,  the  slayer  of  Clytem- 
nestra,  and  the  same  intense  belief  that  the  mother  has  no 
relation  to  her  child,  which  is  all  its  father's,  leads  certain 
savages  to  eat  children  born  to  their  own  wives  of  fathers 
who  are  slaves  captured  in  war.1  In  fact,  the  story  of 
Orestes  represents  a  more  primitive  and  unflinching  asser 
tion  of  the  nullity  of  the  mother's  motherhood  and  the 
reality  of  the  father's  than  does  that  of  the  second  birth  of 
Dionysus.  In  this  last  common  sense  has  asserted  itself, 
and  the  child  is  partially  matured  in  Semele's  womb.2  Then 
when  she  has  been  destroyed  before  the  full  period  for 
Dionysus'  birth  has  come,  the  half-formed  babe,  ^/uTeAcorov, 
as  Nonnus  calls  him,  is  transferred  to  the  thigh  of  Zeus. 

And  now,  since,  the  testimony  of  cannibal  customs  has 
been  referred  to,  it  is  high  time  to  put  the  whole  question  at 
issue  in  the  hands  of  the  anthropologists,  who  are  alone  com 
petent.  Fortunately  Dr.  Tylor  has  dealt  with  the  matter  in 
one  of  his  most  recent  papers.3  Indeed  this  very  point,  i.e. 
the  place  and  the  function,  in  the  early  stratification  of 
family  customs,  to  be  assigned  to  the  violent  assertion  that 
a  child's  father  is  his  all,  and  his  mother  has  no  part  in 
him,  is  taken  by  Dr.  Tylor  as  his  especial  theme.  Out  of 
scattered  materials  strewn  like  glacial  boulders  upon  the 

1  For  the  fact  and  a  most  instructive  account  of  the  couvade,  see  Dr. 
Tylor's  Early  History  of  Mankind,  vol.  i.  pp.  287-297. 

2  A  further  proof  of  the  reassertion  of  the  mother's  natural  rights  which 
plays  its  part  in  shaping  this  myth  is  found  in  the  beautiful  affection  of 
Dionysus  for  his  mother  Semele.      This  lovely  trait  is  omnipresent  in  his 
story.      Mr.  Pater  has  been  particularly  happy  in  his  account  of  it. 

3  "On   a  Method   of   Investigating  the  Development    of   Institutions 
applied  to  Laws  of  Marriage  and  Descent. "   Journal  of  the  Anthropological 
Institute,  February  1889. 


1 70  APPENDIX  III 

path  of  civilisation  he  builds  up  a  wonderfully  well-founded 
and  solidly  based  structure  of  scientific  demonstration.  This 
is  in  fact  the  topic  which  he  has  chosen  for  a  treatment  so 
strict  in  method  that  he  may  well  hope  that  its  elucidation 
shall  "  overcome  a  certain  not  unkindly  hesitancy  on  the 
part  of  men  engaged  in  the  precise  operations  of  mathe 
matics,  physics,  chemistry,  biology,  to  admit  that  the 
problems  of  anthropology  are  amenable  to  scientific  treat 
ment." 

A  more  precise  description  of  the  "  quaint  custom " 
called  the  couvade  is  now  desirable,  since  it  is  here  contended 
that  the  same  explanation  will  account  for  that  and  for 
Apollo's  vindication  of  Orestes  as  guiltless  though  he  had 
slain  his  mother,  together  with  the  episode  of  Dionysus' 
second  and  only  real  birth  from  the  thigh  of  Zeus  his 
father.  In  the  couvade,  to  quote  from  Dr.  Tylor,1  "the 
father,  on  the  birth  of  his  child,  makes  a  ceremonial  pre 
tence  of  being  the  mother,  being  nursed  and  taken  care  of, 
and  performing  other  rites  such  as  fasting  and  abstaining 
from  certain  kinds  of  food  or  occupation,  lest  the  new-born 
should  suffer  thereby.  This  custom  is  .known  in  the  four 
quarters  of  the  globe.  How  sincerely  it  is  still  accepted 
appears  in  a  story  of  Mr.  Im  Thurm,  who  on  a  forest 
journey  in  British  Guiana  noticed  that  one  of  his  Indians 
refused  to  help  to  haul  the  canoes,  and  on  inquiry  found 
that  the  man's  objection  was  that  a  child  must  have  been 
born  to  him  at  home  about  this  time,  and  he  must  not 
exert  himself  so  as  to  hurt  the  infant.  In  the  Mediterranean 
district  it  is  not  only  mentioned  by  ancient  writers,  but  in 
Spain  and  France,  in  or  near  the  Basque  country,  it  went 
on  into  modern  times ;  Zamacola  in  1 8 1 8  mentions,  as 
but  a  little  time  ago,  that  the  mother  used  to  get  up  and 
the  father  take  the  child  to  bed.  Knowing  the  tenacity  of 
these  customs,  I  should  not  be  surprised  if  traces  of  couvade 
might  be  found  in  that  district  still." 

The  place  of  this  custom  in  the  early  history  of  man- 

1  P.  254  in  the  journal  above  quoted. 


SECOND  BIRTH  OF  DIONYSUS  171 

kind  hangs  together  with  the  more  or  less  well-established 
fact  that  there  were  three  stages  of  successive  develop 
ment  in  family  and  tribe  organisation.  In  the  first 
and  earliest  of  these,  sometimes  called  the  matriarchal 
stage,  descent  and  inheritance  had  only  to  do  with  the 
mother.1  Here  then  was  the  absolute  contradiction  of 
Apollo's  dictum  in  the  Eumenides  of  Aeschylus.  Here  the 
child  is  as  solely  and  exclusively  his  mother's  as  he  after 
wards  was  maintained  to  be  solely  his  father's.  Between 
these  two  strata  there  was  an  intermediate  stage  wherein 
both  customs  struggled  for  predominance.2  Now  the  most 
startling  confirmation  of  this  order  for  the  development  of 
early  customs  is  given  by  Dr.  Tylor's  discovery — which  he 
makes  doubly  impressive  by  a  sort  of  geological  diagram — 
that  the  couvade  is  unknown  in  the  lower  or  matriarchal 
stratum,  begins  after  the  middle  of  the  transitional  stratum, 
and  spends  itself  early  in  the  upper  or  patriarchal  stratum. 
Thus  the  couvade  was  a  visible  symbol,  a  practice  by  the 
adoption  of  which  the  father's  authority  was  finally  and 
definitely  asserted.  As  soon  as  this  victory  was  won  the 
custom  by  which  it  gained  the  day  became  a  mere  curiosity, 
a  survival. 

The  curious  thing  is  that  the  Greek  power  to  trans 
mute  all  things  and  to  beautify  whatever  came  into  the 
Greek  consciousness  should  have  conquered  even  the 
stubborn  material  afforded  by  this  graceless  struggle  for 
mastery  within  the  primitive  human  family,  and  should 
have  associated  its  dimmed  and  mysterious  record  with 
those  masterpieces  of  the  high  poetic  genius  of  man,  the 
Oresteia  of  Aeschylus,  and  the  Bacchanals  of  Euripides. 

1  See  in  the  seventh  annual  report  of  the  trustees  of  the  Peabody  Museum, 
Cambridge,  Mass.  (1884),  Mr.  Lucien  Carr's  able  paper  on  "The  Social 
and  Political  Position  of  Woman  among  the  Huron-Iroquois   Tribes." 
These  tribes  and  many  of  those  of  the  Pueblos  in  Arizona  are  still  at  the 
matriarchal  stage. 

2  It  is  perhaps  fanciful  to  suggest  that  the  rival  pretensions  of  Clytem- 
nestra  and  Agamemnon  to  dispose  of  Iphigenia  are  a  record  of  this  middle 
stage. 


172  APPENDIX  III 

Indeed  I  cannot  more  suitably  bring  to  an  end  these  notes 
upon  a  survival  in  Greek  tragedies  themselves  of  primeval 
customs  than  by  referring  to  a  thoughtful  though  a  brief 
account  recently  given  1  of  the  manner  in  which  the  Greeks 
performed  their  tragedies  as  a  similar  survival — I  mean  their 
use  of  masks  in  acting.  "No  one  of  the  early  tragedians  .  .  . 
did  in  fact  invent  masks,  but  .  .  .  these  existed  as  survivals 
of  the  paraphernalia  of  the  Greek  rites  from  remote  and 
uncivilised  times.  .  .  .  Indeed  the  use  of  masks  is  wide 
spread  among  uncivilised  peoples  ;  it  begins  apparently  with 
a  dim  notion  of  terrifying  or  deceiving  demons,  and  soon 
becomes  a  formula  of  worship.  It  was  from  this  state  that 
the  custom  appears  to  have  entered  the  Greek  drama.  .  .  . 
While  the  mask  is  common  among  nearly  all  savage  races, 
we  may  find  it  surviving  in  the  dramatic  performances  of 
the  Chinese  and  Japanese."2  Interesting  though  the 
Chinese  and  Japanese  drama  is,  and  not  devoid  of 
the  genuine  power  that  belongs  to  an  art  which  has 
its  definite  traditions,  the  difference  between  its  appoint 
ments — not  to  speak  of  essentials — and  those  of  the 
Greek  stage  is  very  great,  and  on  the  score  of  beauty 
of  course  is  all  in  favour  of  the  Greeks.  Starting 
apparently  from  the  same  or  practically  the  same  barbaric 
ritual  which  is  the  background  of  Chinese  and  Japanese 
theatrical  performances,  the  Greeks  were  guided  to  beauty 
by  an  instinct  which  was  all  their  own,  and  which  has  made 
them  the  sponsors  of  all  that  is  best  in  dramatic  literature. 
As  Mr.  Perry  has  admirably  said,  behind  the  perfected 
Greek  drama  "was  a  past  that  had  triumphed  successfully 
over  the  barbarism  which  left  its  rites,  so  to  speak,  as 
the  raw  .material  to  be  worked  by  art  and  enthusiasm  into  a 
thousand  charming  forms.  The  savage  survivals  were,  like 
the  physical  geography  of  the  land,  tamed,  smoothed, 
cultivated,  made  inhabitable,  not  destroyed."  3  In  the  realm 

1  A  History  of  Greek  Literature,  by  T.  S.  Perry  :    New  York,  Henry 
Holt  and  Co.,  1890. 

2  Ibid.  pp.  229,  230.  3  Ibid.  p.  224. 


SECOND  BIRTH  OF  DIONYSUS  173 

of  the  drama,  as  in  all  other  regions  of  literature  and  art 
which  the  Greeks  knew,  they  and  they  alone  possessed  the 
art  and  the  enthusiasm  which  could  deal  with  stubborn 
and  primitive  materials — the  only  ones  at  hand.  Accord 
ingly  each  newly  -discovered  trace  in  Hellenic  work  of 
prehistoric  man  and  his  ugly  ways  is  but  a  new  occasion  for 
marvelling  at  the  transcendent  genius  of  Hellas. 


THE  GODS  AT  ELEUSIS 

IN  the  previous  chapter  it  has  been  assumed,  according 
to  abundant  testimony,1  that  Dionysus  in  some  shape  or 
other  very  early  associated  himself  at  Eleusis  with  the 

1  The  presence,  as  an  object  of  early  Eleusinian  worship,  of  a  mystical 
da.ifj.wv  is  denied  by  none.  But  because  there  is  no  mention  of  Dionysus- 
lacchos  in  the  Homeric  Hymn  to  Demeter,  and  because  the  Zagreus 
legend,  which  finally  summed  up  the  nature  of  the  specifically  Eleusinian 
Dionysus,  emanated  from  Onomacritus  and  the  new  Orphic  sect  at  Athens, 
some  maintain  that  Dionysus  was  an  utter  stranger  to  the  Eleusinian  cult 
until  the  days  of  Pisistratus.  This  interpretation  of  the  facts,  which  are 
unfortunately  too  few  to  speak  very  clearly  for  themselves,  fails  to  cor 
relate  with  the  mystical  5a.ifj.wv  of  early  Eleusis,  the  Zagreus  -  Dionysus 
who,  in  the  later  Mysteries  under  the  surname  of  lacchos,  yearly  visited 
Demeter  and  Persephone,  and  who  was  variously  represented  as  a  brother 
and  a  son  of  Persephone.  The  traditional  connection  between  Thrace 
and  the  mythical  Eumolpus,  whose  very  name  has  a  touch  of  the  music 
of  Apollo  and  Dionysus  in  it,  and  the  theory  among  late  Greeks  that 
Eumolpus  had  to  do  with  the  worship  of  lacchos  at  Eleusis,  lead  towards 
the  conclusion  that,  after  all,  the  early  mystical  8a.ifj.wv  is  a  proto-Thracian 
Dionysus  under  some  sort  of  Eleusinian  disguise — the  old  netherworld 
god  of  Thrace,  brought  by  that  early  influence  from  the  north,  represented 
by  the  name  Eumolpus  to  complete  the  group  of  divinities  worshipped 
at  Eleusis.  The  characteristically  Eleusinian  epithets  of  Pluto  and 
Eubouleus  suit  well  this  primitive  divinity  when  once  he  is  far  from 
Thrace  and  under  the  softening  influences  of  Demeter.  As  for  the  absence 
of  any  mention  of  Dionysus-Iacchos  in  the  Hymn,  to  Demeter,  it  must  be 
remembered  that  neither  is  the  great  Eleusinian  hero  Triptolemus  there 
mentioned,  except  among  others  represented  as  of  equal  importance  with 
him  in  the  establishment  of  the  Eleusinian  rites.  In  spite  of  this  circum- 


v  THE  COMING  OF  DIONYSUS  TO  ELEUSIS         175 

two  goddesses  of  the  Mysteries.  So  far  as  chronology 
applies,  it  is  evident  that  this  first  of  his  comings  to 
Eleusis  was  thought  of  as  having  taken  place  not  long  after 
the  day  of  Demeter's  arrival  at  Eleusis  and  of  Dionysus' 
visit  to  Icarius.  We  may  safely  take  it  for  granted,  how 
ever,  that  his  status  in  those  early  days  was  far  inferior  to 
that  to  which  he  subsequently  attained  through  the  reforms 
and  innovations  made  first  by  Epimenides  and  Solon,1  and 
then  by  Pisistratus  and  Onomacritus,  whereby  he  was  enabled 
to  participate  in  Eleusinian  observances  from  the  vantage- 
ground  of  an  independently  organised  Athenian  ritual  in  his 
honour.  After  Epimenides  had  suitably  organised  the  Lesser 
Mysteries  at  the  Athenian  Eleusinion,  the  god  could  in  due 
time  become  the  leader  of  the  mystae  in  their  yearly  procession 
to  Eleusis,  and  under  the  name  of  lacchos,  which  perhaps 
had  not  attached  to  him  in  his  early  days  at  Eleusis,  when 
he  was  merely  a  TrdpeSpos — an  associate  divinity,  was  there 
welcomed  as  the  coequal  of  the  two  great  goddesses. 

In  a  sense  therefore  the  coming  of  Dionysus,  as  an 
independently  recognised  divinity,  to  take  his  share 
in  the  worship  of  those  who  thronged  to  the  Greater 
Mysteries,  was  prepared  by  Epimenides  and  brought  to 
pass  by  Pisistratus  and  Onomacritus ;  and  the  first  move 

stance  testimony  from  other  sources  assigns  a  prominent  place  to  Trip- 
tolemus  in  the  local  cult.  Indeed  we  may  consider  the  presence  of  a 
representative  of  Dionysus  in  the  early  legend  a  thing  assured,  since  there 
was  anciently  more  than  one  version  of  the  story  of  the  Hymn  to  Demeter 
(see  Pausanias,  I.  xxxviii.  3,  where  Pamphos  is  followed  as  to  the  number 
and  names  of  the  daughters  of  Celeus),  and  since  the  Demophoon  incident, 
the  only  point  where  Dionysus-Iacchos-Triptolemus  could  be  concerned, 
plainly  does  not  hang  together  with  its  surroundings.  See  note  i,  p.  194. 
1  I  thus  couple  Solon  with  Epimenides,  because  Plutarch  (as  quoted 
note  i,  p.  123)  says  that  the  latter  prepared  the  way  for  Solon's 
legislation,  and  also  because  one  of  Solon's  laws  distinctly  applied  to 
the  concerns  of  the  mysteries  set  in  order  by  Epimenides.  Andocid.  de 
My  sterns,  110-112. 


1 76  THE  GODS  AT  ELEUSIS  v 

was  brought  about  from  Athens.  But  had  not  earlier 
influences  already  made  some  place  for  the  new-comer  at 
Eleusis,  the  great  Eleusinian  alliance  of  three  coequal 
divinities  would  not  so  easily  have  come  to  pass.  It  is 
an  undoubted  fact  that  the  popular  legends  and  unauthor 
ised  observances  at  Eleusis  began  to  recognise  the  Thracian 
god  at  some  earlier  time  while  he  still  bore  plain  marks  of 
being  king  of  the  underworld.  This  view  is  in  agreement 
with  the  traditions  of  Eumolpus  and  the  Eumolpidae,  while 
any  other  makes  it  difficult  to  understand  why  Dionysus 
attached  himself  in  just  the  way  he  did  l  to  a  group  of  gods 
where  Hades  played  a  part  not  unimportant,  though  to  us 
obscure.  The  coming  of  Dionysus  to  Eleusis  evidently 
enhanced  the  importance  of  Hades,  and  took  away  some 
thing  2  from  Demeter's  overpowering  predominance.  But 
by  this  limitation  she  apparently  gained  in  effectiveness 
what  she  lost  in  exclusiveness. 

1  See  F.  A.  Voigt  (art.  "  Dionysus  "  in  Roscher's  Lexicon.},  where  various 
epithets  of  Hades  are  shown  to  belong  to  Dionysus,  particularly  that  of 
E#/3ovXevs.     The  name  Eubouleus  is  especially  connected  with  the  Hades 
legend  at  Eleusis,  both  in  the  Athenian  and  the  Argive  tale.     There  was 
undoubtedly  a  more  or  less  definite  distinction  drawn,  in  the  Eleusinian 
and  cognate  worships,   between  two  male  divinities  worshipped  in  con 
junction  with  Persephone,  one  of  which  may  have  been  more  especially 
identified  with  Dionysus  than  the  other  ;   but  it  is  more  than  likely  that 
they  represented  the  two  types  of  Dionysus-Dendrites  and  Dionysus-Hades. 
The  chief  authority  for  this  distinction  is  hardly  earlier  than  the  fourth 
century  B.C.     It  is  an  inscription  found  on  a  tablet  in  a  tomb  near  the 
ancient  Sybaris.     The  deceased,  one  of  the  Kadapoi,  i.e.  initiate,  writes: 
"Epxo/J-ai  eK  Kadapuv,  Kadapa  xQoviwv  j3a<ri\eia  EwX-^s,  Ei^ouXetfs  re  ... 
See  Pausanias,  I.  xiv.  1-4,  and  Miss  Harrison  (Mythology  and  Monuments 
of  Athens,  pp.  95-101).     See  also  Chr.  Scherer  (art.  "  Hades  "  in  Roscher, 
pp.  1783  and  ff.),  where  the  euphemistic  epithets  of  the  god  are  discussed, 
and  the  softening  of  the  sterner  aspects  of  Hades  through  contact  with  the 
cult  of  Persephone  and  Demeter  is  noted. 

2  She  lost  a  touch  of  vindictiveness,  which  in  the  legend  at  Hermione 
led  her  to  burn  Colontas  in  his  house  (Pausanias,   II.  xxxv.  4),   and  a 
gloom  which  gave  her  the  surname  Erinys  at  Thelpusa  (Pausanias,  VIII. 
xxv.  4). 


v  THE  ALL-WELCOMING  DEMETER  177 

Like  the  worship  of  Apollo  at  Delphi,  that  of  Eleusinian 
Demeter  did,  however,  owe  its  increasing  importance  to  a 
hospitality,  which  welcomed  new-coming  divinities  with  no 
thought  of  curtailing  their  traditionary  powers.  Dionysus 
came  to  Eleusis  and  took  his  place  there  by  the  side  of 
Hades,1  so  that  Heraclitus  in  one  of  his  dark  words  declares 
this  identification  to  be  a  proof  that  life  and  death  are  one.2 
The  original  Thracian  conception  of  Dionysus,  based  as  it 
was  upon  the  belief  that  death  was  life,  was  in  this  manner 
reasserted. 

Besides  the  bond  of  kingship  in  the  netherworld,  Hades 
and  Dionysus  were  affiliated  by  their  relation  to  the  treasures 
concealed  in  the  bowels  of  earth.  Control  over  these  came 
by  right — so  ancient  piety  argued — to  the  lord  of  the  world 
below.  Hence  Hades  and  Pluto,3  the  god  of  riches,  were 

1  See  Voigt  (art.  "  Dionysus"  in  Roscher's  Lexicon,  p.  1047)  on  Diony 
sus'  bringing  of  Semele  from  Hades  to  Olympus,  which  he  compares  to  the 
Assumption  of  the  Virgin  Mary.     Certainly  Pindar's  tone  (in  the  third 
Pythian)  about  Semele  justifies  some  such  parallel.      Enthusiastic  wor 
shippers  of  Dionysus  attributed  to  him  power  over  the  life  to  come,  and  wel 
comed  his  use  of  it  to  lead  Semele  into  the  assemblage  of  gods  on  Olympus. 

2  Heraclitus,   quoted  by  Clement  of  Alexandria,    Protreptica,   p.   30  : 
wurds  5£  'Ai'5r?s  KO.I  &ibvv<ros  br£($  fiaivovrai  Kail  \rjvdfovcrii'.      (Cf.  Ritter 
and  Preller,  Hist.  Phil.  I.  39  a,  who  say,  after  quoting,  as  of  Heraclitus, 
the  celebrated  $dos   Zrjvl   cn<6Tos  'At'Sfl,   0dos   'Aidy  o-/c6ros   Tiyvi    </>otTg 
/cat  /j.€TaKi.v^€Tat  Kelva  &df  Kal  rdSe  Ket<re,  Traffyv  wprjv  dLaTrptjff<r6fjt.€va  Keivd 
re  TO,  ruvSe,  raSe  5'  ct5  rd  Keivuv,  "  'At'STjy,  quern  eundem  deum  esse  cum 
Libero  Patre  dicebat  [scil.  Heraclitus],  significat  vim  humidam  tenebricosam 
telluris,  lupiter  lucidam  et  ignitam  coeli.")     See  also  Scherer  (as  above)  on 
a  relief  found  at  Locri. 

3  It  might  be  hazarded  as  a  conjecture  that  the  coming  of  Dionysus  to 
Eleusis  brought  with  it  for  Hades  the  surname   Pluto.      Certainly  the 
epithet  HXovruv  first  appears  for  Hades  in  the  Attic  poets  of  the  fifth 
century.     Aeschylus,  Prom.  806  ;  Soph.  Antig.  1200  ;  Euripid.  Alcestis, 
360,   Here.   Fur.   808  ;  Aristoph.  Plut.  727.      See  also  at  the  beginning 
of  the  eighth  book  of  Plato's  Laws  a  passage  where  Pluto  is  named  alone 
for  all  the  Chthonian  gods.     Preller,  commenting  on  this  fact,  attributes 
the  epithet  Pluto  to  the  Eleusinian  worship.     Chr.  Scherer  (art.  ' '  Hades  "  in 
Roscher,  p.  1786)  inclines  to  agree  with  Preller  as  to  the  epithet  Plouto, 
but  objects  that  the  other  euphemistic  names  must  have  come  from  tradi- 

N 


i;8  THE  GODS  AT  ELEUSIS  v 

one,  and  hence  Dionysus,  to  the  extent  that  he  was  originally 
a  netherworld  god,  was  in  his  own  person  called  Father  of 
Gold,  and  to  him  were  dedicated  the  gold-bearing  floods  of 
Phrygian  Pactolus.  Demeter,  Persephone,  Aidoneus-Pluto, 
lacchos-Dionysus,  and  Rhea-Cybele — these,  the  five  divini 
ties  of  Eleusinian  worship,  become  three  before  the  eyes, 
as  it  were,  of  their  worshippers.  lacchos-Dionysus  and 
Aidoneus-Pluto  mysteriously  melt  into  one,  while  Rhea- 
Cybele  and  Demeter  are  similarly  fused.  This  would  leave 
just  three — one  Demeter-Rhea-Cybele  gave  the  feminine 
element.  The  second,  Hades-Iacchos-Dionysus,  represents 
the  male  element,  and  finally  the  third  is  Persephone.  It 
has  been  abundantly  shown  how  Demeter  and  Persephone 
were  regarded  as  one,  being  so  filled  with  mutual  love  that 
all  barriers  between  them  melted  away.  A  similar  identifica 
tion  of  Dionysus  and  Persephone  is  shadowed  forth  by 
legends  of  their  marriage.  Hence  what  we  may  call  the 
first  of  the  Eleusinian  mysteries,1  since  it  deals  with  the 
hidden  nature  of  all  the  gods  at  Eleusis,  is  not  without  a 
modern  parallel.  It  presented  itself  to  the  pious  mind  in 
terms  and  with  difficulties,  most  of  which  recur  in  one 
statement  or  another  of  the  mystery  of  the  Holy  Trinity. 
Eight  names,  four  of  goddesses  and  four  of  gods,  came 
finally  to  stand  for  two  persons  in  whom  was  presented  one 
great  fact — the  course  of  nature.  Demeter  was  Persephone ; 
both  and  each  were  Rhea,  who  was  Cybele.  Aidoneus 

tions  preserved  among  the  people.  Suppose  that  Dionysus  brought  this 
golden  contribution,  and  that  the  other  mild  epithet  of  Eubouleus  came 
from  Greek,  and  especially  Eleusinian,  tradition,  then  the  softening  influ 
ence  which  gathered  these  kindly  qualities  around  forbidding  Hades 
belongs  still  to  Eleusis. 

1  I  am  intentionally  using  the  word  Mystery  in  the  modern  sense, 
because  it  is  noticeable  that  a  religious  conception  very  nearly  approaching 
it  is  characteristic  of  the  Orphic  writings,  and  was  familiar  to  Euripides. 


v  XENOPHANES  AND  HERACLITUS  179 

was  Pluto;  while  both  and  each  were  Dionysus,  who  was 
lacchos,  and  also,  in  some  sense,  Triptolemus. 

The  two  divine  persons  around  whom  these  abundant 
names  and  attributes  gathered  at  Eleusis  were  in  the  highest 
sense  not  two  but  one.  They  were  one  as  concave  and 
convex  are  one ;  they  represented  the  active  and  the  pas 
sive  aspects  of  the  great  and  universal  all.  Nor  is  it  fanci 
ful  to  add  that  they  represented  two  typically  Greek  ways l 
of  understanding  the  world  and  all  that  is  therein  ;  the  one 
way  was  that  of  Demeter  and  Xenophanes,  the  other  way 
was  that  of  Heraclitus  and  Dionysus. 

Dionysus  all  flash,  all  heat,  all  motion  flowing  and  grow 
ing,  living  and  dying,  dancing  and  flying,2  was  a  fit  incarna 
tion  of  the  philosophy  of  those  whom  Plato  laughingly  calls 
the  "  Streamers,"  men  who  with  Heraclitus,  the  dark  philo 
sopher,  talked  of  the  course  of  nature  as  being  that  of  a 
swift  and  shifting  stream  or  a  fitfully  burning  conflagration. 

1  For  a  somewhat  fuller  account  of  these,  see  sections  3-7  of  my  Intro 
duction — based  upon  Dr.  Crons's — to  Plato's  Apology  :  Ginn  and  Heath, 
1886. 

2  See  Pausanias,  III.  xix.  6,  for  Dionysus  worshipped  at  Amyclae  as 
^Xa£,  or  winged,  and  cf.  E.  Thraemer  (art.  "  D.  in  der  Kunst,"  Roscher, 
p.  1152).     Dr.  Braun  {Kunstvorstellungen  des gefliigelten  Dionysos,  Munich, 
1839)  first  called  especial  attention  to  this.      Speaking  of    the  winged 
Dionysus  at  Amyclae,  Pausanias  makes  a  somewhat  forcible-feeble  remark, 
to  the  effect  that  the  god  of  wine  may  well  have  wings,  since  under  wine's 
influence  men  flutter,  and  are  uplifted  as  by  wings.    There  is  a  merry  French 
song  in  praise  of  Dionysus, — "  Vive  Denis  notre  bon  pere  !  "  is  the  gist  of  it ; 
but  the  last  verse  gives  to  Dionysus  Liber  both  wings  and  song,  as  follows  : 

' '  Ce  Liber  pere  des  repas 
Qu'on  adore  au  siecle  ou  nous  sommes, 

En  tre"passant  ne  mourut  pas 
Ainsi  qu'on  voit  mourir  les  hommes  ; 

Un  assoupissement  vineux 

Poussa  son  esprit  lumineux 
Dans  un  doux  repos  de  vingt  heures, 

Apres  quoi  ce  dieu  s'envola 
Dans  les  eternelles  demeures 

Chantant  ut  r6  mi  fa  sol  la." 


i8o  THE  GODS  AT  RLE  US  IS  v 

Certainly  the  poetic  genius  of  man  never  conceived  any 
personality  better  suited  than  that  of  Dionysus  to  represent 
the  ever-moving  stream,  the  ever-living  and  ever-dying  fire 
of  Heraclitus.  Those  minds  whom  this  doctrine  confused 
and  alarmed  could  take  the  very  different  view  of  nature 
and  divinity  presented  by  Xenophanes ;  and  Demeter's 
personality  gives  most  admirably  the  aspect  of  divinity 
which  they  would  chiefly  worship.  Demeter  is  peace 
bought  with  the  price  of  sorrow,  love  mingled  with  sad 
ness  ;  hers  is  a  constant  soul,  unswerving  and  unselfish  in 
her  boundless  love  for  sweet  Persephone.  Let  Demeter 
then  stand  for  the  new  aspect  of  divinity  proclaimed  and 
justified  by  Xenophanes. 

Tired  of  the  tales  that  the  charming  Homer  told,  shocked 
and  pained  at  the  wickedness  of  gods  who  were  human  at 
heart  and  only  superficially  divine — magnified  men  freed  from 
death  and  age  but  not  from  sin — Xenophanes  declared  that 
god  was  one,  even  so  Demeter  and  Persephone  were  one ;  he 
said  that  god  was  infinite,  even  so  was  the  love  and  long 
ing  of  Demeter  for  Persephone.  Indeed  it  has  been  often 
remarked  that  a  new  spirit  came  into  Greek  religion  and  life 
with  the  new  worship  of  non-Homeric  divinities  at  Eleusis ; 
and  this  new  spirit  was  just  what  Xenophanes  longed  for. 

In  the  unknown,  or  at  most  half- known,  spirit  of  the 
Eleusinian  mysteries,  one  virtue  reigned  with  living  power, 
which  some  think  has  in  our  days  vanished  from  all 
Christendom.  This  virtue  is  much  lauded  by  the  pious 
Plutarch ;  it  is  the  virtue  of  silence.  Indeed  all  the  rites 
of  Eleusis  would  have  been  in  vain  if  it  were  possible  to 
describe  minutely  the  Eleusinian  ritual  after  the  confident 
fashion  of  the  author  of  The  Divine  Legation  of  Moses.1 
1  Warburton,  Bishop  of  Gloucester. 


v  THE  HOLY  SILENCE  OF  E  LEU  SIS  181 

Worshippers  were  bound  by  every  fear,  and  lured  by 
every  hope,  touching  their  fate  after  death1  to  reveal  no 
word  of  what  was  said,  and  to  withhold  the  least  hint  of 
what  was  done  in  the  Eleusinian  Holy  of  Holies.  What  is 

1  The  scholiast  on  line  158  of  the  Frogs  of  Aristophanes  says  :  "The 
opinion  prevailed  at  Athens  that  whoever  had  been  taught  the  Mysteries 
would,  when  he  came  to  die,  be  deemed  worthy  of  divine  glory.  Hence 
all  were  eager  for  initiation."  This  would  sometimes  take  place  when  a 
man  was  near  his  death.  See  Aristoph.  Peace,  v.  374  f. ,  where  Try- 
gaeus,  sure  of  approaching  death,  tries  to  borrow  three  drachmas  to  buy  a 
bit  of  a  porker  (for  an  offering  to  the  gods  below),  and  says,  "  You  know 
I've  got  to  be  initiated  or  ere  I  die."  A  curious  ray  of  light  is  thrown  upon 
the  whole  question  of  the  mysteries,  and  the  comfort  which  they  gave  by 
assuring  to  the  initiated  especial  privileges  in  the  life  beyond,  by  four 
Orphic  fragments  found  in  Southern  Italy  (three  at  Sybaris  and  one  at 
Petelia).  The  date  of  the  tombs  wherein  they  were  found  on  thin  plates 
of  gold  is  the  third  century  B.C.  ;  but  Comparetti,  in  his  account  of  them 
(Journal  of  Hellenic  Studies,  vol.  iii.  p.  112),  says  the  Orphic  fragments  go 
back  to  the  time  of  Euripides,  and  he  refers  to  the  well-known  passage  in 
Plato's  Republic  about  the  Orpheotelestae  (ii.  364  B).  In  the  preceding 
chapter  I  have  spoken  of  the  first  Orphic  doctrines  promulgated  by  Ono- 
macritus  at  Athens  ;  Mr.  Cecil  Smith,  "Orphic  Myths  on  Attic  Vases" 
(Journal  of  Hellenic  Studies,  vol.  xi.  p.  346),  gives  the  following  summary 
of  doctrine  (derived  from  the  three  inscriptions  in  question)  from  later  Orphic 
poems,  and  from  a  vase-painting  of  great  and  almost  unique  interest  that 
goes  back  to  a  date  earlier  than  480  B.C.  : — 

' '  In  the  cosmogony  of  the  Orphic  teaching  there  are  two  great  cosmic 
elements — Zeus,  the  omnipotent  all  in  all,  and  his  daughter  Kore,  who 
combines  in  her  personality  the  characteristic  features  of  Persephone, 
Artemis,  and  Hekate  ;  from  the  union  of  Zeus  in  serpent  form  with  Kore, 
Zagreus  is  born,  and  to  him,  essentially  in  his  character  of  x^ojuos,  the 
kingdom  is  given  of  this  world.  Zagreus  is  the  allegory  of  the  life  and 
death  and  resurrection  of  Nature.  In  the  generally  accepted  version,  he 
is  brought  up  as  the  Zeus-child,  and  from  fear  of  Hera,  is  sent  on  earth  to 
be  warded  by  the  Kouretes.  Hera  sends  the  Titans,  who  surprise  Zagreus 
at  play,  tear  him  in  pieces,  and  eat  him  all  except  the  heart.  Zeus  destroys 
the  Titans  with  his  thunderbolts,  and  out  of  their  ashes  the  human  race  is 
born.  Since  the  Titans  had  swallowed  Zagreus,  a  spark  of  the  divine 
element  forever  permeates  the  human  system.  The  heart  is  carried  by 
Athene  to  Zeus,  who  either  gives  it  to  Semele  in  a  potion  or  swallows  it 
himself,  and  thus  is  born  another  Zagreus,  the  '  younger  Dionysus, '  6  v^os 
&i6vv<ros. "  For  the  initiated  death  is  a  piece  of  good  luck,  and  on  one  of 
the  Sybaris  tablets  the  departed  soul  exults,  saying  to  the  gods  :  /cat  yap 
fyijjv  vfj.Q>v  ytvos  6\{3oi>  etf^ojucu  eli/cu.  Having  atoned  for  the  sin  of  the 
Titans  by  mystic  ceremonies,  the  initiated  claim  the  heritage  of  Zagreus, 
which  is  life  everlasting.  He  is  in  their  members,  and  through  his  death 
their  immortality  has  been  won. 


182  THE  GODS  AT  ELEUSIS  v 

sometimes,  with  a  too  ineffable  self-complacency,  called  the 
"  modern  mind,"  might  learn  a  lesson  from  the  novices  at 
Eleusis ;  and  it  is  perhaps  good  for  us  all  to  ponder  over 
this  ancient  recognition  of  the  unutterableness  of  the  un 
utterable.  This  ground  of  holy  reserve,  not  always  respected 
to-day,  was  kept  intact  both  in  Greece  and  at  Rome  by  the 
Mysteries. 

If  silence  is  the  chief  lesson  and  culminating  grace1 
derived  from  Eleusis,  it  may  be  asked  why  there  is  more  to 
say  ?  But  even  the  secret  of  those  Mysteries  has  been  in  a 
certain  fashion  laid  open,  and  their  noble  spirit  breathes  from 
many  masterpieces  of  the  Greek  genius.  Such  was  the 
speaking  power  of  Greek  art,  that  the  sculptors  and  the 
poets  have  almost  revealed  the  secret  in  the  beauty  of  their 
work.  Certain  statues  of  Eleusinian  divinities  bear  the 
impress  of  the  Mysteries,  as  do  indeed  the  eyes  of  many  a 
saint  pictured  by  Christian  art.  Even  in  Botticelli's  awkward 
and  mysterious  grace  we  read  this  same  unnamed  and  un- 
nameable  constraint  and  mystery. 

The  first  and  most  delicate  manifestation  of  this  shows 
in  the  peaceful  and  enigmatical  beauties  of  Demeter  and 
Persephone.  Give  to  this  constraining  power  something  of 
manly  force,  and  it  constrains  no  longer  to  repose.  The 
universe  whirls  onward  then  in  Dionysus'  wake.  The 
trees  are  drawn  to  follow  Orpheus  and  his  Thracian  lyre. 
With  Dionysus  all  nature  floods  forward  and  onward  to  a 

1  For  fear  of  having  been  misled  into  a  one-sided  statement,  I  give  the 
following  graphic  summary  of  the  spirit  in  which  the  faithful  were  invited 
to  the  Mysteries,  which  I  abbreviate  from  Mannhardt's  Demeter  essay  : 
"Come,  whosoever  is  clean  of  all  pollution,  and  whose  soul  hath  not 
consciousness  of  sin.  Come,  whosoever  hath  lived  a  life  of  righteousness 
and  justice.  Come,  all  ye  who  are  pure  of  hand  and  of  heart,  and  whose 
speech  can  be  understood."  Almost  every  Athenian  sought  out  the 
celebration,  and  from  time  to  time  communed  with  the  gods  of  Eleusis  for 
the  ease  of  his  soul. 


v       DIONYSUS  AN  ENHANCEMENT  OF  DE METER    183 

goal,  which  is  neither  named  nor  known,  and  yet  is  the  first 
cause  of  an  irresistible  impulse.  The  intensity  of  calm 
which  is  sometimes  to  be  seen  in  Demeter's  grief,  and 
sometimes  even  gives  to  her  joy  a  sober  hue,  allies  itself 
with  one  aspect  of  the  annually  recurring  tragedy  of  life 
decaying,  and  of  growth  on  earth.  Forsaken  in  her  grief 
she  is  the  spirit  of  loneliness,  the  genius  of  home-sickness ; 
and  even  in  her  appeasement  she  still  seems  alone.  Per 
sephone,  her  joy,  is  with  her  truly,  but  she  brings  to  her 
mother  that  nameless  tremor,  half  of  peace  and  half  of 
unutterable  oppression,  which  comes  to  a  lingerer  musing 
in  the  fields  of  spring. 

The  more  boisterous  joy  of  Dionysus  is  this  tremor 
raised  to  a  higher  power,  and  contains  its  oppression  and 
its  gladness  both  intensified.  The  promise  written,  half  in 
sadness,  first  upon  the  hesitating  face  of  spring  comes  to 
its  uttermost  fulfilment  in  an  ecstasy  of  joy  which  is  near 
to  downright  madness  and  fraught  with  death.  The 
crescendo  of  growth  and  vigour  drives  away  and  utterly 
dispels  the  outward  show  of  mystery,  because  the  mystery 
itself  lies  hidden.  It  is  the  god  himself  who  enters  in 
and  fills  his  worshippers  and  all  the  world  with  his  con 
straining  power.  He  is  in  all  things,  and  he  leads  all 
things  on  the  way  of  his  choice.  From  flash  to  flash,  from 
flame  to  flame,  the  scale  of  bright  and  fluid  being  is  run 
through  with  the  whirl,  as  it  were,  of  a  devouring  fire  that 
darts  across  fields  of  yellow  grain.  Demeter  is  no  longer 
there,  nor  yet  are  we  who  have  been  swept  along  by  Diony 
sus  in  his  fluid  train.  Yet  this  is  not  the  last  word  of  the 
mysterious  power  that  shapes  the  varying  course  of  nature. 
The  learned  and  truly  pious  Strabo  somewhere  says  that  it 
was  but  right  for  the  Eleusinian  worshippers  to  guard  most 


1 84  '    THE  GODS  AT  ELEUSIS  v 

jealously  a  mystical  secret.  How  otherwise  could  their 
ritual  have  shown  forth  the  nature  of  the  gods  at  Eleusis  ? 
Their  secret  always  eluding  inquiry  was  like  their  godhead 
for  ever  eluding  the  grasp  of  our  senses,1  for  ever  streaming 
on  beyond  reach  of  our  straining  eyes.  The  Streaming 
philosopher,  Heraclitus,  declared  solemnly  that  you  could 
not  twice  step  into  the  same  river,  and  Strabo  would  have 
us  apply  this  to  divinity,  and  mark  how  the  same  Dionysus 
is  never  met  with  twice.  This  may  be  called,  and  was 
sometimes  meant  as,  a  Pantheistic  doctrine ;  but  sometimes 
it  was  of  higher  import,  and  Dionysus  was  thought  of  as  a 
spirit  moving  in  all  things,  whose  worshippers  must  not 
attach  themselves  to  any  one  manifestation  of  him,  but 
must  worship  him  in  spirit  and  in  truth. 

And  yet  this  fast  and  furious  race  from  shape  to  shape 
was  thought  of  only  as  a  final  paroxysm,  like  the  fortissimo 
that  comes  near  the  end  of  a  musical  composition.  Then 
nature  reaches  fulness,  fruits  are  shining  where  lately  were 
the  buds  of  spring,  while  the  dancing  Maenads  whirl  across 
the  face  of  the  earth,  moving  in  Bacchic  revelry  their 
gleaming  feet,  tossing  their  necks  into  the  dewy  air.  This 
is  the  Maenads'  hour  of  triumphant  freedom.  Now  let 
them  sing  while  they  may  the  victorious  refrain  in  Euri- 
pidean  numbers.  "What  is  the  wisdom,  what  among 
mortals  the  boon  of  heaven  that  is  fairer  than  waving  the 
hand  victorious  over  a  fallen  foe  ?  What  is  glorious,  that 
is  always  dear." 2 

Dearly  bought  indeed  is  this  Bacchanalian  victory,  for 
there  is  a  mystery  revealed  in  sadness  when  the  ecstasy 

1  Strabo,  X.  iii.  9  (467)  :  77  re  Kpfyis  77  IJLVGTLKT]  T&V  lep&v 
rb  Oetov,  /uu[ji.ov/ji{i''r}  rty  (f>v<ni>  avrov  (petiyovcrav  TJ/J.UV 
-  Euripides,  Bacchanals,  877-881,  and  897-901. 


v  THE  PITILESS  HUNTSMAN  Z ACRE  US  185 

of  joy  is  past.  A  frenzied  impulse  overtakes  the  revelling 
Maenads,  and  lo  !  their  nearest  and  their  dearest  lies 
before  them  hideously  slain.  In  overpowering  their 
foeman  they  have  unspeakably  harmed  their  own.  The 
huntsman  from  whom  they  thought  to  escape  was  none 
other  than  their  own  Dionysus,  the  pitiless  huntsman 
Zagreus.  They  thought  to  be  swift  and  go  from  him 
free  when  he  had  really  entered  in  and  possessed  them 
utterly.  Winter  is  at  hand,  there  are  no  buds,  no  blossoms, 
no  fruits,  and  no  joys.  The  sad  awakening  comes — Diony 
sus  is  dead.  Is  he  not  buried  within  the  temple  of  Apollo 
at  Delphi  ?  And  now  the  worshipper  is  left  alone !  And 
yet  not  quite  alone,  for  he  has  for  his  comrade  in  grief 
Demeter,  the  all-welcoming — Demeter,  in  her  lonely  trial 
longing  and  grieving,  seeking  and  finding  not — Demeter 
whose  only  comfort  is  in  doing  deeds  of  sweet  and  unpre 
meditated  love.1 

Let  us  now,  while  we  still  are  under  the  spell  of  De- 
meter's  sorrowing  godhead,2  enter  into  the  holy  place  at 
Eleusis  and  consider  reverently  its  broken  stones  and  buried 
walls.  Here  is  a  place  consecrated  by  eight  hundred  years 
of  pious  usage  and  spoiled  by  centuries  of  neglect.  At  last 

1  The  way  in  which  allegiance  to  the  spirit  of  the  Mysteries  begins 
with  Persephone  and  Demeter,  transfers  itself  for  a  climax  to  lacchos,  and 
then  dies  down  to  a  calmer  loyalty  again,  chiefly  to  Demeter  and  the  high 
standard  of  right  living  associated  with  her,  is  best  seen  in  the  passage  of 
the  Frogs  of  Aristophanes  summarised  at  the  end  of  this  chapter.     See 
w.  372-459. 

2  This  phase  of  Demeter  is  characteristic  when  her  divinity  stands  in 
contrast  to  that  of  Dionysus.     Dionysus  also  when  taken  alone  has  his 
sad  and  subdued  aspects.       For  both   these  divinities  alone  were  con 
ceived  of   as  covering  the  whole  ground  more  completely  but  not  less 
really  occupied  when  they  each  supplemented  the  other,  and  both  made 
room  for  Persephone.     Demeter  as  the  productive  Earth  (Eur.  Bacch.  274- 
276)  was  conceived  of  as  going  through  in  her  own  person  all  the  stages 
and  phases  of  vegetation,  and  of  the  husbandry  by  which  earth  was  culti 
vated.     See  Lenormant,  art.  "  C£res"  in  Daremberg  and  Saglio. 


1 86  THE  GODS  AT  ELEUSIS  v 

a  time  for  the  re-awakening  of  glorious  pagan  memories  has 
come  at  Eleusis,  since  the  present  condition  of  its  site  is  the 
result  of  much  careful  excavation. 

From  Athens  to  Eleusis  is  not  far,  though  it  is  more  than 
a  Sabbath  day's  journey.  In  more  accurate  measurement 
the  distance  to  Eleusis  is  slightly  over  twelve  miles.  The 
first  excavations  at  Eleusis  were  made  early  in  this  century 
by  the  London  Dilettanti  Society.1  From  these  labours 
came  a  good  account  of  the  site  and  of  the  two  ceremonial 
gates  or  Propylaea — both  of  the  latter  belonging  to  the  days 
of  Roman  supremacy  at  Eleusis.  Of  these  first  excavations 
an  account  is  given  in  the  Unedited  Antiquities  of  Attica^ 
published  in  1817.  The  Dilettanti  Society  could  not  cause 
the  modern  village  of  Levsina  to  be  removed  from  the  site 
most  important  for  excavation,  and  therefore  obtained  little 
or  no  knowledge  of  the  Hall  of  Initiation.  This  forced 
omission,  and  nearly  all  others,  have  been  made  up  for  by 
Greek  excavations  which  were  ended  only  in  1887.  At  the 
request  of  the  Greek  Archaeological  Society,  Dr.  Dorpfeld 
made  out  in  1887  the  full  ground-plan  of  all  buildings  whose 
foundations  were  left  on  the  site  when  the  village  houses 
had  been  removed.  The  plan  published  in  1888  will  never 
receive  any  important  modifications,  though  details  may  still 
be  forthcoming ;  and  I  desire  to  give  my  warmest  thanks  to 
my  friend  Dr.  Dorpfeld  for  allowing  me  to  publish  it  here. 
The  enthusiasm  and  ability  of  Dr.  Philios,  the  commissioner 

1  This  chapter  is  so  especially  concerned  with  Eleusis  rather  than  with  the 
approach  to  it  from  Athens  that  Franpois  Lenormant's  admirable  work  in 
excavation  and  publication  has  no  great  prominence  in  my  presentation  of 
Eleusinian  religious  antiquities.  His  work,  however,  and  his  account  of 
the  Sacred  Way,  demand  the  fullest  recognition,  and  his  Grande  Grtce 
also  contains  much  invaluable  information  about  Dionysus  in  Greater 
Greece.  From  his  articles  in  Daremberg  and  Saglio's  Dictionary  I  have 
constantly  derived  enlightenment. 


v  DESCRIPTION  OF  ELEUSIS  187 

for  years  in  charge  of  the  Eleusinian  excavations,  have  abund 
antly  justified  the  confidence  reposed  in  him  by  the  Greek 
Archaeological  Society,  and  earn  the  gratitude  of  all  students, 
who  may  now  see  in  Dr.  Dorpfeld's  plan  a  record  of  the 
results  due  to  Dr.  Philios'  learning,  energy,  and  ability. 

Eleusis  lies  upon  and  around  a  group  of  rocks  which 
separate  the  south-eastward  breadths  of  the  Thriasian  plain 
from  the  smaller  Rarian  plain,  which  is  north-west  of  it. 
Towards  the  south  and  east  spreads  the  beautiful  Bay  of 
Eleusis,  and  beyond  rise  the  purple  heights  of  Acamas l  on 
Salamis — Salamis  looming  up  as  if  to  shut  out  all  view  of  the 
Gulf  of  Aegina  and  distant  Cyllene.  The  best  description 
of  Eleusis  is  perhaps  that  given  in  the  Unedited  Antiquities 
of  Attica,  as  follows : 

"  The  south-eastern  extremity  of  a  low  rocky  hill  about 
300  yards  from  the  sea  was  chosen  by  the  Eleusinians  for 
their  citadel — their  acropolis.  The  declivity  of  this  hill 
facing  the  south-east  being  formed  into  an  artificial  terrace, 
and  the  rock  having  been  cut  away  from  the  front  to  the 
rear,  a  level  area  was  obtained  for  the  sacred  enclosure  of 
the  mystic  temple.  This  magnificent  structure,  built  by 
Pericles,  stood  a  bold  and  prominent  feature  in  a  picture 
whose  background  was  formed  by  the  walls  and  towers  of 
the  impending  acropolis.  In  front  the  villas  and  gardens 
of  the  Eleusinians  complete  the  picture,  spreading  them 
selves  around  the  foot  of  the  rock  and  along  the  borders  of 
the  Bay  of  Salamis — called  also  the  Bay  of  Eleusis,  since 
Eleusis  is  on  its  northern  shore — while  the  sea-girt  heights 
of  Salamis  lock  it  in  towards  the  south.  As  accessories  in 


1  For  the  authority  upon  which  I  use  this  name  for  the  Salaminian 
mountains,  see  Appendix  VIII.  end,  on  "The  river  Bocarus  and  John 
Meursius,"  after  chap.  vii.  below. 


1 88  THE  GODS  AT  ELEUSIS  v 

the  composition  of  this  grand  design,  the  lofty  gates  or 
Propylaea,  with  the  temple  of  Artemis  Propylaea,  were 
worthy  of  admiration." 

Such  is  the  picture  of  ancient  Eleusis  skilfully  drawn 
in  1814.  In  it  we  see,  vaguely  indeed  but  really,  some 
thing  of  the  later  magnificence  at  Eleusis.  This  was  the 
Eleusis  of  Roman  days,  for  the  great  Propylaea — not  to 
speak  of  the  upper  gateway  or  Lesser  Propylaea  of  Appius 
Claudius  Pulcher1 — were  built  in  Roman  days  and  not 
visible  to  the  eyes  of  Ictinus.  This  was  the  Eleusis  which 
came  to  destruction  in  the  year  of  our  Lord  396  at  the 
hands  of  monks  who  followed  in  the  wake  of  Alaric 
and  his  Ostrogoths.  Eunapius  calls  these  worse  than 
Ostrogoths  "the  black -robed  crew,"  and  their  icono- 
clasm,  no  doubt,  merited  his  most  "vinegar  epithets";  but 
still  even  these  zealots  did  less  harm  probably  than  the 
poverty  and  sloth  caused  by  the  intolerable  and  continuous 
oppression  weighing  heavily  upon  generations  who  after 
wards  lived  near  and  on  the  site.  Various  churches  and  a 
whole  village  got  themselves  built  within  the  boundaries 
of  the  holy  precinct  by  a  process  utterly  destructive  of  all 
manner  of  architectural  remains,  and  particularly  of  the 
precious  statues  left  in  fragments,  but  still  no  doubt  left  by 
those  Ostrogothic  monks  who  would  hardly  have  been  able 
to  demolish  everything  of  the  kind.  Heads,  arms,  and 

1  Cicero  to  Atticus  (vi.  i),  "  Unum  etiam  velim  cogites.  Audio  Appium 
TTpoTrtiXaiov  Eleusine  facere.  Num  inepti  fuerimus,  si  nos  quoque  Academiae 
fecerimus?  Puto,  inquies.  Ergo  id  ipsum  scribes  ad  me.  Equidem  valde 
ipsas  Athenas  amo.  Volo  esse  aliquod  monumentum.  Odi  falsas  in- 
scriptiones  statuarum  alienarum.  Sed  ut  tibi  placebit."  Very  interesting 
remains  of  this  ceremonial  gate  of  Appius  still  lie  upon  its  site.  A  certain 
originality  is  shown  in  its  composite  capitals  and  in  the  decorative  use  of 
wheaten  ears  and  the  vaguely  known  instruments  of  the  mysteries  upon 
its  entablatures.  For  photographs  of  these  remains  see  Appendix  XL  i. 
43-  44- 


THE  ONLY  CHURCH  OF  ANTIQUITY  189 

beautiful  draperies  fashioned  delicately  in  marble  seemed  to 
the  clumsy  and  half-barbarised  Albanian  builders  of  Levsina 
to  exist  for  nothing  else  than  the  fire  whose  burning  gave 
them  lime  for  building  their  unsightly  huts.  After  these 
devastations  no  hope  could  be  entertained  that  any  full 
knowledge  of  the  temples  and  statues  of  the  gods  at  Eleusis 
should  ever  be  rescued  from  its  ruins.  We  are  forced  in 
fact  to  make  many  a  conjecture  before  the  results  of  the 
most  patient  and  painstaking  excavations  will  yield  any  clear 
notion  of  that  unique  structure  the  Eleusinian  Telesterion 
or  Hall  of  Initiation.  This  was  called  in  the  description 
quoted  above  the  Mystic  Temple,  but  is  more  accurately 
designated  by  Aristophanes  as  the  "  Home  that  welcomed 
the  Mystae,"  Strabo's  phrase  for  it  being  o  ILVGTIKQS  cry/cos, 
the  holy  enclosure  of  the  Mystae?-  It  was  in  fact  not  as  other 
Greek  temples  were,  for,  as  Strabo  directly  implies,  it  was 
not  the  dwelling-place  of  any  god,  and  contained  there 
fore  no  holy  image.  It  is  unique  because  on  no  other 
Greek  site  has  there  been  found  a  meeting-house  built, 
as  this  one  was,  for  the  celebration  of  a  definite  ritual. 
The  Thracian  worship  of  one  of  the  Eleusinian  gods, 
Dionysus,  seems  to  have  required  meeting  places  or  houses 
of  some  kind,  but  there  is  only  the  vaguest  record  of  them. 
The  truth  therefore  is  that  the  Eleusinian  Hall  of  Initiation 
is  the  only  known  church  of  antiquity,  if  by  church  we 
mean  not  so  much  the  house  of  the  deity  as  the  meeting 
house  for  worshippers,  a  place  where  they  may  congregate 
for  worship. 

This  Hall  of  Initiation,  if  we  would  know  it  as  it  stood 

1  He  plainly  distinguishes  it  from  the  temple  of  Demeter.  See  IX.  i. 
12  (395)  Elr'  'EXewris  7r6\u,  h  r?  rb  TTJS  ATj/rrjrpos  iepbv  TT)S  'JSXevffivtas  Kal 
6  /iucm/cos  o"r)K6s,  6v  KareffKevaafv  'IKTWOS  (t^Xov  dearpov 


190  THE  GODS  AT  ELEUSIS  v 

in  the  days  of  Athenian  greatness  and  power,  must  be  shorn 
of  the  Roman  fagade  and  the  porch  of  Philo.  These  must 
disappear  with  the  walls  that  go  with  the  lower  or  most 
northern  Propylaea,  a  ceremonial  gate  built  in  the 
Emperor  Hadrian's  day.1 

Suppose,  then,  Hadrian's  grand  gate  of  entrance  is 
removed ;  take  away  also  the  outer  wall  (indicated  by  salmon 
colour  on  the  large  map)  that  this  gate  pierced.  Then  you 
have  thrown  open  a  considerable  space  between  a  sacred 
building — the  temple  of  Artemis  Propylaea  (Artemis  at  the 
gate) — and  an  older  gateway  piercing  an  older  outside  wall. 
This  is  the  gateway,  already  talked  of  above,  built  by  Appius 
Claudius  Pulcher.  The  report  of  it  moved  Cicero  to 
propose  that  he  and  his  rich  friend  Atticus 2  should  build 
something  of  the  kind  for  the  Academy.  Supposing  our 
selves  in  Eleusis  before  Appius  Claudius  and  his  workmen, 
then  in  place  of  his  gate  we  should  have  found  something 
of  wholly  Greek  antiquity — something  to  show  forth  the 
earlier  history  of  the  shrine  and  sanctuary.  Here  anciently 
was  a  strong  gate  which,  with  the  wall  that  it  pierced, 
could  be  defended  against  all  enemies  of  the  gods  and  of 
Eleusis.3  Having  passed  through  this  gate  and  hastened  to 
the  Hall  of  the  Initiated,  we  might,  supposing  our  visit  fell 

1  It  was  the   irony  of  fate   which   afforded   money  and   to  spare   in 
Hadrian's  time  for  completing  at  Eleusis  an  imitation  of  the  masterpiece  ol 
Mnesicles,  itself  left  unfinished  for  the  lack  of  moneys  in  the  coffers  of 
imperial  Athens. 

2  The  passage  at  the  end  of  the  first  letter  in  the  sixth  book  of  their 
correspondence  is  given  above,  p.  188,  note  i.     Atticus  apparently  did  not 
encourage  Cicero,   perhaps  because  he  reflected  that  Cicero  would  have 
contributed  more  beautiful  discourse  than  hard  cash  to  any  joint  under 
taking  of  the  kind. 

3  That  there  was  fortification  in  the  early  days  cannot  be  doubted,  in 
view  of  recorded  attacks.     Certain  remains  of  old-time  masonry,  together 
with  the  fixed  position  of  the  Sacred  Way,  make  it  practically  sure  that 
here  was  a  fortified  gate. 


v  ELEUSIS  FORTIFIED  IN  EARL  Y  DA  YS  191 

after  310  B.C.,  find  an  important  feature  which  was  not  known 
to  the  worshippers  in  the  days  of  the  Peloponnesian  war — I 
mean  the  porch  of  Philo,  built  at  the  expense  of  Demetrius 
the  Phalerean  in  310  B.C.  Ictinus  must  have  planned  either 
this  porch  or  something  like  it,  but  it  certainly  was  not  built 
in  his  day.  And  finally,  if  we  returned  to  the  site  four 
centuries  or  so  later,  we  should  discover  an  enlargement 
and  remodelling  of  the  Hall  as  built  by  Ictinus.  The  site 
as  it  existed  before  310  B.C.  was  enclosed  by  a  defensible 
wall,  and  approached  by  a  fortified  gate  on  the  site  of  the 
ornate  and  unfortified  Propylaea  of  Appius.  The  Telesterion 
or  meeting-house  consisted  of  two  narrow  rooms,  had  no 
front  porch,  and  was  not  quite  so  large  as  Roman  recon 
struction  subsequently  made  it. 

It  is  very  easy  to  forget  the  little  or  nothing  known  about 
certain  small  temples  and  treasure-houses  of  uncertain  date. 
These  grouped  themselves  about  the  great  meeting-house 
of  the  Mystae,  and  like  it  had  the  living  rock  of  the  Eleusinian 
Acropolis  as  their  background.  This  rock  towards  the  north 
exhibits  two  remarkable  cave-like  arches  in  the  living  stone. 
Such  was  the  site  before  310  B.C. — six  hundred  years,  that 
is  to  say,  before  the  sanctuary  was  ravaged  and  destroyed.1 

1  Cf.  Strabo,  quoted  above,  p.  189,  note  i,  and  Vitruvius,  Praef.  vii.  16, 
17,  Schneider.  Plutarch  (Pericles,  xiii.)  gives  a  rather  detailed  account  of 
the  various  architects  and  builders  who  apparently  carried  out  the  plans  of 
Ictinus,  though  the  words  of  Plutarch  alone  might  lead  one  to  think  he  did 
not  connect  Ictinus  with  the  work,  but  rather  considered  its  building  to  have 
been,  like  the  Parthenon,  under  the  general  supervision  of  Phidias.  He 
says  that  Coroebus  began  to  build  it,  proceeding  so  far  as  to  set  the 
columns  up  on  the  foundations,  and  adding  the  architrave.  Coroebus 
died,  and  Metagenes  of  the  deme  Xypeta  continued  the  work,  adding  the 
5idfw/xa  (is  this  to  be  translated  frieze,  or  has  it  the  meaning  of  prae- 
cinctio,  a  narrow  upper  gallery,  for  access  to  upper  seats,  which  at  Eleusis 
would  mean  a  ledge  hewn  out  of  the  rock,  to  allow  access  to  the  upper  story  ?) 
and  the  columns  of  the  upper  story.  Xenocles  of  Cholargia  finished  the 
faalov,  whatever  that  may  here  be  supposed  to  mean. 


i92  THE  GODS  AT  ELEUSIS  v 

Now  let  us  approach  the  remains  as  they  are.  Neglect 
the  Roman-  remains  of  triumphal  arches  on  your  left ;  look 
for  a  moment  at  the  site  of  the  temple  of  Artemis  at  the 
gate ;  consider  the  intense  misunderstanding  of  the  Doric 
capital  that  led  Hadrian's  builders  to  give  such  a  stiff 
and  lifeless  curve  as  that  shown  in  huge  examples  that 
cumber  the  ground  on  the  site  of  Hadrian's  gate  of  Cere 
mony.  There  is,  if  only  it  were  worth  the  looking  at,  a 
monstrous  lump  of  white  marble  here.  It  was  a  huge 
medallion  tastelessly  injected  into  the  gable  or  pediment  of 
the  Propylaea  aforesaid.1  Some  think  a  mysterious  person 
figured  here — a  priest,  say;  but  others  more  prosaically 
claim  that  Hadrian  himself  somewhat  awkwardly  presided 
over  this  rule-of-thumb  Doric  architecture  for  which  he 
is  responsible.  But  let  us  get  inside  this  gate  and  forget 
everything  about  it  save  only  that  it  faced  north-east. 
Following  now  the  Sacred  Way  which  trends  to  the  left  and 
ascends,  we  may  now  pass  the  remains  of  the  smaller  gate 
of  Appius,  which  faces  due  north.  To  those  who  think 
they  can  solve  the  riddles  ot  all  religions  by  accumulating 
facts  about  the  orientation  of  temples  it  will  be  of  import 
ance  to  note  that  the  four  corners  of  the  great  meeting-house 
at  Eleusis  point  respectively  north,  south,  east,  and  west. 
Before  reaching  this  northern  gate  of  Appius  we  are  not 
yet  on  the  ground  of  old  deemed  holy ;  but  this  gate  once 
passed,  we  are  where  the  yearly  procession  from  Athens  first 
felt  that  its  goal  was  reached.  A  long  journey  it  was  for 
those  burdened  with  offerings — this  twelve  miles,  the  last 
nine  of  which  were  without  shade,  if  one  may  rashly  suppose 
the  distribution  of  trees  always  to  have  been  what  it  is  to-day. 

i  See  for  the  photograph  of  it  published  by  the  Hellenic  Society, 
Appendix  XI.  i.  43. 


v  THE  PRESENT  CONDITION  OF  ELEUSIS          193 

Here  we  are  at  last  within  the  sanctuary  rdlv  Oeaiv,  of 
the  goddesses  twain.  Before  looking  about  us  within,  let 
it  be  stated,  for  the  benefit  of  geometers,  that  this  sacred 
ground  enclosed  by  walls  and  rock  is  in  shape  an  irregular 
pentagon.  Of  the  five  enclosing  sides,  the  longest  is  the 
line  of  overhanging  rock — the  Acropolis.  At  the  northern 
end  of  this  rock  wall,  which  runs  from  west  to  north,  are 
the  two  caves.  Just  north  of  the  northernmost  of  the  two 
caves  this  longest  side  meets  the  shortest  side — a  wall  run 
ning  north-eastward  from  the  Acropolis  rock  to  the  gate 
way  just  entered.  Of  the  three  remaining  sides  one  is  a  wall 
parallel  to  the  Acropolis  rock,  and  the  other  two,  also  walls, 
connect  this  parallel  side  respectively  with  the  gate  of  Appius 
and  the  western  end  of  the  Acropolis  side.  Such  are  the 
boundaries,  through  the  ruins  of  which  we  suppose  ourselves 
to  have  walked. 

And  now  we  may  well  begin  with  a  curious  examination 
of  the  ground  we  tread,  over  which  so  many  pious  feet  have 
passed.  Beyond  the  lesser  or  Appian  gateway  traces  appear 
of  the  Holy  or  Processional  Way,  but  under  the  disappointing 
guise  of  a  Roman  pavement — slabs  of  stone  made  fast  with 
mortar  upon  the  native 'rock.  In  Grecian  days  the  bare  rock 
was  probably  not  improved  upon  either  here  or  in  the  much 
and  piously  travelled  roadway  leading  up  to  the  Athenian 
Parthenon  through  the  Mnesiclean  Propylaea.  Various 
traceable  pedestals  indicate  that  many  monuments  lined 
this  processional  road,  which  so  far  resembled  many  others. 
Between  the  gate  of  Appius  and  the  overhanging  Acropolis 
rock  is  the  small  precinct  of  Pluto,  which  is  approached  by 
a  step  or  two  and  an  entrance-gate.  This  small  corner, 
belonging  to  Demeter's  self-constituted  son-in-law,  is  remark 
able  rather  for  the  striking  configuration  of  the  natural  rock 

o 


1 94  THE  GODS  AT  ELEUSIS  v 

that  shuts  it  in  on  the  west  than  for  the  slight  traces  of  a 
very  small  temple  which  it  contains.  The  finding  here  first 
of  a  bust  representing  Eubouleus,1  the  Eleusinian  Hades, 
and  then  of  a  bas-relief  representing  Demeter,  Perse 
phone,  and  Hades,  establishes  the  proprietary  right  of 
Aidoneus-Pluto  to  this  spot,  included  though  it  be  within 
the  sanctuary  walls  of  the  dread  twain  goddesses. 

Within  this  small  precinct  facing  north,  and  just  south  of 
the  intersection  of  the  longest  (or  Acropolis)  side  with  the 
shortest  side  of  the  sacred  pentagon,  is  a  hole  in  the  rock, 
raised  higher  than  man's  stature  above  the  general  level. 
In  the  rock  below,  and  north  of  this  aperture,  are  steps2 
roughly  hewn  leading  to  a  height,  and  a  foothold  from 
which  it  is  easy  for  any  one  to  climb  through  the  hole  and 
enter  the  arched  cave-like  space  beyond.  This  cave,  as  it 
may  be  called,  together  with  a  larger  one  much  resembling 
it  just  south  of  it,  would  have  seemed,  and  apparently  did 
seem,  in  myth-growing  days,  the  very  spot  where  Aidoneus 
on  his  chariot  might  have  swept  with  Persephone  into  his 
nether  abode.  The  rock  overhanging  the  Cnidian3  precinct 

1  See  note  3  above,  p.  177  ;  note  i,  p.  174  ;  and  Chr.  Scherer,  article 
"Hades"  in  Roscher's  Ausfuhrliches  Lexicon.     I  hear  of  an  article  (not 
procurable  before  going  to  press)  in  the  Mittheilungen,  wherein  Dr.  Kern 
successfully  maintains  that  this  beautiful  head  represents  not  Eubouleus  but 
Triptolemus.     This  would  tend  to  confirm  my  contention  (see  chapter  ii. 
above)  that  Demophoon  is  an  interloper  in  the  story  of  Demeter,  and  that 
Triptolemus  was  the  real  nursling.     Furthermore,  it  would  tend  to  connect 
Triptolemus  with  this  precinct  of  Pluto,  and  to  affiliate  his  worship  with 
that  of  Hades  and  lacchos.      Cf.  Daremberg  et  Saglio,  p.  634,  col.  z. 

2  Dr.  Dorpfeld  kindly  calls  my  attention  to  the  possibility  that  these 
very  roughly  cut  steps  may  have  belonged  to  the  arrangements  for  a 
modern  house. 

3  What  Attic  tradition  records  as  the  coming  of  Demeter  to  Eleusis 
gains  in  significance  if  we  find  reason  to  suppose  this  new  departure  of 
nature-worship  in  Attica  to  have  been  prompted  from  the  north, — if  Demeter 
came  from  Thessaly  as  did  Aesculapius  and  as  Dionysus  came  from  Thrace. 
A  very  definite  tradition  asserts  that  the  Cnidian  sanctuary  of  the  two  god 
desses  was  founded  from  Thessaly,  as  were  the  Coan  and  the  Epidaurian  rites 


v  THE  PLUTONIAN  PRECINCT  195 

of  Demeter  and  Persephone  was  found  by  Sir  Charles 
Newton  to  have  similar  peculiarities  to  these  of  the  Eleu- 
sinian  Acropolis. 

But,  to  return  to  the  precinct  of  Hades-Pluto,  nearly  in 
front  of  the  two  caves  are  unmistakable  traces  of  a  very 
small  "cella"  or  temple  of  Pluto.  The  foundations  show 
it  to  have  been  ten  feet  broad  by  sixteen  feet  long.  The 
head  of  Eubouleus-Triptolemus,1  found  near  it,  very  closely 
resembles  one  which  was  long  called  Virgil,  and  which  is  to 
be  found  in  the  Capitoline  Museum.  It  is  none  the  less 
beautiful  because  Professor  Benndorf  is  almost  alone  in 
attributing  it  to  the  Phidian  age. 

Emerging  from  the  Plutonian  precinct,  and  passing 
a  few  steps  southward  on  the  Processional  Way,  turn 
again  westward,  and  there  find  the  more  or  less  uncertain 
foundation-stones  to  which  probably  corresponded  two 
buildings.  These  are  identified  respectively,  but  only  the 
farther  one  confidently,  with  the  two  treasure-houses  men 
tioned  in  Eleusinian  inscriptions.  One  may  have  been  the 
treasure-house  of  Demeter ;  and  if  this  be  so,  the  other,  in 
case  it  was  anything,  was  that  of  Persephone.  The  import 
ance  of  the  treasuries  which  these  foundations  may  represent 
is  abundantly  shown  in  the  accounts  of  the  temple- 
funds  so  plentifully  forthcoming  of  late  years,  and  so  well 

in  honour  of  Aesculapius.  See  above,  ch.  ii.  and  below,  ch.  vi.  The  com 
ing  of  Demeter  from  Pyrasus  and  the  Dotian  plain  of  Thessaly  is  by  no 
means  inconsistent  with  an  aboriginal  Eleusinian  nature -worship  and 
nature-goddess.  So  too  the  Thracian  Dionysus  coming  to  Icaria  absorbed, 
and  was  absorbed  by,  an  indigenous  worship  of  a  kindred  nature  to  him 
self.  The  known  facts  plainly  require  such  an  explanation.  See  above, 
ch.  iii. 

1  For  a  very  good  reproduction  of  it,  see  Miss  Harrison  and  Mrs. 
Verrall's  Mythology  and  Monuments,  p.  105.  Brunn  has  it  in  his 
Monuments  of  Ancient  Art,  It  gains  much  in  interest  and  importance  if 
Dr.  Kern  can  show  that  it  represents  Triptolemus. 


196  THE  GODS  AT  ELEUSIS  v 

edited.  The  reason  why  such  scanty  foundation  remains 
were  here  found  is  that  the  solid  rock  lies  close  to  the 
surface,  and,  accordingly,  all  traces  of  such  buildings  as 
existed  were  most  readily  obliterated.1  Therefore  very 
little  concerning  these  would-be  treasure-houses  can  be 
ascertained. 

Whether  each  of  the  two  goddesses  had  an  especial 
temple  for  her  own  abiding  can  also  never  be  ascertained 
with  certainty  from  anything  that  has  been  discovered  on 
the  spot,  and  accordingly  what  may  possibly  be  traces  of 
two  small  temples  are  only  doubtfully  to  be  described  as 
such.  The  facts,  such  as  they  are,  may  be  stated  as  follows  : 
traces  of  a  smaller  temple,  which  might  be  attributed  to 
Persephone,  are  near  the  northern  angle  of  the  Hall  of 
Initiation  to  the  east.  The  plainer  traces  of  a  temple, 
larger,  though  still  small,  are  visible  at  the  northern  end  of 
the  raised  terrace  which  runs  between  the  Hall  of  Initiation 
proper  and  the  overhanging  north-westward  rock.  Here 
may  conceivably  have  stood  Demeter's  temple  on  a  higher 
level,  to  the  north  of  the  same  north  angle.2 

And  now  I  have  mentioned  the  chief  among  the  lesser 
buildings,  about  none  of  which,  excepting  perhaps  the  first, 
there  can  be  reasonable  certainty,  (i)  The  small  precinct 
and  small  temple  or  "  cella  "  of  Plutus ;  (2)  and  (3)  the 
supposed  treasuries,  one  for  each  goddess  ;  (4)  the  very 
problematical  temple  or  cella  of  Persephone ;  (5)  the 
equally  doubtful  temple  or  cella  of  Demeter.  Besides 

1  The  poor  Albanians,  in  giving  themselves  and  their  animals  various 
rudimentary  comforts,  have  played  fast  and  loose  with  the  rock  here. 

2  Strabo's  words  (see  note  i,  p.  189)  make  it  certain  that  a  temple  of 
Demeter  formed  one  of  the  conspicuous  features  of  the  sanctuary  ;  and 
the  comparative  insignificance  in  size  and  prominence  of  the  remains  on 
this  site  leave  the  whole  matter  in  doubt. 


v  THE  HALL  OF  INITIATION  197 

this  there  was  in  some  place,  not  determined,  within  the 
precinct  a  Neocorion,  i.e.  quarters  for  the  neocoroi — those 
in  executive  charge  of  the  buildings  and  minor  concerns  of 
the  sanctuary.1 

As  for  the  great  Initiation  Hall,  the  most  interesting  by 
far  of  all  the  features  on  this  site,  let  us  admit  to  start  with 
that  its  study  is  a  matter  of  great  perplexity,  in  spite  of  an 
absolute  certainty  with  regard  to  the  most  important  leading 
facts. 

Your  first  feelings,  as  you  wander  up  and  down  across 
this  Eleusinian  wilderness  of  stones,  are  confusion  and  help 
lessness.  Before  you  lies  what  seems  to  be  an  incongruous 
crowd  of  foundations  for  the  bases  of  columns,  no  two  of 
which  seem  to  be  part  of  one  scheme.  A  closer  examina 
tion  shows  in  effect  that  there  are  many  kinds  of  foundations, 
bases,  and  traces  of  columns  belonging  by  their  manner  of 
construction  to  many  epochs  of  building.  These  puzzling 
and  overlapping  traces  are  multiplied  especially  at  the 
eastern  angle  of  the  Hall.  That  quarter  of  the  ground 
occupied  by  the  whole  Hall  which  lies  nearest  this  eastern 
angle  contains  fifty-six  bases  or  traces  of  columns,  while 
upon  all  the  remainder  only  thirty-seven  can  be  found. 

This  curious  fact  leads  to  a  closer  examination  of  the 
column -foundations  where  they  are  most  numerous,  and 
here  a  wishing-cap  is  necessary.  Put  on  this  cap,  while 
looking  at  these  shapeless-seeming  ruins,  and  wish  for  all 
the  knowledge  of  the  various  masonry  of  various  epochs 
possessed  by  Dr.  Dorpfeld,  director  of  the  German  Institute 
at  Athens,  or  by  Dr.  Philios,2  the  indefatigable  excavator  at 

1  See  Appendix  I.  above,  for  some  account  of  the  later  history  of  this 
word  i/eco/cdpos. 

2  See  his  pamphlet   (Athens,    1889,   Ch.  Wilberg)  Fouillcs  d'  &  leu  sis, 
1882-87. 


I98  THE  GODS  AT  ELEUS1S  v 

Eleusis.  Then  you  would  note  how  the  various  traces 
gradually  group  themselves  as  follows:  (i)  twenty -five 
small  square  foundations,  coloured  red  on  the  plan,  about 
10  feet  apart,  these  being  wholly  confined  to  this  eastern 
quarter  of  the  site;  (2)  twenty  places  for  round  columns 
15  feet  apart,  requiring  (in  the  northern  corner)  one 
more,  of  which  no  trace  can  be  found,  to  make  up  the 
symmetrical  tale, — twenty-one ;  (3)  six  large  square  founda 
tions  requiring  the  addition  of  two  more  than  the  remains 
found  to  complete  the  necessary  eight.  Besides  these 
three  fashions,  there  is  a  fourth  fashion  of  column  founda 
tion.  These  are  distributed  very  curiously  over  the  whole 
space, — forty-two  bases  in  a  square  of  a  hundred  feet  more 
or  less. 

To  the  smaller  hall,  destroyed  by  the  Persians,  belong 
the  first  mentioned  square-column  foundations,  discoverable 
exclusively  within  the  eastern  quarter  of  the  site,  and 
coloured  red  on  the  plan.  This  smaller  hall  may  be  called 
the  Hall  of  Pisistratus,  though  what  is  certainly  known 
about  it  is  that  it  was  destroyed  by  the  Persians  after  Xerxes' 
defeat  at  Salamis.1  Traces  of  their  destructive  fire  have 
come  to  light,  giving  the  confirmation  of  our  own  eyes  to 
what  Herodotus  reports  in  general  terms.  In  the  year 
479  B.C.  Mardonius  burned  and  overturned  the  Initiation 
Hall  of  Pisistratus.  Its  building  is  attributed  to  the  age  of 
Pisistratus  because  the  foundation  walls  of  it  are  practically 
identical  both  in  the  materials  used,  the  order  of  their 


tl  Before  the  Persians,  King  Cleomenes  of  Sparta  seems  to  have  de 
vastated  the  sanctuary  (Herod,  vi.  75,  cf.  64,  and  v.  74  and  ff. )  Here  is  a 
confirmation  of  the  notion  suggested  by  the  nature  of  the  remains  of 
the  earlier  Greek  walls  enclosing  the  precinct.  They  must  have  been  a 
fortification,  otherwise  a  King  of  Sparta  would  never  wantonly  have 
attacked  them,  or  the  sanctuary  which  they  enclosed. 


After  DT  Dorpfelds  Plan  in  the  Journal  of  the 

Greek  Archaeological  Society  1887.  and  reproduced  here. 

by  kind  permission 


PLAN    OF     EXCAVATIONS 


Dotted  halchmv  <t*n<-t<*fa  ver*  earliest 

Red,  colour  shows  the  t.iwl*  destroy  <'<!  /n: 

Grey  wlour  shows  tf,,<  vork  of  Ictmus  &  later  additions 

#  Alterations 


NS 


ELEUSIS,   1885-1887, 


v  HALLS  OF  PISISTRATUS,   CIMON,  ICTINUS       199 

superposition,  and  the  manner  of  their  putting  together,  with 
buildings  known  to  have  been  built  by  Pisistratus,  the  great 
advocate  of  Dionysus,  the  defender  of  faith  in  the  gods  at 
Eleusis.  The  ceiling  of  this  Hall  was  supported  by  twenty- 
five  interior  columns,  it  was  entered  from  the  south-east 
through  a  portico,  and  it  can  hardly  have  failed  to  justify  the 
twenty-five  supporting  columns  by  a  large  upper-story  room 
approached  no  doubt  from  the  level  of  the  upper  terrace. 
The  Acropolis-rock  had  not  of  course  been  cut  away  at  that 
time. 

After  this  Hall  of  Pisistratus  a  Hall  of  equal  frontage  but 
twice  as  deep  was  built,  by  cutting  into  the  rock  for  more 
room.  This  Hall  was  supported  by  twenty -one  interior 
columns.  To  this  important  structure  (which  we  may  call 
the  Hall  of  Cimon,  since  it  was  built  in  his  day)  must  be 
allotted  those  twenty  columns  15  feet  apart,  more  than  half 
of  their  number  being  hopelessly  entangled  among  the  older 
remains  of  the  Hall  of  Pisistratus. 

Now,  therefore,  the  wilderness  at  the  east  angle  of  the 
hall — the  columns,  bases,  and  traces,  fifty-six  in  number — 
perplexes  us  no  longer.  The  eastern  angle  of  the  Hall 
proper,  in  Ictinus'  plan,  was  the  eastern  corner  of  the 
porch  of  the  earlier  Hall  of  Pisistratus.  This  early  building 
with  its  porch  covered  slightly  more  than  one-fourth  of  the 
ground  allotted  to  the  Hall  of  Ictinus,  and  about  half 
of  that  occupied  by  the  Hall  of  Cimon.  Of  the  fifty-six 
bases  and  traces  huddled  together  in  this  corner  of  the 
whole  space,  twenty-five  belong  to  the  Hall  of  Pisistratus, 
fifteen  to  that  of  Cimon,  and  sixteen  to  the  last  and 
Roman  refashioning  of  the  building.  There  are  no 
columns  or  bases  of  columns  here  which  belong  to  the 
Hall  of  Ictinus,  for  the  Hall  of  Ictinus  consisted  of  two 


2oo  THE  GODS  AT  ELEUSIS  v 

chambers,  a  new  one  added  by  Ictinus  to  the  already 
existing  Hall  of  Cimon.  This  will  be  made  clear  by  con 
sulting  the  small  plan  here  given  and  by  comparing  it  with 
the  large  one. 

Before  turning  to  a  consideration  of  the  Hall  of  Ictinus, 
it  is  worth  while  to  seek  confirmation  for  the  facts  in  the 
architectural  history  of  the  spot  thus  far  obtained.  Cor 
responding  to  some,  if  not  to  all,  of  these  successive 
temples,  there  must  have  been  various  walls  of  enclosure. 
Around  the  wilderness  of  column  bases,  at  various  dis 
tances  and  in  various  directions,  extend  foundations  of 
walls  built  in  the  most  various  manners  of  masonry.  There 
is  the  old  fashion  of  wall  called  Cyclopean,  there  is  the 
wall  which  belongs  to  the  day  of  Pisistratus,  built  after 
the  manner  of  the  upper  foundation-courses  of  the  building 
just  south  of  the  Erechtheum,  between  it  and  the  Parthenon 
at  Athens.1  Then  come  walls  of  later  and  better  Greek 
workmanship,  belonging  to  the  days  of  Cimon  and  Pericles. 
Finally  there  are  abundant  traces  everywhere  of  Roman 
building. 

The  upshot  of  competent  examination  here  gives  us 
traces  of  a  building  earlier  even  than  the  Hall  of  Pisistratus, 
of  some  building  dating  back  perhaps  beyond  history, — a 
building  too  around  which  ran  a  protecting  wall.  What 
remains  of  it  is  indicated  by  the  dotted  hatching  on  the  plan. 
The  whole  space  thus  pre-historically  pre-empted  was  much 
smaller  than  the  later  precinct.  All  this  may  be  con 
veniently  named  the  Cyclopean  Hall,  if  it  be  remembered 
that  Cyclopean  means  almost  anything,  and  that  there  is 
nothing  to  show  whether  this  early  building  was  a  hall  or  a 

1  For  photographs  showing  this  foundation,  see  Appendix  XL  i.  14. 
Cf.  Mr.  Leaf's  photograph  of  Eleusinian  foundations,  ibid.  ii.  40. 


v     CYCLOPEAN  TRACES,  ROMAN  RESTORATIONS     201 

fortress  simply.  This  phrase  shall  commit  us  only  to  the 
vaguest  recognition  of  the  antiquity  of  building  upon 
Demeter's  Eleusinian  place  of  worship,  and  will  give  as  a 
background  for  the  successive  Halls  of  Pisistratus,  Cimon, 
Ictinus,  and  the  Roman  Hall,  the  dim  vision  of  a  primi 
tive  place  of  refuge,  and  perhaps  of  a  worship  primitively 
lodged  after  the  fashion  of  the  so-called  "  cave  temple  " l  of 
Apollo  on  Delian  Cynthus. 

But  after  these  Cyclopean  remains  and  the  three  Halls 
above  mentioned  came  the  Romans  and  their  buildings  and 
repairings.  They  enlarged  the  space  occupied  by  the  Hall 
of  Ictinus,  especially  towards  the  Acropolis-rock ;  apparently 
they  tore  down  the  wall  by  which  Ictinus  separated  his 
addition  from  the  Hall  of  Cimon,  and  they  suppressed  the 
eight  heavy  columns  (represented  on  the  large  plan  by  six 
large  square  foundations,  and  two  dotted  squares)  of  Ictinus' 
Hall,  as  well  as  the  twenty-one  columns  of  Cimon's  Hall. 
In  place  of  these  they  put  in  forty-two  columns  of  their  own 
more  or  less  symmetrically  distributed  over  the  whole  space. 
Upon  the  Greek  walls  enclosing  the  whole  sanctuary  traces 
of  Roman  repairing  are  tolerably  clear.  The  restoration 
of  the  upper  and  lesser  ceremonial  gate  by  Appius  Claudius 
probably  amounted  to  a  rebuilding,  and  the  lower  wall, 
joining  the  Lower  and  Greater  Propylaea  to  the  Greek 
fortification  walls,  is  wholly  Roman.  To  sum  up  the 
history  of  building  on  this  spot:  Behind,  everything  we 
have  (i)  the  Cyclopean  Hall;  next  came  (2)  the  Hall 
of  Pisistratus  ;  then,  after  (3)  the  Hall  of  Cimon,  came 
(4)  the  Hall  of  Ictinus,  which  was  succeeded  by  (5)  the 
Roman  Hall. 

And  now  it  becomes  necessary  to  give  such  account  as 

1  For  some  account  of  this,  see  below,  chap.  viii. 


202 


THE  GODS  AT  E  LEU  SIS 


may  be  possible  of  the  Hall  of  Ictinus.     I  owe  to  the  kind 
ness  of  Dr.  Dorpfeld  the  sketch-plan  here  given. 

OUTLINE-SKETCH    OF   THE   GROUND-PLAN    OF 
THE   HALL   OF  ICTINUS 


O 

o 

o 

V 

lol     lol 

-/ 

O 

o 

0 

0 

|ol     [pi 

1  —  1 

0 

o 

0 

JX 

K 

1 

o 

0 

o 

t-^-i          rrr-i 

U          U 

0 

0 

0 

0 

0 

0 

\Q\      \Q] 

1 

0 

o 

o 

I, 

X 

I.     Chamber  added  by  Ictinus. 

K.   Hall  of  Cimon. 

A.    Foundations  which  later  served  for  Philo's  Porch. 


This  Hall  of  Ictinus  was  really  two  Halls,  the  old  one  of 
Cimon,  which  it  was  apparently  necessary  to  respect,  and 
the  new  one  of  Ictinus  doubling  the  available  space.  It 
seems  likely,  however,  that  the  new  was  separated  from  the 
old  by  a  wall  which  was  changed  when  Ictinus  built  from 
the  north-west  outside  wall  of  Cimon's  Hall  into  a  partition 
wall  between  the  two  large  chambers.  The  forty-two  founda 
tions  in  the  Roman  refashioning  of  the  whole  form  so  con 
spicuous  a  feature  on  the  large  plan  that  it  is  difficult  to 


v  THE  HALL  OF  1CTINUS  203 

disentangle  from  them  the  work  of  Ictinus  and  Cimon. 
The  late  character  of  these  forty-two  bases  is  proved  by  the 
comparative  poorness  of  their  construction  and  by  the 
presence  of  a  Roman  inscription  among  materials  built 
into  the  most  easterly  of  these  forty -two  bases.  The 
Romans  made  a  slight  gain  in  size  by  hewing  a  few  more 
inches  out  of  the  native  rock,  and  hence  their  Hall  was  up 
wards  of  170  feet  in  length  by  about  169  in  width.  The 
Hall  of  Ictinus,  including  both  compartments,  was  ap 
parently  of  the  latter  dimension  both  ways.  That  it  had 
a  partition  wall  is  made  probable  by  the  necessity  of  respect 
ing  what  Cimon  built,  and  still  more  nearly  certain  by  the 
survival  in  the  Roman  plan  of  two  front  doors  of  entrance. 
These  doors  were  necessary  before  they  remodelled  the 
interior  and  removed  the  partition.  Afterwards  one  large 
one  would  have  been  in  every  way  more  effective  and 
useful,  but  so  serious  a  departure  from  the  original  plan 
was  not  made.  Nothing  is  definitely  known,  however,  of 
what  and  where  were  the  doors  used  in  Greek  days.  Since 
the  Hall  of  Ictinus  had  this  partition,  its  lower  story  would 
not  have  been  suited  for  the  largest  meetings.  Hence  the 
necessity  of  an  upper  story  where  the  space,  divided  below, 
was  thrown  into  one  large  square  room.  In  this  room, 
entered  from  the  upper  terrace  where  lay  the  larger  Temple 
— supposably  that  of  Demeter — we  imagine  services  to  have 
been  held  in  the  greatest  days  of  Athens. 

There  is  no  room  for  doubt  as  to  the  existence  of  an 
upper  story,  because  Plutarch's  testimony  is  explicit,1  and  is 
confirmed  by  various  features  of  the  site.  The  numerous 
columns  of  the  lower  hall  require  it,  as  before  said;  and 
moreover  a  carefully  wrought  terrace,2  hewn  out  of  the 

1  See  above,  note  i,  p.  191.  -  Plutarch's 


204  THE  GODS  AT  E  LEU  SIS  v 

rock  high  up,  on  the  level  probably  with  its  ceiling,  calls 
for  it.  Ictinus,  following  no  doubt  the  practice  used  in 
building  the  Hall  of  Cimon,  utilised  the  receding  con 
figuration  of  the  rock-hill,  for  the  first  story,  by  hewing 
from  a  point  high  enough  above  the  level  of  the  Hall  of 
Pisistratus  to  give  his  ground  floor  sufficient  height,  and 
thus  placed  its  north  and  west  corners  in  the  solid  rock ;  he 
was,  in  beginning  again  from  that  high  point,  so  near  the 
rounding  top  of  the  hill  that  he  could  easily  cut  out  this 
upper  ledge  from  which  to  step  into  the  second  story  of  his 
great  hall.  To  this  ledge  broad  rock -cut  stairs  lead  up 
from  a  point  near  the  western  angle  of  the  lower  hall,  while 
narrower  steps  are  found  at  the  north  angle.  Mount  these 
stairs  and  you  are  on  the  artificial  terrace  just  described, 
and  find  that  it  is  a  continuation,  beyond  the  outer 
walls  of  the  Hall,  of  the  floor  of  the  upper  story.  More 
than  the  fact  of  its  existence  at  the  level  of  this  still 
existing  ledge  we  do  not  know  of  Ictinus'  upper  chamber  or 
hall,  except  that  Metagenes  of  Xypeta,  who  fashioned  the 
outside  ledge,  set  up  the  upper  story  columns.  There  is, 
of  course,  very  little  doubt  that  a  similar  upper  chamber 
existed  in  the  Hall  of  Pisistratus,  all  of  whose  essential 
features  were  reproduced  four  times  as  large  by  Ictinus, 
who  here,  as  at  Bassae,  included  the  site  of  the  ancient 
sanctuary  in  his  new  building.  At  Bassae  the  old  temple 
of  Apollo,  facing  east,  was  incorporated  by  Ictinus  into 
the  great  Temple  facing  north  and  south.  The  configura 
tion  of  the  Eleusinian  Acropolis  did  not  force  the  architect 
to  make  any  change  in  orientation,  and  Ictinus  faced  his 
hall  south-east  just  as  the  smaller  hall  of  Pisistratus,  its 
predecessor,  had  faced.  Only  by  a  skilful  use  of  new  space 
and  of  the  upward  slope  of  the  rock  Ictinus  made  his  Hall 


v  THE  PORTION  OF  HADES  205 

look  far  longer  in  proportion  to  its  width  than  it  really  was, 
and  therefore  his  building  assumed  something  more  near  the 
ordinary  proportions  of  such  a  temple  as  the  Parthenon. 
The  device  above  alluded  to  was  the  suggestion  that  the 
back  of  the  building  was  hidden  in  the  rock,  which  was  in 
fact  to  some  extent  the  case.  Something  in  the  finished 
surface  of  a  vertical  course  of  rocks  along  the  north-west 
or  Acropolis  side  of  the  raised  ledge  (or  Sid^w/jia),  already 
talked  of,  makes  it  seem  probable  that  this  impression  of  a 
temple  to  the  gods  of  nether  earth,  partly  hid  by  that  jealous 
element  and  partly  showing  in  the  light  of  day,  was  still 
further  enhanced  by  the  suggestion  of  a  portico  towards  the 
Acropolis.  It  is  conceivable  that  the  roof  of  the  upper 
story  should  have  covered  more  than  the  space  above  the 
lower  hall,  by  the  extent  of  this  ledge.  Thus  would  the 
demands  of  symmetry,  always  listened  to  by  Greek  artists, 
and  heeded  by  none  more  than  by  the  architect  of  the 
Parthenon,  have  been  satisfied.  This  single  story  would 
have  looked  like  the  top  of  a  larger  porch  whose  lower 
parts  were  concealed.  This  would  have  been  such  a 
counterpoise  to  the  porch  of  Philo  on  the  opposite  or  front 
end  as  was  allowed  by  the  requirements  of  the  ritual  and 
the  configuration  of  the  precinct  of  the  gods  at  Eleusis. 
Other  considerations  of  proportion  make  this  notion  that  a 
porch  was  suggested  at  the  back  seem  plausible  to  me. 
One  of  these  is  that  the  space  between  the  front  wall  of  the 
initiation  chamber  and  the  front  of  the  colonnade  of  Philo's 
porch  is  just  about  the  width  of  the  upper  ledge  extending 
from  the  back  wall  of  the  same  chamber  (the  wall  hewn  out 
of  the  rock)  to  the  point  where  the  hill  unhewn  resumes  its 
natural  upward  slope.  The  second  argument  from  propor 
tion  is  simply  the  Greek  habit,  and  especially  that  of  Ictinus, 


206  THE  GODS  AT  ELEUSIS  v 

which  would  seem  to  have  abhorred  undisguised  and  un 
blushing  squareness  in  a  building.  Ictinus  was  willing  to 
increase  at  Bassae  the  difference  between  length  and 
breadth  which  he  had  made  so  markedly  in  the  Parthenon 
(as  compared  with  previous  buildings),  and  there  was  still 
more  reason  for  him  to  do  something  of  the  kind  at 
Eleusis,  since  the  total  height  of  Philo's  portico — some 
equivalent  to  which  was  contemplated  in  Ictinus'  plan — 
must  have  been  that  of  the  two  stories  behind  it.  This 
must  have  seemed  to  come  dangerously  near  l  the  dimen 
sions  of  the  hall  itself.  Therefore,  without  some  such 
device  he  would  have  had  a  building  of  very  obviously 
monotonous  proportions,  not  far  from  a  cube  in  the  effect 
of  its  mass  upon  the  eye.  With  the  aid,  however,  of  Philo's 
porch  and  of  the  corresponding  porch  suggested  at  the  back, 
the  whole  building  viewed  from  without  gains  the  appearance 
of  a  lofty  structure  about  half  as  long  again  as  it  is  wide.  The 
back  of  this  building  showed  only  indistinctly  because  its 
base  was  wholly  entangled  and  involved  in  the  hill  out  of 
which  the  front  portion  had  completely  disengaged  itself, 
standing  forth  with  its  portico  of  fourteen  lofty  columns. 
Nether  earth  clung  to  her  fraction  of  this  Hall  of  the 
Eleusinian  gods,  just  as  she  claimed  her  fraction  of  fair 
Persephone's  life. 

Such  then  was  the  Telesterion,  the  great  house  of  wor 
ship  at  Eleusis,  the  enlarged  and  embellished  substitute 
for  that  once  built  according  to  the  commands  given  by 
Demeter  to  Triptolemus,  and  to  his  father  and  brethren. 

1  I  do  not  of  course  mean  to  suggest  the  possibility  of  its  really  having 
been  even  half  as  high  as  the  hall  was  long.  But  the  height  was  of  course 
exceptional,  and  to  it — so  far  as  the  eye  of  one  approaching  Philo's  porch, 
or  the  building  as  Ictinus  left  it,  was  concerned — had  to  be  added  that  of 
steps  leading  up  to  it. 


v  GREATER  AND  LESSER  MYSTERIES  207 

She  commanded  that  it  be  built  above  the  well — Callichoros 
by  name — where  Celeus'  daughters  found  her  sorrowing  as 
she  sat.  This  well  and  the  Agelastos  Petra  or  Laughless 
Stone  close  by  it,  upon  which  the  goddess  sat,  cannot  now 
be  found.  Such  was  not  the  case  in  ancient  days,  as  is 
shown  by  the  following  record  found  in  one  of  the  inscribed 
temple  accounts  unearthed  upon  the  spot. 

"  Paid  for  the  transport  of  25,000  bricks  to  the  Eleusinian 
temple  in  Athens  from  the  (Agelastos)  Laughless  Stone, 
120  drachmae."  From  these  same  accounts  it  further 
appears  that  the  Rarian  plain — memorable  because^  it  was 
there  that  Triptolemus  first  taught  men  the  cultivation  of 
grain — was  owned  and  farmed  by  the  Eleusinian  temple 
administration.  One  entry  shows  that  the  grain  from  this 
field  came  to  the  priest  at  Eleusis,  and  another  entry  runs  : 
"  To  Nicon,  for  removing  a  corpse  from  the  plain  and 
for  its  purification."  An  especial  item  allowed  in  this 
account  is  for  a  pig  used  in  the  cleansing  or  purification 
just  mentioned.  The  account  above  alluded  to,  of  25,000 
bricks  transported  from  Eleusis  to  the  Eleusinian  precinct 
in  Athens  at  the  expense  of  the  administrators  of  the 
sanctuary  at  Eleusis,  indicates  that  the  two  holy  precincts 
were  under  one  management,  and  recalls  the  story  of  the 
institution  of  the  Lesser  Mysteries  at  Athens  and  their  close 
affiliation  with  the  Eleusinian  festival.1 

This  lesser  and  affiliated  festival  was  celebrated  at 
Athens  in  the  month  of  flowers  (February-March),  and  the 
Athenian  precinct  where  these  Lesser  Mysteries  took  place 
was  appropriately  called  the  Eleusinion.  Of  this  and  of  its 
two  temples,  one  of  Demeter  and  Persephone,  and  one  of 
Triptolemus,  I  have  already  had  occasion  to  speak,  and 

1  See  above,  ch.  iv. 


208  THE  GODS  AT  E  LEU  SIS  v 

also  of  the  statue  which  it  contained  representing  that  great 
religious  mediator,  Epimenides  of  Crete,  who  consecrated 
this  Eleusinion,  and  through  the  Lesser  Mysteries  made 
of  the  worship  at  Eleusis  a  bond  of  union  between  those 
old-time  rivals,  Athens  and  Eleusis.  To  the  former  and 
more  or  less  exclusively  Eleusinian  and  patrician  worship 
of  Demeter,  Persephone,  and  Hades,  a  new  and  popular 
feature  attached  itself  in  honour  of  the  Icarian  and  Athenian 
Dionysus,  and  the  yearly  visit  of  Dionysus  lacchus  to  the 
gods  at  Eleusis  was  made  a  cardinal  and  carnival  feature  in 
the  Eleusinian  Mysteries,  which  thus  became  a  pledge  of 
brotherhood  and  equality  for  all  the  tribes  of  Attic-born 
men.1 

The  Lesser  Mysteries  at  Athens  were  a  sort  of  preface 
to  the  greater  ones  of  Eleusis,2  and  the  time  of  their  cele 
bration  was  earlier  in  the  year.  The  Greater  or  Epoptical 
Mysteries  did  not  come  until  the  month  Boedromion 
(August-September),  six  months  later  than  the  flower-month 
(Anthesterion)  of  the  Lesser  Mysteries. 

The  ordinary  progress  of  initiation  was  as  follows  : — In 
the  flower  month  at  Athens  an  applicant  could  become  a 
mystes — a  novice,  let  us  say — by  participation  in  the  Lesser 
Mysteries  at  the  Athenian  Eleusinion.  Thus,  and  appar 
ently  only  thus,  was  a  man  qualified  to  take  part,  six  months 
later,  in  the  Greater  Mysteries  at  Eleusis.  But  even  then  he 
appeared  at  Eleusis  only  as  a  mystes  or  novice,  and  could 
not  join  in  all  the  acts  of  worship  or  see  all  the  ceremonial. 
After  a  year  had  elapsed,  however,  our  mystes  became  an 

1  Mannhardt  very  truly  says  that  the  inspiration  drawn  from  initiation 
at  Eleusis  by  the  noblest  spirits  of  Greece  has  illustrated  the  sanctuary 
there  as  a  place  where  humanity  made  one  of  its  most  ennobling  forward 
steps  (Posthumous  Essays,  "Demeter"). 

2  The  apxujj'  f3a<ri.\evs  with  four  commissioners  (eTri/xeA^rcu)  appears  to 
have  controlled  the  whole  cult  at  Athens  and  Eleusis. 


v  THE  GREATER  MYSTERIES  209 

epoptes,  and  as  such  saw  with  his  own  eyes  and  heard  with 
his  own  ears  all  that  the  Greater  or  Epoptical  Mysteries 
afforded.  The  religious  privileges  of  the  completely  initiated 
are  reached  at  Eleusis  by  two  qualifying  stages,  as  who 
should  say  by  baptism  at  the  Lesser,  and  confirmation  at  the 
Greater  Mysteries.1  The  vague  and  unprecise  terms  in  which 
the  full  ceremony  is  described  are  terms  of  sight.  The 
Epoptes  or  Viewer  is  said  to  have  Autopsy,  or  sight  with  his 
own  eyes  —  Real  Vision.  These  hints,  with  others,  such  as 
the  connection  between  showing  light  and  the  title  of  the 
leader  of  the  mystic  ceremonial  who  was  called  Hierophant, 
persuade  some  that  after  a  period  of  darkness  the  initiated 
saw  a  great  light. 

Little  as  we  know  of  the  unrevealed  ceremonial  that 
took  place  within  the  Eleusinian  precinct,  we  know  that 
enormous  numbers,  as  many  perhaps  as  30,000,2  gathered 
from  various  parts  of  Greece  went  from  Athens,  and  we 
know  something  in  detail  of  the  preliminaries  at  Athens  and 
of  the  observances  on  the  Sacred  Way  to  Eleusis.  The 
whole  festival  lasted  about  twelve  days.  Several  days  before 
it  began  there  was  a  preliminary  meeting  at  Athens.  Just 
one  day  before  it  there  was  also  at  Athens  the  day  of  puri 
fication.  All  the  mystae3  and  every  creature  and  thing 

1  Plutarch  in  his  life  of  Demetrius  Poliorcetes  (26)  says  of  Demetrius  :  r6re 
8'o$v  dvafcvyvvuv  els  ras  'Adrjvas  cypa-^ev  6'n  /SouXercu  ira.pa.yev6/ji.evos 


evdbs  /j.vr]drjvaL  /cai  rty  reXeT-rji/  (Liracrav  airb  T&V  fjuicp&v 
eTTOTTTLKijjv  trapa.\af3ew,  TOVTO  5"  ou  Otfjurov  ty  ovfte  yeyovbs  irpbrepov 
.  .  .  eTTUTrrevov  de  TouX(£%i0"ro»'  eviavrbv  dtaXelTrovres.  The 
intimidated  governors  at  Athens  did  not  dream  of  refusing  this  unparalleled 
demand,  but  resorted  to  a  juggle  with  their  calendar  that  the  letter  of  the 
sacred  law  might  be  observed. 

2  See  Herod,  viii.   65,  where  the  visionary  procession  on  the  Sacred 
Way  is  seen  Kovioprbv  •x.upeovra.  O.TT  'EXewnVos  ws  dvdp&v  /txdXtora  Kr)  rpiff- 


3  See  ibid,  the  account  of  the  freedom  of  all  to  be  initiated.     Dicaeus 
there  describes  the  Eleusinian  festival  for  the  benefit  of  king  Demaratus, 

P 


210  THE  GODS  AT  ELEUSIS  v 

that  was  to  play  a  part  in  the  great  ceremonial  underwent 
purgation  by  washing  in  the  sea. 

Sea-surges  dash  all  human  harms  away,  says  Euripides 
somewhere,1  expressing  a  belief  well-nigh  universal  in 
ancient  Greece.  Truly  the  sea  entered  into  Greek  worship, 
with  its  suggestions  of  infinite  space  and  calm,  of  limitless 
motion,  its  mighty  and  tumultuous  heart-beat.  At  Eleusis 
and  elsewhere  the  ever -sounding  sea,  whose  surge  still 
echoes  through  the  most  beautiful  and  pious  masterpieces 
of  the  tragic,  the  lyric,  and  the  epic  muse,  was  present  with 
worshippers  whose  frequent  footfall  reverberated  through 
the  precincts  and  the  dwellings  of  the  gods  in  Greece. 

The  first  two  days  of  the  Eleusinian-Athenian  festival 
were  spent  in  Athens  after  these  ceremonies  of  purgation. 

Solemn  preparations  were  there  and  then  completed  for 
the  great  ceremonial  procession  from  Athens  to  Eleusis 
along  the  Sacred  Way  and  through  the  sacred  gates  into  the 
precinct  and  its  Great  Hall  of  Initiation.  By  means  of  all 
this  pomp  Dionysus-Iacchos  was  associated  with  Demeter 
and  Persephone  at  Eleusis,  and  Dionysus  became  one 
of  the  gods  at  Eleusis,  under  the  name  of  lacchos 
which  was  chanted  by  the  mystae  all  through  the  day 
while  they  brought  him  to  Eleusis,  and  again  during  the  day 
spent  in  bringing  him  back  to  his  home  in  the  Athenian 
laccheion,2 — within  the  Eleusinion  already  much  spoken  of. 

Underlying  all  the  light-heartedness  shown  by  those  who 

saying  :  TT^V  8£  opTyv  ralJTrjv  &yov<ri  'Adyvaiot  dva  irdvra  Urea  rr/  M.r)rpl 
Kal  TTJ  Koi/p?7  (the  regular  phrase  for  Demeter  and  Persephone),  Kal 
avruv  re  6  j8ouA6/uej'os  Kal  TWV  &\\uv  'EXXTyj'WJ'  /iveiTcu'  Kal 
TT}v  (fiuvriv  r?7S  a/comets  kv  ratr-fl  ry  bprrf  laKxdfrovffi.  This  last  word 
describes  the  cry  "  lacch ',  Oh  lacchos,"  and  thus  brings  into  prominence 
the  part  of  Dionysus  in  these  yearly  observances. 

1  /.  T.  1193. 

2  Plutarch,  Arist.  27.     See  above  the  citations  from  Herod,  viii.  65. 


v  YEARLY  PROCESSION  OF  THE  MYSTAE          211 

joined  this  procession  was  an  incommunicable  solemnity 
shadowed  forth  in  that  strangely  awe-inspiring  chapter 
where  Herodotus  tells  of  a  vision  of  floating  dust  and  of 
echoing  cries  from  a  ghostly  choir  of  disembodied  celebrants 
on  the  Sacred  Way  from  Athens.1  This  host  from  the 
world  beyond  led  lacchos  to  the  rescue  at  a  time  when 
Attic  sanctuaries  had  been  devastated  and  Athenian  altars 
overturned,  when  none  dared  longer  to  walk  in  the 
deserted  streets  of  Athens  or  visit  her  ruined  temples.  By 
this  portentous  apparition  the  doom  of  the  Persian  invader 
was  foreshadowed  and  the  coming  of  a  brighter  day  was 
assured  by  the  gods  at  Eleusis.  They  sent  forth  in  the 
darkest  hour  of  danger  and  despair  a  rescuing  band  of 
hope  to  lead  fainting  Hellas  from  martyrdom  to  peace. 
After  the  dust  and  the  sound  as  of  many  voices,  a  mist  arose 
and  floated  off  towards  Salamis,  foreboding  the  destruction 
of  the  Persian  fleet.  Thus  in  the  eyes  of  Herodotus — a  fit 
spokesman  here  as  elsewhere  for  all  the  faithful — the  Holy 
Alliance  at  Eleusis  of  Demeter,  Persephone,  and  Dionysus 
— the  Mother,  the  Daughter,  and  the  Son — was  the  com 
forter  and  the  saviour.  Through  it  were  assured  knowledge 
and  maintenance  of  Greek  laws  and  religion,  progress  in 
learning,  and  union  of  heart  with  all  the  divinities  of  Hellas.2 
No  wonder  then,  if  the  yearly  procession  of  the  living 
mystae  was  often  thought  of  as  a  foretaste  of  the  life  beyond, 
a  dim  vision  of  happiness  to  be  hereafter  in  the  islands  of 
the  blest,  a  rehearsal  or  promise  in  this  world  of  the 
performance  in  the  world  to  come.  No  wonder  that 

1  The  first  reading  suggests  that  Herodotus  is  thinking  of  a  procession 
from  Eleusis  to  Athens  ;  that  is  not,  however,  necessarily  the  case.      For 
the  passage,  see  above,  note  2,  p.  209. 

2  rbv  rpo(f>{a,  rbv  ffWTrjpa  di'  8v  eldov  v6fjiovs"E\\r)i>as, 

0eoij. 


212  THE  GODS  AT  ELEUSIS  v 

Aristophanes  puts  away  for  a  moment  his  cap  and  bells 
when,  having  brought  down  into  the  world  below  his 
caricature  of  the  god  Dionysus  accompanied  by  Xanthias,  a 
type  of  the  boisterous  clown  in  old  comedy,  he  suddenly 
confronts  these  two  jesters  with  the  march,  the  music,  and 
the  song  of  a  mystic  chorus  of  the  initiated,  who  are  repeating 
in  the  world  below  the  yearly  procession  from  Athens  to 
Eleusis.  They  are  bringing  home  the  god  lacchos.  Here 
is  a  striking  and  unstudied  homage  paid  to  the  solemnity 
of  Eleusinian  worship  in  the  sudden  cessation  of  boisterous 
fooling  at  the  approach  of  the  mystae.1  Breathless  and  all 
in  a  tremor  they  finally  hear  the  mystic  cry  from  afar : 
"  lacch',  oh  lacchos  !  lacch',  oh  lacchos  !  "  2  Then  they 
know  that  the  band  of  the  faithful  is  coming,  and  are 
abashed,  and  for  the  first  time  they  hold  their  peace. 

Meanwhile  the  mystae  draw  near  and  enter  the  orchestra 
with  a  song  to  the  god  in  their  midst : 3  "  Stir  thou  the  fire- 
flakes  of  torches,  whirling  them  with  thy  hands,  lacch',  oh 
lacchos,  fire-bearing  star  of  night  and  of  our  mystic  rites. 
Look,  the  meadow  is  aflame  with  fire ;  old  men's  knees  are 
lithe  for  dancing  now,  they  shake  off  all  their  pains  and  all 
the  time-long  weariness  of  hoary  years  in  the  rites  of  holy 
observance ;  but  thou,  flaming  with  thy  torch,  lead  on  the 
forward  march  to  the  blossoming  meadows  by  the  stream." 

Here,  as  a  preface  to  the  solemn  invocation  of  each  of 
the  three  gods  at  Eleusis  in  turn,  the  Hierophant  bids  the 
profane  and  uninitiated  to  depart :  "I  forbid  them,  I  forbid 
them  again,  and  again  a  third  time  I  forbid  them ;  let  them 
make  way  for  the  initiated."  Turning  to  the  latter,  he  then 

1  Aristoph.  Frogs,  312,  cf.  154  ff. 

2  Ibid.  316-459. 

3  Only  the  substance  is  here  given,  except  where  every  word  has  its 
important  bearing. 


v  THE  " FROGS"  OF  ARISTOPHANES  213 

says :  "  Raise  ye  the  voice  of  song,  begin  your  night-long 
revels  that  beseem  the  festival  we  keep." 

Then  follows  a  solemn  processional  song  in  honour  of 
Persephone,  which  is  full  of  the  cheer  of  glad  spring. 
Manfully  each  is  advancing  towards  the  flowering  nooks 
of  fair  meadows,  dancing,  gibing,  frolicking,  and  railing 
cheerily.  For  verily  each  has  had  his  fill  of  fasting  and 
purification.  "  March  on  in  cadence,  and  take  care  to  exalt 
right  heartily  the  saviour  goddess  with  your  voice  in  song 
uplifted,  for  she  it  is  who  saith  that  she  is  the  country's 
salvation  forever  assured." 

Now  the  Hierophant  bids  them  invoke  and  give  thanks 
to  the  harvest-queen  Demeter,  and  thus  their  song  begins : 
"  Demeter,  queen  of  hallowed  services,  join  now  in  help  of 
thine  own  to  save  them.  Suffer  me  to  dance  the  livelong 
day  in  unendangered  jollity,  and  let  me  utter  much  in  jest, 
and  much  in  earnest  too.  Make  my  doings  worthy  of  thy 
festival,  and  when  my  frolic  hour  is  past,  and  all  my  fooling 
done,  victorious  let  me  crown  my  brow." 

And  now  the  Hierophant  calls  for  a  song  in  honour  of 
the  last  of  the  three,  the  last-come  god  at  Eleusis,  who  is 
first  in  the  hearts  of  those  whom  he  has  led  and  who  have 
brought  him  to  the  Eleusinian  merry-making  mysteries. 
"  lacchos,"  they  fervently  sing,  "most  precious  to  my  heart, 
make  this  the  sweetest  moment  of  the  feast,  follow  along 
with  me  to  the  dwelling  of  the  goddess,  and  show  thee  a 
stranger  to  weariness  though  long  is  the  journey  that  thou 
art  making.  Oh  lacchos,  lover  of  the  dance,  come  thou  with 
me  to  help  me  on  the  way.  Yea,  in  merrily  tattered  garb 
with  thy  sandals  recklessly  torn  thou  canst  discover  the 
way  to  let  us  dance  and  play  and  pay  no  penalty.  Oh 
lacchos,  lover  of  the  dance,  come  thou  with  me  to  Jielp  me  on 


2i4  THE  GODS  AT  ELEUS1S  v 

the  way.  For  verily  I  gave  a  sidelong  glance  at  a  bit  of  a 
girl  just  now,  and  through  a  rent  in  her  bosom's  array  I 
caught  sight  of  my  beauteous  playmate's  charms.  Oh 
lacchos,  lover  of  the  dance,  come  thou  with  me  to  help  me  on 
the  way" 

The  unruly  element,  associated  chiefly  with  a  certain 
phase  of  the  cult  of  Dionysus,  has  asserted  itself  more  and 
more  in  this  last  song,  and  at  the  end  it  reaches  such  a 
pitch  of  license  that  the  two  jesters  quite  recover  their 
balance  of  mind,  while  the  chorus  of  the  initiated  yields 
wholly  to  a  headstrong  impulse.  Trusting  in  lacchos  for 
impunity,  they  fling  wide  the  floodgates  of  ribaldry,  raining 
alike  upon  the  unjust  and  the  just  their  jibes,  sacred  and 
profane,  mentionable  and  unmentionable.  "Now  if  you 
choose,"  they  say,  "  let  us  join  one  and  all  in  scoffing  at 
Archedemus.  Why,  he  has  lived  to  be  seven  years  old 
before  he  cut  a  single  Athenian  grinder,1  you  know,  but 
still  he's  in  business  in  the  demagoguing  line  up  in  the  world 
among  the  living  corpses, — a  captain  in  the  knavery  of  the 
world."2  And  so  the  mud-throwing  goes  merrily  from  bad 
to  worse  and  worst,  giving  a  wonderfully  telling  and  ideally 
realistic  picture  of  scenes  that  were  yearly  enacted  by  the 
real  procession  from  Athens  to  Eleusis.  This  feature  in 
the  day's  doings  was  connected  with  a  bridge  over  the 
Attic  river  Cephissus.  Just  out  of  the  gates  of  Athens, 
just  after  various  solemn  preliminaries  at  the  city  shrines 
had  ended,  the  mystae  halted,  and  took  their  revenge  for 
days  of  purification  and  fasting.  Here  they  let  their  pent- 
up  jollity  have  its  full  fling,  and  these  jibes  at  the  bridge 
have  also  to  take  their  part  in  the  netherworld  celebration 
described  by  Aristophanes. 

1  O$K  tyvtre  (pparepas. 


v  THE  MEASURE  LED  BY  THE  FATES  215 

At  last  there  is  a  pause,  the  Hierophant  bids  them  re 
sume  their  march,  and  gradually  as  they  go  the  thought  of 
the  woes  of  Demeter  shows  itself  first  in  their  choice  of 
words,  and  finally  in  the  return  of  serious  thoughts  and 
solemn  aspirations*  which  form  the  pious  burden  of  their 
closing  strain.  "  Onward  we  go  to  the  flower-faced  meadows 
where  abundant  roses  grow.  On  we  go,  in  our  own  merry 
fashion  dancing,  dancing  more  than  well l  the  measure  led 
off  by  the  glorious  Fates.  On  us  alone  in  very  truth  the 
sun  doth  shine,  we  only  know  the  light  of  gladness,  as 
many  of  us  as  have  passed  through  the  rites  of  the  Mysteries, 
and  lead  our  lives  in  piety  among  the  native  born  and 
the  strangers  within  our  city's  gates." 

Here  we  have  a  case  where  all  the  essentially  religious 
features  of  the  yearly  holiday-making  in  honour  of  the  gods 
at  Eleusis  were  enacted  before  the  eyes  of  those  who  under 
ordinary  circumstances  would  yearly  take  part  in  the  cele 
bration  themselves — the  whole  Athenian  public.  But  the 
circumstances  were  extraordinary ;  and  although  the  scene 
is  in  a  comedy,  there  is  throughout  a  pervasive  seriousness 
which  alone  would  require  us  to  assign  a  religious  motive 
for  its  performance — that  motive  was  the  satisfaction  of  a 
fervent  desire  to  propitiate  the  gods  at  Eleusis  felt  uni 
versally  at  Athens.  The  play  was  brought  out  in  January 
of  the  year  405,  just  after  the  dearly  bought  victory  of 
Athens  at  Arginusae.2  Ever  since  the  Spartans  had  taken 
the  advice  of  the  exiled  Alcibiades  and  maintained  a  garri 
son  in  Attica  at  Decelea,  the  merriment  of  the  yearly 

1  rbv  Ka\\LXopuTa.Tov  (rpdirov]  has  an   unmistakable  reminiscence  in 
meaning  and  sound  of  the  KaXXixopov  (fiptap,  where  Demeter  sat  until 
she  was  comforted  there  by  dances.      Homer,  Hymn  to  Demeter,  272  ; 
Pausanias,  I.  xxxviii.  6. 

2  See  Kock's  introduction  to  his  edition  of  the  Frogs. 


2i 6  THE  GODS  AT  ELEUSIS  v 

procession  to  Eleusis  had  been  greatly  interfered  with,  even 
though  it  be  supposed  to  have  had  the  protection  of  a 
sacred  truce.1  Thus  at  a  time  when  the  Athenians  were  in 
a  gloomy  and  thoroughly  discouraged  state  of  mind,  the 
religious  consolations  and  assurances  of  salvation  gained  by 
the  normal  celebration  at  Eleusis  and  the  processions  before 
and  after  it  were  so  curtailed  that,  instead  of  expecting  help 
from  Eleusis,  the  pious  Athenian  must  have  feared  that  the 
Eleusinian  Alliance  of  gods  was  offended  by  years  of  com 
parative  neglect.  Then  it  was  that  Aristophanes,  inspired 
perhaps  by  the  story  recorded  in  Herodotus  of  a  visionary 
celebration  held  by  the  departed  in  the  upper  world,  hit 
upon  the  idea  of  having  something  like  it  celebrated  by  his 
living  mimes  in  a  stage  counterfeit  of  the  world  below. 
The  resulting  success  of  his  play  was  overwhelming  and 
unexampled.  The  Frogs  was  acted  a  second  time  without 
alteration  that  same  year,  and  over  and  above  all  the 
ordinary  marks  and  rewards  of  victory,  the  poet  had  given 
to  him  a  branch  of  the  sacred  Athenian  olive-tree.  This 
last  honour  was  a  rare  and  high  mark  of  popular  gratitude 
for  great  help  to  the  State,  paid  afterwards  to  Thrasybulus. 
It  was  equivalent  to  the  public  bestowal  of  a  crown  of  gold, 
which  fell  to  the  lot  of  Demosthenes.  Only  the  propitiation 
of  the  favour  of  the  gods  at  Eleusis,  a  thing  quite  independ 
ent  of  the  merits  of  the  play,  could  have  warranted  such  a 
mark  of  favour  to  Aristophanes. 

To  his  audience  the  presentation  which  Aristophanes 
gave  of  the  procession  and  the  rites  before  and  after  up  to 
the  moment  when  the  real  and  most  unutterable  mysteries 
began  was  in  fact  eminently  consoling,  for  the  poet,  without 

1  The  willingness  of  Sparta  to  interfere  with  Eleusinian  worship  had 
been  amply  shown  by  the  conduct  of  Cleomenes  referred  to  above. 


v  STATIONS  OF  THE    YEARLY  PILGRIMAGE        217 

dwelling  upon  details  known  familiarly  to  all,  reproduced 
the  spirit  and  the  truth  of  the  observance.  Indeed  as  the 
mystae  leave  his  stage  they  are  proclaiming  the  justice  and 
loving-kindness  of  Athens  to  all  within  her  gates.  In  order 
to  complete  in  detail  the  picture  given  by  our  poet,  certain 
facts  familiar  to  his  audience,  or  else  presented  to  them 
visually  on  the  stage,  must  be  rehearsed  and  heeded.  Some 
of  them  are  as  follows :  The  Hierophant  or  leader  and 
marshaller  of  the  procession  had  other  names  such  as 
laccha^ogos,  the  Vauntcourier  or  Leader  of  lacchos. 
Through  the  Holy  Gate  they  passed  crowned  with  parsley 
and  with  ivy  where  fruits  were  intertwined.  In  their  hands 
they  carried  Bacchus'  lighted  torches  or  else  Demeter's 
sheaves,  and  thus  their  mere  array  was  eloquent  of  har 
mony  between  the  goddess  and  the  god.  Many  were  the 
stations  required  by  immemorial  custom  for  this  procession. 
Harvest  usages  observed  in  the  intervening  villages 
naturally  grouped  themselves  around  the  passage  of  these 
pilgrims,  who  formed  the  annual  escort  of  the  farmer's  god, 
Dionysus,  to  Eleusis.1  Before  the  Attic  Cephissus  was 
reached  the  district  of  Lakkiadae  required  a  pious  pause. 
Phytalus  there  had  played  host  to  Demeter,  and  his  reward 
had  been  the  gift  of  figs.  With  jibes  and  jollity  the 
Eleusinian  band  crossed  the  memorable  bridge  and  ap 
proached  the  altars  of  most  gentle  Zeus  beyond  the 
Cephissus.  Reminiscences  of  heroic  Theseus  detained 
them  then,  and  after  this  they  halted  at  the  shrine  of 
Cyamites,  giver  of  beans.  Much  as  the  Bean -giver  was 
here  revered,  the  beans  which  he  gave,  and  all  beans,  were 
strictly  excluded  from  the  Eleusinian  precinct  of  Demeter. 

1  For  a   more   complete  account   see   F.   Lenormant's  La  -vote  sacrle 
already  alluded  to. 


218  THE  GODS  AT  ELEUSIS  v 

This,  says  Pausanias,  is  a  mystery  known  to  those  initiated 
at  Eleusis.1 

In  the  pass  where  stands  the  modern  cloister  and  church 
of  Daphne  they  stopped  at  a  temple  originally  Apollo's,  but 
where  later  on  Demeter  and  Athena  shared  the  sanctuary 
with  Apollo.  Then,  just  after  a  forward  glimpse  of  the  bay 
of  Eleusis,  halt  was  made  at  Aphrodite's  temple,  and  the 
tomb  of  Eumolpus  was  reached.  This  great  Eleusinian 
hero  welcomed  them  and  lacchos  to  the  Eleusinian  plain, 
all  the  more  heartily  perhaps  because,  like  Dionysus  himself, 
he  was  from  Thrace.2 

A  whole  book  was  written  in  antiquity  by  Polemon  on 
this  processional  progress,  and  Pausanias  repeats  from  this 
source  many  interesting  details,  willing  all  the  more  to  give 
information  on  the  preliminaries  of  the  great  Eleusinian 
festival  because  divine  warning  has  sealed  his  lips  about 
the  Mysteries  themselves. 

It  does  not  suit  the  present  theme,  however,  to  dwell 
further  upon  details ;  enough  has  already  been  said  to 
show  how  great  a  complexity  of  ritual,  what  an  enormous 
variety  of  local  customs,  attended  the  annual  progress  of 
Dionysus  to  Eleusis.  If  the  ceremonial  used  upon  his 
arrival  were  known  to  us,  we  should  doubtless  marvel  still 
more  at  the  power  of  growth  and  of  fusion  inherent  in  the 
local  religions  of  Attica,  at  the  way  in  which  Demeter  and 
Persephone  tamed  the  wildness  of  Thracian  Dionysus,  while 
all  three  counteracted  the  bloodless  gloom  of  Hades  and 
were  united  with  him  at  Eleusis  in  the  time-honoured 
observance  of  eight  hundred  years. 

1  The  bean  seems  by  long  familiarity  to  have  fallen  into  contempt,  so 
that  we  no  longer  shudder  at  it,  and  are  only  amused  at  Pythagorean 
scruples  which  led  men  to  die  rather  than  pass  through  a  field  planted 
with  this  tragic  vegetable.  2  Strabo,  VIII.  vii.  i  (383). 


VI 
AESCULAPIUS  AT  EPIDAURUS  AND  ATHENS 

ONE  of  the  great  features  of  the  Greater  Mysteries  com 
memorated  the  Eleusinian  initiation  of  the  god  Aescul 
apius.  In  the  fabulous  past  this  god  of  healing  had  crossed 
the  Saronic  gulf  and  associated  himself  and  his  Epidaurian 
worship  with  Athens  and  Eleusis,  and  this  mythical  arrival 
prefigured,  as  it  were,  the  introduction  of  Aesculapian  wor 
ship  at  Athens,  and  the  renown  of  his  Athenian  shrine 
founded  from  Epidaurus  in  historical  times. 

In  the  days  of  his  widest  influence  Aesculapius,  the  god 
of  healing,  was  looked  upon — in  spite  of  various  records 
making  him  a  son  of  Zeus — as  the  son  of  Apollo.  So  com 
pletely  was  he  associated  at  one  time  or  another  with  his 
father  Apollo  Epicures  or  Epicurios, — the  supporter  of 
health,— that  we  may  if  we  choose  look  upon  him  as 
Apollo's  plenipotentiary  in  the  comparatively  late  legend 
that  connects  him  with  Eleusis.1  A  tie  of  more  than  com- 

1  The  influence  of  Pythagoreanism  upon  the  beginnings  of  medicine 
is  not  less  abundantly  proven  than  the  close  tie  that  bound  Pythagoras 
and  his  school  to  Apollo.  Seven  of  the  Alexandrine  doctors  (see  Darem- 
berg's  list)  bore  the  name  Apollonius.  This  denoted  one  whose  spirit 
was  under  Apolline  guidance.  Such  a  son  of  the  healing  spirit  of  Apollo 
was  Aesculapius,  the  divine  exemplar  to  whom  Greek  doctors  looked  up. 


220      AESCULAPIUS  AT  EPIDAURUS  AND  ATHENS     vi 

mon  strength  seems  to  attach  him  to  Demeter  and  Perse 
phone,  for  he  is  associated  with  them  in  certain  bas-reliefs. 
His  connection  with  Dionysus  is  vaguer  by  far,  but  not  less 
real. 

Ultimately  it  may  be  possible  to  make  out  with  some 
clearness  the  precise  nature  of  this  tie  binding  Aesculapius 
to  the  gods  at  Eleusis.  But,  in  the  present  state  of  know 
ledge,  the  closest  scrutiny  of  Aesculapius  and  his  worship 
only  reveals  uncertain  associations  and  resemblances. 
Like  the  gods  at  Eleusis,  Aesculapius  was  not  recognised 
in  the  fulness  of  his  subsequent  godhead  by  the  Greeks 
of  Homer's  epoch.  Aesculapius,  like  those  same  gods, 
but  far  more  vaguely  and  uncertainly,  is  a  nature -god. 
Like  Dionysus  and  Demeter  he,  or  at  least  his  character 
istic  element,  came  to  an  ultimate  and  more  southerly 
birth-place  from  the  north.1  Perhaps  he  may  be  looked 
upon  as  a  netherworld  nature-divinity,  the  same  in  many 
respects  with  Dionysus,  but  without  his  tragic  intensity. 
To  this  residuum  add  something  of  the  Olympian  mildness 
of  Zeus,  and  you  have  a  being  who  may  with  equal  appro 
priateness  be  classed  with  the  netherworld  brother  of  Zeus 
or  be  called  a  son  of  Zeus.  Dionysus  came  from  Thrace 

In  later  days  Apollonius  of  Tyana  (see  Appendix  IV.  below),  the  favourite 
of  Aesculapius,  stood  for  the  same  Apolline  perfection.  In  prayers  and 
offerings  at  Aesculapius'  shrine  Apollo  was  commonly  named  first  and  then 
Aesculapius,  and  furthermore,  according  to  the  rule  of  Hippocrates,  every 
doctor  qualified  as  such  by  an  oath  ' '  in  the  name  of  Apollo  the  healer, 
and  of  Aesculapius,  of  Hygieia  and  Panacea,  and  of  all  the  gods  and 
all  the  goddesses." 

1  I  have  not  found  it  either  possible  or  advisable  to  go  into  the  theory, 
a  plausible  one,  that  the  Thracians  may  have  been  the  Pelasgian  pre 
decessors  of  what  we  call  the  Greek  civilisation,  and  that  we  should  not 
talk  of  an  invasion  from  the  north,  but  rather  of  a  survival  or  inheritance. 
According  to  this  view  the  most  ancient  traces  of  building  on  the  Acropolis, 
whether  of  Eleusis  or  of  Athens,  might  be  attributed  to  these  Pelasgo- 
Thracians. 


vi  NORTHERN  ORIGIN  OF  THE  GOD  221 

and  from  Macedonia  just  beyond  the  range  of  Mount 
Olympus,  Aesculapius  came  from  that  part  of  Thessaly 
which  is  closest  to  these  mountains,  and  Demeter's  Thes- 
salian  origin  was  from  Pyrasus  not  far  away.  At  Pyrasus 
was  Demeter's  first  home,  and  among  the  mountain 
tribes  near  by  Aesculapius  originated.  He  was  the  tribal 
god  of  the  half  mythological  people  called  Phlegyae  and 
Lapithae.  In  the  wake  of  northern  tribes  this  god  Aescu 
lapius — a  more  majestic  figure  than  the  blameless  leech  of 
Homer's  song — came  by  land  to  Epidaurus  and  was  carried 
by  sea  to  the  eastward  island  of  Cos.  With  him  perhaps 
was  borne  from  her  Thessalian  home  the  goddess  Demeter, 
who  found  her  Cnidian  shrine  not  far  from  the  Coan  home 
of  Aesculapius.  This  southward  journey  is  the  counterpart 
of  that  by  which  Dionysus  is  supposed  to  have  reached, 
with  bands  of  Thracian  invaders,  Attica  and  Boeotia  as  well 
as  his  island  home  on  Naxos.  Our  knowledge  of  these 
invasions,  Thracian  and  Thessalian,  is  so  misty  that  it  is 
well-nigh  absurd  to  attempt  to  say  which  preceded  the 
other,  or  indeed  to  maintain  with  any  vigour  that  all  these 
various  divinities  cannot  have  been  brought  in  by  one  and 
the  same  southward  movement  of  mountain  tribes ;  for  the 
boundary  line  separating  Thessalians  from  Macedonians, 
and  the  distinction — never  clear — between  Macedonians 
and  Thracians,  are  not  strictly  applicable  to  these  prehistoric 
days. 

Arrived  from  the  north,  Aesculapius  grew  in  importance 
with  the  growth  of  Greece,  but  may  not  have  attained  his 
greatest  power  until  Greece  and  Rome  were  one.  At  all 
events  every  stage  of  his  power  and  prestige  connects  itself 
so  closely  with  the  various  phases  of  secular  medicine  that, 
in  order  to  understand  the  results  of  recent  excavations  at 


222     AESCULAPIUS  AT  EPIDAURUS  AND  ATHENS     vi 

Athens  and  Epidaurus, — made  at  both  places  in  sanctuaries 
of  Aesculapius, — something  must  be  said  about  the  position 
of  Greek  doctors  and  the  history  of  Greek  medicine,  sacred 
and  secular.  This  last  distinction  was  certainly  not  made 
until  after  the  fabled  siege  of  Troy,  as  is  shown  by  the 
earliest  record  of  Greek  opinion  about  doctors  which  is  to 
be  found  in  the  Iliad.  One  of  the  sons  of  Aesculapius, 
Machaon,  was  seriously  wounded  in  a  melee.  When  he 
fell  disabled,  consternation  seized  the  Greeks  until  Nestor's 
timely  aid  was  invoked.  Nestor,  the  personification  of 
respectable  tradition  in  those  days,  bears  off  the  healer 
Machaon,  declaring  roundly,  as  he  does  so,  that  a  doctor  is 
far  better  worth  saving  than  many  warriors  unskilled  in 
leechcraft.  Plainly  a  doctor,  as  Nestor  understood  the 
word,  meant  not  a  secular  but  a  sacred  person,  and 
medicine  was  both  sacred  and  secular.  Here  is  a  half 
superstitious  and  wholly  generous  admiration  for  skill 
in  medicine  that  may  be  called  a  typically  Greek  senti 
ment  ;  since  it  never  has  died  out  in  Greece,  and  is  found 
intact  among  the  Homerically  simple-minded  peasants  of 
to-day.1 

It  will  eventually  be  necessary  to  analyse  this  typically 
Greek  sentiment  of  Nestor's  and  to  appeal  to  the  Homeric 
poems  at  large  about  doctors  and  medicine.  Thus  we  shall 
understand  on  the  one  hand  the  worship  of  Aesculapius  as 
a  wonder-worker,  and  on  the  other  the  non-miraculous  pro 
fessional  skill  possessed  by  Greek  doctors  who  pursued  the 
art  of  healing  and  perfected  the  science  of  medicine  inde 
pendently  of  the  god.  Let  us  begin  our  survey  at  a  time 
when  doctors  had  a  considerable  knowledge  of  medicine. 
This  unmiraculous  and  scientific  profession  may  be  traced 

1  See  Appendix  V.  on  the  status  of  Greek  doctors  in  modern  times. 


vi  HIPPOCRATES  AND  DEMOCEDES  223 

back  to  a  correspondingly  positive  and  unsuperstitious 
aspect  of  Homeric  medicine.1 

Let  us  consider  the  condition  of  Greek  medicine  as  it 
was  in  the  days  of  Hippocrates  of  Cos,  ordinarily  miscalled 
the  father  of  Greek  medicine.  A  casual  glance  at  Littre's 
complete  edition  of  all  works  handed  down  under  the  name 
of  Hippocrates  shows  that  among  them  are  monuments  of 
sound  medical  labours  carried  out  before  his  day.  Indeed 
Hippocrates  dealt  with  a  large  body  of  ascertained  medical 
facts,  and  Greek  medicine  was  far  advanced  when  he  began. 
To  make  this  apparent  we  need  only  consider  the  career  of 
Democedes  as  related  by  Herodotus. 

Two  generations  before  Hippocrates,  in  the  second  half 
of  the  sixth  century  B.C.,  this  Democedes  lived,  and  enjoyed 
an  Asiatic  renown,  equivalent  in  those  days  to  what  our 
doctors  call  a  European  reputation.  Starting  from  the  far 
west,  Croton  in  Italy,  where  Pythagoreanism  had  given  a 
great  impetus  to  the  study  of  medicine,2  the  alert  and 
ready-witted  Democedes  went  to  Aegina  and  distanced  all 
competitors  in  the  race  for  appointment  there.  This  was 
the  more  brilliant,  says  Herodotus,  because  he  was,  when  he 
entered  the  lists,  without  the  instruments  freely  used  by  his 
fellow-candidates.  A  year's  service  on  Aegina  as  public 

1  Daremberg,  Histoire  des  Sciences  Mtdicales,  I.  ch.  iii. 

2  The  influence  of  many  philosophers  may  be  traced  in  the  method  and 
opinions  of  Hippocrates,  but  probably  no  school  affected  the  beginnings 
of  medicine  so  much  as  that  of  Pythagoras.     See,  among  the  letters  at 
tributed  to  Apollonius  of  Tyana,  number  xxiii. ,    ' '  Pythagoras  said  that 
medicine  came  most  near  to  divinity,  and  inasmuch  as  this  was  the  case, 
medicine  should  care  for  the  soul  as  well  as  for  the  body;    or  else  the 
whole  living  being  would  fail  of  full  health  from  having  his  higher  element 
diseased."    The  Pythagorean  Alcmaeon  of  Croton  was  addicted  to  anatomy 
— he  dissected  animals  ;    and  these  studies  contributed  to  give  a  specially 
useful  bent  to  the  school  of  medicine  at  Croton,  of  whose  renown  Hero 
dotus  makes  admiring  mention.     It  must  be  remembered  that  Croton  was 
the  centre  of  Pythagoreanism. 


224      AESCULAPIUS  AT  EPIDAURUS  AND  ATHENS     vi 

practitioner  paid  by  the  state  so  increased  the  reputation  of 
Democedes  that  Athens  offered  him  the  same  duties  and  an 
increase  of  salary.  He  was  no  sooner  settled  at  his  work  in 
Athens  than  Polycrates,  the  too-fortunate  tyrant  of  Samos, 
succeeded  in  getting  him  by  doubling  his  salary.  A  call 
still  further  east  soon  came  to  Democedes, — a  promotion 
under  the  disguise  of  complete  disaster.  The  flood  of 
fortune  that  had  so  long  upborne  his  patron  Polycrates 
ebbed  suddenly  away.  Democedes,  captured  and  enslaved 
during  the  sack  of  Samos,  was  hurried  into  the  far  interior, 
to  the  palace  of  Darius.  Long  a  despised  and  unnoted 
captive,  he  was  at  last  terrified  by  a  summons  to  the 
king.  The  rest  of  the  story,  beginning  with  his  refusal 
to  acknowledge  his  own  skill,  repeated  itself  at  the  court 
of  the  Duke  of  Savoy  in  the  sixteenth  century,1  and  is 
travestied  in  one  of  Moliere's  best  farces.  Threats  over 
came  the  great  doctor's  scruples;  and  Darius'  sprain, 
which  had  only  been  aggravated  by  the  treatment  of  the 
accredited  Egyptian  doctors,  was  quickly  and  completely 
cured. 

The  story  of  Democedes'  career  proves  at  least  that 
Herodotus,  writing  and  living  early  in  the  days  of  Pericles 
and  Hippocrates,  believed  that  skilled  doctors  of  Greek 
training  had  been  in  request  as  such  for  a  century  and  a 
half  at  least.  To  show  how  advanced  was  the  condition  of 
medical  science  in  the  day  of  Herodotus,  there  is  all  manner 
of  undoubted  testimony.  Socrates  for  instance,  who 
was  only  eight  years  older  than  the  great  Coan  doctor, 
throws  light  upon  this  point  in  his  chaff  of  a  friend  named 
Euthydemus,  who  undertook  to  make  a  stir  in  the  world  by 
having  many  books.  "  Of  course  you  who  have  so  many 

1  See  Malgaigne's  Chirurgie  grecque  avant  Hippocrate, 


vi  THE  CO  AN  SANCTUARY  225 

books  are  going  in  for  being  a  doctor,"  says  Socrates,  and 
then  he  adds,  "  there  are  so  many  books  on  medicine,  you 
know."  Euthydemus  repudiates  this  inference  with  indigna 
tion.  Whatever  the  quality  of  these  books  may  have  been, 
their  number  must  have  been  great  to  give  point  to  this 
chaff.  Xenophon  is  nearly  contemporary  though  some 
what  later,  and  his  testimony  may  be  added.  The  liberal 
provision  of  medical  care  for  his  retreating  army,  the 
matter-of-course  way  in  which  the  most  suitable  remedies  get 
themselves  promptly  applied  on  occasion — all  this  tells  of 
an  established  system  of  military  practice,1  and  proves  again 
how  little  sense  there  ever  was  in  saying  that  Greek  medicine 
began  in  460  B.C.  with  Hippocrates  of  Cos.  There  can  be 
no  doubt  that  Hippocrates  was  not  bred  under  the  shadow 
of  the  great  Coan  temple  of  Aesculapius  for  nothing.  His 
own  writings  prove  that  he  heeded  well  the  lore  of  the 
priests  at  Cos.  For,  although  he  is  not  the  author  of  the 
compilation  made  from  materials  accumulated  at  the  Coan 
temple  and  included  among  his  supposed  work,  he  plainly 
used  that  compilation,  and  was  guided  by  the  traditions 
which  it  embodies.  It  is  equally  certain  that  he  gathered 
in  the  fruits  of  many  generations  of  zealous  labour  in  surgery, 
and  it  seems  possible  that  much  of  the  surgery  before  his 
time  had  been  developed  quite  independently,  without 
knowledge,  so  to  speak,  or  connivance  of  Aesculapius,  the 
god  of  medicine.  There  is  at  all  events  little  or  nothing  to 
show  that  Aesculapius  was  worshipped  in  Magna  Grecia 
and  at  Croton,  while  Democedes,  who  was  trained  there, 
certainly  was  specially  qualified  as  a  surgeon.  Malgaigne 

1  See  in  the  Gazette  Hebdomadaire  de  Mddecine  et  de  Chirurgie  for 
June  2oth  1879,  Dr.  Corlieu's  "  6tude  m^dicale  sur  la  retraite  des  10,000 
prec£d£e3de  considerations  sur  la  me'decine  militaire  dans  les  armies 
grecques." 

Q 


226     AESCULAPIUS  AT  EPIDAURUS  AND  ATHENS     vi 

goes  too  far  if  he  claims  that  Democedes  had  not  a  full 
acquaintance  with  the  remedies  in  use  when  he  flourished.1 
It  is  best,  however,  not  to  linger  over  the  question 
whether  or  not  the  surgery  known  to  Hippocrates  had 
become  as  intimately  associated  with  the  shrines  of  Aescu 
lapius  and  the  guild  of  the  Asclepiadae  as  the  other  tradi 
tions  and  practices  which  made  up  early  Greek  medicine.2 
One  thing  at  least  is  certain,  that  necessity  was  the  mother 
of  this  invention  \  it  had  a  secular  origin  in  constant  war 
fare.  To  the  bickerings  of  earliest  Greece  science  owes  a 
greater  debt  than  is  often  recognised.  The  fullest  record  of 
the  way  in  which  this  debt  was  incurred  is  found  in  the  Iliad^ 
that  poem  of  glorified  bloodshed.  Here  is  the  positive,  the 
secular,  the  scientific  aspect  of  Homeric  medicine.  There 
can  be  no  reasonable  doubt  that  all  the  minute  descriptions 
given  in  the  Iliad  of  wounds,  thrusts,  and  contusions  were 
listened  to  by  men  of  Homer's  time  with  a  breathless  interest. 
Everybody  finds  them  more  or  less  trying  now.  To  know  in 
one  of  these  battles  just  where  the  man  was  struck,  just  how 
far  the  weapon  went,  and  exactly  what  was  the  behaviour 
of  the  striker  and  the  struck  when  the  blow  was  given, 
seems  of  slight  interest  or  of  none  at  all  to-day.  The 
wonderful  thing  to  us  is  that  there  should  ever  have  been  a 

1  Malgaigne  and  Daremberg  are  at  variance  here.     See  Ch.  Darem- 
berg,  £tat  de  la  Mtdecine  entre  Hom&re  et  Hippocrate,  p.  52. 

2  For  an  astonishingly  unsubstantiated  claim  that  Democedes  and  the 
medical  school  of  Croton  were  absolutely  outside  the  domain  of  Aescu 
lapius  and  uninfluenced  by  his  worship,  see   Guardia's  La  Mtdecine  a 
travers  les  Sibcles.     This  book  would  perhaps  not  be  one  to  mention  if 
it  had  not  given  the  authority  of  its  writer's  scholarship  to  the  useless 
theory  that  in  ancient  as  in  modern  times  there  was  a  recognised  conflict 
between  "science  and  religion."     A  particularly  misleading  reproduction 
of  Guardia's  arbitrary  and  baseless  account  of  the  worship  of  Aesculapius 
and  its  relation  to  sound  medicine  can  be  found  in  an  essay  read  before  the 
Birmingham  Speculative  Club  by  Balthazar  W.  Foster,  M.D.,  and  pub 
lished  in  1870. 


vi  THE  MEDICINE  OF  HOMERIC  DA  YS  227 

popular  interest  in  these  slaughterous  minutiae  of  the  human 
frame.  But,  for  all  that,  Homer's  careful  accuracy  is  better 
art,  and  of  more  enduring  interest,  than  the  loose  and 
laughable  anatomical  absurdities,  the  braggart  atrocities  so 
frequently  admired  in  chansons  de  gestes  and  in  various 
utterances  of  the  age  of  medieval  chivalry.  The  tiresome 
minuteness  of  Homer  has  always  the  merit  of  accuracy  and 
truth.  Competent  judges  in  matters  medical  have  pro 
nounced  Homer  a  marvel  of  clearness  and  precision.  That 
his  account  should  be  trustworthy  was  absolutely  required 
by  his  hearers.  They  had  a  personal  knowledge  in  the 
matters  whereof  he  sang,  and  demanded  of  him  not  simply 
such  precision  as  they  could  attain  themselves,  but  hoped 
no  doubt  also  to  glean  from  his  descriptions  hints  for  future 
combat.  They  knew  anatomy  chiefly  that  they  might,  when 
fighting,  put  in  each  blow  where  it  would  do  the  utmost 
harm.  They  wished  to  kill  rather  than  to  cure ;  and  yet, 
like  the  heroes  in  whose  life  the  poet  mirrored  their  own, 
they  had  some  knowledge  of  surgery — enough  to  help  a 
wounded  comrade  in  danger  of  his  life. 

With  the  incentive  supplied  by  a  breathlessly  interested 
audience  it  is  not  surprising  that  Homer,  or  the  Homeric 
bards,  should  have  been  extraordinarily  painstaking  in 
matters  anatomical.  Among  scores  of  wounds  described  by 
him,  -only  seven,  it  is  said,1  are  given  so  vaguely  that  the 
skilled  anatomist  cannot  determine  very  nearly  where,  and 
in  some  degree  also  how  serious,  a  wound  is  meant.  As 
for  the  defects  in  Homer's  anatomy  they  are  few,  and  such 
as  may  more  fairly  be  cloaked  with  the  poet's  mantle  than 

1  For  a  competent  specialist's  account  of  the  facts  upon  which  this 
appreciation  of  Homeric  anatomy,  surgery,  and  medicine  is  based,  see 
Charles  Daremberg  in  the  Revue  Archeologique  for  September,  October, 
and  November,  1865.  "  Etudes  d'arche"ologie  me'dicale  sur  Homere." 


228      AESCULAPIUS  AT  EPIDAURUS  AND  ATHENS     vi 

the  shortcomings  discoverable  in  Ariiadis  de  Gaule,  or  even 
in  our  own  Spenser's  Faery  Queene. 

The  same  detailed  knowledge  of  anatomy  which  Homer 
possessed,  and  which  was  possessed  by  his  most  critical 
audiences,  is  very  naturally  attributed  to  his  great  heroes. 
To  be  convinced  of  this,  hear  Odysseus  when  a  desperate 
situation  prompts  the  thought  of  suicide.  He  does  not  think 
vaguely  of  self-destruction,  he  knows  the  exact  and  most 
vulnerable  spot  where  he  will  strike  himself ;  and  it  is  the 
same  when  he  has  the  giant  Polyphemus  in  drunken  sleep 
before  him.  You  can  fancy  a  warrior  of  Homer's  day 
teaching  his  son  by  Odysseus'  example  the  duty  of  knowing 
the  human  frame  in  every  least  detail.  You  can  fancy  the 
same  anxious  father  taking  the  miserable  case  of  Pandarus 
to  bring  home  to  his  boy  the  fatal  consequences  of  incom 
petence  and  inaccuracy.  Foolish  Pandarus  thought  that  a 
mere  shoulder-wound  inflicted  by  his  arrow  on  Diomede 
had  killed  him,  and  not  brought  him  merely  to  a  faint. 
Therefore  the  reappearance  of  Diomede,  after  recovery,  so 
unmans  this  ignorant  would-be  slayer  that  he  loses  nerve  and 
is  slaughtered  ignominiously. 

A  second  and  strong  impulse  to  this  minute  anatomical 
knowledge  of  Homer's  day  was,  as  already  said,  the  need 
of  such  knowledge  to  succour  a  wounded  comrade.  This 
further  involved  a  rough  knowledge  of  surgical  aids  and  of 
certain  simple  remedies.  A  good  man  of  war,  a  real  hero, 
was  bound  to  know  the  surgery  of  his  day.  Rough  and 
rudimentary  as  this  was,  it  involved  a  knowledge  of  bandag 
ing,  the  respectability  of  which  is  proved  from  early  pictures 
representing  the  process.  Combined  with  this  heroic  surgery 
was  a  certain  familiarity  with  drugs.  Powdered  herbs,  for 
instance,  were  used  to  staunch  the  flowing  blood,  and  also 


vi  THE  DEBT  OF  DEMOCEDES  TO  HOMER         229 

to  ease  pain.  It  may  in  fact  be  said  most  truly  that  the 
reader  of  Homer,  to  be  ideally  qualified,  so  far  as  medicine 
is  concerned,  must  know  anatomy  rather  well,  should  have 
seen  some  simple  processes  of  surgery,  and  should  know  the 
medical  properties  of  several  common  herbs. 

Thus  it  gradually  grows  plain  that  the  anatomy  of  Homer 
had  a  very  considerable  bearing  on  the  subsequent  develop 
ment  of  medicine.  To  the  Homeric  infatuation  for  minutely 
clear  accounts  of  the  give  and  take  of  sword-thrusts,  spear- 
thrusts,  arrow-wounds,  and  of  all  the  awful  bruises,  fractures, 
and  contusions  caused  by  such  jagged  stones  as  still  cover 
the  fields  of  Greece,  modern  science  owes  tools  without 
which  its  early  course  would  have  been  hampered  and  its 
vision  constantly  befogged.  The  Homeric  heroes  won 
more  than  their  own  victories  where  they  fought  and  con 
quered  with  such  desperate  skill ;  they  won  a  victory  for  us 
as  well.  They  fought  strenuously  that  we  might  think 
clearly,  since  a  vast  proportion  of  the  anatomical  terms  in 
scientific  use  to-day  are  words  whose  meaning  became  de 
finite  as  those  heroes  grew  more  skilful  in  fighting,  and 
learned  to  use  their  weapons  with  a  deadlier  knowledge. 

The  chief  inheritors  of  the  almost  scientific  and  wholly 
unmiraculous  surgery  and  surgical  skill  of  the  Homeric  age 
were  professional  doctors,  such  as  those  who  competed  at 
Aegina  and  Athens  with  the  skilful  Democedes.  These 
men,  often  in  the  employ  of  the  state,  made  possible,  and 
kept  in  successful  operation,  large  public  establishments 
which  really  deserved  the  name  of  hospitals.1 

This  must  be  insisted  upon,  because  there  is  a  growing 
danger  of  calling  by  the  name  of  hospitals  institutions 

1  On  this  difficult  question  see  Dr.  Vercoutre,  "  La  mddecine  publique 
dans  I'antiquite"  grecque,"  in  the  Revue  Archtologique,  1880. 


230      AESCULAPIUS  AT  EPIDAURUS  AND  ATHENS     vi 

which,  in  spite  of  certain  resemblances  to  hospitals,  have  a  very 
different  character — the  temples  of  Aesculapius.  A  sever 
ance,  gradually  indeed  but  very  early,  took  place  between 
secular  and  Aesculapian  healing.  It  is  not  easy  to  recognise 
this  fact,  because  anciently  there  never  was  in  any  field, 
least  of  all  in  the  field  of  ancient  medicine,  the  modern 
antagonism  between  science  and  religion.  Let  those  who 
wilfully  misinterpret  the  past  in  order  the  more  completely 
to  misunderstand  the  present  say  that  this  was  so  because 
science  was  unscientific,  or  because  religion  was  an  empty 
show.  The  fact  remains  that,  in  spite  of  the  severance 
above-mentioned,  the  doctors  kept  in  touch  with  the  worship 
of  Aesculapius,  and  the  priests  in  his  temples  did  not 
scorn  such  secular  knowledge  as  they  could  gain  from  lay 
practitioners. 

Perhaps  the  difference  in  temper  between  these  two 
schools,  if  the  word  school  may  be  so  far  misused,  is  best 
understood  by  a  backward  glance.  Let  us  again  apply  to 
Homer  and — forgetting  this  time  that  he  had  facts  to  deal 
with — let  us  ask  him  for  fancies.  In  contrast  to  what  I  have 
said  concerning  the  definite  knowledge  implied  by  the 
Homeric  anatomy,  there  was  a  fairy-land  in  the  medical 
world  of  the  heroic  age,  and  within  its  borders  ruled  a 
spirit  which  knew  not  accuracy,  and  was  but  faintly  and 
distantly  acquainted  with  facts.  The  two  sons  of  the 
noble  leech  Aesculapius,  named  Machaon  and  Podalirius, 
together  with  an  unspecified  number  of  doctors,  not  only 
had  in  a  more  perfect  degree  the  knowledge  of  anatomy 
which  the  Homeric  heroes  possessed,  but  also  a  general 
claim  to  infallibility  was  popularly  made  for  them.  They 
were,  as  has  been  abundantly  shown,  surrounded  by  a  defer 
ence  not  shown  to  ordinary  men.  A  superstitious  regard 


vi  THE  SCHOOLING  OF  AESCULAPIUS  231 

for  Aesculapius  and  his  two  sons  allied  itself  to  a  child-like 
belief  in  the  existence  of  miracle-working  drugs.  These 
drugs  were  either,  like  the  moly  given  by  Hermes  to 
Odysseus,  procurable  only  by  an  immortal  god,  or,  like 
Helen's  Egyptian  nepenthe^  they  came  from  some  far-off  and 
unvisitable  place.  Just  so  it  was  with  the  miraculous  lotus 
blossom.  Such  too  were  the  herbs  of  marvellous  and 
uncanny  effect  known  to  Circe  and  Medea,  who  both  had 
learned  of  them  from  their  father  Aee'tes,  to  whom  the 
knowledge  descended  from  his  father  the  Sun.  From 
Paean  (who  came  later  to  be  identified  with  the  Sun  and 
Apollo)  were  descended,  so  Homer  says,  the  Egyptians,  and 
all  Egyptians  had  wonderful  knowledge  of  herbs.  Aescu 
lapius  himself  was,  as  his  worshippers  finally  agreed,  the 
offspring  of  Apollo,  who  was  Helios,  this  same  Paean,  the 
sun-god.  One  more  touch  of  Homer's  must  here  be  men 
tioned.  His  Aesculapius,  although  Apollo  is  his  father  and 
protector,  had  Coronis,  a  mortal  maiden,  for  his  mother, 
and  had  to  gain  by  mortal  means  his  more  than  mortal 
skill  in  medicine.  This  brings  us  to  a  whole  cycle  of  early 
legends,  touched  upon  more  or  less  fully  by  Homer,  where 
medicine  becomes  further  involved  in  the  mists  of  uncertain 
mythology  and  early  superstition. 

The  schooling  of  Aesculapius  in  medicine  was  not  different 
from  that  of  many  other  heroes.  The  master  common  to 
them  all  was  Chiron,  in  whose  nature  the  irrepressible  bestial 
ity  of  his  fellow-centaurs  has  been  transformed  into  a  wise  and 
genial  power  of  sympathy.  The  gentle  Chiron  possessed  a 
power  of  insight  into  nature,  was  so  at  one  with  the  hearts 
of  men  and  beasts,  that  although  by  nature  he  was  below, 
by  knowledge  he  was  above  mere  human  kind.  Chiron's 
strange  name  and  nature,  half  human  and  half  of  lower  origin, 


232      AESCULAPIUS  AT  EPIDAURUS  AND  ATHENS     vi 

may  stand  as  a  link  between  the  spirit  of  man  and  the  useful 
essence  of  plants,  just  as  the  lower  animals  connect  man's 
bodily  frame  with  the  shapes  of  the  vegetable  kingdom.1 
Chiron  embodied  for  the  Homeric  understanding  what  we 
prefer,  after  our  more  abstract  fashion,  to  call  the  earliest  of 
all  early  stages  of  medicine.  This  prehistoric  medicine 
consisted  of  a  well-defined  though  superficial  knowledge 
of  the  human  frame,  by  no  means  equal  to  that  which  may 
fairly  be  attributed  to  Odysseus,  and  of  a  limited  acquaint 
ance  with  nature's  most  obvious  simples.  So  far  as  this 
last  point  is  concerned,  Chiron  embodied  all  the  knowledge 
of  Homeric  days,  which  was  by  no  means  incompatible  with 
that  superstitious  belief  in  the  efficacy  of  certain  unpro 
curable  roots  and  herbs  of  which  Homer  is  full,  and  the 
like  of  which  survives  to-day  in  various  tales  of  the  mad-dog 
stone.  This  skill  of  Chiron  the  centaur  in  the  medicine  of 
herbs  is  medicine  reduced  to  its  simplest  terms,  and  in 
this  were  versed  those  who  bore  the  greatest  names  upon 
the  Heroic  roll  of  honour — Aesculapius  and  Amphiaraus, 
the  Boeotian  Aesculapius  about  whom  much  has  been 
recently  discovered  at  Oropus  in  Attica,  Achilles  and 
Theseus,  Jason  and  Aeneas,  Castor  and  Pollux,  Nestor  and 
Odysseus,  Peleus,  Telamon,  Meleager,  and  many  others. 
Hence  it  is  that  all  of  them  are  spoken  of  as  pupils  of 
Chiron. 

The  fact  that  Aesculapius,  although  under  the  especial 

1  Just  here  the  distinction,  much  insisted  upon  above,  stands  within 
the  danger  of  confusion  and  threatens  to  break  down,  for  Chiron  ends  by 
representing  the  reasonable  and  unmiraculous  aspect  of  early  medical  lore. 
We  find  in  the  fantastic  centaur  a  spirit  of  serene  science  and  right  reason 
which  defies  every  attempt  to  draw  a  sharp  line  dividing  the  fanciful  and 
fairy-like  from  the  positive  and  practical  in  Homer's  poetical  account 
of  heroic  medicine.  Here  in  a  new  case  we  feel  the  incommunicable 
charm  and  subtle  creative  power  of  Greek  fancy. 


vi  EARLY  GREEK  MEDICINE  233 

favour  of  the  god  of  healing,  was  yet  classed  among  the  other 
illustrious  pupils  of  Chiron,  shows  that  the  Homeric  age 
was  hardly  more  appreciative  of  the  divinity  of  Aesculapius 
than  of  the  divine  character  and  importance  of  Demeter  and 
Dionysus.  Aesculapius  and  his  sons  are  thought  of  by 
Homer  as  divinely  perfected  men — leeches  whose  skill  is 
human,  though  of  an  excellence  all  but  divine.  Plainly  this 
Homeric  Aesculapius  is  not  the  great  god  of  the  Thessalian 
Lapithae  and  Phlegyae.  Only  an  echo  of  his  power  and 
helpful  kindness  reached  the  early  Greeks,  sounding  through 
the  Iliad  and  the  Odyssey.  In  order  that  the  divine  pre 
tensions  of  Aesculapius  might  ally  themselves  to  the  gentler 
and  more  human  aspect  which  he  wears  in  Homeric  story, 
a  radical  change  was  required.  All  this  is  brought  to  pass 
in  the  story  of  the  birth  of  the  god  at  Epidaurus,  where 
Coronis,  a  daughter  of  the  Thessalian  king  Phlegyas, 
brought  him  to  birth.  The  accident  of  Phlegyas'  tem 
porary  sojourn  in  Argolis  and  Epidaurus,  so  the  Epidaurian 
legend  runs,  made  Aesculapius  an  Epidaurian;  but  upon 
this  accident  his  latter-day  majesty  depends. 

But,  before  pursuing  this  Epidaurian  theme,  let  us 
summarise  the  early  course  of  medicine  in  Greece.  Even 
in  Homer's  account,  where  the  whole  field  of  medicine  is 
small,  and  where  there  are  no  clear  subdivisions,  certain 
divergent  tendencies  may  be  dimly  distinguished.  There  is 
the  positive  practical  tendency,  and  this  is  perhaps  the  pre 
ponderating  one.  There  is  also  the  poetically  superstitious 
tendency,  which  shows  itself  in  tales  of  marvellous  cures  by 
Aesculapius  and  others,  of  wonderful  drugs  procured  by 
heroes  under  the  especial  protection  of  heaven,  and  of 
wonderful  skill  and  knowledge  possessed  and  taught  by 
Chiron.  From  the  former  and  more  positive  tendency 


234     AESCULAPIUS  AT  EPIDAURUS  AND  ATHENS     vi 

sprang  Greek  anatomy  and  surgery,  the  medicine  of  Demo- 
cedes,  Hippocrates,  and  the  school  that  sprang  up  under  the 
shadow  of  the  Coan  sanctuary,  together  with  a  fair  propor 
tion  of  the  sayings  and  doings  of  conscientious  priests  in  the 
sanctuaries  of  Aesculapius  scattered  over  Greece.  From  the 
less  positive  and  more  superstitious  aspect  of  medicine  as 
known  in  early  legends,  Homeric  and  others,  nothing  per 
haps  would  have  come  without  the  help  of  the  Thessalian 
deification  of  Aesculapius.  When  the  Thessalian  cult  of 
the  god  of  healing  came  into  contact  with  the  conceptions 
of  medicine  embodied  in  the  Iliad,  it  apparently  exercised 
little  or  no  influence  upon  the  positive,  but  absorbed  into 
itself  the  vague  and  the  miraculous.  All  the  wondering 
terror  with  which  Chiron's  skill,  Circe's  sorcery,  and  Medea's 
knowledge  of  simples  had  been  regarded  was  soon  garnered 
into  the  treasure-houses  of  Aesculapius.  His  temples 
became  centres  of  miracles,  as  well  as  places  for  the  practical 
study  of  medicine.  Of  course  there  was  this  latter  side  to 
the  worship  of  Aesculapius,  or  else  Hippocrates  would  not 
have  spoken  as  he  did,  and  in  later  days  Galen  would  not 
have  had  such  close  commerce  with  the  priests  of  Aescu 
lapius.  Indeed,  superstitious  as  the  worship  of  Aesculapius 
was,  the  most  irrefutable  proof  that  it  was  neither  wholly 
nor  intolerably  so  is  the  more  than  toleration  of  it  by  the 
most  admirable  men  of  Greek  and  Roman  medicine. 

There  is  one  point  of  view  common  to  the  most  mar 
vellous  of  Homer's  fairy  tales,  to  the  practice  of  medicine 
by  the  priests  of  Aesculapius,  and  to  certain  most  and 
least  approved  aspects  of  modern  medical  procedure. 
This  is  the  notion  of  affecting  the  mind  through  the  body. 
That  wonder-working  Egyptian  drug,  Helen's  nepenthe,  and 
also  the  fatal  flower  of  the  lotus,  cast  a  spell  upon  the  mind. 


vi  MIND  AND  BODY  235 

In  like  manner,  after  the  worship  of  Aesculapius  had  run 
its  course  through  centuries  and  reached  its  final,  perhaps 
its  most  useful,  form  in  the  days  of  Galen  and  the  Antonines, 
this  same  belief  was  most  vigorous.  "  It  was  an  age  of 
valetudinarians,"  says  a  competent  authority,  "in  many 
cases  of  imaginary  ones;  but  below  its  various  crazes  concern 
ing  health  and  disease  ...  lay  a  valuable,  because  partly 
practicable,  belief  that  all  the  maladies  of  the  soul  might  be 
reached  through  the  subtle  gateways  of  the  body."1  The 
man  who  understood  drugs  was,  in  Homer's  day,  and  during 
the  age  of  the  Antonines,  as  he  is  now,  a  healer  of  all 
curable  illnesses  whether  of  body  or  mind.  Then  as  now 
power  through  the  body  over  the  mind  was  attributed  to 
him. 

The  priests  of  Aesculapius,  however,  were  far  from  taking 
a  materialistic  view  of  the  soul.  They  supplemented  the 
notion  that  an  unsound  mind  can  be  cured  through  the 
body  by  another  to  which  they  attached  every  importance,  i.e. 
that  the  sound  mind  can  and  should  completely  control  the 
sound  body.  The  prescriptions  of  Aesculapius  were  some 
times  given  to  the  purified  and  expectant  sufferer  in  dreams. 
Often  Aesculapius  himself  appeared  in  a  dream  and  touched 
the  sick ;  sometimes  a  messenger  came,  a  voice  as  it  were 
through  the  gateways  of  sleep  would  tell  what  herb  or  what 
treatment  was  necessary.  Sometimes  healing  came  from 
the  nocturnal  touch  of  serpents  or  of  dogs  sent  by  the  god 
to  his  suppliants. 

The  prescribed  process  by  which  the  possibility  of 
dreaming  an  inspired  dream  was  attained  was  one  which 

1  Marius  the  Epicurean,  by  Walter  Pater,  M.  A.,  chap.  iii.  The  whole 
of  this  admirable  chapter  well  repays  careful  study.  I  know  no  other 
adequate  modern  presentation  of  the  sweetness  and  sanctity  of  the  service 
of  Aesculapius. 


236     A ESCULAP1  US  A  T  EPIDA  UR  US  AND  A  THENS      v  i 

necessarily  stilled  the  mental  alarms  of  the  sufferer.  His 
condition  had  to  be  one  of  passivity,  such  as  doctors  some 
times  impose  upon  those  who  suffer  from  nervous  prostration. 
Not  in  a  moment  of  excitement,  but  during  the  calm  hours 
of  unstirred  sleep  came  these  divine  dreams.  They  might 
visit  men  anywhere,  but  for  the  most  part  they  came  only  in 
the  hallowed  seclusion  of  the  Aesculapian  Sanctuary.  Since 
all  who  were  at  the  point  of  death  and  of  child-birth  were 
rigorously  excluded,  panics  and  excitements  were  the  less 
possible ;  the  patient  had  to  conform  to  the  law  of  purification 
prescribed  in  the  temple,  and  then  to  lie  down  within  the 
temple  itself  or  a  porch l  near  by,  and  within  the  precinct. 
This  process  of  lying  down  in  the  temple  for  the  purpose  of 
dreaming  gets  itself  called  by  a  Latin  name  which  means 
literally  sleeping  in,  and  we  hear  much  of  the  practice  of 
Incubation  in  the  ancient  temples  of  Aesculapius.  Men  of 
pious  minds  resorted  to  it  in  order  to  hatch  out  dreams 
whereby  knowledge  of  needful  remedies  came  to  them. 
The  dream  was  more  or  less  consciously  thought  of  as 
having  a  being  apart,  like  the  dream  in  the  Iliad  sent  by 
Zeus  to  Agamemnon, — only  the  dream  in  the  temple  of 
Aesculapius  came  to  enlighten,  not  to  deceive.  What  such 
dreams  were  supposed  by  the  pious  to  accomplish  is  best 
shown  by  the  prayer  which  Aristides  addressed  to  them. 
"  Endue  my  body,"  prays  the  grateful  worshipper  of  dreams, 
"  with  such  measure  of  health  as  may  suffice  it  for  the  obeying 
of  the  spirit,  that  I  may  pass  the  day  unhindered  and  in  quiet 
ness."  2  The  body  was  cured  in  order  that  through  it  the 
spirit  might  gain  self-command  and  rule  the  whole  man. 

1  The  word  stoa  or  portions  is  strictly  required.      Neither  of  these  being 
English,  I  have  preferred  to  stretch  the  meaning  of  the  word  porch. 

2  Mr.  Pater's  translation. 


vi  HARMONY  OF  RELIGION  AND  SCIENCE          237 

I  have  said  that  Greek  secular  medicine  sprang  from  the 
more  positive  and  surgical  side  of  the  earliest  pursuit  of 
medicine.  I  have  also  said  that  all  the  extravagances  and 
miracles  believed  in  from  the  earliest  days  centred  gradually 
around  the  worship  of  Aesculapius.  But  in  fact  the  line 
between  secular  and  sacred  is  hardly  more  easy  to  draw 
for  these  later  days  than  for  Homeric  times.  Surgery,  of 
all  things,  ought  to  have  been  the  exclusive  province  of  the 
secular  practitioners,  and  yet  inscriptions1  found  at  Epidaurus 
within  the  precinct  of  Aesculapius  show  that  operations  were 
sometimes  performed  by  the  servants  of  the  god  and  under 
his  inspiration,  though,  to  be  sure,  the  particular  cases  there 
described  would  appear  to  have  been  most  unsurgically 
dealt  with.  On  the  other  hand,  if  the  distinction  in  question 
is  pressed  too  far,  or  too  sharply  drawn,  a  secular  practitioner 
like  Herophilus  ought  to  have  been  quite  free  from  the 
Homeric  point  of  view  about  the  superhuman  efficacy  of 
drugs.  And  yet  this  Herophilus,  a  celebrated  physician 
who  flourished  during  the  first  years  of  the  third  century 
B.C.,  speaks  of  all  medicines  as  gifts  from  the  gods,  and  calls 
them,  when  rightly  used,  "the  hands  of  the  gods."  *  This 
appeal  to  the  healing  hands  of  the  gods  in  everyday 
practice  is  a  beautifully  enlightened  modification  of  Homer's 

1  The  best  concise  account  of  these  inscriptions  known  to  me  is  Dr.  Mer- 
riam's  referred  to  below.  Their  discovery  and  elucidation  is  one  of  the  first 
of  the  useful  achievements  of  the  distinguished  M.  Kabbadias,  Ephor-in- 
chief  of  Antiquities.  Strabo  speaks  of  them,  and  adds  that  similar  ones 
were  to  be  found  at  Cos  and  Tricca  :  i.  e.  inscriptions  that  give  record  of 
the  manner  of  each  cure,  VIII.  vi.  16,  p.  375. 

-  See  the  first  sentence  of  the  dedication  addressed  by  Scribonius 
Largus  to  Caius  Julius  Callistus.  Compare  the  use  of  this  quotation  by 
Erasmus  in  his  ingenious  comparison  of  the  Gospel  of  St.  Luke  to  a  healing 
medicine.  This  is  in  the  dedication  of  his  paraphrase  of  that  Gospel,  and 
contrast  the  sense  attached  to  an  analogous  expression  in  the  Philoctetes  of 
Sophocles  :  speaking  of  Philoctetes  writhing  with  pain,  the  chorus  cries  out 
w  TraXd/icu  deuv  (177). 


238     AESCULAPIUS  AT  EPIDAURUS  AND  ATHENS     vi 

notion  that  the  root  moly  could  only  be  digged  from  its 
secret  hiding  place  by  a  god.  And  indeed  this  utterance 
of  Herophilus  was  quite  in  agreement  with  the  view  of 
Hippocrates,  who  said  long  before  the  day  of  Herophilus, 
with  reference  to  divine  intervention  and  healing,  "  Medicine 
inclines  to  do  honour  to  the  gods  as  concerning  symptoms 
or  sickness,  and  doctors  give  way  before  them,  since  medical 
lore  has  no  superabundance  of  power." * 

And  yet  this  harmony  between  science  and  religion,  this 
pious  deference  of  physicians  to  the  god  of  physic  and  their 
respect  for  the  miracles  worked  in  his  name,  left  a  difficult 
question  for  the  decision  of  laymen.  When  should  there 
be  appeal  to  the  god  and  his  divine  skill,  and  when  should 
the  counsel  of  human  doctors  be  resorted  to  ?  The  doctrine 
of  Socrates  may  well  be  taken  to  represent  the  mind  of  the 
most  enlightened  men.  "  Seek  as  far  as  you  may  to  help 
yourself  before  asking  the  gods  for  help  and  counsel."  This 
was  the  view  of  Socrates  about  consulting  oracles  in  general, 
and  no  doubt  he  would  have  applied  it  to  the  most 
primitive  and  wide  -  spread  of  all  Greek  ways  of  consult 
ing  oracles,  the  dreaming  of  dreams  in  the  sanctuary  of 
Aesculapius,  as  well  as  to  other  appeals  to  Aesculapian  skill. 
"Exhaust  human  skill  and  resource  before  appealing  to 
the  god,"  he  would  have  said.  Theoretically  this  view  of 

1  See  the  sixth  paragraph  of  the  treatise  on  professional  honour  (jrepl 
tvff-XyfJ-offtivqs}.  This  treatise  is  by  some  considered,  though  on  purely 
negative  grounds,  of  doubtful  authorship.  I  am  convinced  with  Darem- 
berg  that  it  is  by  Hippocrates  ;  some  doctor  certainly  wrote  it,  and 
it  certainly  represents  a  typical  point  of  view.  This  spirit  of  pious  de 
ference  to  divine  power  is  by  no  means  confined  to  one  treatise  of 
Hippocrates.  There  is  a  solitary  quotation  upon  which  the  ingenious 
Wilamovitz  -  Moellendorf  founds  his  otherwise  baseless  assertion  that 
Hippocrates  was  free  from  any  belief  in  Aesculapius.  The  passage  occurs 
in  the  treatise  on  airs,  waters,  and  climates,  and  is  a  protest  against  a 
gross  Scythian  superstition.  Moellendorf  s  very  strained  reading  of  it  can 
be  refuted  by  other  undoubted  sayings  of  Hippocrates. 


vi  AESCULAPIUS  A  FRIEND  OF  MAN  239 

duty  was  above  reproach,  but  practice  was  another  matter. 
Many  motives  led  the  faithful  to  consult  Aesculapius  more 
frequently  than  this  principle,  strictly  adhered  to,  would 
allow  ;  and  among  them  the  most  decisive  one  was  his 
approachability.  A  feeling  of  familar  comradeship  was 
inspired  in  all  his  worshippers  by  Aesculapius,  and  in  this 
Socrates  certainly  shared,  since  his  dying  words  were :  "Crito, 
we  owe  a  cock  to  Aesculapius."  The  meaning  of  this 
solemnly  smiling  farewell  of  Socrates  would  seem  to  be  that 
to  Aesculapius,  a  god  who  always  is  prescribing  potions  and 
whose  power  is  manifest  in  their  effects,  was  due  that 
most  welcome  and  sovereign  remedy  which  cured  all  the 
pains  and  ended  all  the  woes  of  Socrates — the  hemlock, 
which  cured  him  of  life  which  is  death,  and  gave  him  the 
glorious  realities  of  hereafter.  For  this  great  boon  of 
awakening  into  real  life  Socrates  owed  Aesculapius  a  thank- 
offering.  This  offering  of  a  cock  to  Aesculapius  was 
plainly  intended  for  him  as  the  awakener  of  the  dead  to 
life  everlasting. 

In  the  story  which  makes  Aesculapius  incur  the  wrath  of 
Zeus  in  order  to  recall  to  life  one  who  was  dead,  and  further, 
in  the  minds  of  all  worshippers,  this  god — standing  before 
Zeus  as  divine  yet  also  human — is,  like  Prometheus,  a  loving 
and  indulgent  friend  of  man  even  when  other  deities  frown. 
Apollo  intercedes  for  him  with  angered  Zeus  much  as  he 
might  for  a  man.  Something  of  the  mortality  attributed  to 
him  in  the  Homeric  poems,  a  half- humanity,  clung  to 
Aesculapius  throughout  antiquity;  and  the  latter  Greeks 
never  quite  banished  from  their  worship  of  this  god  the 
notion  that  he  was  a  hero  or  demigod  only.  How  natural 
it  was  in  Athens  to  think  of  a  healing  power  under  this 
aspect  is  shown  by  the  dim  knowledge  that  we  have  of  an 


240     AESCULAPIUS  AT  EPIDAURUS  AND  ATHENS     vi 

Athenian  temple  dedicated,  not  to  Aesculapius,  but  to  the 
"  Hero  physician."  Even  after  his  full  divinity  came  into 
general  recognition,  therefore,  Aesculapius  bore  marks  of 
his  previous  condition.  He  was  worshipped  and  besought 
not  always  under  the  name  of  a  god,  but  most  frequently 
under  the  designation,  familiar  to  Christian  ears,  of  the 
Son  of  God.  Filius  dei  was  in  fact  the  habitual  and  un 
qualified  manner  of  addressing  Aesculapius  in  his  temples 
at  Rome.  Partly  human  of  birth,  he  was  wholly  so  in 
sympathy ;  but,  in  his  perfect  power  to  help  and  heal,  he 
was  divine. 

This  halo  of  humanity,  if  the  expression  be  allowed, 
was  worn  by  Aesculapius  with  all  the  better  grace  because 
he  was  by  no  means  foremost  in  the  Olympian  hierarchy — 
since  our  minds  condemn  us  to  talk  of  a  hierarchy  when 
there  was  none.  The  god  of  healing,  with  all  his  train  of 
abundant  divinities,  Health,  Panacea,  Convalescence  (Teles- 
phorus),  and  the  many  others, — kindly  presences  all  of  them, 
called  into  being  solely  to  ease  men's  pain, — may  be  thought 
of  as  dwelling  somewhere  midway  between  the  gods  above 
and  men  below.  There  they  dwelt  in  order  perhaps  to 
be  near  at  hand  when  the  calamities  of  men  required  their 
instant  aid.  So  human  were  the  beginnings  of  Aesculapius 
that  he  depended  upon  the  power  and  presence  in  Olympus 
of  Apollo  his  father.  Just  as  we  may  imagine,  if  we 
choose,  that  Aesculapius  was  the  vicar  of  Apollo  on  earth 
to  represent  him  at  the  Eleusinian  Mysteries,  so  we  know 
that  Apollo  was  the  heavenly  presence  whose  Olympian 
power  sustained  and  increased  the  divine  efficacy  of  all  the 
works  of  his  son  Aesculapius.  The  words  filius  dei  apply 
to  Aesculapius  as  the  son  of  Apollo  the  god. 

"  Save  me,  and  heal  my  grievous  gout,  O  blessed  and 


vi  THE  SON  OF  APOLLO  241 

most  mighty  presence,  I  adjure  thee  by  thy  father,  to  whom 
I  loudly  pray."  Such  is  the  prayer  addressed  to  Aesculapius, 
"Son  of  Leto's  son,"  by  Diophantus,  an  attendant  at  his 
shrine,  which  has  been  lately  uncovered  on  the  southern 
slope  of  the  Athenian  Acropolis.  This  inscribed  prayer  is 
that  of  Diophantus  born  in  the  Athenian  township  of 
Sphettus.  Its  faults  are  many  in  versification,  and  it  lacks 
poetic  delicacy  of  phrase,  but  still — Diophantus  having  been 
an  attendant  in  the  Athenian  temple  of  Aesculapius — it  pre 
serves  for  us  an  official  view  of  the  relation  between 
Aesculapius  the  divine  son  and  the  divine  father  Apollo. 
In  their  eyes  a  prayer  to  Aesculapius  was  also  a  prayer 
to  Apollo,  and  the  god  of  healing  was  thought  of  by  them 
as  a  pitiful  and  indulgent  mediator  between  man  and 
the  Holiest  and  Mightiest.  "  No  one  of  mortals,"  Diophantus 
continues  in  this  same  inscription,  "  can  give  a  surcease  from 
such  pangs.  Thou  alone,  divinely  blessed  one,  hast  the 
power ;  for  the  supreme  gods  bestowed  on  thee,  all  pitying 
one,  a  rich  gift  for  mortals.  Thou  art  their  appointed 
deliverer  from  pain."1  Thus  Aesculapius  was  not  mortal 
though  he  was  under  inspiration  from  above — he  was  the 
well-beloved  saviour  from  suffering,  the  comforter  sent  by 
Apollo. 

A  curiously  close  relation  between  Apollo  the  father  and 
the  son  Aesculapius  is  shown  by  the  Apolline  epithet  Paean 

1  See  in  the  May  number  of  Gaillard's  Medical  Journal  (vol.  xi.  No.  5), 
published  in  New  York,  an  article  on  "  Aesculapia  as  revealed  by  In 
scriptions,"  recast  from  a  paper  read  before  the  New  York  Academy  of 
Medicine,  igth  March  1885,  by  Augustus  C.  Merriam,  A.M.,  Ph.D. 
Professor  Merriam  has  there  given  the  most  concise  account  of  all  the 
facts  bearing  upon  the  worship  of  Aesculapius  at  Athens  and  at  Epidaurus, 
and  his  account  of  the  inscriptions  is  not  only  exhaustive  but  most  enter 
taining.  He  also  gives  abundant  references  to  more  detailed  accounts  of 
the  matter  in  hand. 


242      AESCULAPIUS  AT  EPIDAURUS  AND  ATHENS     vi 

which  this  same  Diophantus  bestows  upon  Aesculapius  in 
his  record,  made  in  the  same  place,  of  thanks  for  recovery. 
He  thanks  "  Paean  Aesculapius,"  to  whose  skill  he  attri 
butes  his  deliverance.  This  consummation  devoutly  to  be 
desired  was  promised  him  by  Aesculapius,  who  appeared  in 
a  dream.  This  whole  episode  in  Diophantus'  life  is  a  most 
authentic  and  imperishable  record,  kept  upon  stone,  of  the 
mediating  and  human  divinity  of  Aesculapius,  who  trans 
mitted  the  kindly  will  of  Apollo  to  suffering  men,  and  lent 
them  the  means  of  grace.  It  may  truly  be  said  of  this  god 
of  healing  that  he  and  his  father  are  one,  for  even  the 
dreams  wherein  Aesculapius  himself  appeared  and  wrought 
cure  were  addressed  as  the  "  children  of  Apollo."  "  Oh,  ye 
children  of  Apollo,  who  in  times  past  have  stilled  the  waves 
of  sorrow  for  many  people,  and  lighted  up  a  lamp  of  safety 
before  those  who  travel  by  sea  and  land,  be  pleased  in  your 
great  condescension  ...  to  accept  this  prayer  .  .  ." l  So 
opens  the  collect  of  Aristides  already  alluded  to.  The 
final  source  of  power  is  Apollo;  and  the  accomplishment 
of  cure,  no  matter  what  natural  means  and  medicines  are 
employed,  is  at  the  bidding  of  Aesculapius,  whose  loving- 
kindness  miraculously  brings  healing.  Often,  therefore,  he 
was  looked  upon  as  a  patron  saint  might  be.  He  was  a 
mediator  and  an  elder  brother — a  being  close  to  the  divinity, 
with  whom  the  worshipper  need  not  always  be  on  terms  of 
the  most  ceremonious  observance. 

This  nearness  to  man  involved  what  we  might  call 
humility  in  some  sort,  or  self-subordination  in  regard  to  the 
other  gods ;  but  both  of  these  terms  are  far  too  exclusively 
modern  to  be  used  very  strictly  of  any  Greek  divinity. 
Aesculapius  was  not  a  jealous  god,  and  when  his  holy  pre- 

1  Mr.  Pater's  translation. 


vi  THE  HIE  RON  OF  EPIDAURUS  243 

cinct  was  set  apart  near  to  the  ancient  places  of  worship 
sacred  to  other  and  older  gods,  there  was  not  room  perhaps 
for  showing  him  all  due  honour.  Therefore  Aesculapius, 
more  than  most  of  the  gods  in  Greece,  required  for  his 
cult  a  district  all  his  own — a  country  sacred  to  him,  where 
his  worship  should  be  the  centre  of  religious  and  also  of 
social  life.  Such  a  country,  dedicated  to  his  worship,  was 
the  district  of  Epidaurus. 

On  the  eastern  coast  of  Argolis,  full  in  view  from  the 
islands  of  Salamis  and  Aegina,  over  against  Athens  and  the 
Piraeus,  lies  the  town  of  Epidaurus,  with  the  volcanic  and 
picturesque  peninsula  of  Methana  just  to  the  south  of  it, 
beyond  a  fertile  seaward  plain.  This  plain  and  the 
mountain  heights  beyond  it  form  the  district  of  Epidaurus. 
In  the  town  itself  were  minor  sanctuaries  of  the  god — not 
only  a  temple  for  himself,  but  one  for  his  wife,  Gentleheart 
or  Epione  they  called  her ;  but  beyond  this  name,  and  the 
existence  of  her  Epidaurian  temple,  which  disappeared  with 
the  town  of  Epidaurus,  little  more  is  known  about  her. 
Fortunately  it  is  otherwise  with  the  great  centre  of  all 
Aesculapian  worship  in  antiquity — the  Epidaurian  Hieron 
or  Holy  ground.  This  lies  higher  up  and  farther  inland 
than  the  town  of  Epione's  shrine.  From  this  Hieron  of 
Epidaurus  went  forth  to  the  east  and  west  those  who 
established  the  great  centres  of  Aesculapian  worship  else 
where.  They  claimed  to  have  founded  the  Coan  *  temple, 
near  which  Hippocrates  was  born,  and  the  sanctuary  sacred 

1  Some  sort  of  Aesculapian  worship  at  Cos,  of  an  earlier  date  than  any 
possible  foundation  from  Epidaurus,  must  be  allowed.  Indeed,  success 
and  pre-eminence  at  a  comparatively  late  date  probably  made  the  Epi- 
daurians  claim  to  have  founded  various  temples  quite  as  old  as  their  own. 
The  Eleusinians  certainly  claimed  the  same  sort  of  precedence  over 
Peloponnesian  shrines  of  less  note  than  theirs,  but  of  equal  or  greater 
antiquity. 


244      AESCULAPIUS  AT  EP1DAURUS  AND  ATHENS     vi 

to  Aesculapius  on  the  island  in  the  Tiber  at  Rome  certainly 
derived  from  them. 

Suppose  we  have  landed  at  ancient  Epidaurus  and  are 
bound  for  this  beautiful  upland  health  resort.  First  our 
course  lies  southward  till,  at  half  a  mile's  distance,  the 
inland  road  turns  to  cross  the  fertile  but  narrow  Epidaurian 
plain,  which  is  about  a  quarter  of  a  mile  in  width.  The  way 
then  follows  a  mountain  torrent  for  a  time,  and  goes  inland 
two  miles  and  more.  Here  at  a  crossways  the  pilgrim  to 
the  shrine  of  Aesculapius  leaves  the  high  road  to  ascend  the 
side  and  cross  the  shoulder  of  Mount  Titthion.  Two 
downward  miles,  and  you  are  at  last  on  consecrated  ground. 
A  semicircle  of  gentle  and,  for  those  parts,  well-wooded 
slopes  hems  in  the  Hieron  to  the  northward,  the  southward, 
and  the  eastward,  while  towards  the  north-west  the  valley 
leans  downward  into  a  wider  valley,  through  which  extends 
the  carriage-road  that  goes  to  Nauplia.1 

The  Epidaurian  birth-legend  of  Aesculapius  has  already 
been  alluded  to.  When  her  father,  the  Thessalian  King 
Phlegyas,  visited  Epidaurus  to  spy  out  the  land  with  a  view 
to  conquest,  Coronis  was  with  him.  She,  fearing  discovery  by 
him  when  her  time  came,  caused  the  new-born  babe  Aescu 
lapius,  her  son  by  Apollo,  to  be  exposed  on  the  upland  slopes 
of  Mount  Titthion.  The  existence  of  this  babe  remained 
unknown  of  Phlegyas,  and  would  perhaps  long  have  been 
unheard  of,  had  it  not  been  for  what  befell  a  mountain- 
ranging  shepherd,  just  when  the  babe  was  exposed.  This 
shepherd  missed  a  faithful  dog  and  also  one  of  his  flock. 
Aresthanas — for  such  was  the  shepherd's  name — hastened 
to  make  thorough  search,  and  after  wandering  through 
many  mountain  places,  found  the  missing  goat  giving 

1  The  usual  approach  is  by  this  excellent  road. 


vi  FINDING  OF  THE  INFANT  GOD  245 

sustenance  to  a  new-born  babe,  while  the  faithful  dog  was 
keeping  careful  guard  over  the  two.  To  commemorate 
this  beautiful  and  miraculous  episode  the  name  of  that 
mountain  became  Titthion — the  mountain  of  the  nursing- 
goat. 

When  Aresthanas  sought  to  lift  up  the  babe  a  great  light 
streamed  from  it,  as  it  were  the  flash  of  lightning.  This 
was  a  sign  from  Heaven ;  therefore  he  left  the  infant  god 
where  he  had  found  him,  not  lifting  him  up  nor  bearing 
him  away.  Soon  the  fame  of  this  and  other  wonders  that 
followed  it  was  noised  abroad  over  land  and  sea,  and  people 
knew  that  the  infant  Aesculapius  was  skilled  in  all  manner 
of  devices  for  the  sick,  and — most  wonderful  of  all — people 
were  made  aware  that  in  him  was  the  miraculous  power  to 
raise  from  the  dead  whomsoever  he  would. 

This  later  story  just  given  from  Pausanias  is  very 
different  from  Pindar's  earlier  one.  When  Pindar  wrote, 
Aesculapius  had  not  yet  definitely  changed  his  abode,  and 
was  still  sometimes  thought  of  as  living  in  Thessaly.  The 
general  course  of  events,  as  well  as  the  names  Phlegyas  and 
Coronis,  are  common  to  both  stories,  and  prove  them  to  be 
one;  but  on  the  whole  the  earlier1  one — Pindar's,  which 
knows  not  Epidaurus  but  unfolds  itself  in  Thessaly — is  the 
more  tragical.  There  was  on  the  part  of  Coronis,  whom  Apollo 
had  wedded,  a  faithlessness  so  flagrant  that  it  brought  her 
destruction.  The  righteous  indignation  of  Apollo,  whose 
sister  Artemis  slew  the  guilty  maid,  made  him  forget  the  child 
that  was  to  be,  until  Coronis,  not  yet  a  mother  though  a 
guilty  spouse,  lay  stretched  upon  the  flaming  funeral  pyre. 

1  See  the  XVIth  Homeric  Hymn,  where  the  Dotian  plain  of  Thessaly  is 
given  as  Aesculapius'  birthplace.  "  To  Asclepios,  healer  of  sickness, 
begins  my  song  ;  to  Apollo's  son,  whom  heavenly  Coronis  bare  on  the 
Dotian  plain,  and  she  was  Phlegyas'  daughter." 


246     AESCULAPIUS  AT  EPIDAURUS  AND  ATHENS     vi 

Snatched  from  the  flames,  Aesculapius  is  given  to  the  care 
of  Chiron,  of  whom  he  learns  the  art  of  healing.  Pindar's 
tale  keeps  Aesculapius  near  to  Tricca,  his  most  ancient  and 
original  place  of  worship,  not  far  from  Pyrasus,  the  earliest 
and  Thessalian  abode  of  Demeter.  Homer's  account  of 
Aesculapius  differs  from  both  of  the  above  legends  in  its 
more  matter-of-fact  tone.  The  miracles  are  fewer  in  Homer's 
version,  but  he  agrees  with  Pindar  in  making  Thessaly  the 
birthplace  of  the  god.  As  before  insisted  upon,  Homer's 
Aesculapius  was  scarcely  a  god  —  he  was  the  hero  who 
came  to  parry  and  make  unavailing  the  thrusts  of  all 
manner  of  diseases. 

Turning  now  from  the  god  Aesculapius  to  his  chief 
dwelling  place,  from  mythology  to  archaeology,  let  us  go  up 
to  his  holy  place  in  the  valley  overlooked  by  his  Epidaurian 
birthplace  Mount  Titthion,  and  also  by  Mount  Cynortion, 
sacred  from  of  old  to  his  father  Apollo.  Once  arrived 
there,  we  cannot  fail  to  notice  the  health-giving  purity  of 
the  air  and  a  kindly  cheerful  smile  that  meets  us  in  the 
landscape.  But  soon  the  most  surprising,  the  only  surpris 
ing,  feature  in  the  landscape  lays  hold  upon  the  eye  and 
engrosses  the  mind — the  theatre  of  Polycletus.  Many  ancient 
theatres  have  been  excavated  in  Greece,  in  Greater  Greece, 
and  Grecian  Asia  Minor,  but  the  Epidaurian  theatre  is  the 
most  perfectly  preserved  and  the  most  beautiful  of  them  all. 
This  theatre  of  Dionysus,  and  also  the  exquisite  and  unique 
Rotunda,  which  lies  within  the  sacred  enclosure  of  Aescu 
lapius,  are  architectural  masterpieces  by  Polycletus,  a  native 
of  Argolis,  where  Epidaurus  lies.  Although  there  were 
two  artists  of  this  name,  the  elder  and  the  younger,  and 
there  has  consequently  been  a  discussion  of  the  point,  the 
theatre  and  the  Rotunda  at  Epidaurus  are  now  generally 


vi  A  PRECINCT  OF  AESCULAPIUS  247 

credited  to  the  younger  Polycletus.  Of  the  Rotunda  I  shall 
presently  speak.  Of  the  theatre  Pausanias  declares  his  high 
opinion:  "Roman  theatres  may  be  finer,"  he  says,  "and 
those  of  latter-day  Greece  may  be  larger,  but  still  the  Epi- 
daurian  masterpiece  of  Polycletus  is  peerless  for  harmony 
of  proportion  and  charm  of  aspect." l 

From  various  sources,  but  chiefly  from  the  minute  pains 
given  to  results  of  excavation  at  Epidaurus  and  Athens,  it 
appears  that  certain  features  characterised  any  and  every 
precinct  of  Aesculapius  in  the  days  when  his  worship  was 
finally  organised.  First  a  small  temple  for  the  god  himself 
to  dwell  in  was  required.  Aesculapius  was  too  generously 
scrupulous  about  any  curtailment  of  comfort  for  the  sick 
who  resorted  to  him  ever  to  require  a  large  temple  for 
himself.  In  this  modest  building  was  the  statue  of  the 
god,  and  there  were  hung  or  disposed  in  some  satisfactory 
way  the  smaller  and  more  valuable  votive  offerings  made  to 
him  by  grateful  convalescents.  The  one  thing  needful  was 
room  for  long  and  commodious  porches  with  the  right 

1  It  was  not  accident  which  grouped  together  on  the  Athenian  Acropolis 
as  well  as  in  the  Hieron  at  Epidaurus  the  temple  of  Aesculapius  and  the 
theatre  of  Dionysus.  Convenience  certainly  had  something  to  do  with  it, 
and  at  Athens  the  comfort  of  the  sick  required  just  the  exposure  of  the 
theatre.  Moreover,  the  inspiration  and  amusement  afforded  to  invalids 
by  ready  access  to  theatrical  performances  were  numbered  among  their 
curative  resources  by  the  priests  of  Aesculapius.  The  well-known  case  of 
Aristides  leaves  no  doubt  on  this  point.  Still,  beyond  these  more  prosaic 
reasons  religious  ones  might  be  assigned.  Aesculapius  and  Dionysus 
were  associated  in  ritual  by  their  connection  with  the  Greater  Mysteries  at 
Eleusis.  Their  common  origin  in  the  northward  regions  of  Thessaly  and 
Thrace  left  its  mark  in  certain  touches  common  to  the  legends  of  their 
birth.  There  was,  furthermore,  a  part  assigned  to  the  god  Aesculapius  in 
the  festivals  of  Dionysus.  What  this  was  can  only  be  guessed.  Perhaps 
it  may  have  connection  with  a  need  for  Aesculapius,  the  upraiser  of  the 
dead  to  life  ;  for  Dionysus,  typifying  by  his  yearly  death  the  winter  of  each 
year,  had  to  be  quickened  every  spring,  and  thus  could  profit  by  the  near 
presence  of  the  healing  god,  the  well-beloved  son  of  his  brother  and  ally, 
Apollo. 


248     AESCULAPIUS  AT  EPIDAURUS  AND  ATHENS     vi 

exposure.  In  these  the  wards  of  Aesculapius  were  housed. 
There  and  in  the  temple  too  they  slept,  awaiting  visits  of 
healing  from  the  inspired  dreams. 

This  feature  is  more  clearly  made  out  in  the  precinct  at 
Epidaurus  than  in  the  Athenian  sanctuary,  though  both  of 
them  certainly  were  amply  provided  with  such  accommoda 
tion  for  patients.  The  capitals  of  the  Epidaurian  porches1 
were  Ionic,  those  at  Athens  seem  to  have  been  Doric.  The 
Epidaurian  porch  was  upwards  of  120  feet  long.  The 
length  of  the  porches  at  Athens  is  not  easy  to  make  out, 
since  there  appear  to  have  been  different  buildings  at 
different  times.  The  long  porch  at  Epidaurus  was  really 
two  and  not  one.  Though  the  two  unite  into  one  con 
tinuous  stretch,  one  has  a  lower  story  which  is  absent  from 
the  other.  This  is  largely  due  to  a  clever  use  of  the  natural 
slope  of  the  ground.  Both  at  Athens  and  at  Epidaurus  the 
god  could,  for  the  purpose  of  visitation  by  night,  emerge 
from  his  temple  and  immediately  find  his  expectant  sup 
pliants  sleeping  in  the  porch  close  by. 

The  temple  at  Athens  has  not  been  clearly  made  out. 
There  seem  to  have  been  two  temples  not  now  easy  to 
disentangle  and  attribute  rightly  to  the  right  period  for  the 
building  of  each.  Both  were  small ;  of  that  little  we  may 
be  perfectly  sure.  The  temple  at  Epidaurus  was  about 
eighty  feet  long  and  upwards  of  forty  feet  in  width.  It  was 
of  the  Doric  order,  and  fine  fragments  of  its  sculptured 
ornamentations  are  preserved  in  the  museum,  a  farm-house 
close  to  the  theatre  of  Dionysus.  There  was  apparently  a 
very  fine  frieze  of  lions'  heads  upon  it,  much  like  the  one 
which  we  shall  find  upon  the  Rotunda  near  by.  The 

1  Here  again  I  am  using  the  word  porch  to  describe  a  stoa  or  porticus  ; 
in  fact  I  plead  guilty  to  doing  this  throughout  this  chapter  and  elsewhere. 


vi  THE  STATUE  OF  IVORY  AND  GOLD  249 

eastern  pediment  contained  a  group  representing  the  defeat 
of  the  Amazons,  while  the  western  pediment  was  filled  by 
sculptures  representing  a  Thessalian  tale  of  the  victory  in 
Thessaly  by  the  Lapithae  (near  friends  of  Thessalian 
Phlegyas)  over  the  Centaurs.  The  three  angles  of  the 
gables  were  surmounted  by  delicate  statues  of  victory, 
whose  more  or  less  marred  remains  are  visible  now  in  the 
Central  Museum  at  Athens.  Within  was  the  great  statue  of 
all-pitying  Aesculapius  wrought  in  ivory  and  gold  by  Thrasy- 
medes.  Of  this  statue  coins  were  the  only  record  until  the 
beautiful  bas-reliefs  of  Epidaurus  came  to  light.1  His  brow, 
like  that  of  Zeus,  has  all  the  serenity  and  unfathomable 
peace  that  glows  upon  the  noonday  firmament  in  cloudless 
summer  time.  There  is  no  trace  here  of  sternness ;  all  that 
the  face  of  Aesculapius  discloses  well  behooves  the  gentle- 
hearted  husband  of  (Epione)  Gentleheart.  Aesculapius  sits 
not  too  majestic  in  benign  repose.  One  upraised  leg  is 
resting  on  the  other,  and  he  gazes  with  eyes  overflowing 
with  health-giving  wisdom  not  far  away,  and  not  upward 
but  forward  as  if  kindly  to  entreat  with  welcome  all  those 
who  suffer  and  are  heavy  laden.  To  him  let  them  confide 
their  woes,  on  him  let  them  lay  their  burdens  of  suffering 
and  their  forebodings  of  despair.  He  sits  calm  and  most 
divinely  competent  to  counsel  and  to  guide.  This  attitude 
of  reposeful  capability  was  given  to  the  god  of  healing  on 
many  a  tablet  inscribed  with  grateful  names,  and  bearing  on 
its  sculptured  surface  a  picture  of  the  god  at  the  moment 
when  offerings  and  supplications  were  made  to  him.  Many 
such  have  been  found  at  Athens,  and  these  votive  bas-reliefs 
form  some  of  the  most  interesting  features  in  the  too  little 

1  See  Brunn's  beautiful  photograph  of  the  best  of  the  two,  Denkmiiler 
der  ant i ken  Kunst. 


250     AESCULAPIUS  AT  EPIDAURUS  AND  ATHENS     vi 

frequented  chamber  on  the  Acropolis,  which  is  devoted  to 
fragments  from  the  Asclepieium  at  Athens. 

With  the  exception  of  the  Rotunda,  a  temple  of  Artemis 
near  the  entrance  was  the  only  other  important  building  in 
the  Epidaurian  precinct.  Dogs'  heads,  which  took  the  place 
of  the  lions'  heads  conspicuous  on  the  temple  of  Aesculapius 
and  the  invalids'  porch,  ornamented  the  frieze  of  this  temple 
of  Apollo's  twin  sister.  This  interesting  detail  suggests  that 
Artemis  was  here  worshipped  as  huntress  or,  if  you  choose, 
as  mistress  of  the  hounds.  The  temple  of  the  god  Aescu 
lapius  being  small,  that  of  Artemis  was  smaller  still ;  and  as 
little  is  known  of  it  in  detail,  we  may  with  a  clear  conscience 
neglect  it  in  order  to  devote  attention  to  the  most  marvellous 
building  of  them  all — the  Rotunda  or  Tholos  of  the  younger 
Polycletus. 

This  remarkable  structure  was  famous  throughout  anti 
quity,  for  it  was  one  of  the  most  perfect  examples  of  the 
graceful  and  efflorescent  style  which  came  into  favour  after 
the  day  of  severely  perfect  architecture  and  sculpture.  Its 
round  shape  invited  flowery  ornament,  and  the  genius  of 
the  younger  Polycletus  showed  itself  here  at  its  best.  A 
new  delicacy  and  life  was  given  to  traditional  forms  of 
ornamentation.  They  left  the  hand  of  Polycletus  so 
quickened  and  transformed  that  they  seem  to  have  come 
to  him  fresh  from  the  flowering  meadows.  He  took  the 
massive  Doric  column  and  lent  to  it  for  his  purpose  a  deli 
cate  outline,  while  he  preserved  the  significance  and  charm 
of  modulated  curves  in  its  capital.  Such  was  the  external 
circle  of  columns  which  he  placed  upon  triple  steps  of 
ascent.  Thus  we  have  a  less  massive  seeming  basement 
than  the  three  steps  of  the  stylobate  of  the  Parthenon,  three 
concentric  circles,  two  within  and  one  at  the  outermost 


vi       THE  THOLOS  OF  POLYCLETUS  THE  YOUNGER    251 

verge.  Footed  upon  the  innermost  round  of  this  triple 
support  a  circle  of  twenty-six  Doric  columns  arose  to  bear 
the  most  beautiful  of  burdens — an  entablature  composed  of 
three  harmonious  parts.  First,  and  resting  directly  upon 
the  columns,  came  the  architrave — a  smooth  marble  beam 
running  without  interruption  around  the  whole  circum 
ference.  Resting  upon  this  was  the  frieze,  a  very  broad 
band  of  alternating  rosettes  and  triglyphs.  The  triple 
vertical  line  in  the  triglyph  framed  most  exquisitely  the 
square  slabs,  each  bearing  a  central  rosette.  Rosettes  have 
been  found  on  Mycenae  vases,  and  are  generally  said  to  be 
an  inheritance  from  Assyria;  they  appear  on  Egyptian 
monuments  of  the  eighteenth  dynasty.  Still,  in  the  hands 
of  the  younger  Polycletus,  they  appear  as  something  new. 
They  seem  to  have  been  gathered  by  him  from  the  fields  of 
Greece  in  the  loveliest  meadows  of  spring.  There  they  still 
grow  to-day,  and  begem  the  tanglewood  on  every  sheltered 
slope  with  dots  of  pure  and  incandescent  red.  The  shapely 
form  of  that  bright  red  anemone  was  commonly  set  upon 
memorial  slabs  above  funereal  inscriptions,  and  now  we  find 
it  idealised  and  complicated  but  still  the  same  simple  flower- 
form  taken  to  be  the  heart  and  essence  of  the  frieze  in 
Polycletus'  entablature.  The  third  and  crowning  portion 
of  it,  the  last  perfection  surmounting  this  most  exquisite  of 
buildings,  was  its  cornice.  This  was  a  band  with  beautifully 
sculptured  lions'  heads  surmounted  by  acroteria,  which  in 
this  case  are  flower-like  points  surmounting  graceful  leaf-like 
curves. 

Pass  now  between  the  Doric  columns  on  the  eastward 
side  where  the  doorway  opened  towards  the  temple  of 
Aesculapius.  There  the  door  pierces  the  cella-wall,  and 
beyond  is  the  interior.  The  roof  within  was  supported  by 


252     AESCULAPIUS  AT  EPIDAURUS  AND  ATHENS     vi 

fourteen  of  the  most  exquisite  Corinthian  columns  that  the 
mind  of  man  has  ever  dreamt  of.  From  the  same  meadows 
where  grew  the  red  anemones  Polycletus  took  the  delicate 
daisy — I  have  seen  just  such  growing  profusely  upon  the 
battle-field  of  Marathon l — and  set  it  upon  his  capital  above 
the  acanthus  leaf.  The  curving  tendrils,  not  too  profusely 
clustering  around  the  summit  of  the  column,  seem  with 
gladsome  upward  swing  to  tempt  the  eye  still  on  until  the 
delicate  daisy  crowns  this  creation  woven  together  out  of 
the  graceful  spirit  of  grasses  that  wave  in  the  field,  of 
tendrils  that  cling,  and  of  flowers  that  bloom  by  the  way 
side.  Here  is  the  earliest  existing  and  the  best  of  all  known 
Corinthian  capitals.2 

When  the  delicate  grace  of  these  exquisite  architectural 
blossoms  gathered  itself  into  final  form  before  his  mind, 
Polycletus,  the  great  architect,  knew  perhaps  that  Pausias 3 
was  to  decorate  the  circle  of  its  walls  with  paintings.  Com 
bining,  perhaps  for  the  first  time,  two  great  discoveries, 
perspective  and  encaustic,  which  were  partly  his,  Pausias 
decorated  the  interior  curves  of  the  Rotunda  with  beautiful 


1  See  a  most  beautiful  photograph  of  it  taken  by  Mr.  Elsey-Smith  and 
published  by  the  Hellenic  Society.      A  very  beautiful  enlargement  of  this 
is  most  opportunely  published  in  the  Proceedings  of  the  Royal  Institute  of 
British  Architects,  vol.  vi.  New  Series,  1890. 

2  Of  this,  Mr.  Francis  C.  Penrose  writes,  see  p.  67  of  the  Proceedings  re 
ferred  to  in  the  preceding  note.    "  To  me  the  cap  from  Epidaurus  is  extremely 
interesting,  because  it  is  very  similar  to  the  capitals  of  the  columns  of  the 
temple  of  Jupiter  Olympius  at  Athens — a  temple  to  which  I  have  paid 
much  attention.     The  forms  of  the  leaves  of  the  two  examples  greatly  re 
semble  one  another,  and  the  ornaments,  namely,  both  the  central  flower 
and  that  figure    somewhat  resembling  a  fleur-de-lis  which  occupies  the 
corner  of  the  volute,  have  their  counterparts  in  the  central  flowers  of  the 
caps  of  the  Athenian  temple,   and   thoroughly  confirm   my  opinion  that 
the  columns  of  the  latter  are  Greek  and  not  Roman  work." 

3  Not  Pausias  but  Pauson  is  unfavourably  compared  by  Aristotle  with 
Polygnotus.      His  influence  was  the  less  pure  because  of  his  preoccupation 
with  technical  matters,  partly  too  because  of  the  subjects  which  he  chose. 


vi  THE  PAINTING  OF  PA  US  I  AS  253 

pictures.  One  represented  Eros,  whose  weapons  were  flung 
away,  while  he  grasped  a  lyre  upon  which  he  discoursed 
sweet  music ;  another,  of  less  high  inspiration  but  most 
celebrated  for  its  technique,  was  his  allegorical  figure  of 
Methe  or  drunkenness.  Like  Benozzo  Gozzoli,  Pausias 
excelled  in  the  painting  of  children,  little  boys  especially, 
and  none  could  rival  his  painting  of  flowers.  Polycletus 
had  already  framed  this  building  out  of  the  glowing  shapeli 
ness  of  anemones  and  the  delicate  loveliness  of  the  pale 
and  golden  daisies  of  the  fields.  Pausias,  in  the  chaplet  of 
Eros,  no  doubt  justified  the  words  used  of  his  skill  by 
Pliny,  who  says  that  "  he  brought  the  much  practised  art  of 
painting  flower-garlands  to  the  climax  of  harmonious  varie 
gation."  Here  then  was  the  gist  of  his  flower-like  building 
in  a  boy  embowered  with  blossoms,  who  was  in  act  of 
choosing  the  better  part — music  instead  of  mischief-making 
arrows.  The  lyre  is  better  far,  since  music  charms  the 
highest,  the  deepest,  and  the  inmost  soul,  and  therefore 
best  symbolises  Eros,  the  awakener  of  unstinting  and  ex- 
haustless  love. 

But  we  have  not  exhausted  the  wonders  of  the  Tholos. 
The  very  centre  of  its  beautifully  tesselated  floor  had  a 
downward  exit.  Here  was  most  artfully  constructed  a 
labyrinth,  traceable  still  in  all  its  windings.  These  made  it 
necessary  to  pass  forwards  and  backwards,  going  three  times 
completely  around  the  circle  before  the  lower  door  of  the 
exit  could  be  reached.  The  use  of  this  subterranean 
labyrinth  is  no  easier  to  make  out  than  that  of  the  whole 
building.  Some  think  the  harmless  snakes  sacred  to 
Aesculapius  had  quarters  here,  and  issued  hence  to  play 
their  part  in  healing  visions.  But  this  is  pure  conjecture, 
as  is  also  the  suggestion  that  the  Tholos  was  built  around 


254      J ESCULA PIUS  A  T  EPIDA  UR US  AND  A  THENS     v i 

a  well  or  a  spring  whose  healing  waters  have  wholly  dis 
appeared. 

Here  is  no  room  for  speaking  of  the  part  in  miraculous 
cures  played  at  Epidaurus  by  the  venomless  serpents  which 
still  abound  in  those  parts,  and  are  made  very  prominent  by 
Aristophanes  in  his  famous  burlesque  account  of  a  night  in  the 
Athenian  sanctuary  of  Aesculapius.  We  cannot  enter  here 
into  the  history  of  the  serpent  impostures  practised  by  that 
arch-mountebank  Alexander  of  Abonotichus.  The  considera 
tion  of  all  the  miracles  commemorated  by  inscriptions  at 
Epidaurus,  and  of  the  trial  which  they  were  to  the  faith  of 
pious  believers,  would  be  too  long.  The  distinction  between 
Roman  and  Greek  Aesculapia,  as  well  as  the  whole  question 
of  the  relation  between  sound  practice  and  that  of  the  priests 
of  Aesculapius,  of  Serapis,1  and  of  a  host  of  divinities  who 
sprang  up  in  the  latter  days  of  paganism  to  cure  all  diseases, 
must  be  left  without  discussion  here.  It  is  sufficient  to  know 
that  men  of  reasonable  minds  continued  even  in  later  days  to 
resort  to  the  various  shrines  of  healing,  and  frequently  found 
restoration  and  consolation  by  that  means. 

Pain  of  whatsoever  kind  moved  the  benign  hero-physician, 
the  divine  Aesculapius.  His  aid  therefore  was  granted  to 
all  those  needing  it  if  they  only  could  receive  it.  The 
possibility  of  receiving  it  depended  in  one  sense  not  upon 

1  Egypt,  from  which  it  is  supposed  many  features  of  the  earliest  worship 
of  Aesculapius  were  borrowed,  sent  forth  in  later  times  the  healing  god 
Serapis,  a  powerful  rival  to  Aesculapius.  How  closely  his  original  Egyptian 
character  clung  to  Serapis  even  in  his  shrine  at  Delos  may  be  gathered  from 
an  absurd  story  preserved  by  Aelian.  We  are  told  that  Serapis  granted 
the  restoration  of  an  eye  to  a  horse  who  was  brought  in  distress  to  his 
temple.  The  horse  was  of  course  a  thoroughbred,  and  naturally  made  his 
appearance  in  the  temple  with  thankofferings.  This  last  touch  recalls  the 
sayings  and  doings  of  the  deathless  steeds  of  Achilles,  and  the  whole 
episode,  like  the  Homeric  account  of  the  wounding  and  fall  of  Nestor's 
horse,  is  based  upon  a  commiseration  for  suffering  beasts  which  finds  ex 
pression  in  modern  times  more  substantially  but  less  poetically. 


vi  APOLLONIUS  THE  PURE  255 

him  at  all  but  solely  on  them.  They  had  to  have  faith, 
and  such  faith  that  it  blossomed  into  purity.  The  pre 
liminary  laving,  usually  in  sea -water,  required  before 
entering  the  porch  to  await  the  coming  of  inspired  dreams, 
symbolised  outwardly  the  inner  obedience  of  the  faithful  to 
a  command  inscribed,  as  we  well  know,  upon  the  doorway 
of  entrance  to  the  Epidaurian  sanctuary — 

"  None  but  the  pure  shall  enter  here." 

I  have  used  the  word  faith  because,  in  addition  to  purity, 
there  was  a  deeper  tie  involved,  a  personal  compatibility 
between  the  suppliant  and  the  divinity  supplicated :  to  be 
healed  by  the  god  it  was  needful  to  be  pleasing  in  his  eyes, 
otherwise  he  failed  to  appear. 

Here  was  a  religious  idea  capable  of  many  abuses,  but 
useful  and  right  for  controlling  the  self-indulgent  who  stood 
between  themselves  and  health.  An  especial  oracle  from 
the  god  could  not  intervene  at  every  meal  prescribing  each 
disobedient  patient's  meat  and  drink.  For  this  duty  as  well 
as  for  an  example  some  one  especially  accredited  by  the 
favour  of  the  god  and  qualified  by  the  rigour  of  his  own 
life  was  needed.  Such  a  divinely  chosen  guide  for  the 
weak  and  erring  was  young  Apollonius  of  Tyana,1  during 
his  monastic  seclusion  in  the  temple  of  Aesculapius  at 
Aegae.  He  was  especially  called  to  the  Pythagorean  life 
and  discipline,  his  revival  of  which  begins  with  his  recourse 
to  Aesculapius  and  his  rejection  of  the  teachings  of  Epi 
curus.  "  Wouldst  thou  but  talk  with  Apollonius,  thy  relief 
is  sure,"  said  the  oracle  at  Aegae  to  an  unruly  and  self- 
indulgent  youth  whose  much  eating  and  drinking  prevented 
his  cure. 

1  See  Appendix  IV. 


256  '  AESCULAPIUS  A 7"  EPIDAURUS  AND  ATHENS     vi 

This  idea  of  the  necessity  of  some  one  whose  life  should 
be  purity  incarnate,  and  who  should  intercede  with  Aescu 
lapius  (himself  thought  of  in  Homeric  days  as  an  inter 
cessor),  became  more  prominent,  and  one  of  the  very  last 
glimpses  given  us  of  the  persistent  worship  of  Aesculapius 
upon  the  Athenian  Acropolis  is  in  the  life  of  Proclus,  of 
whom  we  hear  as  one  of  those  holy  men  whose  intervention 
was  all-powerful  with  the  god  of  healing. 

Thus  as  we  bid  farewell  to  Aesculapius  he  seems  himself 
in  act  of  bidding  farewell  to  earth  and  is  withdrawing  him 
self  from  men  to  the  far-off  dwellings  of  the  careless 
Olympian  gods. 


APPENDIX    IV 
APOLLONIUS   OF  TYANA 

THIS  Tyanaean  master  of  miracles  attached  his  teachings 
and  his  philosophy  to  Pythagoras — a  name  to  conjure  by — 
and  his  miracles  received  a  certain  divine  sanction  from  his 
acceptableness  in  the  eyes  of  Aesculapius.  In  his  early 
youth,  at  the  age  of  sixteen,  a  sort  of  inner  light  irresistibly 
prompted  him  to  leave  Tarsus  and  his  first  master  who  was 
an  Epicurean,  and  to  resort  to  the  temple  of  Aesculapius 
at  Aegae,  a  Cilician  town  not  far  away.  Neither  Tarsus 
nor  Aegae  in  fact  was  very  far  from  Tyana  of  Cappadocia, 
his  birthplace.  During  four  years  Apollonius  lived  in  a 
monastic  seclusion  at  Aegae,  increasing  in  stature  and  in 
favour  with  the  god,  loving  with  unspeakable  love  and 
strictly  living  the  ascetic  Pythagorean  life.  An  account  of 
these  years  was  written  by  his  contemporary,  probably  also 
a  sharer  in  his  life  of  self-denial  and  self-devotion, 
Maximus  of  Aegae. 

The  more  adventurous  and  most  miraculous  career  of 
Apollonius  upon  his  travels,  and  during  his  trial,  was 
chronicled  in  rough  notes  by  his  companion  Damis  the 
Ninivite,  a  remarkably  credulous  person,  who  seems  in 
vented  for  the  purpose  of  believing  more  than  the  utter 
most  possible.  The  Grammar  of  this  Ninivite's  Assent  in 
deed  makes  exceptions  into  rules,  and  leaves  nothing  that 
can  surprise  except  the  normal  and  natural  course  of  events. 


258  APPENDIX  IV 

A  third  account  of  Apollonius  was  written  contemporarily 
by  Moeragenes.  Possibly  this  was  the  same  Moeragenes 
who  figures  as  ah  Athenian  elsewhere,1  but  whoever  he 
was,  he  wrote  in  four  books  a  life  of  Apollonius  of  Tyana. 
These  four  books  of  Moeragenes  were  read  2  by  that  great 
champion  of  Christianity  against  Paganism,  Origen,  and  his 
estimate  of  the  Tyanaean  ascetic  and  worker  of  miracles 
was  evidently  derived  from  them. 

So  far  I  have  named  writers  on  the  career  of  Apollonius 
whose  books  have  perished.  The  book  which  has  not 
perished  is  that  of  a  fourth  biographer  who,  unlike 
Maximus,  Damis,  and  Moeragenes,  was  not  a  contem 
porary.  His  name  was  Philostratus,  and  he  has  most  aptly 
been  called  "Romancer -in -ordinary"3  to  her  Imperial 
Majesty  Julia  Domna.  When  he  had  culled  marvellous 
incidents  from  Maximus,  and  gathered  in  romantic  ad 
ventures  and  incredible  miracles  from  Damis,  Philostratus 
found  the  narrative  of  Moeragenes — full  of  wonders  though 
it  apparently  was — too  tame,  and  therefore,  choosing  to  think 
that  it  betrayed  ignorance  of  his  hero,  he  neglected  the 
best  material  at  hand.4  In  place  of  this  he  added  to  what 
was  already  untrustworthy  popular  rumours  and  traditional 
records  of  miracles — a  mass  of  mythology  which  had 
gathered  around  Apollonius  during  the  century  separating 
his  death  from  the  day  of  Philostratus  and  his  protectress, 
the  Empress  Julia.  Among  these  tales  were  no  doubt 
many,  if  not  all,  of  the  features  which  Philostratus'  work  has 
in  common  with  the  four  Gospels  and  the  career  of  St. 
Paul  as  set  forth  in  the  Acts  of  the  Apostles.  That  nothing 
might  be  lacking,  Philostratus  contrived  to  use  in  one  way 
or  another  various  favourite  passages  of  his  borrowed  from 


1  Pseudo- Plutarch,  Quaest.  Conv.  book  iv.  end.     It  is  to  be  hoped  that 
he  is  no  kinsman  of  the  bandit  Moeragenes  who  ranged  the  Taurus  a  few 
generations  before.     Cf.  Cic.  Ad  Att.  v.  15  and  vi.  i. 

2  Origen,  Con.  Cels.  vi.  41. 

3  Essays  and  Studies,  by  B.  L.  Gildersleeve. 

4  E.  Miiller,  Eine  culturhistorische  Untersuchung. 


APOLLONIUS  OF  TYANA  259 

Xenophon,1  an  author  upon  whom  he  formed  his  style, 
and  to  adorn  the  already  overloaded  travellers'  tales  of 
Damis  with  elegant  extracts  from  such  records  of  history 
and  travel  as  were  accessible  to  him.  He  is  evidently 
indebted  here  and  there  to  Lucian's  True  Story,  a  charming 
caricature  of  the  marvellous-absurd  yarns  which  had  so  long 
a  vogue  at  Rome.  The  way  in  which  this  last  indebtedness 
is  contracted  abundantly  convicts  Philostratus  of  an  utter 
lack  of  humour.  Indeed  this  weakness  is  the  pith  and 
marrow  of  the  whole  biography,  as  may  most  agreeably  be 
revealed  to  readers  of  the  delightfully  humorous  summary 
of  the  work  given  by  Professor  Gildersleeve  in  his  essay  on 
"  Apollonius  of  Tyana." 

Philostratus  was  not  called  to  the  office  of  writing  the 
"  Evangel  of  Apollonius  "  by  an  inner  light,  as  Apollonius 
was  to  the  Pythagorean  way  of  living.  The  first  suggestion 
of  it  came  from  the  Empress  Julia  Domna,  to  whose  re 
markable  literary  circle  Philostratus  belonged.  We  know 
just  enough  of  this  circle  to  see  that  it  contained  many 
cleverer  people  than  Philostratus,  and  it  seems,  from  admis 
sions  in  his  preface  to  the  book,  that  Philostratus  wrote, 
or  if  you  choose  compiled,  the  life  of  Apollonius  with  a 
hampering  desire  to  suit  the  tastes  of  this  coterie.  There 
was  Moesa,  Julia's  sister,  a  particularly  domineering  person, 
as  history  was  soon  to  show  ;  and  Moesa  had  two  daughters 
who  well  knew  their  own  minds.  Supposing  these  ladies 
interested  in  having  some  one  write  an  ideal  presentation  of 
the  life  of  Apollonius,  to  whom  could  they  turn  ?  They 
might  think  of  their  legal  friend  Ulpian,  like  themselves  of 
Phoenician  descent,  but  he  was  out  of  the  question ;  nor 
could  they  expect  literary  skill  of  Papinian,  as  they  well 
knew,  since  he  was  a  kinsman  of  theirs.  Of  their  circle, 
however,  were  Aelian  the  honey-tongued,  and  Philostratus. 
The  ridiculous  credulity  of  Aelian  makes  it  possible  that 
he  would  have  done  worse  even  than  Philostratus.  Julia 
perhaps  chose  Philostratus  because  he  represented  Greek 

1  See  a  dissertation  by  C.  von  Wulfften  Pathe.     Berlin  1887. 


26o  APPENDIX  IV 

culture  and  had  no  convictions  of  his  own.  It  was  his 
clever  style  that  she  especially  appreciated.  To  him  the 
empress  gave,  as  he  is  careful  to  relate  in  beginning, 
the  notes  roughly  made  by  Damis  the  Ninivite  and  en 
trusted  to  a  kinsman  from  whom  she  got  them  in  the  pro 
cess  of  collecting  books,  which  occupied  much  of  her 
attention.  Out  of  these  notes  she  thought  the  skill  of 
Philostratus  could  fashion  a  record  which  should  embody 
the  Way,  the  Truth,  and  the  Life.  Being  daughter  to  a 
Phoenician  who  was  high  priest  in  the  temple  of  the  Sun 
at  Emesa,  she  naturally  felt  that  impulse  toward  intensi 
fying  and  reforming  men's  faith  which  characterised  in 
those  days  her  native  corner  of  the  world.  Thence  had 
come  not  only  Apollonius  of  Tyana,  but  Paul  of  Tarsus 
and  Simon  Magus,  as  well  as  the  beginnings  of  Christianity, 
and  out  of  Julia's  own  house  was  soon  to  come  Elagabalus, 
whose  execrable  reign  was  a  Nightmare  of  Religious 
Reformation. 

We  may  well  shudder  to  think  what  might  have  been 
the  result  if  our  records  of  the  life  of  Christ  had  fallen 
into  the  hands  of  a  Philostratus.  What  if  we  had,  instead 
of  the  four  Gospels,  a  smoothed  and  would-be  racy  narra 
tive  written  in  the  vein  of  your  Parisian  feuilletonist,1  to 
suit  the  tastes  of  a  circle  far  more  definitely  restricted  than 
the  modem  "  tout  Paris  "  ?  An  irritating  impression  of  un 
reality,  which  forces  itself  upon  the  reader  of  Philostratus' 
Life  of  Apollonius,  makes  it  hard  to  get  through  even 
one  of  its  eight  tedious  books.  No  one  has  more  pithily 
expressed  the  feelings  of  all  upon  this  point  than  Erasmus 
in  his  preface  to  St.  Luke  addressed  to  Henry  the  Eighth. 
I  quote  from  an  old  translation  :  "  Who  readeth  the  lyfe 
of  Apollonius  Tyaneus  any  otherwayse  then  as  a  certayne 
dreamel  Yea  or  rather  who  vouchsalueth  to  reade  it  at  all  ?  " ! 

1  G.  Bernhardy  applies  this  word  to  Philostratus  in  the  Allgemeine 
Litteratur  Zeitung,  1839. 

-  Quis  Apollonii  Tyanaei  vitam  non  veluti  somnium  quoddam  legit  ? 
Imo  quis  legere  dignatur  ? 


APOLLONIUS  OF  TYANA  261 

Indeed  the  surprising  fact  is  that  respect,  veneration,  and 
even  worship  should  have  attached  to  Apollonius  after  he 
had  been  so  completely  victimised  by  the  journalism  of  his 
day.  And  this  surprising  fact  is  surely  an  argument  for 
something  made  of  solider  stuff  than  dreams,  something 
really  admirable  in  the  true  Apollonius,  if  anybody  ever 
disentangled  him  from  the  play-acting  personage  of  Philo- 
stratus. 

No  one  has  the  means  of  forming  at  the  present  day  an 
independent  opinion  of  Apollonius.  We  are  obliged  first 
to  reject  the  picture  of  him  given  by  Philostratus,  and  then, 
either  to  have  no  mind  about  him  at  all  or  to  be  of  the 
mind  of  those  who  did  know  the  facts.  These  may  be 
separated  into  three  classes  :  first  come  the  religious-minded 
Christians  who  in  spite  of  their  prejudice  against  a  reviver 
of  paganism  respected  in  Apollonius  a  man  of  saintly  life 
and  religious  inspiration ;  second  come  the  scoffers  like 
Lucian  and  Apuleius,  who  did  not  take  religion  seriously, 
and  thought  such  a  man  necessarily  either  a  knave  or  a 
fool,  or  both  ;  third  comes — in  a  class  all  by  himself — Dio 
Cassius  the  historian,  who  evidently  never  quite  appre 
ciated  what  he  was  talking  about  in  mentioning  Apollonius. 
In  one  place  where  he  is  irritated  he  says  substantially, 
"Caracalla  preferred  the  company  of  freedmen  to  that  of 
men  of  my  mark,  and  was  given  over  to  cheats  such  as  was 
Apollonius  the  Cappadocian."  1  Dio  forgets  the  tremendous 
endorsement  with  which  he  has  previously  accompanied  a 
much  clumsier  account  than  is  given  by  Philostratus  of  one 
of  Apollonius'  great  deeds.  "  No  matter  how  much  men 
may  doubt  it,  this  story  is  as  true  as  truth  itself,"  "2  he  says 
after  relating  that  while  Domitian's  assassination  was  in 
progress  at  Rome  a  certain  man  of  Tyana  named  Apollonius 
went  up  to  a  high  place  and  described  the  event  before  the 
people  as  if  he  were  present,  though  he  and  they  were  in 
Ephesus  "or  some  such  place." 

1  Ixxvii.  1 8. 
~  TOVTO  fj.tv  ourws  tytvero,  K&V  /JLVpiaKis  rts  diriffT^ffr).      Ixvii.  end. 


262  APPENDIX  IV 

Of  the  three  classes  above  mentioned  of  those  who  had 
opportunity  to  know  the  facts  about  Apollonius,  the  third, 
which  contains  only  Dio,  need  occupy  us  no  further, 
because  Dio  plainly  had  never  given  any  serious  thought  to 
the  career  of  the  Tyanaean,  and  did  not  know  that  his  won 
derful  Ephesian  clairvoyant,  Apollonius,  was  Apollonius 
the  Cappadocian,  whom  he  scorned.  The  second  class 
contains  Lucian  and  Apuleius,  each  of  whom  mentions 
Apollonius  just  once. 

Apuleius,  in  his  self-vindication  against  the  charge  of 
using  sorcery  to  win  the  hand  of  the  heiress  Prudentilla, 
says  indignantly,  after  repudiating  the  charge  :  "If  you  can 
but  prove  the  least  colourable  motive  of  self-interest  in  my 
suit  for  Prudentilla,  then  I  grant  you  all.  Then  you  shall 
call  me  a  Carinondas,  a  Damigeron,  a  very  Moses,  an 
Apollonius,  Dardanus  himself,  or  any  magician  of  vaunted 
repute  since  Zoroaster  and  Hostanes."  Here  murmurs 
interrupt  him,  and  he  turns  to  the  magistrate,  saying, 
"Maximus,  you  see  the  hubbub  they  make,  because  I 
named  a  few  magicians'  names.  What  can  be  done  with 
people  so  low  and  uncivilised  ?  "  It  does  not  suit  Apuleius, 
for  the  purposes  of  his  defence,  to  say  more  than  this,  but 
he  plainly  suggests  that  Apollonius,  who  had  not  been  dead 
more  than  forty  years  at  the  time,  was  better  than  popular 
prejudice  would  allow. 

Lucian  of  Samosata  —  a  Voltaire  capable  of  making 
merry  over  the  burning  of  Servetus — bears,  by  his  silence 
more  than  by  what  he  says,  a  similar  testimony  to  the 
quality  of  Apollonius,  whom  he  dismisses  with  a  sneer  at  the 
absurdity  of  the  play-actor's  part  which  he  played.  If  the 
Apollonius  known  to  him  had  been  really  the  Philostratus- 
Apollonius,  Lucian  would  have  made  merry  over  all  his 
adventures  and  pretensions l  instead  of  saying  simply  that 

1  For  a  forcible  presentation  of  this,  see  Bishop  Lloyd's  letter  to 
Bentley,  who  had  consulted  him  about  a  proposed  edition  of  the  life  of 
Apollonius.  After  printing  one  or  two  pages  Bentley  abandoned  this 
project. 


APOLLONIUS  OF  TYANA  263 

the  first  teacher  and  corrupter  of  the  arch-impostor  Alex 
ander  was  a  follower  of  the  "tragic  Tyanaean." 

When  it  is  remembered  that  Lucian  believed  fervently 
in  the  doctrines  of  Epicurus,  against  which  the  whole  life 
of  Apollonius  had  been  a  protest,  an  inclination  to  neglect 
any  insinuations  from  this  source  will  become  irresistible. 
Accordingly  we  have  left  for  our  advisers  only  the  Chris 
tian  opponents  of  Apollonius,  who  seem  the  only  persons 
that  took  the  pains  to  understand  him.  The  general  im 
pression  which  they  had  of  him  is  borne  out  by  the  best  of 
all  possible  witnesses — Apollonius  himself.  There  remains 
an  extract  from  a  work  written  by  the  Tyanaean's  own 
hand,  which  was  so  prized  that  it  was  engraved  at  Byzant 
ium  on  pillars  of  brass. 

This  quotation  is  cited  twice  by  Eusebius  of  Caesarea — 
whose  date  is  330  A.D. — once  in  his  Preparation  for,1  and 
once  in  his  Demonstration  of,2  the  Gospel.  "  Even  the 
well-known  Apollonius  of  Tyana,  whose  name  is  upon  all 
men's  lips  for  praise,  is  said  to  write  much  in  the  same 
strain  in  his  work  on  sacrifice  about  the  first  and  the  great 
God."  Then  follows  a  confirmation  of  the  account  in 
Philostratus  of  the  new  Pythagorean  doctrine  of  his  hero. 
Apollonius  teaches  that  "there  is  one  Highest  God  above 
and  apart  from  the  lower  and  many  gods.  Beyond  the 
reach  of  the  contaminating  world  of  sense  as  he  is,  nothing 
apprehensible  by  any  organ  of  sense,  neither  burnt  offerings 
nor  bloodless  sacrifices,  can  reach  him,  not  even  uttered 
prayers.  He  is  the  substance  of  things  seen,  and  in  him 
plants,  animals,  men,  and  the  elements,  of  which  the  world 
is  made,  have  life  and  exist.  He  is  the  noblest  of  exist 
ences  ;  and  men  must  duly  worship  him  with  the  only 
faculty  in  them  to  which  no  material  organ  is  attached — 
their  speculative  reason."  Eusebius  quotes  this  with 
approval  as  coming  from  the  most  illustrious  philosophers 
of  Greece,  and  thus  pays  to  Apollonius,  by  appealing  to 
him  against  the  baser  sort  of  paganism,  the  highest  tribute 

1  C.  12,  book  iv.  '-  C.  5,  book  iii. 


264  APPENDIX  IV 

that  a  Christian  could.  This  is  the  more  significant,  be 
cause  Eusebius  is  the  one  who  was  at  great  pains  to  refute 
Hierocles  on  behalf  of  the  Christians. 

Hierocles  maintained  that  Apollonius  had  lived  a  more 
exemplary  and  divine  life  upon  earth  than  the  Christ  of  the 
Gospels.  His  miracles  were  more  numerous,  Hierocles 
said,  and  better  vouched  for  than  those  of  Christ,  and  yet 
there  was  no  pretence  that  he  was  god,  but  only  one 
favoured  of  the  gods.  All  this  Eusebius  attacks  and 
refutes  without  pretending,  as  many  carpers  at  the  Philo- 
stratus-Apollonius  have,  that  Apollonius  was  "  a  devil  with 
bat's  wings  and  a  long  tail."1  He  very  rightly  holds 
Philostratus  responsible  for  many  of  the  erroneous  preten 
sions  of  Hierocles.  Origen,  who  wrote  his  defence  of 
Christianity  against  Celsus  before  this  controversy  between 
Hierocles  and  Eusebius  took  place,  and  before  any  attempt 
to  put  Apollonius  on  a  par  with  Christ  had  been  made, 
speaks  with  the  same  temperate  respect  of  Apollonius, 
mentioning  him,  however,  as  a  worker  of  miracles.2 

Other  authors  confirm  some  of  the  better  traits  of  the 
Apollonius  of  Philostratus.  St.  Augustin,  for  instance, 
commends  him  as  exercising  a  larger  measure  of  self-control 
than  Zeus.  Eunapius  speaks  of  his  life  as  proving  him 
better  than  a  philosopher.  "  He  was  something  between 
the  divine  and  the  human,  and  his  is  not  so  much  a  life  as 
the  sojourn  of  a  god  with  men."  Eunapius,  however,  was 
a  pagan,  and  his  testimony  is  of  very  different  weight  from 
that  given  by  Christian  writers.  Moreover  Eunapius 
certainly  had  in  mind  chiefly,  if  not  solely,  the  Apollonius 
of  Philostratus.  Concurrent  testimony  to  the  high  standing 
of  Apollonius  may  be  accepted  with  all  due  reservations 
from  various  pagan  sources.  There  was  doubtless  some 
thing  in  the  facts  about  him  known  to  the  Empress  Julia 
and  her  circle  which  singled  him  out  as  the  best  hero  for 
a  Pagan  reformation.  Less  weight  attaches  to  her  son 

1  More  the  Platonist  puts  this  phrase  into  the  mouths  of  the  detractors 
of  Apollo  and  his  oracles.  2  Contr.  Cels.  vi.  41. 


APOLLONIUS  OF  TYANA  265 

Caracalla's  building  of  a  temple  in  his  honour,  but  the 
same  is  hardly  true  of  the  conduct  of  her  great-nephew, 
the  Emperor  Alexander  Severus.  In  his  Lararium,  says 
Spartianus,  he  had  not  only  the  statues  of  deified  emperors, 
but  also  a  choice  of  the  most  righteous — the  more  especially 
hallowed  souls.  Among  these  was  Apollonius,  and  also, 
according  to  a  contemporary  writer,  Christ,  Abraham,  and 
Orpheus,  as  well  as  his  own  ancestors.  This  is  a  particu 
larly  interesting  passage,  for  it  tends  to  show  how  the 
religious  impulse  which  led  men  to  deify  the  emperors, 
finding  in  their  unsaintly  lives  its  own  corrective,  was  led 
on  to  the  deification  of  any  and  all  conspicuously  noble 
characters.1  Who  shall  say  that  the  state  of  mind  to  which 
this  led  did  not  make  men  better  able  to  reach  the  ideal 
of  the  Imitation  of  Christ  ? 

Under  the  same  category  of  pagan  tributes  to  the 
character  and  sanctity  of  Apollonius  falls  the  tale  of  how 
he  saved  his  native  Tyana  from  destruction  by  turning 
aside  the  wrath  of  Aurelian.  Apollonius  appeared,  and  by 
his  bodily  presence  saved  Tyana,  just  as  Athena  is  said  to 
have  shown  herself  to  the  affrighted  Alaric,  who  spared 
Athens  in  consequence.  These  latest  results  of  the  myth- 
making  power  in  paganism  bear  a  striking  family  resem 
blance  to  the  earliest  lives  of  the  Christian  saints.  In 
them  gods,  demi-gods,  and  heroes  alike  are  assimilated  to 
the  status  of  patron  saints. 

But  to  return  to  the  most  important,  the  only  convincing, 
tributes  to  the  excellency  of  Apollonius  of  Tyana,  those 
from  early  Christian  writers,  the  last,  but  not  the  least, 
which  shall  be  mentioned  is  in  a  letter  of  the  Christian 
Sidonius  Apollinaris  to  a  counsellor  of  Evariges,  king  of 
the  Goths.2  As  Cudworth  (writing  in  a  rather  prejudiced 
vein)  says  of  him  :  "  though  a  Christian,  he  (Sidonius)  was 

1  The  pinnacle  of  sanctity  reached  by  Apollonius  did  certainly  place 
his  claims  to  reverence  on  a  par  with  those  of  the  emperors.     This  accounts 
for  the  well-known  coins  that  bear  his  profile  upon  them. 

2  Letters,  book  viii. ,  third  letter. 


266  APPENDIX  IV 

so  dazzled  with  the  glittering  show  and  lustre  of  his 
(Apollonius')  counterfeit  virtues,  as  if  he  had  been  enchanted 
by  this  magician  long  after  his  death."  The  following  is  a 
sufficient  though  not  minutely  accurate  translation  of  the 
letter  in  question.  "  Read  of  a  man  (with  reverence  to  the 
Catholick  faith  be  it  spoken)  in  most  things  like  yourself, 
sought  after  by  the  rich,  yet  not  seeking  after  riches; 
covetous  of  knowledge,  not  of  money;  abstemious  in  the 
middle  of  feasts,  plainly  cloathed  amongst  the  sumptuous, 
severe  amongst  the  luxurious,  rough  and  unadorned  in  the 
midst  of  delicate  nations,  and  shining  with  a  venerable 
negligence  amongst  the  wanton  nobles  of  Persian  kings. 
And  when  he  made  no  use  of  the  flocks  either  for  food  or 
apparel,  he  was  rather  slighted  than  envied  in  the  kingdoms 
thro'  which  he  travelled,  and  when  the  good  fortune  of 
kings  favoured  him  in  everything,  he  only  asked  those 
favours  which  he  was  more  ready  to  give  than  to  take." 1 

Surely  we  need  not  be  more  prejudiced  against  Apol 
lonius  than  were  his  Christian  antagonists  in  the  days 
when  more  was  known  about  him  than  we  can  know 
to-day.  We  may,  at  least,  leaving  momentarily  out  of 
account  whether  he  was  or  was  not  in  himself  an  impostor, 
be  thankful  that  the  idealised  figure  of  Apollonius  rallied 
around  itself  so  much  of  the  life  and  enthusiasm  of  depart 
ing  paganism ;  for  there  were  points  of  striking  similarity 
between  his  life  and  teaching  and  that  life  the  imitation  of 
which  is  Christianity.  We  may  safely  see  in  the  attempted 
reforms  of  Apollonius  a  preparation  of  the  pagan  world  for 
Christianity,  since  "even  Christians  have  thought  rever 
ently  of  him,  and  believed  that  he  did  his  wonders  by  the 
power  of  God,  or  by  secret  philosophy  and  knowledge  of 
nature  not  revealed  to  other  men." 2 

1  For  this,  version  see  Bayle's  Dictionary,    translated    in   1735,    s.v. 
"  Apollonius." 

2  Meric  Casaubon,  alluding  to  words  attributed  to  Justin  Martyr. 


APPENDIX  V 

THE   STATUS   OF   MODERN    GREEK   DOCTORS 

IN  almost  any  Greek  village  you  choose,  the  man  whom  all 
delight  to  honour  is  pretty  sure  to  be  a  doctor.  The 
mayor,  or  demarch,  whose  courtesy  I  experienced  at 
Thebes,  was  a  medical  practitioner,  and  in  many  other 
places — Kalamata,  for  instance — great  kindnesses  came  to 
me  from  doctors,  men  of  influence  always.  In  one  town 
of  considerable  importance  the  apothecary  had  been  elected 
mayor. 

What  the  Greeks  have  always  admired  is  that  men's 
intellects  should  unerringly  hit  the  mark.  Nowhere  can 
this  unerring  insight  be  better  shown  than  by  the  swift  leap 
of  a  wise  doctor's  mind  to  the  truth  about  disease.  In  the 
United  States  of  America  there  was,  at  the  beginning  of 
this  century,  an  admiration  more  than  Greek  for  doctors, 
since  they  were  credited  with  such  artistic  capacity  that  a 
Dr.  Thornton  was  encouraged  to  compete  with  a  pro 
fessional  architect  for  the  honour  of  building  the  Capitol 
at  Washington.1  Special  causes,  however,  brought  Greek 
doctors  into  the  forefront  of  intellectual  life  after  the 
supremacy  of  the  Turks. 

"  Under  the  Turkish  rule  and  before  the  Greek  revolu- 

1  See  Mr.  Henry  Adam's  interesting  account  of  this  matter  in  his 
History  of  the  United  States  of  America  during  the  first  Administration 
of  Thomas  Jefferson,  vol.  i.  ch.  iv.  p.  in. 


268  APPENDIX  V 

tion,"  my  friend  M.  Demetrios  Bikelas  writes  me,  "most 
of  the  learned  Greeks  who  were  not  clergymen  were  doctors. 
They  could  not  pursue,  with  any  chance  of  profit,  any 
other  branch  of  studies.  Medicine  sometimes  opened 
the  way  to  higher  positions.  Thus  the  celebrated  Alex 
ander  Mavrocordato,  grand  drogman  of  the  Porte  (1636- 
1714),  was  a  doctor  of  Padua.  Italy  was  then  the  place 
whither  such  Greek  students  as  could  leave  their  country 
resorted.  At  the  beginning  of  this  century  a  great  many 
of  the  learned  Greeks  who  helped  the  revival  of  their 
country  were  equally  doctors.  Neither  these  nor  those;  of 
the  preceding  centuries  confined  themselves  to  medicine ; 
some  were  poets,  such  as  Vilaras  of  Janina  (1771-1823). 
Even  Christopoulos,  one  of  the  best  lyric  poets  of  Greece 
(1772-1847),  had  studied  medicine  at  Buda ;  Coray,  the 
famous  scholar,  was  a  doctor  of  Montpellier  (1748-1833). 
Colettis,  who  played  a  great  part  during  the  war  of  the 
revolution,  and  died  Prime  Minister  of  Greece  in  1847, 
began  his  career  as  the  doctor  of  Ali  Pasha  of  Janina. 
The  fathers  of  the  Caratheodori  —  two  cousins,  one  of 
whom  (after  having  been  minister  of  foreign  affairs,  and  the 
representative  of  Turkey  at  the  Congress  of  Berlin)  is 
actually  Prince  of  Samos,  and  the  other  Minister  of 
Turkey  at  Brussels — were  both  distinguished  doctors  at 
Constantinople. 

"  The  calling  of  a  doctor  was  so  honoured  that  custom 
had  attached  to  it  the  title  of  '  Your  Excellence,'  ^o^rarc. 
Doctors  are  no  more  addressed  by  this  title  in  Athens,  but 
I  am  not  sure  that  there  are  not  still  Greek  countries 
where  they  enjoy  this  title,  generally  applied  to  diploma 
tists.  Since  the  revolution  and  the  institution  of  the 
Greek  university,  medicine  is  so  far  from  being  the  only 
branch  of  knowledge  pursued  in  Greece,  that  the  faculty  of 
law  draws  the  greatest  number  of  students  at  the  Athenian 
University.  And  yet  there  has  been  the  natural  increase, 
a  very  considerable  one,  in  the  number  of  Greek  doctors. 
It  would  be  invidious  to  choose  names  among  the  Greek 


STATUS  OF  MODERN  GREEK  DOCTORS 


269 


doctors  who  have  distinguished  themselves  in  Greece  of 
late  years.  I  may,  as  an  instance  of  the  aptitude  of  Greeks 
for  that  science,  mention  the  names  of  the  late  M.  Damas- 
chino  and  of  M.  Panas,  who  had  both  attained  to  the 
Professorship  in  the  Faculty  of  Paris.  I  may  also  mention 
the  well-known  author,  Dr.  Paspati,  who  studied  in 
America.  Dr.  Cavafy,  too,  one  of  the  physicians  of  St. 
George's  Hospital  in  London,  is  a  Greek  by  birth." 


VII 

APHRODITE   AT   PAPHOS 

THE  statue  of  Aphrodite  found  by  excavation  on  ground 
sacred  to  Epidaurian  Aesculapius  represents  that  smiling 
goddess  as  the  wearer  of  a  sword.1  Here  is  an  ancient 
reproduction  of  that  beautiful  statue  fashioned  by  the 
Athenian  Alcamenes,2  a  statue  whose  cheeks  and  full  front 
face  were  lauded  by  Lucian  of  Samosata.  Of  this  early 
masterpiece  the  Epidaurian  statue,  though  mutilated,  gives  a 
charming  suggestion.  But,  alas  !  Lucian  tells  of  a  pulse 
of  sweet  harmony  in  its  rounded  and  dainty  wrist,  and  a 
light  movement  of  charm  and  delicacy  in  the  fingers.  These 
perfections  we  shall  never  see,  for  the  Epidaurian  copy  is 
bereft  of  hands. 

But  why  should  a  statue  of  Aphrodite  grace  the 
sanctuary  at  Epidaurus?  The  same  reason,  though  hard 
for  us  to  give,  suffices  which  established  her  worship  at 
Athens  on  the  south  side  of  the  summit  of  the  Acropolis, 
not  far  above  the  Athenian  precinct  of  Aesculapius.3  There 

1  See  a  beautiful  photograph  in  Brunn's  Denkmaler  series. 

2  See  Furtwangler,  in  Roscher's  Ausfiirliches  Lexicon,  p.  413. 

3  Still  nearer  to  this  Asclepieium  was  a  shrine  probably  of  Aphrodite 
Pandemos,  i.e.  that  aspect  of  the  goddess  which  brings  her  into  the  worst 
company. 


vii          APHRODITE  AT  VARIOUS  SANCTUARIES         271 

she  was  worshipped  as  Sosandra,1  a  rescuer  of  men,  and 
there  stood  the  wonderful  and  very  ancient  statue  by 
Calamis.  At  Athens,  furthermore,  was  the  great  original  by 
Alcamenes  just  spoken  of,  the  Aphrodite  of  the  garden. 
Within  the  Cnidian  precinct  of  Demeter,  and  not  far  from 
her  own  home  at  Cnidus,  a  head  of  Aphrodite  was  found, 
which,  though  mutilated,  shows  in  its  expression  the 
influence  of  Demeter;  and  Aphrodite's  shrine  upon  the  pass 
near  Daphne  was  one  of  the  halting  places  of  the  procession 
to  Eleusis.  Thus  was  Aphrodite  often  associated  in 
neighbourly  fashion  with  Demeter  and  with  Aesculapius. 

At  Corinth,  Elis,  Sparta,  Delphi,  in  various  places  of 
Crete,  on  the  island  Cythera,  in  most  centres  of  Sicilian  life, 
all  over  Asia  Minor  —  everywhere,  in  fact,  where  Greek 
religion  had  its  footing,  near  Apollo's  temple  on  Delos,  on 
the  Olympian  hill  of  Cronos,  high  above  the  temple  of  great 
Zeus  himself,  the  Greek  goddess  Aphrodite  was  worshipped, 
and  statues  of  her  were  placed  near  the  special  shrines  of 
all  the  other  gods,  while  the  very  throne  of  Phidian  Zeus, 
in  his  Olympian  temple,  was  adorned  with  a  sculptured 
representation  of  her  birth  at  Paphos. 

The  influence  for  good  and  for  harm  of  the  ideal  repre 
sented  by  this  goddess  was  one  of  the  most  widely  felt  in 
antiquity,  nor  was  her  hold  upon  men's  minds  always  super 
ficial.2  According  to  the  ancient  view  of  her  power,  Aphrodite 
swayed  the  fate  of  gods  as  well  as  of  men.3  This  power  she 
exercised  not  only  to  cover  the  earth  with  fair  flowers  and 
fruits,  but  also  to  bewitch  men,  to  allure  women,  and  to 
enchant  the  mind  of  Zeus  himself,  and  high  Apollo.  The 
impulse  of  love  which  she  inspired  was  stronger  even  than 

1  See  Lucian,  Imag.  4,  and  cf.  Furtwangler  as  above,  pp.  411  and  412. 
-  See  below,  p.  303.  3  Horn.  Hymn,  iv.  247  ff. 


272  APHRODITE  AT  PAPHOS  vn 

Fate,  or  as  her  Athenian  worshippers  reverently  put  it,  she 
was  the  oldest  of  the  Fates. 

It  will  repay  us  well  to  go  and  find  this  marvellously 
strong  power  in  the  place  where,  as  an  influence  over 
Greeks,  it  had  its  birth.  Let  us  therefore  turn  to  the 
western  coast  of  Cyprus,  an  island  which  is  probably  more 
Greek  to-day  than  it  ever  was  in  the  past.1  Aphrodite  rose 
from  the  sea  at  Old  Paphos  in  Cyprus,  and  as  a  Greek 
goddess  of  strong  power,  she  ruled  Hellas  from  her  Paphian 
precinct.  This  is  significant,  for  the  fact  that  Cyprus  and 
Paphos  were  not  originally  of  Greece  prepares  us  for 
another  fact,  which  the  Greeks  themselves  well  knew,  that 
Aphrodite  came  to  them  from  many  parts  of  the  world, 
but  chiefly  from  Phoenicia,  and,  through  Phoenicia,  ulti 
mately  from  Assyria.2 

Having  her  home  in  Paphos,  aloof  from  the  centre  01 

1  See  F.  W.  Barry,  "Report  on  the  Census  of  Cyprus  "  (London,  1884). 
Of  186,173  souls  in  Cyprus  (1881)  140,793  speak  Greek,  and  42,638 
speak  Turkish,  while  Arabic,  English,  Maltese,  Armenian,  and  French  are 
spoken  by  not  inconsiderable  groups.  The  Greek  Church  has  137,631 
members,  and  there  are  45,458  Mohammedans  in  Cyprus. 

2  If  this  statement  be  so  worded  as  to  exclude  originally  Greek  charac 
teristics  from  the  early  goddess,  it  is  undoubtedly  false.  Aphrodite, 
daughter  to  Zeus  and  Dione,  appears  as  an  almost  purely  Greek  concep 
tion  (free  from  any  oriental  touch)  every  now  and  again  in  the  early  poets 
(Homer  and  Sappho),  and  there  is  an  echo  of  such  a  report  of  her  in 
Euripides ;  but  after  all  that  Holwerda  and  Engel  have  said  to  prove  her  to 
be  Greek,  and  to  derive,  so  far  as  she  was  not  Greek,  from  Asia  Minor, 
Aphrodite  remains  in  her  most  characteristic  qualities  of  Semitic  origin. 
See  Duncker's  History,  i.  p.  274.  As  a  learned  authority  says  (Robert's 
Preller,  p.  352)  ' '  Diese  Verschmelzung  der  Leben  gebenden  und  vernicht- 
enden  Macht  in  einer  weiblichen  Gottheit  finden  wir  nicht  nur  in 
Babylonien  und  Assyrien,  sondern  auch  bei  den  Semitischen  Stammen 
des  Westens,  bei  den  Syrern  den  Phoenizern  und  Carthagern,  bei  denen 
die  Baaltis,  die  Astarte-Aschera,  die  Dido- Anna  abwechselnd  Segen  und 
Frucht,  Tod  und  Verderben  sendet."  See  Plautus,  Mercator,  near  the 
beginning  :  ' '  Diva  Astarte  hominum  deorumque  vis,  vita  salus  ;  rursus 
eadem  quae  est  |  Pernicies,  mors,  interitus."  On  the  theory  of  her  Hittite 
origin,  see  Appendix  VII. 


vii  EASTERN  CHARACTERISTICS  273 

Greece,  Aphrodite  was  less  transformed  by  Hellenism 
than  any  other  of  the  figures  and  ideas  brought  to  Greece 
from  the  East.1  Here,  then,  in  Cyprus,  may  be  studied  a 
visible  contact  between  the  Semitic  and  the  Hellenic  genius 
in  religion.  Aphrodite,  as  the  Greeks  knew  and  worshipped 
her,  was  neither  Semitic  nor  Greek,  but  a  curious  compli 
cation  akin  both  to  Greece  and  to  the  East  far  and  near. 
She  bears  traces  of  the  armed  Ishtar  of  Assyrian  mythology, 
as  for  instance  in  the  above  mentioned  Epidaurian  copy 
from  the  statue  of  Alcamenes,  where  she  has  put  on  a  sword, 
and  again  in  statues  known  to  have  existed  at  Sparta,  Corinth, 
and  elsewhere.  In  fact  under  the  smiles  and  blandishments 
of  golden  Aphrodite  in  her  sunniest  Grecian  days  lurked 
always  the  jealous  wrath  of  a  divinity  who  would  have 
none  other  before  her.  Here  for  the  first  time  in  Greek 
mythology  we  have  clearly  set  before  us  a  jealous  and 
revengeful2  omnipotence  asserting  itself,  not  as  in  Dionysus' 
case  over  men  only,  but  also  over  the  gods.3  Aphrodite  is 
much  more  than  a  deified  incarnation  of  the  powers  of 
growth  and  increase  in  nature,  and  although  the  same  be 
true  of  the  Eleusinian  gods,  yet  in  a  certain  oriental  sense 
her  power  has  a  wider  field  than  that  either  of  Dionysus  or 
of  Demeter.  '  No  doubt  her  influence  over  typically  Greek 
minds  was  the  more  superficial  on  this  account.  But 
where  and  while  she  ruled,  her  sway  was  absolute,  and 
admitted  of  no  questioning.  Her  will,  whether  for  evil  or 
for  good,  was  always  law,  and  the  worshippers  of  her 

1  The  Cyprians  well  knew  and  freely  admitted  that  Aphrodite  came 
from  the  East.      See  Herod,  i.  105  :    TT)S  Su/j^s  ev  'Acr/cdXon'i  7r6Xt  rtJ'ej 
viro\ei(j)dfrTcs   effvX-rjcrav  TT/S  Ovpavit^s  '  A(f>po8lTrjs  rb  ip6v  tffri  8£  TOVTO 
rb  Ipbv    iravT&v  apx^rarov  ipQ>v    6<ra  TCWTTJS  TT}S  Qeov'     /cat  yap  TO  tv 

Ipbv  evdevrev  ey&ero,  cbs  avroi  \tyov<ri 

2  Iliad,  iii.  413  and  ff. 

3  Horn.  Hymn,  iv.  247  and  ff. 

T 


274  APHRODITE  AT  PAPHOS  vn 

predilection  prostrated  themselves  before  it,  like  slaves  be 
fore  a  sultan.  They  all  trembled  and  bowed  down  in  fear  of 
the  dreadful  visitation  of  her  wrath.  This  is  an  aspect  of  the 
goddess  of  which  those  who  try  to  understand  her  never  must 
lose  sight,  but  lest  we  dwell  upon  it  so  long  as  to  forget  the 
more  genial  and  graceful  traits  in  her  character,  let  us  rather 
contemplate  Cyprus,  her  island  home,  her  refuge  in  supreme 
moments  alike  of  sorrow  and  of  joy,  Cyprus,  the  island  made 
glad  with  dances — a  mother  of  winsome  loves.1 

This  is  the  isle  of  many  names^called  by  the  prophets  of 
Israel  Chittim ;  by  the  poets  of  Greece  surnamed,  from  its 
many  jutting  headlands  on  the  north,  the  Horned  Isle;  and, 
from  the  lowland  flats  upon  its  southern  and  eastern  sides, 
the  Hidden  Isle.  Approach  the  south-western  coast,  if  you 
will — as  did  the  Knights  Hospitallers  of  old — from  Rhodes. 
Let  the  morning  be  breezy  and  hazy  so  that  clouds  may  rest 
upon  the  hills  that  cluster  at  the  feet  of  the  higher  ranges 
behind,  crowned  by  Mount  Troodos,  anciently  Mount 
Olympus.2  Then  there  will  be  a  mist  rising  up  from  the 
sea  and  meeting  these  low-lingering  vapours.  At  such  a 
moment  Cyprus  is  the  hidden  isle,8 — hidden  until,  at  the 

1  Insula  laeta  choris,  blandorum  et  mater  amorum,  Claudian. 

2  Many  modern  maps  put  the  Olympus  elsewhere,  but  with  no  good 
reason.     See  Appendix  VII. 

3  "  Varia  autem  nomina  illius  erant,  in  quibus  Cryptus,  Plin.  V.  xxxi. 
Vocatam  ante  Acamantida  tradit  Philonides,  Cerastin  Xenagoras,  et  Aspel- 
iam  et  Amathusiam;  AstynomusCrypton,  et  Coliniam.  Causam  addit  Steph- 
anus,    quod  sub    mare   occultata    merit.      'A(rriW/iios   8£  (pTjffi,    lUpvirrbv 
K€K\T)(rdai  diet,  TO  Kptiirrecrdai  TroAAd/as  virb  T?}S  daXdaffrjs,      Ita  enim  hie 
emendo:   quippe  perperam  editur,  Kvirpov,  Cryptum  .  .  .  turn  Eustathius, 
in  Dionysium.    ot  5£  Kpvirrov  irore  K\T]0TJvai  avrrjv  \{yov<ri  5td  rb  K€Kpixf>6ai 
virb  daXacra-rjs.     Aliam  causam  videtur  assignare  Phurnutus,  De  nat.  Deor. 
ubi    de   Venere  :  e/c  roirrou   8e  KOI  iepa  rijs  '  A^podir^  ^  r&v  Kvd^pojv 
vrjcros  elvai  SoKel,   rd%a  de  Kal  i)  KI^TT/JOS,  ffW^Sowrd  TTWS  /card  rotfvojjut 
Ty  Kpv\f/ei, .  .  .  nomine  suo  occultationem  referens. "     From  an  unpublished 
MS.  (autograph)  of  John  Meursius,  Cyprus,  Lib.  I.  chap.  vi. ,  in  St.  Mark's 
Library,  class  x.  cod.  sec.  xvii.  a,  214,  i,  156. 


vii  THREE  REGIONS  OF  CYPRUS  275 

last  moment,  the  view  of  it  springs  upon  you.  Then  after 
crossing  the  tempestuous  gulf  of  Sathalia  l  one  may  feel 
perhaps  like  shipwrecked  Odysseus,  who — 

.  .  .   caught  a  glimpse  of  shore  close  at  hand 
Giving  a  sharp  glance  forward  upborne  on  the  crest  of  a 
billow. 

If,  instead  of  landing  at  Limassol,  the  port  lying  nearest 
to  the  temple  of  Paphian  Aphrodite,  we  had  approached 
Cyprus  from  the  north,  a  far  more  picturesque  impression 
would  have  been  received.  Close  to  the  northern  coast 
line  from  end  to  end  runs  a  bold  chain  of  picturesque 
peaks,  of  the  same  formation  with,  and  parallel  to,  the 
Taurus  range  in  Asia  Minor.  Cyprus  consists  first  of  these 
two  mountain  ranges,  the  northern  and  southern;  and 
secondly  of  an  alluvial  plain,  called  Mesorea  or  Mid- 
mountainland,  that  stretches  between  and  connects  the  two. 

A  geologist  of  the  airy  and  positive-poetical 2  sort  would 
say  that  this  midmountain  plain  was  but  a  film  of  yesterday,3 

1  Brother  Stephen  Lusignan  describes  it  as   "  1'espouvantable  gouffre 
de  Sathalie,"  and  the  legend  connecting  St.   Helena  (the  empress)  with 
Cyprus  tells   of  her  stilling  its   stormy  uproar  by  dropping  into  it  a  nail 
from  the  true  cross.     Curiously  enough  the  cross  upon  which  the  penitent 
thief  was  crucified  was  discovered  in  Cyprus  by  Lazarus  and  St.   Mary 
Magdalene  !     Meursius,  Cyprus,  cap.  28  of  Book  II. 

2  I  need  not  say  that  a  very  different  sort  of  geologist  has  been  found 
in  the  gifted  M.  Albert  Gaudry,  to  give  an  admirable  account  of  Cyprus. 
See  his  paper  presented  to  the  Soci^te"  Ge"ologique  de  France,  on  November 
14,  1853  (vol.  xi.  in  the  second  series  of  that  Society's  publications) ;   also 
his   "  Recherches  scientifiques  dans  I  Orient,"    published  with  a  beautiful 
geological  map  at  Paris  1855. 

3  How  far  this  is  really  the  case  largely  depends  upon  the  as  yet  unde 
termined  age  of  the  northern  range  of  mountains.      This  whole  question  is 
dealt  with  by  Dr.  Oberhummer  of  Berlin  in  two  most  thorough  papers 
"  Aus  Cypern  "   and   "Die  Insel  Cypern,"  published  respectively  in  the 
7.eitschrift  der  Gesellschaft  fiir  Erdknnde  zu  Berlin  (xxv.  Band  1890),  and 
the  Jahresbericht  der  Geographischen  Gesellschaft  zu  Miinchen  (Heft  13, 
1890).       Dr.    Oberhummer  speaks   of  the  probability  that   Herr  Alfred 
Bergeat  will  soon  solve  the  difficult  questions  at  issue  by  careful  reports 


276  APHRODITE  AT  PAPHOS  vn 

and  would  tell  you  of  a  recent  era  when  there  was  no  visible 
plain,  but  only  the  submerging  sea  stretched  between.  Then 
Cyprus  was  not  one  but  two  islands — the  northern  and  the 
southern.  The  northern  island  was,  at  that  imagined  and 
not  impossible  time,  a  mere  backbone  of  bold  peaks,— an 
outwork,  so  to  speak,  protecting  the  parallel  mountains  of 
the  Asiatic  coast  in  Cilicia  or  Caramania  only  fifty  miles 
distant  to  the  north.  Look  southward  and,  supposing 
yourself  (always  in  company  of  the  poet  geologer)  to  have 
gone  backward  a  few  aeons, — you  see  first  a  gulf  of  thirty 
miles  expanse,  and  then  the  southern  or  Troodos  island  of 
Cyprus.  Of  these  two  mountain  parts  of  Cyprus  the  history 
has  been  as  different  as  are  their  scenery  and  their  climate. 
On  the  southern  side  of  the  island  the  slopes  of  Mount 
Troodos  are  the  one  and  only  refuge  from  the  fever- 
exhalations  and  terrific  heats  of  the  height  of  summer,  and 
thither  all  fly  who  are  able  to  do  so,  when  the  dog-days 
impend.  Of  this  southern  side  of  the  island  Martial's 
saying  holds  true,  "  infamis  calore  Cyprus,1— Cyprus  is 
decidedly  hot"  Such  harbours  as  Cyprus  boasts— there  is 
not  one  on  the  island  where  an  ordinary  steamship  can 
find  safe  anchorage — are  with  one  unimportant  exception 
on  the  southern  side.  This  southern  Cyprus  is  the  Cyprus 
of  history,  and  the  midmountain  plain2  belongs  to  it.  In 

from  the  spot.      The  interesting   and  instructive  account  given  by  Dr. 
Oberhummer  of  the  table-mountains  in  the  midmountain  plain  of  Cyprus, 
and  his  comparison  of  them  with  features  in  the  Sahara,  is  one  of  his  many 
contributions  to  knowledge  of  Cyprus.     His  papers  came  to  me  unfor 
tunately  as  I  was  going  to  press,  and  I  therefore  profited  too  little  by  them. 
1  Infamem  nimio  calore  Cypron 
Observes,  moneo  precorque,  Flacce, 
Messes  area  cum  teret  crepantes 
Et  fulvi  iuba  saeviet  leonis. — Epigr.  ix.  91. 

2  This  Mesorea,  one  end  of   which  is  fertilised   by  the  Pediaeus, — a 
Cypriote  river  Nile,   Gaudry,   Recherches,  p.   96,— M.  Gaudry  calls   "  un 


vii  MEETING  POINT  OF  EAST  AND  WEST  277 

what  remains,  the  mountain  backbone  of  northern  Cyprus, 
no  great  centre  either  of  religious,  political,  or  commercial 
interests  (unless  Cerynia 1  be  forced  to  do  duty  as  such)  has 
ever  been  established ;  and  therefore  this,  its  most  beautiful 
and  healthful  portion,  may  be  said  never  to  have  belonged 
to  Cyprus  in  any  real  fashion.  This  whole  northern  reach 
is  little  more  than  an  elegant  extract  from  the  obscurer 
portions  of  Asia  Minor,  whereas  the  great  plain  and  the 
southern  mountains  of  Cyprus  had  a  physical  character  of 
their  own.  Here  was  a  natural  meeting  -  ground  for  all 
peoples  of  the  East  and  West  —  where  the  tongues  of 
Syria,  Assyria,  Phoenicia,  and  Egypt  from  the  East,  and 
the  Greek  and  Roman  vernaculars  from  the  West,  could 
be  heard.  Even  when  the  centre  of  commercial  exchange 
between  the  East  and  West  had  long  passed  away  from 
Southern  Cyprus,  still  the  island  remained  a  place  of 
congress,  a  point  of  contact  and  impact  for  eastern  and 
western  religious  influences. 

The  present  status  of  Cyprus  is  in  fact  its  whole  history 
in  a  nutshell.  Nothing  for  Cyprus  and  everything  for  all 
the  world  besides.  Governed  by  the  English  on  ordinary 
terms,  Cyprus  would  have  every  chance  of  prosperity,  if  we 
may  judge  from  the  Ionian  islands  under  British  adminis 
tration.  But  these  western  rulers  in  Cyprus  are  now 
administering  Eastern  laws,  not  English  nor  even  Cypriote, 
but  Turkish  laws.  The  English  in  Cyprus  have  been 
described  somewhat  bitterly  by  one  of  themselves  as  Turkish 
tax-gatherers.  And  these  taxes  are  not  even  levied  for  the 
Turk's  own  use,  but  go  to  assure  a  pittance  to  Turkish 

des  lieux  les  plus  fertiles  du  mondc."     In  the  season  when  I  saw  it  nothing 
of  this  richness  appeared. 

1  Before  the  days  of  Cerynia  (Cerines),  Lapethus  came  very  near  to 
attaining  a  real  importance. 


278  APHRODITE  AT  PAPHOS  vn 

bondholders  in  France,  England,  and  elsewhere.1  So  it  has 
been  always  with  Cyprus.  Before  the  Turks,  the  more 
rapacious  Venetians  exploited  the  island  most  mercilessly ; 
they  got  it  by  a  trick  the  very  nature  of  which  showed  that 
Venice  had  lost  every  imperial  instinct,  and  could  only 
oppress.  Before  the  Venetians  came  the  French  dynasty  of 
the  Lusignans,2  who  established  a  feudalism  efficient  perhaps, 
but  certainly  corrupt  and  monstrously  cruel  from  the  first. 
Lusignans,  Genoese,  Venetians,  all  these  western  potentates 
and  powers  lost  every  moral  quality  the  moment  they 
touched  unhappy  Cyprus,  so  that  the  islanders  might  well 
regret  the  days  when  Rome  and  Byzantium 3  ruled  with  even 

1  In  his  paper  (pp.  98  and  ff. )  "  Die  Insel  Cypern ' '  referred  to  above,  Dr. 
Oberhummer  gives  an  extremely  clear  account  of  the  financial  impossibilities 
under  which  the  English  administration  of  Cyprus  is  now  and  has  been 
labouring.     The  yearly  tribute  ultimately  payable  to  the  bondholders  is 
£92,686.     This  was  fixed  on  the  basis  of  an  average  account  of  revenues 
for  the  five  years  preceding  1878.      Those  familiar  with  the  incapacity 
for  administration  possessed  and  prized  by  the  unspeakable  Turk  will  not 
wonder  that  the  calculations  of  1878  were  all  wrong.      Dr.   Oberhummer 
gives  the  following  figures  of  yearly  income  and  outgo  in  Cyprus,  and  they 
speak  for  themselves — 

Income.  Outgo. 

1879-80                     £148,360  £117,445 

1880-81                      156,095  119,416 

1881-82                      163,732  157,672 

1882-83                      189,000  120,000 

1883-84                      194,051  111,685 

2  On  this  whole  chapter  in  Cypriote  history  the  greatest  authority,  as 
well  as  the  most  entertaining,  is  the  gifted  M.  de  Mas  Latrie.      For  a 
most  readable  book  on  Cyprus  in  the  Middle  Ages,  and  in  modern  times 
down  to  the  English  occupation,  see  his  L'ile  de  Chypre,  Paris,  1879,  which 
is  dedicated  to  Sir  Henry  Layard.     See  also  his  monumental  Histoire  de 
I'ile  de  Chypre  sous  le  regne  des  princes  de  la  maison  de  Lusignan,   Paris, 
1861. 

3  A  flourishing  condition  of  Cypriote  industry  and  a  high  repute  for 
Cypriote  stuffs  in  Byzantine  days  is  implied  in  a  letter  written  by  the  holy 
Epiphanios  to  the  Bishop  of  Jerusalem.      In  the  midst  of  a  tirade  against 
the  heresy  of  Origen,  Epiphanios  alludes  to  a  high-handed  proceeding  of 
his  own.     He  had  entered  a  church  and  found  there  a  rich  cloth,  with  a 
representation  of  the  face  of  Christ  or  of  that  of  some  saint.      He  ordered 


vii  NO  CYPRIOTE  NATIONALITY  279 

hand,  assured  protection,  and  moderate  imposts  over  their  ill- 
fated  country.  And  to-day,  when  their  rulers  are  the  best 
and  most  upright  of  men,  ready  and  eager  to  help,  a  fatality 
prevents  the  island  from  gaining  what  it  should,  and  still 
the  hardships  of  Cyprus  win  her  no  hope. 

So  far  back  as  history  reaches  there  has  been  no  inde 
pendence  for  Cyprus,  and  there  never  has  been  a  Cypriote 
nationality  or  national  enthusiasm.  It  is  all  the  more  sur 
prising,  therefore,  that  a  Greek  enthusiasm  has  seized  upon 
the  larger  portion  of  the  natives  to-day.  The  vivacity  and 
strength  of  the  modern  Greek  nationality  is  nowhere  more 
apparent  than  in  this  peaceful  and  almost  complete  conquest 
of  the  allegiance  of  Cyprus. 

For  the  theatre  of  a  religious  evolution  where  the  western 
spirit  of  Hellenic  beauty  and  independence  should  meet  the 
eastern  spirit  of  blind  submission  and  comprehension  of 
divine  omnipotence,  and  combine  on  equal  terms,  a  place 
which  belonged  to  no  one  and  to  every  one — Cyprus — was 
required.  I  have  enumerated  the  various  occupants  of 
Cyprus  since  the  days  of  Rome  only  as  a  prelude  to  a 
similar  enumeration  which  shall  go  back  as  far  as,  and  per 
haps  a  step  farther  than,  the  undimmed  record  of  history 
may  warrant.  Such  a  backward  survey  is  plainly  needed  in 
order  that  there  may  be  some  knowledge  of  the  background 
upon  which  the  composite  international  worship  of  Paphian 
Aphrodite  was  sketched. 

Before  the  Romans  were  the  Greeks,  before  the  Greeks 
the  Phoenicians.  Before  the  Phoenicians  were  the  Lycians — 

this  to  be  cut  down,  and  gave  word  that  a  poor  man  should  have  it  for 
grave-clothes.  There  were  murmurs  ;  the  people  thought  he  should  have 
replaced  it.  This  he  promised,  and  now  apologises  for  delay.  He  had 
delayed,  ' '  wishing  to  get  a  good  one  from  Cyprus."  Now,  under  pressure 
he  sends  the  best  he  can  get. 


280  APHRODITE  AT  PAPHOS  vn 

or  were  they  the  Hittites? — and  a  Semitic  race — so  thinks 
Mr.  Max  Duemmler— was  displaced  by  the  Lycians.  Perhaps 
these  were  the  Hittites.  The  whole  subject  of  the  pre- 
Phoenician  inhabitants  of  Cyprus  is  beset  with  unusual 
difficulty.1  Were  these  Lycians,  for  instance,  the  writers  of 
the  strange  Cypriote  characters  found  all  over  the  island  ? 
Or  are  these  Cypriote  letters  to  be  attributed  to  Max 
Duemmler's  Semitic  aborigines  ?  Other  authorities  declare 
that  whoever  first  used  these  Cypriote  characters — whether 
you  call  him  Lycian  or  Semitic  or  aboriginal  —  was  a 
Hittite  of  the  Hittites. 

Plainly  we  shall  hardly  escape  without  offending  various 
people  of  various  minds  if  we  undertake  to  have  an  opinion 
about  the  earliest  takers  and  makers  of  Cyprus.  There 
are  those  who  find  traces  in  Lycia  and  Phrygia,  through 
out  the  modern  Caramania,  of  an  early  invasion  from  Thrace 
and  the  West,  and  who  claim  that  this  invasion  swept  over 
Cyprus  as  well.  Certainly  these  Hittites  or  Thracians  or 
Semites  or  Lycians  could  easily  cross  the  fifty  miles  of  sea 
and  come  over  from  Cilicia.  In  fact  the  crossing  has  many 
times  been  made  by  invading  grasshoppers,2  feared  to-day 
more  than  Hittites,  Thracians,  or  Lycians. 

1  Mr.  J.  Arthur  R.  Munro  writes  me  :   "As  to  the  primitive  population 
I  think  the  evidence  is  tending  to  show  that  there  was  a  great  immigration 
of  a  semi-Greek  stock  from  Asia  Minor  where  they  had  passed  under 
oriental   influences,   and   only   a  slight   immigration  (if  any)   direct  from 
Greece.      The   connection   with  Arcadia   in   language   (v.    Meister)   and 
traditions,  names,  etc.,  seems  to  me  best  explained  not  by  colonisation, 
but  by  supposing  the  first  tide  of  Greek  peoples  flowed  southward  in 
parallel  streams  (i)  into  Greece,  and  (2)  over  the  Hellespont,  the  Arcadians 
being  a  remnant  of  the  pre-Dorian  people  maintaining  themselves  in  the 
Highlands." 

2  See  Gaudry's  Recherches,  p.  147,  also  the  excellent  book,  published 
in   1878,   on  Cyprus  by  R.    Hamilton    Lang  of  the   Imperial  Ottoman 
Bank.       Chapter    x.     deals    with   droughts     and     grasshoppers.       The 
very   remarkable   sympathy   of    Mr.    Lang   for    the   Cypriote   peasantry 


vii  PHOENICIANS  IN  CYPRUS  281 

The  Phoenician  invasion  itself  was  marvellously  early, 
for  their  first  Cypriote  possession  was  apparently  their  first 
foothold  on  foreign  soil.  Anxious  to  enlarge  the  trade  of 
their  coast  towns  of  Tyre  and  Sidon,  they  founded  a  trading 
post  on  the  nearly  opposite  strand  of  Cyprus,  at  the  place 
now  called  Larnaca,  and  used  in  modern  and  in  medieval 
times  as  the  sea -port  of  the  inland  cathedral -town  of 
Nicosia,  also  called  Lefcosia.  The  ancient  name  bestowed 
here  by  the  Phoenicians  was  Kittim,  preserved  by  the 
Greeks  and  Romans  in  the  form  of  Citium.  The  site  chosen 
first  by  the  Phoenicians  in  Cyprus  was  wisely  selected,  for 
it  remains  to-day  the  important  port  of  the  whole  island. 
Now  its  importance  comes  chiefly  from  relations  with  the 
West ;  anciently  it  was  wholly  identified  with  the  fortunes  of 
Phoenician  Tyre  and  Sidon. 

"The  burden  of  Tyre,  howl,  ye  ships  of  Tarshish,"- 
says  Isaiah  the  prophet, — "  for  it  is  laid  waste  so  that  there 
is  no  house,  no  entering  in ;  from  the  land  of  Chittim  it  is 
revealed  to  me.  Be  still,  ye  inhabitants  of  the  isle,  thou 
whom  the  merchants  of  Zidon  that  pass  over  the  sea  have 
replenished." 

The  other  two  places  for  colonising  selected  by  the 
Phoenicians  were  Amathus  and  Old  Paphos,  neither  of  which 
has  preserved  a  shadow  even  of  that  sometime  glory. 
Even  in  ancient  times  it  was  apparently  the  religion  rather 
than  the  commerce  of  Phoenicia  that  kept  its  foothold  in 
these  two  places,  which  were  never  of  any  great  extent  or 
importance  outside  of  their  respective  sanctuaries.  Hence 

makes  his  book  an  invaluable  one,  and  its  worth  is  increased  by  the 
appreciative  use  he  has  made  of  an  unpublished  report  on  Cypriote 
agriculture  made  in  1844  by  M.  Fourcade  of  the  French  Consular  Service. 
Mr.  Lang's  book  well  deserved  to  be  translated  into  French,  as  it  was 
in  1879. 


282  APHRODITE  AT  PAPHOS  vn 

there  is  a  plentiful  lack  of  evidence1  from  the  soil  itself 
and  its  contents  to  show  that  Phoenicians  ever  found  lodg 
ment  in  either  place. 

The  Phoenicians  were  the  common  carriers  of  antiquity, 
and  their  genius  was  so  purely  for  expediting  exchange  and 
promoting  commerce  that  they  had,  even  in  their  great 
centres  of  commerce — let  alone  such  places  as  Paphos  and 
Amathus — little  or  no  energy  left  for  building.  If  this  was 
true  of  Phoenicians  in  the  country  of  their  birth  and  pre 
ference,  how  much  more  true  must  it  have  been  of  the 
Phoenicians  in  Cyprus, — an  alien  land  where  no  monuments 
of  any  epoch,  saving  tombs,  have  survived ; 2  with  the  sole 

1  In  his  admirable  publication  Devia  Cypria,  Mr.  D.  G.  Hogarth  raises 
the  whole  question  of  Phoenician  influence  in  Cyprus,  saying  ' '  Indeed  as 
research  has  tended  more  and  more  to  minimise  the  part  played  by  the 
latter  (Phoenicians)  in  Cyprian  economy,  and  to  reject  their  claim  to  be 
the  importers  even  of  the  great  goddess,  or  the  founders  of  her  temples, 
so  western  influence  must  be  relegated  to  the  days  of  Evagoras."     To 
this  Mr.  Hogarth  adds  in  a  note, — where  he  speaks  with  the  utmost  know 
ledge  of  the  facts,  since  he  superintended  the  excavations  at  Old  Paphos — 
"  It  will  be  remembered  that  we  found  no  Phoenician  relics  at  Old  Paphos 
at  all  ;  nor  have  any  been  found  at  Amathus,  Salamis,  Lapethus,  or  indeed 
(except  in  isolated  instances)  anywhere  but  at  Citium  and  Idalium."     This 
is  all  very  strong  negative  evidence,  nor  is  the  late  use  of  the  Ionian  alphabet 
more  positive.    Really  positive  evidence  seems  required  to  refute  Herodotus 
and  others  who  put  Phoenician  cults  at  Old  Paphos,  and  who  doubtless 
had  before  them  avenues  of  information  for  ever  closed  to  us.     They  con 
stitute  the  only  evidence  at  hand  to  show  what  intelligent  men  believed  in 
those  days  about  the  origin  of  the  worship  of  Aphrodite  at  Paphos.    We  are 
beginning  to  have  a  certain  amount  of  evidence  which  they  lacked,  and  it 
is  becoming  plain  to  me  that  Aphrodite  united  to  the  Phoenician  strain, 
that  fixes  her  character,  peculiarities  derived  neither  from  the  Phoenicians 
nor  from  the  Greeks.     The  only  possible  way  in  which  these  new  and  im 
portant  facts  can  eventually  achieve  due  recognition  is  to  have  them  stated 
at  this  stage  of  the  investigation  by  one  who  denies  the  Phoenician  origin 
of  the  goddess.    Such  a  statement  Mr.  Hogarth,  who  inclines  to  that  view, 
has  been  good  enough  to  make  in  a  letter  to  me  full  of  valuable  suggestions 
on  many  points.      For  his  account  of  the  Hittite  origin  of  Aphrodite,  see 
Appendix  VII.  below. 

2  The  condition  in  which  the  ruins  of  the  temple  of  Paphian  Aphrodite 
have  been  found  is  so  lamentable  that  it  would  be  a  mockery  to  speak  of 
it  as  a  surviving  monument.     See  below,  Appendix  VI. 


vii  LACK  OF  GREEK  MONUMENTS  283 

exception  of  a  few  bits  from  Roman  times  and  some  few  very 
noteworthy  medieval  churches  and  ruined  castles.  A  century 
will  hardly  be  required,  at  the  rate  of  dilapidation  now  observ 
able  ;  in  less  than  that  time  even  these  will  have  disappeared, 
and  there  will  be  no  reading  of  the  past  elsewhere  than  in 
tombs  used  many  times  and  rifled  by  many  generations  of 
men  who  were  for  the  most  part  ignorant  and  superstitious. 
That  the  Greeks,  who  did  so  much  building  of  durable 
sort,  should  have  left  no  monuments  in  Cyprus  worth 
mentioning  is  a  matter  of  surprise,  because  of  the  early  date 
of  their  first  occupation  of  various  parts  of  the  island. 
But,  in  spite  of  this  early  occupation,  it  cannot  be  denied 
that  the  Greeks  have  never  been  justified  at  any  time  in  the 
past  as  much  as  they  are  now  in  calling  Cyprus  a  Greek  island. 
The  monuments  in  this  land  were  sure  to  perish.  They 
were  never  built  by  a  people  rooted  in  the  soil,  and  some 
new  master  always  came  who  did  not  know  and  had  no  care 
for  the  buildings  left  by  those  whom  he  dislodged.  The 
site  of  Salamis  and  Famagosta  best  illustrates  this  curious 
state  of  things.  Farthest  north  lay  the  Greek  colony  of 
Salamis.  This  was  abandoned  for  the  medieval  walled  city 
about  five  miles  to  the  south  of  it,  Famagosta — a  town  with 
churches  innumerable  and  a  cathedral,  as  well  as  the  most 
wonderfully  complete  fortifications.  Famagosta  is  now  an 
untenanted  simulacrum  of  the  commercial  glories  of  the 
days  of  Italian  supremacy  in  Cyprus,  and  not  quite  a  mile 
farther  south  lies  Varoschia,  where  the  living  successors  of  the 
Salaminians  and  Famagostans  of  old  now  congregate.  It  is 
as  if  we  found  a  ruined  city  at  the  battery  end  of  Manhattan 
Island  and  deserted  tenements  at  23d  Street,  because  the 
traffic  of  New  York  had  been  transferred  to  the  neighbour 
hood  of  the  Harlem  River. 


284  APHRODITE  AT  PAPHOS  vn 

The  moral  to  be  drawn  from  this  digression  is  that  the 
circumstances  of  Cyprus,  and  the  native  pursuits  of  the 
Phoenicians  there  and  elsewhere  alike,  will  prepare  us  for 
the  scantiest  yield  of  Phoenician  remains  on  sites  where 
they  are  known l  to  have  lived  for  many  years. 

I  have  said  that  Cyprus  was  just  the  place  where  Greek 
and  Phoenician  influences  could  meet  and  mingle  without 
the  interference  of  surrounding  circumstances  to  give  pre 
dominance  to  either.  This  is  so  true  in  essential  matters 
that  even  the  date  of  the  Greek  colonisation  of  Cyprus 
stretches  far  enough  back  to  come  near  the  earliest  occupa 
tion  of  Cyprus  by  Phoenicians.  Kittim  (Larnaca-Citium) 
preceded  the  Greek  occupation  ;  but  there  are  stories  which 

1  This  knowledge  is  neither  derived  from  monuments  nor  confirmed  in 
any  appreciable  degree,  as  regards  Paphos  and  several  other  Phoenician 
abiding  places,  by  archaeological  discoveries.  See  above,  note  i  p.  282, 
where  Mr.  Hogarth's  striking  testimony  is  quoted.  But  still  undoubted 
traces  of  a  Phoenician  occupation  have  been  found  at  Citium  and  Idalium, 
and  their  presence  at  this  last  place  gives  a  trace  of  their  close  connection 
with  the  goddess  there  worshipped.  Furthermore,  the  oldest  portion  of 
the  temple  of  Aphrodite  at  Paphos  (see  Appendix)  might  perhaps  prove  to 
be  of  Phoenician  construction,  if  we  were  fortunate  enough  to  know  any 
thing  sufficiently  definite  about  Phoenician  building.  These  slight  clues 
derived  from  excavation  and  archaeological  research  help  us  toward  cer 
tainty  when  the  early  traditions  connecting  Cyprus  with  the  Phoenicians 
are  duly  taken  into  account.  To  begin  with  Homer,  the  Odyssey  mentions 
Cyprus  five  times,  and  the  Iliad  only  once ;  but  Aphrodite  is  mentioned  as 
Cypris — a  name  unknown  to  the  Odyssey — five  times  in  the  Iliad.  Thus 
Cyprus  was  well  known  to  the  Homeric  mind  ;  and  Aphrodite,  in  spite  of 
her  Greek  father  Zeus  and  mother  Dione,  was  habitually  severed  from  her 
Olympian  peers,  and  associated  with  Cyprus.  Now  Cyprus  is  plainly 
associated  in  the  Odyssey  with  Phoenicia  and  Egypt  (iv.  83),  and  in  the 
Homeric  Hymn  to  Dionysus,  the  Tyrsenian  (Phoenician)  pirate  talks  of 
taking  his  captive  to  Egypt  or  to  Cyprus  (v.  28),  and  thus  bears  testimony 
to  the  intimate  home  relations  between  Cyprus  and  the  Phoenicians. 
Plainly  if  the  Homeric  mind  had  been  as  deeply  interested  in  the  history  as  in 
the  worship  of  Aphrodite,  we  should  have  had  from  Homer  what  Herodotus 
finally  gives  us,  a  statement  that  the  Phoenicians  brought  the  goddess's 
worship  to  Cyprus.  Homer  knows  nothing  of  the  goddess's  birth  in 
Cyprus.  This  came  in  the  Homeric  Hymns  and  in  Hesiod.  The  mention 
of  Cyprus  in  connection  with  Aphrodite  by  Herodotus  and  later  writers  is 
familiar,  and  is  often  alluded  to  in  other  portions  of  this  chapter. 


vii         AGAPENOR  AT  PAPHOS  AND  IN  ARCADIA       285 

leave  it  doubtful  if  Old  Paphos  may  not  have  known  Greeks 
as  well  as  Phoenicians  in  its  earliest  days. 

The  Arcadian  chieftain  Agapenor  —  naively  qualified 
by  a  Lusignan  chronicler1  as  "Lieutenant-general  des 
Navires  du  Roy  Agamemnon,"  "  King  Agamemnon's  lord 
high  admiral " —  founded  Paphos  on  his  way  home  from 
the  sack  of  Troy.  This  Paphos  of  Agapenor  is  probably 
the  later  New  Paphos, — a  town  commercially  far  more  suc 
cessful  than  the  older  and  Phoenician  Old  Paphos  on  the 
hill  not  far  away ;  but  still,  our  confused  informant  Pausanias 
declares  that  Agapenor  established  the  worship  and  the 
temple  of  Aphrodite  at  Paphos,  and  adds  by  way  of  making 
confusion  worse  confounded  that  this  Paphian  worship 
instituted  by  Agapenor  had  until  then  been  maintained  at 
Golgoi,  another  place  in  Cyprus.  But  testimony  of  superior 
weight  to  that  of  Pausanias  tells  us  that  the  first  worship  of 
Aphrodite  at  Old  Paphos  was  Phoenician.  Only  it  evidently 
came  very  early  and  very  closely  in  contact  with  the  adven 
turous  Greeks.  Much  or  little  as  Agapenor  may  have  done 
for  the  worship  of  the  goddess  at  Paphos,  he  certainly 
did  as  much  for  it  in  Arcadia, — the  temple  of  Aphrodite 
Paphia  at  Tegea  in  south-western  Arcadia  was  of  Agapenor's 
founding.  It  is  worth  noticing  carefully  how  inevitable  was 
the  acceptance  by  Greece  of  this  originally  Assyrian  goddess. 
Arcadia  is  at  the  centre  of  the  mountainous  Peloponnesus, 
and  yet  to  this  heart  of  the  Highlands  of  Greece  came  from 
opposite  quarters  the  same  eastern  and  alien  goddess.  The 

1  Brother  Stephen  of  Lusignan  in  the  very  much  fuller  French  version 
published  by  himself  of  his  villainously  printed  Chorograffia  et  breve 
historia  universale  dell  isola  di  Cipro  princ ipiando  al  tempo  di  Nocperin- 
sino  al  1572,  per  il  R.  P.  Lettore  Fr.  Stephano  Lusignano  di  Cipro  dell' 
ordine  de  Predicatori.  Bologna,  1573.  For  a  further  account  of  this 
book  see  Appendix  VIII. 


286  APHRODITE  AT  PAPHOS  vn 

temple  at  Tegea  was  founded  from  Cyprus  (Paphos) l  in  the 
East,  whereas  in  the  opposite  and  north-western  corner  of 
Arcadia  a  temple  of  Aphrodite  was  founded  in  the  inacces 
sible  district  of  Psophis,  and  founded  too  from  what  then 
were  the  uttermost  parts  of  the  West.  The  Psophidian 
temple  was  due  to  the  zeal  of  worshippers  of  the  goddess 
at  the  ancient  shrine  on  Mount  Eryx,  in  north-western 
Sicily.2  A  third  Phoenician  foundation  (where  recent 
diggings 3  have  discovered  next  to  nothing  at  all)  was  far 
nearer  to  the  heart  of  Greece  than  eastern  Paphos  or 
western  Eryx, — the  island  of  Cythera  from  which  Aphro 
dite  drew  one  of  her  sweetest  names,  Cytherea.4  Strange  to 

1  Paus.  VIII.  liii.  3  ;  cf.  ibid.  v.  2. 

2  Paus.  VIII.  xxiv.  i  and  6. 

3  These  were  made  with  his  accustomed  skill  by  Dr.  Schliernann. 

4  This   epithet   of  Aphrodite   has  had   a   surprising   effect  upon  the 
beautiful  name  of  that  beautiful  corner  of  Cyprus  named  Cythrea.     Brother 
Stephen  of  Lusignan  and  many  others  transform  it  by  a  slight  change  in 
its  last  syllable  but  one  into  Cythera.     A  more  forbidding  spot,  Mount 
Cithaeron,  has  also  attempted  to  usurp  the  place  of  Cythera  in  Aphrodite's 
affections  (cf.  Bartolommeo  da  li  Sonetti  on  Cythera).     See  Boccaccio's 
Teseide,  VII.  stanza  43. 

O  bella  Iddea  del  gran  Vulcano  Sposa 
Per  cui  s'allegra  il  monte  Citerone. 

This  is  reproduced  by  Chaucer,  Knight's  Tale,  1363, 

Fairest  of  faire,  o  lady  myn  Venus, 
Daughter  of  Jove  and  spouse  to  Vulcanus, 
Thou  goddess  of  the  mount  of  Citheroun. 

Again  Dryden,  in  his  Palamon  and  Arcite,  following  Chaucer,  gives  a  de 
scription  of  beautiful  paintings  where  the  Cithaeron  again  does  duty  for 
Cythera, 

For  there  th'  Idalian  Mount  and  Citheron, 
The  court  of  Venus,  was  in  colours  drawn. 

Last  of  all,  in  a  recent  writer  we  find  the  old  confusion  reasserting  itself  in 
a  description  of  Paris  society  :  "It  was  a  charming  world  of  fancy  and 
caprice  ;  a  world  of  milky  clouds  floating  in  an  infinite  azure,  and  bearing 
a  mundane  Venus  to  her  throne  on  a  Frenchified  Cithaeron."  It  is  need 
less  to  say  that  no  place  less  belongs  to  Venus  Aphrodite  than  the  gloomy 
wilderness  of  Boeotian  Cithaeron  where  the  Maenads  tore  Pentheus  in 
pieces,  and  whither  Oedipus  was  sent  as  a  babe  to  be  exposed. 


vii  APHRODITE  ALWAYS  PAPHIAN  287 

say,  there  is  no  record  of  any  considerable  influence  exer 
cised  from  Cythera  upon  the  worship  of  Aphrodite  in 
Greece.  From  the  West,  beyond  the  foundation  at  Psophis 
and  some  probable  influence  upon  Greeks  in  Sicily,  little 
or  no  influence  reached  Greece,  and  therefore  Paphos  in 
Cyprus  was  the  centre  of  the  worship  of  Grecian  Aphrodite, 
first  and  last. 

Accordingly,  if  we  would  come  under  the  Grecian  spell 
of  Phoenician  Aphrodite,  we  must  leave  Greece  proper  and 
go  to  the  southward  Cyprus,  which  is  neither  in  Europe 
nor  in  Asia ;  neither  of  the  East  nor  of  the  West ;  not 
Greek  nor  yet  Phoenician.  There  we  must  go  from  the 
harbour  of  Limassol  by  the  shortest  road  to  Old  Paphos, 
for  although  not  wholly  Greek  and  not  wholly  Phoenician, 
Aphrodite  is  Paphian  l  always  and  entirely. 

Let  us  approach  the  island  and  the  way  from  Limassol 
to  Paphos  with  no  undue  anticipations.  Rifled  tombs, 
broken  fragments,  foundations  half  effaced  are  the  reward 
which  students  of  Cypriote  antiquity  chiefly  receive.  Bear 
ing  this  in  mind,  we  may  well  linger  a  moment,  when  we 
are  four  miles  from  Limassol  on  the  road  to  Old  Paphos,  at 
Colossi  to  admire  a  splendid  square  castle,2  not  utterly  ruined 
as  yet.  Here  the  Knights  Commanders  dwelt  for  many 
years.3 

1  Od.  viii.  363,  Horn.  Hymn,  iv.  59  ff.  ;   Eurip.  Bacch.  385  and  406,  etc. 

2  See  a  photograph  of  it  published  by  the  Hellenic  Society,  Appendix 
XI.  i.  89. 

3  Nothing  seems  to  me  more  discouraging  than  the  attempt  to  photo 
graph,  or  rather  to  orthograph,  the  shades  of  popular  mispronunciation 
where  a  name  has  its  established  form.     Accordingly,  I  adhere  with  M.  de 
Mas  Latrie    to  Colossi,  leaving  others  to  choose    between  Colossois  and 
Colossin.     The  mystifying  name  of  Colossi — given  as  Colosso  on  the  un 
published  map  made  by  Leonida  Attar  in  1542,  and  now  in  the  Museo 
Correr   at   Venice  —  is   said    to   derive   from    the  Colossos   of    Rhodes. 
This  may  be  the  case,  but  the  name  certainly  has  nothing  whatever  to  do 
with  the  Rhodian   Knights  Hospitallers,  through  whom  Colossi  became 


288  APHRODITE  AT  PAPHOS  vn 

The  character  of  the  coast-shapes  and  landscapes  as  you 
journey  on  past  Colossi  to  Old  Paphos  is  noteworthy, 
because  of  a  contrast  with  what  comes  when  Paphos  is 
reached.  Around  the  monastery  of  St.  Nicholas — on  the 
promontory  of  Curias,  before  Colossi  is  reached — may  be 
found  vast  low-lying  areas  close  to  the  beach  of  the  sea. 
Here  the  strangest  shapes  may  be  seen  fashioned  out  of 
very  soft  limestone  by  the  action  of  the  rains  sometimes, 
and  sometimes  of  the  waves.  Vast  arches  reach  out  to  dim 
and  shapeless  buttresses,  while  the  sea  dashes  up  under  the 
arch  and  often  covers  wholly  its  seaward  support.  This  is 
nature's  architecture,  and  the  like  of  it  may  be  seen  in 
various  lands,  as,  for  instance,  on  the  Californian  coast  near 
Santa  Cruz.  So  friable  is  the  limestone  of  Cyprus  that  it 

famous.  The  Knights  Templars  had  the  whole  of  Cyprus  for  part  of  a 
year,  built  the  cathedral  at  Nicosia  in  part,  and  made  strong  castles 
at  Larnaca  and  elsewhere,  but  they  never  held  Colossi.  Before  its 
grant  to  the  Knights  of  Rhodes,  a  Frenchman  named  Nicholas  Garin 
owned  it,  and  he  adopted  from  it  the  feudal  title  of  Garin  de  Kolossi 
or  du  Colos.  The  only  analogy  that  suggests  itself  to  confirm  the 
derivation  from  the  Colossus  is  an  insufficient  one.  Padre  Coronelli's 
book  on  Morea  and  Negroponte  has  on  its  title-page  "Si  vende  alia 
Libreria  del  Colosso  sul  Ponte  di  Rialto."  Nicolas  Garin  may  after  all 
have  got  the  title  "  Du  Colos"  by  trade  associations,  and  then  have 
named  his  estate  in  order  to  derive  a  feudal  title  from  it.  After 
purchasing  all  outstanding  rights,  Hugues  I.  of  Lusignan  was  the 
final  grantor  to  the  Knights  of  these  estates  in  entirety,  his  father 
or  uncle  having  already  given  some  portions.  The  Hospitallers  finally 
appointed  a  commandery  for  Colossi,  and  made  it  the  centre  of  manage 
ment  for  all  their  rich  possessions  in  the  island.  These  they  held  for 
three  centuries — until  the  Turks  came  in  1571.  The  Sultan  appropriated 
most  of  the  Hospitallers'  lands.  He  had  good  reason  to  covet  Colossi,  for 
its  commandery  wine — so  called  from  the  Knights  and  their  commandery 
(preceptorery  was  the  earlier  polysyllable) — still  enjoys  a  well-deserved 
celebrity.  To  be  sure  it  is  now  made  unpalatable  to  some  by  the  tarred 
skins  which  serve  for  bottles.  On  this  subject  see  Gaudry,  Recherches,  p. 
330.  See  also  in  Mr.  Lang's  Cyprus ,  chap.  ix. ,  where  he  speaks  of  the 
Commanderia  wine  as  used  in  France  and  Italy.  Whoever  has  tasted  a 
good  old  vintage  of  this  wine  will  rejoice  that  Madeira  has  been  restocked 
since  \hephylloxera  from  the  district  of  Colossi.  Nothing  more  like  Madeira 
exists  than  the  wine  of  Colossi. 


vii  FROM  LIMASSOL   TO  PAPHOS  289 

seems  ever  ready  to  vanish  away.  There  is,  in  fact,  a 
melting  process  continually  in  progress  all  over  the  island. 
Hence  vast  caves  near  the  earth's  surface  which  invited  the 
successive  occupants  of  Cyprus  to  use  them  for  places  of 
burial.1  Hence  also  the  habit  of  hollowing  out  in  the 
willing  stone  all  manner  of  tombs  and  passages  leading 
from  subterranean  vault  to  hidden  chamber.  The  ground 
of  Cyprus  sounds  hollow  everywhere  under  the  hoofs  of  the 
mules, — the  whole  landscape  sometimes  seeming  but  a  mask 
covering  over  the  bones  of  men  long  dead. 

Close  to  Colossi  and  the  castle  of  the  Knights  Com 
manders  is  Episcopia  or  Piscopia,  associated  with  a  branch 
of  the  Venetian  Cornaro  family  and  with  one  of  Titian's 
masterpieces.2  But  it  is  time  to  draw  near  to  Paphos,  which 
is  after  all  only  a  day's  journey  on  muleback  from  Limassol. 
As  we  approach,  the  limestone  hills  cover  themselves  with  a 
brighter  green.  Near  Pissouri  and  Old  Paphos,  now  called 
Couclia,  are  sudden  whims  in  the  surface  of  the  soil.  Green 
slopes  lead  more  or  less  precipitously  downward,  sometimes 
to  a  field  shaped  like  an  amphitheatre,  sometimes  to  a 
meadow  nearly  square.  These  lower  places  cover  them 
selves  with  shrubs,  and  if  the  early  day  brought  showers,  as 
it  did  when  we  were  riding  that  way,  the  golden  slanting 
sun  of  eventide  shoots  pleasantly  through  their  dripping 
leafage,  making  them  sparkle,  as  it  were,  with  the  winning 
smile  of  golden  Aphrodite.  Cheered  by  this  glimmering 
show  of  welcome,  we  turned  around  the  jutting  foot  of  a 
wooded  hill,  and  came  upon  the  last  stream  to  be  forded 

1  More  than  once  I  have  seen  daylight  at  the  farther  end  of  a  cave 
near  the  top  of  a  ridge,  which  had  evidently  been  melting  away  by  this 
process.     These  caves  are  often  used  for  sheep-folds,  which  go  by  the  old 
name  of  mandra. 

2  In  the  Frari  Church  at  Venice. 

U 


290  APHRODITE  AT  PAPHOS  vn 

that  day.  Then  in  a  half-ruined  castle  called  the  Tschiflik 
or  farmstead,  which  surmounted  the  upward  swerve  of  the 
Paphian  hill,  we  were  most  warmly  welcomed  by  the  English 
excavators,  already  beginning  to  wind  up  their  business  at 
Paphos,  and  preparing  to  hand  over  the  site  to  the  charge 
of  Government. 

The  next  morning  we  had  before  us  the  site  of  the  temple 
of  Aphrodite  at  Old  Paphos,  discovered  and  fully  under 
stood  for  the  first  time  since  the  day  when  St.  Barnabas l 
called  down  upon  it  the  wrath  of  heaven,  and  by  a  judgment 
of  earthquake  and  fire  put  an  end  to  many  abominations. 
To  discover  precisely  what  earthquake  the  saint  inflicted 
would  be  vain ;  there  have  been  so  many  before  and  since 
upon  this  luckless  spot.2  There  is  no  lack  of  founders  and 
builders  for  this  home  of  Paphian  Aphrodite ;  they  are  as 
numerous  almost  as  are  the  writers  who  tell  of  the  temple's 

1  The  martyrdom  of  St.  Barnabas  took  place  in  Cyprus.     See  in  John 
Meursius1  unpublished  MS.  revision  of  his  well-known  book  on  Cyprus, 
a  note  which  refers  to  this  at  the  end  of  chap.  v.  in  book  II.  :    Nee  fecit 
tantum  martyres  Cyprus,  sed  et  dedit.     In  his  erat  Barnabas  apostolus 
de  quo  dixi. 

2  There  was  a  serious  earthquake  here  under  Augustus,  in  the  twenty- 
seventh  year  of  his  reign,  cf.  Eusebius  copied  by  Abbas  Uspergensis  and 
Marianus  Scotus.     Meursius,  Cyp.  II.  xx.  and  I.  xviii.,  quotes  Seneca,  Epist. 
xci.,  Nat.  Quaest.  vi.  26,  and  also  Sibylline  oracles,  III.  and  IV.      Under 
Vespasian,  in  the  ninth  year  of  his  reign,  three  towers  of  the  temple  fell. 
Hieronymus  (note  on  the  lost  text  of  Eusebius,  lib.  II.  Chron.  given  by  Meur 
sius,  I.  xviii. )  quotes  Bartholomaeus  Saligniacus,//z'«^rar?V  iv.  cap.  6  :  Paphos 
ruinis  plena  videtur,  templis  tamen  frequens.     This  is  a  century  after  the 
Acts,  where  see  chap.  xiii.  ;  see  also  Bede  on  the  Acts.     Fortunately  of 
late  years  there  has  been  a  cessation.     In  1830  the  unlucky  Dr.  Ludwig  Ross 
was  badly  shaken  while  in  Cyprus,  and  has  given  an  account  of  the  adven 
ture.     He  was  at  Dalin  at  the  time,  and  he  says:   "Am  andern  Morgen 
(February  21)  erwachte  ich  schon  um   fiinf   Uhr  durch  ein   langes   an- 
haltendes  Erdbeben  das  mit  leisen  zitternden  Schwingungen  anfing  und  mit 
einer  heftigen  Erschiitterung  endigte,  und  dessen  ganze  Dauer  wohl  eine 
halbe  Minute  betrug. "      Later,  while  at  Hieroskipou  (Strabo's  lepoK-rjiria  ; 
see  Hogarth's  Devia  Cypria,  p.  41),  two  miles  east  of  New  Paphos,  he 
says:   "iiberhaupt  soil  dieser  Theil  der  Insel  bis  Limissos  den  Erdstossen 
sehr  ausgesetzt  sein." 


vir  CINYRAS  A  T  PAPHOS  291 

foundation.  The  first  builders  up  of  the  Paphian  sanctuary 
appear,  in  fact,  to  have  been  as  plentiful  as  the  earthquakes 
that  shook  the  temple  down. 

With  the  legendary  name  of  Cinyras — the  most  widely 
recognised  of  these  first  founders  —  connects  itself  the 
Eastern  pedigree  of  Aphrodite.  Aphrodite  came  originally 
from  Assyria,  and  it  is  therefore  no  surprise  to  hear  from 
one  source  that  Cinyras,  usually  regarded  as  from  Syria,  was 
king  of  the  Assyrians.  This  wonderful  Cinyras  was  the 
great  ancestor  of  the  priestly  family  or  guild  that  long  ruled 
over  the  district  and  sanctuary  at  Old  Paphos.  The  collec 
tive  name  applied  here  was  that  of  the  Cinyradae  or  de 
scendants  of  Cinyras.  The  most  interesting  of  the  sons  of 
Cinyras  is  the  beautiful  and  ill-fated  Adonis,  who,  among 
other  things,1  is  the  Assyrian  Tammuz  brought  to  the  western 
coasts  of  Syria  and  Phoenicia.  In  various  legends  this  son 
of  Cinyras,  Tammuz-Adonis,  plays  the  part  of  Dionysus  and 
Persephone  in  one.  Around  him  gather  the  sorrow  and  the 
joy  of  the  yearly  death  and  yearly  revival  of  nature's  growth. 

The  beautiful  legend  of  Aphrodite's  love  for  Adonis — 
how  he  was  wounded  when  engaged  in  the  chase,  and  how 
Aphrodite,  filled  with  grief  on  hearing  the  sad  tidings, 
made  haste  to  the  Idalian  fields  of  Cyprus  and  found  there 
her  dying  love — appears  to  be  a  story  of  later  growth. 
It  was  connected  with  the  inland  sanctuary  of  Aphrodite 
in  Cyprus,  having  little  or  nothing  to  do  with  Old  Paphos. 
Amathus,  the  third  Cyprian  centre  for  the  goddess's  worship, 
was,  like  Old  Paphos,  on  the  sea  coast.  There,  we  hear, 
was  a  temple  where  Aphrodite  was  worshipped  in  common 

1  I  am  well  aware  that  it  is  quite  possible  to  bear  too  exclusively  in 
mind  the  eastern  affinities  of  Adonis.  He  is  as  close  to  the  Phrygian 
worship  of  Atthis  and  Cybele  as  he  is  to  the  Phoenician  Ishtar  and  Tammuz, 
— closer  perhaps. 


292  APHRODITE  AT  PAPHOS  vn 

with  Adonis,  the  two  forming  a  godhead,  both  male  and 
female.  Perhaps  this  dual  worship  at  Amathus  was  of 
distinctly  Phoenician  origin,  and  it  is  possible  that  the  Adonis 
of  Amathus  was  an  earlier  aspect — nearer  to  what  is  found 
in  the  annals  of  Ninivite  religion — of  the  charming  and 
tragical  Tammuz-Adonis. 

But  the  father  of  Adonis,  Cinyras,  is  by  no  means 
disposed  of  in  the  slight  mention  made  above.  Pindar 
speaks  of  "  Cinyras  whom  the  golden-haired  Apollo  dearly 
loved,  the  dutiful  servant  of  Aphrodite."  The  men  of 
Cyprus,  so  Pindar  says,  were  never  tired  of  telling  various 
stories  about  Cinyras.  This  native  tendency  to  tell  any 
thing  that  came  into  your  head  about  Cinyras  maintained 
itself  even  till  late  Christian  days.  Malevolence  and 
fanaticism  then  combined  with  a  justified  reprobation  of 
Paphian  licentiousness  to  frame  a  story  that  Aphrodite 
was  only  a  criminally  beloved  paramour  of  Cinyras.  In 
honour  of  this  mortal  and  misguided  woman  Cinyras 
built  a  temple  at  Old  Paphos,  where  her  tomb  and  his  are 
still  shown  to-day.  The  only  religious  fact  to  be  derived 
from  this  story  is  an  important  one,  as  will  appear  later, 
i.e.  that  in  the  precinct  of  Paphian  Aphrodite  at  Old  Paphos 
the  tomb1  of  that  goddess  and  of  Cinyras  were  con 
spicuous. 

In  giving  some  idea  of  the  medley  of  legends  and  tales 
that  centred  about  the  name  of  Cinyras  we  must  resign 
all  hope  of  gaining  a  consistent  idea  of  his  personality. 
Avalanches  of  mythology  have  swept  down  upon  him  from 
all  quarters,  so  that  he  is  but  a  remembered  name  and  not 

1  On  the  various  aspects  of  the  Greek  Aphrodite  that  make  of  her  a 
goddess  of  the  land  of  the  dead,  see  Dr.  W.  H.  Roscher,  in  his  own 
Ausfurliches  Lexicon,  p.  402. 


vii  UBIQUITY  OF  CINYRAS  293 

a  person.  He  is  connected  with  the  Trojan  war,  during  which 
he  was  king  of  Cyprus.  Thus  Homer's  conception  of  him 
lends  him  no  preternatural  powers,  no  divine  attributes. 
He  was  Agamemnon's  distant  admirer  and  benefactor  who 
sent  him  armour : 

Tidings  he  heard  in  Cyprus  and  fame  that  Achaeans 
Planned  to  set  sail  in  their  ships  for  Troy  and  its  capture, 
Wherefore  he  gave  the  breastplate  in  order  to  please  Agamem 
non. 

All  this  is  creditable,  but  the  tale  which  arose  shortly 
after  Homer's  day  was  less  creditable  to  Cinyras'  heart 
than  to  his  head.  Bound  by  oath  to  supply  a  certain 
number  of  ships  against  Troy,  the  crafty  Paphian  monarch 
did  supply  just  the  number  promised ;  only,  as  their  size 
and  equipment  were  not  nominated  in  the  bond,  it  pleased 
his  royal  thriftiness  to  send  ships  of  microscopic  dimensions 
fashioned  out  of  clay.1 

But  after  all  the  real  Cinyras,  founder  of  Old  Paphos, 
lived  long  before  the  Trojan  war,  and  was  far  more  than 
the  obsequious  admirer  of  Agamemnon,  far  higher  minded 
than  this  crafty  evader  of  the  solemn  terms  of  a  treaty. 
Ages  before  the  Trojan  war  he  lived,  first  founder,  priest, 
and  king,  not  only  of  the  sanctuary  of  Aphrodite  in  Old 
Paphos,  but  of  one  at  Byblus  in  Syria — Byblus,  where  the 
goddess  was  revered,  as  at  Paphos,  under  the  form  of  a 
cone  of  white  stone.2  Byblus  is  quite  as  closely  identified 
in  fact  with  the  goddess  in  the  history  of  Syrian  religion, 
as  is  Paphos  in  that  of  Greek  religion.  A  name  well  known 

1  Probably  some  primitive  record  of  ritual  observance  lurks  beneath 
this  pointless  tale. 

2  Maximus  Tyrius,    Dissert,    viii.    7  :    ITa0/ois    i]    ^v   ' AQpodirr)   raj 

j,a  OUK  &v  et/cd<7<us  AXXy  Tip  ?)  Trvpa.fj.l5i  \evKrj,  i]  8t 


294  APHRODITE  AT  PAPHOS  vn 

of  us  all,  that  of  Pygmalion,  is  connected  with  Cinyras.  The 
two  were  kinsmen;  and  their  bond  of  kinship,  so  far  as 
mythology  has  worked  it  out,  consisted  in  their  both  being 
originators  of  skilled  processes  in  the  various  arts  that  adorn 
the  mind  and  the  life  of  man.  Both  play  a  Promethean 
part  in  dowering  mankind.  Cinyras,  for  instance,  invented 
mining,  bricks,  and  various  agricultural  implements,1  and 
Pygmalion  became  celebrated  for  his  skill  in  sculpture. 

Cinyras  is  so  closely  identified  with  Cyprus,  in  spite  of  the 
tie  binding  him  to  Byblus,  that  he  is  spoken  of  as  coming 
from  almost  every  country  which  contributed  to  the  early 
peopling  of  the  island.  He  was  from  Cilicia  as  well  as 
from  Syria  and  Assyria,  and  he  was  beloved  of  Aphrodite. 
This  glory  he  shared  in  legend  with  Adonis,  and  is  half 
identified  with  him.  Apollo  delighted  in  him,  and  yet 
there  was  a  rivalry  in  musical  performances  wherein  Cinyras 
— taking  for  the  nonce  the  role  of  Marsyas  in  the  Thraco- 
Phrygian  story 2 — was  worsted  by  his  protector  Apollo. 

As  for  this  musical  episode  in  his  career,  it  has  a  greater 
importance,  some  think,  than  would  casually  be  attached  to 
it.  The  theory  is  that  the  name  Cinyras  thinly  disguises 
that  of  a  musical  instrument,  associated  with  the  royal 
founder  and  priest  of  Old  Paphos.  This  name  and  the 
proverbial  wealth  of  a  Midas,  along  with  the  story  of  his 
birth  in  Cilicia,  have  attached  themselves  to  him  from 
Asia  Minor,  the  land  of  the  Phrygian  flute  and  the  home 
of  soft  Lydian  airs. 

If  it  were  possible  to  unravel  aright  all  the  legends  of 
Cinyras,  the  whole  early  and  unknown  history  of  Cyprus 

1  Tegulas  invenit  Cinyras  .  .  .  et  metalla  aeris,  utrumque  in  insula 
Cypro,  item  forcipem,  martulum,  vectem,  incudem  [Pliny,  Nat.  Hist. 
vii.  195).  2  See  p.  92  above. 


vii  THE  TEMPLE  ON  THE  HILL  295 

might  be  understood.  For  just  as  the  three  touches  above 
mentioned  come  from  Asia  Minor,  so  from  Phoenicia  in 
part,  and  largely  again  from  Phrygia  in  Asia  Minor,  comes 
his  close  connection  with  Adonis.  As  the  case  stands  we 
have  only  the  tantalising  satisfaction  of  knowing  how  many 
and  of  what  divergent  origin  were  the  threads  that  time 
and  the  fruitful  invention  of  story-tellers  and  poets  have 
woven  into  the  variegated  yarns  concerning  Cinyras.  Let 
the  whole  of  them  serve  as  a  background  for  the  worship 
and  the  presence  of  the  goddess  Aphrodite  at  Old  Paphos. 
Let  us  fling  upon  the  throne  of  Paphian  Aphrodite  the 
richly  variegated  web  of  early  legends,  and  then  when  she 
shall  have  taken  her  seat,  and  with  smiles  shall  be  admiring 
her  own  loveliness  in  an  upraised  mirror-disc,  we  may  call 
to  her  as  burning  Sappho  called  of  old,  "  Immortal  Aphro 
dite  throned  in  many  hues."1  The  hues  lent  by  early 
legend  are  always  pure  of  quality  and  soft  in  tone,  and  even 
Aphrodite's  beauty  is  enhanced  by  them. 

Now  that  we  know  the  worst  and  the  best  of  the  first 
founder  of  the  Paphian  temple  of  Paphian  Aphrodite,  let  us 
look  at  the  site  2  where  he  builded  her  temple  of  old.  He 
built  it,  after  the  manner  of  the  Assyrians  and  Phoenicians, 
in  a  high  place.  The  hill  of  Old  Paphos  is  distant  about 
one  mile  from  the  sea.  Half  of  this  mile  is  taken  up  by  a 
gradual  slope  from  the  hill's  summit  to  its  base.  The  sea 
lies  south-west  of  the  temple-site,  and  in  the  far  distance,  at 
the  end  of  a  gradual  but  constant  upheaval  of  range  upon 
range  of  hills,  rise  the  heights  around  Mount  Troodos,  itself 
not  visible  from  here. 


'  dOdvar'  'A0p65tra. 
2  For  a  detailed  account  of  the  remains  there  discovered,  see  Appendix 
VI.  below. 


296  APHRODITE  AT  PAPHOS  vn 

A  gentle  slope,-  strewn  with  bits  of  limestone  ranging  in 
colour  from  yellow  to  gray ;  close  to  its  brow  a  small  and 
uncleanly  village,  on  the  east  and  west  of  which  are  the 
beds  of  inconsiderable  streams, — such  is  the  Paphian  site, 
and  such  the  modern  village  of  Couclia  (Couvocles  in  Old 
French).  The  prevailing  tint  in  early  May,  when  we  visited 
it,  was  a  yellow  which  might  have  verged  towards  brown 
had  it  not  been  for  a  spare  crop  of  grain  that  invisibly 
warmed  the  surface  tint.  The  curious  eye  peering  eastward 
is  rewarded  by  a  vision  of  palms.  These  and  the  peculiarly 
picturesque  discomfort  of  the  squalid  village  houses  assure 
you  that  you  are  neither  in  England  nor  in  New  England. 

Just  here  the  friable  limestone,  beaten  on  other  coasts 
near  by  into  fantastic  cliff- shapes,  takes  unnoticeable 
rounded  forms  or  clothes  itself  with  soil.  A  stretch  of 
fertile  bottomland  reaches  from  the  foot  of  the  Paphian  hill 
south-westward  to  the  strand  of  the  sea.  At  a  distance  of 
about  three  hundred  yards  from  the  water's  edge  there  is  a 
drop  of  ten  feet,  and  again  at  half  that  distance  a  second 
drop  of  about  the  same  number  of  feet.  Thus  was  the 
sheer  rock  rounded  and  the  higher  reach  of  meadow-land 
terraced  down  to  the  verge  of  the  sea  for  a  gradual  ascent 
to  the  high  place  of  Aphrodite  at  Old  Paphos.  Across 
this  green  and  flower-strewn  meadow-carpet  moved  the  new 
born  goddess  from  the  surface  of  the  wave.  Thither,  says 
Hesiod,  she  had  been  gently  borne  drifting  eastward  from 
Cythera ;  or,  as  others  might  say,  westward  from  Syria. 
Whichever  of  these  directions  is  assigned  to  the  coming  of 
the  goddess,  one  detail  all  legends  have  in  common,  i.e. 
that  Aphrodite  was  born  here  in  Cyprus,  and  her  name  is 
Cypris  or  Cyprogeneia — Cyprus-born. 

As   a  Greek   goddess,   swaying   the  gods  on  Olympus 


vii  APHRODITE'S  BIRTH  AT  PAPHOS  297 

and  ruling  according  to  her  good  pleasure  the  hearts  of 
all  mortals,  Aphrodite  was  born  at  Paphos  in  Cyprus. 
Speculations  about  her  history  before  her  birth  at  Paphos 
did  not  enter  into  the  devout  minds  of  her  worshippers. 
For  them  she  was  what  the  unhappy  Phaedra  calls  her, 
Cyprian  Aphrodite  of  the  sea — Cypris  Pontia.  Nor  was  this 
Paphian  birth  without  an  enormous  influence  upon  the 
larger  world  outside  of  Greece.  There  was  a  something  in 
the  Paphian  Aphrodite  that  her  eastern  original  lacked,  and 
this  something  can  by  a  process  of  exclusion  be  proved  to 
have  been  a  distinctively  Greek  aspect  of  her  divinity. 
Aphrodite's  birth  at  Paphos  made  her  at  last  accessible  to 
purely  Greek  influences,  and  qualified  her  to  play  a  great 
part  in  the  religious  history  of  the  world.  It  was  perhaps  a 
misfortune  that  Greek  influences  never  wholly  divorced  her 
from  the  manners  and  customs  that  finally  attached  to  her 
in  the  East,  as  a  Syrian  and  Phoenician  divinity.  At  all 
events  much  that  was  outrageously  gross  and  uncivilising 
in  her  latter-day  worship  at  Corinth  and  elsewhere  in  Greece 
is  traceable  to  the  direct  influences  exercised  from  the  East 
by  her  later  and  unclean  worship  there.  We  can,  therefore, 
agree  with  a  gifted  American  scholar,1  who,  under  the 
happy  inspiration  of  words  associated  with  Plato's  name,2 
has  said  that  "the  Greek  received  nothing  from  the  East 
that  he  did  not  make  doubly  his  by  the  beauty  with  which 
he  invested  it,  and  the  Aphrodite  Urania  is  as  far  above  the 
Oriental  personification  of  the  conceptive  principle  of  nature, 
as  the  graceful  image  of  Venus  issuing  from  the  shell  .  .  . 
excels  the  clumsy  merman  Dagon  whose  hands  and  feet 


1  Essays  and  Studies,  by  Basil  Lanneau  Gildersleeve,  1890. 

2  Epinomis,  487  E,  6n  Trep   &v  "EXX^i/es  fiapfiapuv  Trapa\df3w(ri,  /caX- 
\Lov  TOVTO  eis  rcXos 


298  APHRODITE  AT  PAPHOS  vn 

were  cut  off  upon  the  threshold  of  his  temple  when  he  lay  a 
prostrate  deformity  before  the  ark  of  the  Lord  at  Ashdod." 

The  something  which  stamps  Aphrodite  as  a  specifically 
Greek  version  of  the  often  perverted  Eastern  goddess  is 
certainly  that  which  made  her  admirable  above  her  eastern 
originals,  and  secured  her  the  adoration  of  all  the  world.1 
The  Greek  strain  in  Paphian  Aphrodite  is  almost  at  war 
with  the  sombre  jealousy  which  never  really  left  her  because 
it  attached  itself  to  her  Assyrian  nativity.  Greece  gave  to 
her  a  fresh  and  breezy  sanity  and  an  inborn  grace  that 
Euripides  describes  so  perfectly,  calling  her  peerless  among 
the  gods  in  just  this  her  quality  of  qualities.  Eucharis^  he 
calls  her, — gently,  sweetly,  gracefully  charming.  She  had,  as 
Euripides  worshipped  her,  beyond  all  other  goddesses  the 
charm  that  wins  and  never  loses,  the  grace  that  woos  and 
wearies  not,  the  beauty  that  waxes  and  never  wanes,  the 
smile  that  attaches  and  never  repels. 

Somewhere  Euripides  describes  the  only  ray  of  hope 
left  to  lighten  the  darkness  of  despair  as  the  Aphrodite 
of  woes.  That  hold  upon  hope  and  help  which  leads 
a  man  in  the  midst  of  desperate  flames  to  wait  one 
precious  moment  longer  before  he  flings  himself  head 
long  to  destruction  is  the  Aphrodite  of  his  woeful  state, 
whose  power  subdues  the  wildness  of  sudden  impulse 
and  saves  his  life.  Here  is  the  Aphrodite  Sosandra, 
rescuer  of  men,  the  Notre  Dame  de  Bon  Secours,  wor 
shipped  on  the  Athenian  Acropolis,  where  her  statue  by 

1  The  very  transformation  in  meaning  which  Aphrodite's  epithet  Urania 
underwent  before  our  very  eyes,  so  to  speak,  is  full  of  the  Greek  spirit  which 
purified  and  ennobled.  Herodotus  identifies  Aphrodite -Urania  with 
Mylitta  (i.  131  and  199),  with  whose  coarse  rites  Plato's  Urania  has 
nothing  in  common.  The  real  equivalent  of  Mylitta,  so  far  as  these  rites 
go,  would  be  Aphrodite  Pandemos.  See  Robert's  Preller,  pp.  354  ff. 


vii  THE  GREEK  APHRODITE  299 

Calamis  cheered  and  uplifted  the  sinking  hearts  of  those 
who  came  to  worship.  Hers  was  a  graciousness  so  bountiful 
and  subtly  winning  that  its  spell  cannot  and  never  will  be 
broken.  To  this  degree  of  spiritual  nobility  had  the  Greeks 
at  Athens  in  their  pious  meditations  brought  the  charm  of 
Aphrodite,  known  in  simpler  Homer's  day  as  her  charmed 
girdle,  of  which  he  says  : 

Broidered  and  fashioned  therein  were  all  means  of  enchant 
ment, 

Love-longing  was  there  and  commerce  of  love,  and  there  was 
love's  chatter, 

Words  and  cajolements  that  steal  away  wits,  even  men's  that 
are  wisest. 

The  almost  wholly  physical  charm  of  her  Cestus  was  not, 
as  Homer  conceived  it,  absolutely  inseparable  from  Aphro 
dite.  She  alone  possessed  it,  but  if  she  chose  she  could 
lend  it,  and  so  it  was  that  Hera  borrowed  it  of  her.  On 
the  other  hand,  the  later  and  more  spiritual  charm  which 
shone  through  disasters  and  rescued  sorrowing  men  was 
inseparable  from  her.  No  goddess,  not  even  Hera,  Queen 
of  Olympus,  could  borrow  this,  or  rescue  as  she  could  by 
the  gracious  subtlety  of  her  smiling  love.  Not  Euripides 
only  but  Pindar  the  sublime  saw  and  spoke  of  the  pecu 
liarly  Greek  quality  in  Aphrodite.  Only  Pindar,  as  in  so 
many  other  cases,  beheld  the  more  solemn  and  serious 
aspect  of  this  great  truth.  That  principle  of  measure  and 
retributive  compensation — which  bids  despair  despair  not, 
but  lures  a  desperate  man  to  hope,  revealing  in  the  very 
presence  of  woe  glimpses  of  grace  and  showing  the  face  of 
joy  to  come — assumes  another  and  more  solemn  guise  when 
men  are  glad  and  fortune  favours  them.  A  pang  there  is 
in  fullest  moments  of  supreme  joy  which  prompts  us  to 


300  APHRODITE  AT  PAPHOS  vn 

beware, — a  sense  of  penalties  to  come  which  visits  him 
whose  joy  might  grow  excessive,  and  whose  pleasures  make 
him  lose  the  consciousness  of  duty  to  be  done.  This  throb 
of  pain  in  pleasure's  midst,  this  pausing  in  mid-course  of 
the  forward  fling  of  gladness,  is  perhaps  but  the  negative 
pole  that  should  be  coupled  with  the  saving  graciousness  of 
Aphrodite  Sosandra's  smile.  She  saves  man  from  himself 
both  in  sorrow  and  in  joy.  And  so  Pindar,  in  his  ninth 
Pythian  ode,  agrees  with  Euripides,  telling  how  "  Silver-foot 
Aphrodite  tempered  the  loves  of  Apollo  and  Cyrene  with 
awe,"  with  a  sense  of  some  limit  not  to  be  exceeded.  This 
awe-struck  sense  of  shame  at  the  thought  of  too  great  excess 
Pindar  has  called  Aidos,  and  Euripides  has  personified  the 
impulses  which  violate  it  in  speaking  of — 

Eager  Loves  that  past  all  measure  fling. 

From  these  the  chorus  of  women  in  the  Medea  of  Euripides 
beg  Aphrodite  to  deliver  them.  They  pray  to  her  for 
deliverance  from  jealousy  such  as  has  devastated  Medea's 
life  and  happiness.  They  beseech  her  to  "  choose  with 
critic  glance  "  the  "  wedded  happiness  of  women." 

And  so,  with  all  the  purity  of  her  natal  element,  Aphro 
dite  rises  up  before  the  high  moods  of  imaginative  medita 
tion  vouchsafed  to  Euripides  and  to  Pindar.  Yet  Euripides 
does  not  forget  her  original  function  as  a  nature-goddess, 
as  another  beautiful  song  from  his  Medea  well  shows : 

Cypris,    when   from   fair  Cephissus'  downward   streaming 

wave 

Draughts  of  water,  so  'tis  said,  to  quaff  she  sometime  drew. 
Breathed  adown  the  countryside  blithe  winds  that  softly  blew, 
Crowned  her  locks  with  roses  red  that  sweetest  perfume 

gave, 
Sent  to  second  wisdom,  Loves  all  righteousness  to  save. 


vii  THE  ROMAN  AND  ASSYRIAN  GODDESS  301 

These  are  some  of  the  higher  notes,  the  more  purely 
Grecian,  and  therefore  the  more  beautiful  and  nobler 
aspects  which  made  the  cult  of  Aphrodite,  when  it  spread 
over  all  the  Roman  world,  a  benefit  to  civilisation.  Would 
that  only  her  nobler  part  had  ever  found  allegiance  in  the 
pagan  world.  For  better  for  worse,  for  richer  or  for  poorer, 
the  Roman  Emperors  took  Venus  Aphrodite  as  their  own 
and  made  her  mistress  of  the  world.  Their  veneration  sprang 
from  no  idealised  adoration  of  the  highest  range  of  inspira 
tion  derivable  from  her,  they  were  prompted  rather  by  con 
siderations  heraldic,  genealogical,  and  theological.  It  was 
therefore  not  purely  an  accident  that  made  Caesar's  watch 
word  on  the  field  of  Pharsalia  "Venus  victrix,"  for  such  an 
accident,  without  being  deeply  significant,  reveals  the  presence 
of  Aphrodite  in  the  thoughts  of  those  who  fashioned  Rome. 

This  being  the  case,  it  is  no  matter  of  surprise  that  one 
of  the  very  noblest  and  broadest  conceptions  of  the  goddess 
in  the  plenitude  of  her  imperial  power  should  be  that  of  a 
Roman  poet, — Lucretius.  For  him  she  is  more  than  the 
source  of  all  joy  on  earth  and  in  heaven.  She  is  the 
principle  of  life ;  without  her  earth  can  have  no  flowers,  the 
woods  can  have  no  foliage,  and  the  birds  sing  only  at  her 
will.  "  Without  thee,"  he  cries,  "  nothing  that  is  can  ever 
attain  existence." 

This  is  really  not  very  far  from  that  simple  and  touching 
conception  which  was  the  very  primal  foundation  of  immor 
tality  for  the  goddess  Venus  Aphrodite,  the  pure  and  noble 
delineation  of  the  Assyrian  goddess  Ishtar  in  a  poem  far  older 
than  that  of  Roman  Lucretius.  This  poem  was  found  stamped 
upon  tablets  of  burned  clay  in  the  library  of  Sardanapalus 
or  Assurbanipal  at  Niniveh.  The  goddess  left  this  upper 
earth  to  seek  among  the  dead  below  her  Tammuz  dearly 


302  APHRODITE  AT  PAPHOS  vn 

loved  and  lately  lost.  No  sooner  had  the  wondrous  Ishtar 
disappeared  from  earth  than  life  in  the  upper  air  was 
brought  to  a  pause,  and  the  most  unspeakable  distress 
made  the  whole  universe  wail  and  faint.  Ishtar,  waiting  at 
the  dust-bestrewn  portals  of  Urugal,  laments  her  case  in  the 
most  exquisite  strains  of  the  old  Assyrian  poem  : 

All  love  from  earthly  life  with  me  departed, 

With  me  to  tarry  in  the  gates  of  death  .   .  . 

And  chilled  shall  cheerless  men  now  draw  slow  breath. 

I  left  in  sadness  life  which  I  had  given, 

I  turned  from  gladness  and  I  walked  with  woe. 

I  search  for  Tammuz,  whom  harsh  fate  laid  low. 

The  darkling  pathway  o'er  the  restless  waters 
Of  seven  seas  that  circle  death's  domain 
I  trod,  and  followed  after  earth's  sad  daughters 
Torn  from  their  loved  ones  and  ne'er  seen  again. 

Here  must  I  enter  in,  here  make  my  dwelling 
With  Tammuz  in  the  mansion  of  the  dead, 
Driven  to  Tammuz'  house  by  love  compelling, 
And  hunger  for  the  sight  of  that  dear  head. 

O'er  husbands  will  I  weep,  whom  death  has  taken, 
Whom  fate  in  manhood's  strength  from  life  has  swept, 
Leaving  on  earth  their  living  wives  forsaken, 
O'er  them  with  groans  shall  bitter  tears  be  wept. 

And  I  will  weep  o'er  wives,  whose  short  day  ended 
Ere  in  glad  offspring  joyed  their  husbands'  eyes  ; 
O'er  them  shall  tearful  lamentations  rise. 

And  I  will  weep  o'er  babes  who  left  no  brothers, 
Young  lives  to  the  ills  of  age  by  hope  opposed, 
One  moment's  life  by  death  unending  closed.1 

1  For  a  more  complete  version  of  this  with  others,  made  from  a  literal 
version  of  the  cuneiform  text  kindly  supplied  me  by  my  friend  and  former 
colleague  Dr.  Lyon,  see  The  Story  of  Assyria,  where  the  gifted  author, 
Mme.  Ragozin,  does  me  the  honour  of  printing  my  versions  in  an  appendix. 


vii  APHRODITE  AND  DEMETER  303 

Here  is  a  large  and  oriental  luxury  of  grief,  a  tragedy  and 
mystery  of  woe  which  is  ill  suited  to  the  smiling  Aphrodite 
of  the  Greeks,  and  which  stirs  far  deeper  down  the  hearts 
of  men  than  even  the  Venus-Aphrodite  of  Lucretius.1  The 
truth  is  that  there  is  no  parallel  easily  found  to  the  tragical 
shudder  that  runs  through  every  line  of  this  Assyrian  poem. 
Among  Greek  divinities  Demeter2  rather  than  Aphrodite 
represents  this  utterness  of  woe  and  this  fulness  of  heart- 
sympathy  for  the  calamities  of  others. 

The  goddess  Demeter  even,  when  sorrow  had  all  but 
slain  immortality  within  her,  was  rescued  by  the  gracious 
promise  of  hope  in  Aphrodite's  smile ;  and  this  may  justify 
us  in  maintaining  once  more  that  the  Greek  conception  of 
Aphrodite  was  that  of  a  consoling  goddess,  the  enemy  of 
despair,  the  harbinger  of  joys  unlocked  for,  and  the  moder 
ator  of  unruly  pleasures. 

This  noble  and  consoling  figure  has  dominated  the 
destinies  of  nations  since  history  began.  Arising  among 
the  unknown  tribes  of  Accad  in  the  far-off  east,  she  was 
honoured  and  worshipped  at  Niniveh  and  Babylon,  at  Tyre 
and  Sidon  and  Carthage.  Before  the  taint  of  monstrous 
licentiousness  had  brought  corruption  upon  her,  she  found 

1  To  this  extent  and  in  this  sense  I  should  incline  to  dissent  from  Pro 
fessor  Gildersleeve,  who  says,    ' '  the  worship  of  Aphrodite  is  no  less  pro 
found  while  it  is  infinitely  more  graceful  than  the  Oriental."     With  the 
latter  part  of  this  I  agree  too  thoroughly  to  be  able  to  subscribe  to  the 
former.      Demeter's  awkward  and  mysterious  strains  go  together,  and  both 
of  them  make  her  the  Greek  equivalent  of  the  tragic  Ishtar  who  laments 
for  Tarn  muz. 

2  Euripides,  in  the  Helena,  almost  seems  to  have  the  Assyrian   poem 
before  him  in  his  beautiful  song  of  the  sorrows  of  Demeter-Cybele,  who 
falls  fainting  and  lifeless  on  the  snowdrifts  of  mountains  in  Asia, — fainting 
and  lifeless  because  she  finds  not  Persephone.     The  earth  meanwhile  is 
stricken  with   barrenness.      Until   the  Graces   and  Aphrodite  sought   her 
there  and  ministered  to  her,  there  was  no  fruitfulness,  no  springs  would 
flow,  and  feasting  fled  from  earth  and  forsook  the  gods  even. 


304  APHRODITE  A  T  PAPHOS  vn 

from  the  purifying  sea  a  new  birth,  and  came  to  her  rising 
in  Greece  from  Cyprian  Paphos.  When  the  power  of  self- 
repression,  with  which  the  nobler  thought  of  Greece  asso 
ciated  Aphrodite,  began  to  falter  and  fail,  Rome  saved  the 
sea-born  goddess  and  her  higher  worship.  Indeed  the 
Roman  Mother  Venus  of  Virgilian  song  was  as  intensely 
kind,  as  irresistibly  winning,  as  that  most  primally  and 
exclusively  Greek  goddess  the  darling  daughter  of  Zeus 
and  Dione,  the  slender  maid1  who  was  worshipped  and 
feared  by  Sappho.  Nor  does  the  broader  and  more  mature 
as  well  as  more  oriental  conception  of  Paphian  Aphrodite, 
into  which  this  goddess  of  Sappho  was  finally  merged,  reach 
quite  the  heights  and  depths  of  the  all-nourishing  mother 
Venus  to  whom  Lucretius  makes  melodious  prayer. 

Perhaps  the  poor  Cypriote  papissa  or  priest's  wife  was 
not  wrong  in  correcting  the  learned  Ludwig  Ross.  He 
spoke  to  her  of  a  sanctuary  of  Aphrodite.  "  No,"  she  said, 
"  not  Aphroditissa  ;  it  is  the  Holy  Chrysopolitissa."2  To  her 
is  sacred  a  church  on  Aphrodite's  Hill  of  Old  Paphos.  The 
Mother  of  God  surnamed  the  Golden  has  taken  the  place 
once  filled  by  Homer's  golden  Aphrodite. 


fj-arep,   OVTOL  5vva.fj.ai.  Kpticrjv  TOV  'icrrov 
ircudos, 


See  Euripides,  Helena,  1098  :    K6pr]  AIWPTJS  K^TT/H,  ^77  p 

ray>a   TT]V  \a\ov<ri    Xpvcro- 


APPENDIX  VI 
THE   TEMPLE   AT   OLD    PAPHOS 

THE  irregular  trapezium l  which  the  temple  and  its  con 
tiguous  buildings  once  occupied  lies  at  some  distance — a 
stone's  throw,  be  it  said — from  the  brow  of  the  hill.  Be 
fore  this  trapezium  area  is  reached  you  come  upon  the 
most  remarkable  remains  upon  the  site.  I  cannot  describe 
these  better  than  in  the  words  of  Mr.  R.  Elsey  Smith.2 
"  Starting  from  the  south-west  corner,  and  examining  the 
walls  in  detail  as  proposed,  we  find  first  of  all  a  very  large 
massive  wall  extending  for  some  eighty-five  feet  in  a  nearly 
northerly  direction,  with  a  short  return  at  the  south  end. 
It  consists  of  a  basement  of  polygonal  blocks  mostly  of 
massive  proportions  brought  to  a  fairly  even  face,  and  with 
a  carefully  wrought  and  levelled  upper  bed,  on  which  rests 
a  series  of  magnificent  rectangular  blocks,  the  largest  of 
which  measures  seven  feet  by  over  fifteen.  These  blocks 
are  of  limestone,  and  have  been  laid  with  their  beds 
vertical,  so  that  they  have  suffered  severely  from  the  effects 

1  Excavations  in  Cyprus  (1887-1888),  reprinted  from  the  Journal  of 
Hellenic  Studies  (1888).  See  pp.  58  and  ff. ,  where  Mr.  Ernest  Gardner 
conclusively  disposes  of  all  plans  previously  published,  and  clearly  sets 
forth  Mr.  Elsey  Smith's  plan  in  connection  with  ancient  authorities  and 
coins.  For  my  own  detailed  account  of  the  site  and  of  Cesnola's  so-called 
plan,  see  Nation  (Sept.  6  and  13,  1888). 

a  Ibid.  See  Mr.  Elsey  Smith's  discussion  of  his  own  plan  in  the  Journal 
of  Hellenic  Studies  as  cited  above.  The  plan  is  by  his  kind  permission 
here  reproduced. 

X 


306  APPENDIX  VI 

of  weather.  The  stones  both  of  the  basement  and  upper 
parts  of  the  wall  are  pierced  with  holes  for  the  purpose  of 
hauling  them ;  the  larger  stones  have  two  holes,  but  some 
of  the  smaller  ones  are  pierced  with  a  single  hole  only. 
In  the  upper  stones  these  holes  run  from  the  vertical  face  at 
one  end  in  a  quadrant  form  up  and  down  to  the  upper  and 
lower  beds  of  the  stone ;  in  the  basement  stones,  which  of 
course  were  below  the  pavement  level,  these  holes  generally 
run  from  the  face  backward  to  the  vertical  joints.  " 

This  remarkable  and  puzzling  remnant  is  undoubtedly, 
Mr.  Elsey  Smith  and  all  others  conclude,  the  oldest  feature 
discoverable  at  Couclia.  Curiously  enough  it  contains  the 
only  evidence  upon  the  whole  site  that  there  was  any  door 
way  in  any  particular  place.  The  two  socket -holes  for 
door-posts  have  two  steps  leading  down  to  them.  This 
door  was,  however,  much  less  than  ten  feet  wide  from  post 
to  post,  and  can  hardly  have  been  the  principal  entrance 
to  any  court,  such  as  Mr.  Elsey  Smith  thinks  was  possibly 
enclosed  by  this  the  oldest  structure  at  Old  Paphos.1  The 
piercings  in  these  huge  monoliths  are  said  strongly  to 
resemble  what  is  discoverable  upon  the  site  of  temples 
presumably  of  Phoenician  origin  on  the  island  of  Malta. 
Among  the  ruins  of  Sicilian  Selinus,  which  was,  like  Old 
Paphos,  a  meeting  ground  for  Greeks  and  Phoenicians  in 
the  earliest  days,  Ludwig  Ross  saw  stones  similarly  pierced, 
and  he  attributed  them  to  Phoenician  workmanship.  Can 
it  be  amiss  to  recall  here  the  Phoenician  method  of  pre 
paring  for  the  building  of  walls,  as  reported  in  the  account 
of  the  building  of  Solomon's  temple  ? 

"And  the  house  when  it  was  in  building  was  built  of 
stone  made  ready  before  it  was  brought  thither ;  so  that 
there  was  neither  hammer  nor  axe  nor  any  tool  of  iron 
heard  in  the  house  while  it  was  in  building." 

Piercings  of  some  kind  must  have  been  made  in  the 

1  Of  course  the  theory  is  that  all  traces  of  the  main  door  have  dis 
appeared.  Wherever  it  was,  it  certainly  was  not  the  small  door  of  which 
traces  remain. 


THE  TEMPLE  AT  OLD  PAPHOS  307 

quarry  by  Hiram's  builders  and  stone  squarers,  otherwise  it 
would  not  have  been  possible  to  transport  huge  blocks  and 
put  them  in  their  place  without  leaving  some  projection 
upon  the  blocks,  which  must  then  have  been  removed  upon 
the  site  itself  by  chiselling.  This  would  have  involved  the 
sound  of  hammer,  of  axe,  or  some  tool  of  iron.  The 
Greek  l  way  of  getting  stones  in  place  was  entirely  different, 
since  they  had  no  objection  to  hammering  on  the  site.  They 
shaped  the  stones  very  accurately,  but  finished  them  only 
roughly  in  the  quarry.  The  final  smooth  finish,  together 
with  the  chiselling  down  of  square  projections  left  for  a 
hold  in  transportation,  was  done  when  the  walls  were  up. 
The  first  builders  at  Paphos,  then,  may  have  been  Phoeni 
cians,  since  they  probably  used  the  methods  of  Hiram's 
builders,  and  not  those  of  the  Greeks. 

A  change  not  only  in  the  manner  of  building  walls  but 
in  the  way  of  finishing  stones,  and  in  the  methods  used 
for  their  transportation,  took  place  among  the  workmen  at 
Old  Paphos  before  a  stone  of  what  became  the  temple  of 
Aphrodite  was  laid.  The  proof  of  this  is  ready  at  hand  in 
the  remains  of  walls  which  are  close  to  the  supposably 
Phoenician  work  just  described  in  detail.  Here  are  marks 
(A)  of  an  ancient  reconstruction  of  parts  of  walls,  and  (B) 
of  an  addition  made  at  the  same  ancient  date.  All  these 
reconstructed  parts — including  two  rows  of  columns — 
indicate  a  new  manner  of  workmanship.  "The  stones 
employed,"  says  Mr.  Elsey  Smith,2  "are  of  smaller  and 
more  regular  dimensions  .  .  .  and  they  are  very  evenly 
laid  without  mortar ;  each  stone  has  a  broad  draught  along 
the  upper  edge,  and  down  the  two  sides  of  its  outer  face, 

1  To   understand    this  only  a  glance   at   the  unfinished    Propylaea  of 
Mnesicles    at   Athens    is    needful.      The    downward    (westerly)    face    of 
this  building  was   so   far   finished   that  the    square    projections   used   for 
transport  have  been  chiselled  off,  but  the  smooth  chiselling  of  the  outward 
surfaces  has  never  been  completed.     The  upward  (easterly)  face  is  left 
wholly  unfinished,  and  is  still  covered  with  the  curious  square  projections 
intact.     See  Appendix  XI.  i.  8  and  9. 

2  Journal  of  Hellenic  Studies,  1888,  p.  48,  at  the  bottom. 


308  APPENDIX  VI 

leaving  a  rough  panel  in  the  centre."1  This  rough  dressing 
of  the  stone  with  a  draught  around  the  edge — it  commonly 
goes  around  the  whole  edge — is  a  mark  of  workmanship 
which  may  have  been  Phoenician,  and  is  probably  not 
Greek.  The  other  technical  details,  however,  show  a 
nearer  approach  to  Greek  workmanship. 

In  these  early  walls  of  two  kinds  some  discover  an 
earlier  temple  which  became  in  later  and  more  magnificent 
days  an  appendage  to  the  enlarged  sanctuary :  something 
which,  for  want  of  a  better  name,  has  been  called  the  South 
Wing.  The  unmistakably  unique  and  possibly  Phoenician 
character  of  some  of  the  walls  here  not  unnaturally  leads 
one  to  look  for  a  parallel  at  the  monuments  in  Phoenicia. 
Here  is  Professor  Reber's  brief  description  (taken  from  M. 
Renan's  account  of  his  expedition)  of  the  "  Snail's-Tower  " 
at  Marathus,  the  ancient  Amrit,  not  far  from  the  Byblus  of 
which  the  Paphian  Cinyras  was  the  fabled  king.  This 
building,  with  four  others,  forms  the  most  considerable  mass 
of  Phoenician  work  preserved  in  Phoenicia.  The  effect  of 
them  from  afar  is  said  to  be  very  wonderful.  A  closer 
view  shows  a  certain  helpless  heaviness  that  makes  them 
less  interesting.  The  "  Snail's-Tower  "  is  constructed  out 
of  huge  square -hewn  blocks  of  limestone.  It  is  a  cube 

•i  This  rough  dressing  of  the  stone  with  a  draught  around  the  edge  is 
known  as  Rustica,  and  has  been  supposed  to  be  a  mark  of  Phoenician  work 
manship.  But  though  such  chiselling  has  been  found  in  Phoenicia,  it  is 
also  plentiful  elsewhere  and  does  not  prove  Phoenician  handiwork  here. 
Perhaps  it  is  vain  to  seek  any  real  knowledge  of  Phoenician  buildings,  or 
to  identify  anything  as  certainly  Phoenician,  outside  of  Phoenicia  itself. 
Dr.  Franz  Reber  has  truly  said  in  his  Kunstgeschichte  des  Alterthums  (p. 
134)  that  the  characteristically  Phoenician  adornments  of  buildings,  private 
and  public,  were  of  wood.  The  cedars  of  Lebanon  were  so  near  at  hand. 
It  is  therefore  possible — since  Cyprus  of  old  abounded  in  forests — that  the 
first  temple  at  Paphos  was  chiefly  of  wood.  The  tradition  of  abundant 
wood  upon  the  island  existed  in  1532,  when  Ziegler  wrote — "  Silvosa 
primitus  fuit,  sed  quia  metalli  aeris  ferax  esset  silva  excisa  in  caminos  et 
opera  metalli  campos  aperuerunt  arabiles. "  Ortelius,  1573,  gives  Ammianus 
Marcellinus  (book  xiv. )  and  other  authorities  for  forests  and  shipbuilding  in 
Cyprus.  Evagoras  and  Conon  both  resorted  to  these  forests  for  ship 
building,  which  must  have  had  great  importance  from  the  end  of  the  fifth 
century  B.C.  downward. 


THE  TEMPLE  AT  OLD  PAPHOS  309 

eleven  metres  high  and  nine  metres  square,  and  there  are 
traces  of  a  pyramid  that  once  surmounted  this  cube.  Its 
hewn  stones  five  metres  in  length  —  longer,  that  is  to  say,  than 
the  longest  of  the  pierced  monoliths  of  the  south  wing  at 
Paphos  —  are  treated  in  Rustica,  like  the  later  walls  of  the 
same  south  wing  at  Paphos.  The  "  Snail's  -Tower  "  has 
within  it  two  mortuary  chambers,  and  it  was  undoubtedly  a 
tomb,  as  were  also  the  four  smaller  buildings  not  far  away. 
I  venture  to  suggest  that  in  these  remarkably  puzzling 
walls  of  the  so-called  south  wing  we  may  have  the  remains 
of  one  or  more  tombs  built  by  the  Phoenicians.  These 
buildings  were  partly  suggested  by  the  pyramids  of  Egypt, 
no  doubt  ;  the  subterranean  mortuary  chambers  hollowed 
out  in  the  rock  under  all  the  great  tombs  at  Marathus, 
excepting  the  "  Snail's  -Tower,"  are  indeed  an  attempt  to 
reproduce  the  conditions  of  the  chamber  of  the  dead  in  the 
bowels  of  the  Egyptian  pyramid.  In  the  "  Snail's-Tower  " 
and  —  if  it  be  a  tomb  —  also  at  Paphos  we  have  a  departure 
from  this  type.  The  builders  at  last  contented  themselves 
with  massively  built  chambers  above  ground.  What  leads 
me  to  think  of  the  south  wing  here  at  Paphos  as  having 
possibly  been  a  tomb  is  the  matter-of-course  way  in  which 
Clement  of  Alexandria,1  quoting  from  the  first  book  of  a 
work  by  Ptolemy  the  son  of  Agesarchus,  talks  of  the  tombs 
of  Cinyras  and  his  descendants  as  being  at  Paphos  within 
the  precinct  of  Aphrodite.  They  are,  he  as  much  as  says, 
well-known  features  of  the  sanctuary  at  Old  Paphos.  We 
need  not,  in  spite  of  some  traditions  to  that  effect,  assume 
a  tomb  of  Aphrodite,  but  merely  that  there  was  at  Paphos 
one  tomb  or  several  for  the  great  king-priests  of  importance. 
But  if  Clement  the  Roman  were  taken  literally  in  what  he 
says  of  a  tomb  of  Aphrodite  at  Old  Paphos,2  this  might  very 
well  be  paralleled  by  the  statue  of  Aphrodite  surnamed  of 


1  Clemens  Alexandrinus  (Protrept.  cap.  iii.  ad  fin.}  —  nToXe/Kuoy  3£  6 
roO  'Ayrjffdpxov  ev  T£  Trpwro;  r&v  Trepi  rbv  ^lAoTrdro/xz  ev  IId0y  X^*yei  (v 
T$  TTJS  'A0po5tT77s  tepy  Hivvpav  re  Kal  roi)s  Kivupov  airoybvovs  KfKrjdfVffBai. 

2  HomiL  v.  23. 


310  APPENDIX  VI 

the  tomb  («riTv/i/?ta)  l  and  the  tomb  of  Chthonian  Dionysus 
which  were  features  of  the  great  temple  at  Delphi.  At 
Delphi  the  tomb  of  Dionysus  was  a  half-understood  relic 
of  an  order  of  worship  long  superseded.  At  Paphos  the 
Phoenician-built  tomb  of  Aphrodite  would  represent  an 
early  and  perhaps  a  purely  Phoenician  phase  of  worship, 
and  would  commemorate  the  descent  of  Ishtar  —  her  death 

—  into  Urugal. 

But  now,  if  these  oldest  ruins  are  the  remains  of  tombs, 
it  is  certainly  time  to  turn  to  the  temple  itself,  which  is 
only  a  few  steps  north  of  the  northernmost  portion  of  the 
remains  just  mentioned.  I  have  called  the  whole  space 
occupied  by  the  temple  ruins,  as  Mr.  Gardner  does,  an 
irregular  trapezium.  Its  west  and  south  sides  measure 
220  feet,  its  east  side  228  feet,  and  its  north  side  207  feet. 
This  great  quadrilateral  enclosure  is  not  rectangular,  though 
the  Romans,  when  they  repaired  and  rebuilt  it,  did  their 
best  to  make  a  rectangle  out  of  it.  Its  eastern  side  con 
sisted  of  a  range  of  chambers  and  a  way  out  ;  the  whole  of 
its  northern  side  probably,  and  certainly  the  whole  of  its 
southern  side,  consisted  of  a  long  porch.2  These  porches, 
had  their  unmistakable  traces  not  appeared,  would  have 
been  suggested  not  only  by  what  is  known  of  cognate 
temples,  but  also  by  the  heats  of  Cyprus.  These  impera 
tively  demanded  a  shelter  within  the  precinct  for  the  pro 
cessions  wearied  by  the  long  way  from  New  Paphos.3 
These  considerations  suggest  that  upon  the  remaining  side 

—  the  western  one  —  may  also  have  been  a  portico.     Un 
fortunately  there  are  absolutely  no  remains   of  any  kind 

1  Plutarch,  Q.  Rom.  23.      The  other  Clement  (of  Alexandria),  speak 
ing   of    the   Argives,   describes    them    as    ol 


I  use  porch  here,   as  in  chap.  vi.  above,  in  the  sense  ot    stoa  or 
porticus. 

3  A  remarkably  interesting  proof  of  the  great  good  sense  which  guided 
the  learned  authors  Perrot  and  Chipiez  in  their  account  of  Paphos  may  be 
found  in  their  statement  that  such  porticoes  were  required  upon  the  site. 
It  is  not  often  that  subsequent  excavations  on  a  site  bear  such  clear  testi 
mony  in  favour  of  those  who  had  to  speak  without  all  the  facts. 


PLAN  OF  THE  TEMPLE  AT  OLD  PAPHOS 


NORTH 


COURT 


EAST        ENTRANCED 

m^^^^^ 

CENTRAL  CHAMBER 


-  Notes— 
Portion  of  'walls  of  periods  earlier  than. 

Roman  existing  in  site 

Portion*  of  early  walls  supplied  to  complete 

existing  fragments 

Portions  of  walla  of  Roman  note  existing 

tnsite    _ r 

Pert    ,3  of  Rom-an  walls  suppftei  to 
complete  existing  fragments 


Clst't  Smith  Hint.  A  Del.  1888 


To  face  page  310 


THE  TEMPLE  AT  OLD  PAPHOS  311 

upon  the  western  side,  and,  therefore,  only  the  natural 
requirements  of  worshippers,  and  the  not  wholly  convincing 
parallel  of  Solomon's  temple,  can  be  appealed  to  in  favour 
of  a  porch  or  anything  else.1  With  this  general  description 
of  the  temple  after  the  excavation,  let  us  compare  what  was 
supposed  to  be  known  before  excavations  took  place. 
Professor  Reber,  with  admirable  judgment  and  brevity,  says 
— "  Within  a  circular  fence  enclosure  there  was  a  group  of 
buildings  consisting  of  two  lower  wings  overtopped  by  a 
higher  central  nave.  The  wings  were  supported  by  columns, 
perhaps  they  were  porticoes.  Two  columns  shaped  like 
those  of  Egyptian  architecture  stood  disengaged  and  un 
burdened.  It  is  likely  that  these  last  had  no  architectural 
function,  but  were  like  Jachin  and  Boaz."  These  last,  it 
will  be  remembered,  were  either  in  the  front  of  or  in  front 
of  the  entrance  to  the  Hebrew  sanctuary.  This  whole 
description  is  based  upon  representations  of  the  temple  on 
coins  and  upon  an  intaglio. 

Excavation  on  the  site  has  done  away  with  the  notion 
of  a  circular  enclosure.  The  shape  of  the  coin,  Mr. 
Gardner  suggests,  left  no  choice  to  the  artist.  The 
quadrilateral  court  really  has  no  curved  lines  on  any  of  its 
sides.  It  is  hard  to  determine  the  exact  position  of  the 
sanctuary  where  was  the  pyramid  of  white  stone  worshipped 
at  Paphos  as  at  Byblus.  It  was  in  all  probability  one  of 
the  chambers  built  along  the  east  side  of  the  quadrilateral. 
Of  these  chambers  three  are  now  traceable,  and  strange 
to  relate,  by  the  side  of  the  central  one  of  these  three  is 
an  open  way  out.  This  has  been  called  the  east  entrance. 
North  of  this  entrance  are  the  north  chambers,  not  sup 
posed  by  any  one  to  have  been  the  sanctuary.  South  of 
it  is  the  central  chamber,  which  was  almost  certainly  the 
sanctuary  in  later  Roman  days,  because  in  the  open  court 
opposite  its  front  the  Romans  built  a  colonnaded  approach 

1  Many  cross  trenches  were  made  in  the  hope  that  further  excavation  to 
the  west  might  be  justified.  Nothing,  however,  was  found  except,  at  a  few 
inches  below  the  soil,  the  bed  rock. 


312  APPENDIX  VI 

on  a  higher  level  than  the  south  portico.  This  we  may 
call  the  Central  Hall.  Thus  the  sanctuary  of  later  Roman 
days — a  chamber  older  than  the  Roman  Hall  in  front  of 
it — was  approached  through  a  hall  of  Roman  construction. 
When  this  hall  was  built  the  Romans  undertook  to  make 
the  open  court  square,  and  there  resulted  an  irregular 
passage  just  in  front  of  the  chambers  and  east  entrance 
which  they  very  probably  roofed  over. 

In  spite  of  the  fact  that  the  Romans  have  made  it  easy 
to  identify  the  central  chamber  as  the  sanctuary  by  build 
ing  a  hall  in  front  of  it,  some  extremely  minute  observations 
about  the  manner  of  construction  exhibited  by  various  walls 
have  made  certain  visitors  of  the  site  incline  to  think  that 
at  one  time  (before  serious  earthquakes  made  Roman 
rebuilding  necessary)  there  was  one  more  chamber  on  the 
east  side  of  the  court.  At  that  time  the  Roman  sanctuary 
(central  chamber)  was  not  the  sanctuary  but  one  of 
two  lower  wings  flanking  the  sanctuary,  which  was  part  of 
the  present  south  chamber.  Thus  we  should  have  for  the 
earlier  Greek  period  of  the  temple  a  plan  which  the  coins 
continued  to  reproduce  in  late  Roman  days,  although  earth 
quakes  and  alterations  had  sadly  interfered,  and  had  even 
made  it  necessary  to  shift  the  sanctuary.  The  oldest  walls 
belonging  to  the  temple  proper  are  all  of  them  of  an  epoch 
later  than  any  of  the  walls  south  of  the  temple,  and  con 
stituting  the  south  wing.  The  central  and  the  south 
chambers  and  the  north  portico  were  apparently  built  long 
before  the  Romans,  and  were  merely  patched  by  them. 
This  third  species  of  early  work  on  the  site — the  two  earlier 
exemplified  only  in  the  south  wing  being  supposably 
Phoenician — may  be  called  the  Greco -Phoenician  walls. 
A  common  feature  distinguishing  them  all  from  Roman 
work  is  that  there  is  no  mortar  used. 

The  temple  then  consists  of  four  main  parts  whose  exist 
ence  is  established,  and  one  missing  part  whose  existence 
is  problematical  but  may  for  the  moment  be  taken  for 
granted. 


THE  TEMPLE  AT  OLD  PAPHOS  313 

i  st.  The  Chambers  and  Entrance,  which  are  its  east 
side.  Traces  of  a  curiously  modified  Greek  work  (Greco- 
Phoenician)  coexist  with  Roman  mending  in  these  walls. 

2nd.  The  North  Portico,  which  is  the  north  side.  Here 
the  work  is  Roman  with  just  enough  traces  of  what  I  have 
called  Greco-Phoenician  to  make  it  exceedingly  improbable 
that  the  original  ground-plan  was  materially  departed  from. 

3rd.  The  entirely  Roman  Southern  Portico,  which  is 
the  south  side.  Here,  again,  dim  traces  of  an  original  build 
ing  are  visible,  and  confirm  the  notion  that  the  Romans 
did  not  seriously  innovate.  Under  the  mosaic  pavement 
here  were  found  various  fragments  of  undoubtedly  Greek 
workmanship. 

4th.  The  entirely  Roman  Hall,  by  which  the  later 
sanctuary — marked  Central  Hall  on  the  plan — was 
approached.  This  ran  parallel  to  the  two  porticoes,  passing 
through  or  near  the  middle  of  the  great  would-be  quadrangle 
or  court  of  the  temple.  Beyond  the  construction  of  this 
hall  of  approach  no  serious  departure  from  the  original 
ground-plan  was  apparently  made  by  the  Romans.  That 
their  south  portico  should  have  been  a  great  enlargement 
of  what  had  been  there  before,  and  should  have  wiped  out 
nearly  all  vestiges  of  it,  is  natural.  The  analogy  of  the 
Jewish  temple  requires  a  far  more  spacious  portico  here 
than  on  the  opposite  and  northern  side  of  the  court.  The 
temple  built  by  Herod  had  its  royal  portico  on  the  south. 
Finally,  if  there  were  no  other  reason  for  a  more  spacious 
portico  on  the  south  side  of  the  court  at  Paphos,  the  fact 
that  processions  from  New  Paphos  must  have  entered  from 
the  south  would  settle  the  question  as  a  practical  one. 

To  sum  up,  the  temple  ruins  at  Old  Paphos  have  proved 
on  examination  not  to  be  those  of  a  Grecian  temple 
similar  to  those  built  in  honour  of  the  Paphian  goddess  in 
Greece.  We  have  at  Old  Paphos  the  very  interesting  but 
unfortunately  defaced  remains  of  a  temple  resembling  in 
many  ascertainable  points  that  of  Solomon.  It  must  be 
admitted,  however,  that  a  comparison  between  two  things 


3H  APPENDIX  VI 

so  insufficiently  known  as  the  plans  of  these  two  temples 
must  of  necessity  be  somewhat  barren  of  tangible  results. 

The  remains  called  the  South  Wing,  which  have  chiefly 
excited  the  curiosity  of  all  travellers,  have  not,  I  think,  any 
thing  to  do  with  the  temple- building  as  such,  but  are  the 
remains  of  a  Phoenician  tomb,  like  the  "  Snail's  -Tower  "  at 
Amrit. 

Would  that  we  had  more  knowledge.  If  other  temples 
and  more  tombs  built  by  the  Phoenicians  or  on  the  Phoen 
ician  plan  had  been  examined  and  understood,  it  would 
be  perhaps  easy  to  know  all  manner  of  things  about  the 
remains  at  Old  Paphos.  As  it  is,  the  biblical  accounts  of 
temple -building  at  Jerusalem,  supplemented  by  Josephus 
and  his  description  of  Herod's  temple-building,  afford  only 
the  palest  side-lights — the  only  lights  available  for  illustrating 
the  temple  ruins  discovered  at  Old  Paphos.1  And,  as  for 
the  walls  of  the  South  Wing,  it  is  by  no  means  certain  that 
they  can  be  confidently  described  as  the  Tomb  of  Cinyras ; 
but  this  account  of  it  is  the  best  to  be  had,  I  am  sure. 

1  The  plan  just  discussed  results  from  the  most  thorough  and  scientific 
exploration  of  remains  existing  upon  the  traditional  site  of  the  Paphian 
temple.  Yet  it  is  quite  possible  to  explain  all  that  is  shown  upon  this 
plan  without  being  sure  that  any  part  of  it  was  the  real  temple  of  the 
Paphian  goddess.  This  uncertainty  has  been  recently  expressed  by  no 
less  an  authority  than  Mr.  F.  C.  Penrose  (new  series  of  the  Transactions 
of  the  Royal  Society  of  British  Architects,  vol.  vi.  p.  66).  Mr.  Penrose, 
after  recognising  a  possible  connection  between  this  plan  of  the  Paphian 
temple  and  that  of  Solomon's  temple,  says  :  "On  this  account  alone  the 
examination  of  the  Temple  of  Aphrodite  in  Cyprus  has  not  been  labour 
wasted.  The  finds  indeed  have  not  been  very  beautiful  architecturally, 
but  still  the  plan  is  remarkable,  and  its  use,  I  think,  has  yet  to  be  made 
out."  Unlikely  as  any  new  discoveries  near  the  site  already  excavated 
will  seem  to  any  who  know  the  spot,  their  impossibility  has  certainly  not 
been  ocularly  demonstrated.  Until  such  demonstration  shall  have  been 
given,  other  archaeologists  will  be  not  unlikely  to  share  the  uncertainty  of 
Mr.  Penrose.  This  being  the  case,  it  is  a  great  satisfaction  to  know  that 
a  project  for  eventual  excavations  at  Couclia  which  shall  make  uncertainty 
impossible  has  been  discussed.  The  best  guarantee  for  its  efficient  manage 
ment — in  case  it  should  ever  be  carried  out — is  that  those  who  have  been 
moving  in  the  matter  were  the  originators  and  prosecutors  of  the  excava 
tions  just  described. 


APPENDIX  VII 

APHRODITE   OF   THE   GREEKS,    HITTITES,   AND 
PHOENICIANS 

MANY  who  are  learned  in  Greek  have  a  pardonable  bias  in 
favour  of  a  Greek  rather  than  a  Phoenician  origin  for  so 
important  a  goddess  of  the  Greeks  as  Aphrodite.  Others 
who  have  studied  the  monuments  of  Asia  Minor,  and  found 
there  new  and  interesting  materials  about  the  "  Hittites," 
are  beginning  to  suspect  that  Paphian  Aphrodite  was 
originally  a  goddess  of  the  "  Hittites,"  and  to  maintain 
that  neither  in  her  origin  nor  in  her  main  characteristics 
was  she  of  Greek  or  of  Phoenician  derivation.  In  one 
doctrine  only  do  these  two  jarring  sects  agree — both  main 
tain  the  non-Phoenician  birth  and  breeding  of  the  goddess 
worshipped  at  Paphos  and  in  Greece  as  of  Paphos. 

Engel  and  Heffter,  who  are  anxious  to  prove  the  Greek 
origin  of  this  Greek  goddess,  have  the  following  case  : l — 
The  Homeric  poems  contain  many  allusions  to  Aphrodite, 
the  daughter  of  Zeus  and  Dione.  Now  Dione  is  the 
feminine  form  of  Zeus,  and  these  names  both  lead  back  to 
a  primitive  cult  as  far  from  Cyprus  and  the  east  as  any 
thing  on  Greek  soil  possibly  could  be.  Dodona  and  the 
extreme  north-west — not  Cyprus  and  the  east — would 
then  be  the  cradle  of  Aphrodite-worship.  It  may  be  main 
tained  that  a  knowledge  on  Homer's  part  of  Paphos  and 

1  I  am  perhaps  bound  to  say  that  the  statement  of  their  case  is  mine. 


3i6  APPENDIX  VII 

Cyprus  disappears  with  the  application  of  critical  standards 
to  the  text  of  the  Odyssey  and  the  relegation  to  a  later 
date  of  the  Aphrodite- Ares  episode  in  the  eighth  book,  at 
the  end  of  which  the  goddess  betakes  her  to  Paphos. 
Every  one  must,  I  think,  admit  that  the  oldest  portions  of 
the  Iliad  and  Odyssey  know  nothing  of  Aphrodite  as  Cypris 
or  the  Paphian  goddess.  But  to  return  to  the  vindicators 
of  the  purely  Hellenic  origin  of  our  goddess ;  Herodotus, 
they  say,  and  other  Greeks *  who  talk  confidently  of  the 
Cypro-Phoenician  beginnings  of  the  Greek  goddess,  take 
this  view  partly  from  a  perceptible  bias  in  favour  of  the 
east  as  the  place  for  all  origins,  especially  religious  ones, 
and  partly  from  a  lack  of  the  critical  point  of  view. 

This  critical  treatment  of  Homer  must  not,  however,  be 
half  applied,  and  if  it  is  conscientiously  brought  to  bear  on 
all  the  places  where  Aphrodite  is  mentioned  by  Homer,  the 
result  is  not  by  any  means  in  favour  of  her  Greek  origin. 
The  most  undeniably  primitive  portions  of  the  Iliad  and 
the  Odyssey,  the  oldest  groundwork  of  Homeric  song,  con 
tain  no  word  or  words  to  show  that  the  poets  of  primeval 
Greece  knew  at  all  either  whence  or  from  what  parentage 
Aphrodite  came.  All  the  passages  of'  Homer  where 
Aphrodite  is  called  the  daughter  of  Zeus  and  Dione,  like 
all  those  where  she  is  spoken  of  as  Cypris,  or  in  any  way 
connected  with  Paphos,  are,  by  a  remarkably  general  con 
sensus  of  otherwise  discordant  critics,  pronounced  to  belong 
to  a  comparatively  late  period  when  the  Homeric  poems 
underwent  what  may  be  called  revision  and  amplification.2 

1  For  the  various  authorities  see  note  i,  p.  284,  above. 

2  There  are  eleven  passages  of  the  Homeric  poems  where  Aphrodite  is 
mentioned  as  the  daughter  of  Zeus,  or  of  Dione.     They  occur  in  the  Third, 
Fifth,  Fourteenth,  Twentieth,   and  Twenty-first  books  of  the  Iliad,  and 
in  the  Eighth  book  of  the  Odyssey,     They  are  to  be  naturally  grouped  as 
follows  :— (i)  //.  iii.  374  ;   (2)  //.  v.  131,  312,  371,  382,  820  ;   (3)  //.  xiv. 
193,  224  ;  (4)  //.  xx.  105  ;   (5)  //.  xxi.  416  ;  and  (6)  Od.  viii.  308.    A  number 
of  these  passages  cannot  possibly  count  in  favour  of  the  greater  antiquity 
of  a  purely  Greek  Aphrodite,  since  they  associate  her  as  the  daughter  of 
Zeus  and  Dione,  not  with  Dodona  or  any  place  in  Greece,  but  by  calling 
her  Cypris,  or  otherwise,  identify  her  with  Paphos  and  Cyprus.     At  the 


APHRODITE— GREEK,  HITTITE,  PHOENICIAN     317 

As  far  then  as  the  testimony  of  the  Homeric  poems  can 
avail  to  prove  anything  about  the  earliest  Greek  knowledge 
of  Aphrodite,  it  shows  that  her  powers  and  her  charms 
were  known  long  before  her  origin  was  thought  of.  Further 
more  it  is  plain  that  when  the  question  of  her  birthplace 
and  parentage  did  arise,  it  received  a  twofold  answer.  In 
absolute  unconsciousness  of  the  contradiction  involved, 
Homer  tells  us  that  Aphrodite,  the  daughter  of  Zeus  and 
Dione,  was  named  Cypris,  and  had,  for  her  favoured  place 
of  abiding,  Paphos.  Homer  begins  by  knowing  too  little 
and  ends  by  knowing  too  much  about  the  Paphian  divinity. 

same  time  (Od.  viii.  288  and  xviii.  193)  the  Odyssey  indicates  that  this 
Paphian  goddess  was  immediately  identified  with  the  Phoenician  divinity 
of  the  island  of  Cythera  (Cerigo)  by  calling  her  Cytherea  (see  Pausanias, 
III.  xxiii.  i). 

All  the  passages— see  (2)  above — in  the  Fifth  Iliad  are  off-set  by 
others,  330,  422,  458,  760,  883,  where  this  Greek  daughter  of  Zeus  and 
Dione  is  called  Cypris,  i.e.  not  Greek  but  of  Cyprus.  Indeed  this  whole 
episode  of  the  wounding  of  Aphrodite  is  quite  as  certainly  of  late 
composition  as  is  the  tale  of  the  loves  of  Ares  and  Aphrodite  in  the 
Eighth  Odyssey,  where  Aphrodite,  daughter  of  Zeus  and  Dione — (6) 
above — claims  Cyprus  as  her  own,  and  takes  refuge  in  Paphos.  See 
lines  363  ff.  Thus  (2)  and  (6)  are  ruled  out  as  evidence  of  the  greater 
antiquity  of  a  Greek  Aphrodite  for  the  early  Greeks,  and  (i),  (3),  (4),  (5) 
remain. 

As  to  (i),  Bergk  (Litteraturgeschichte,  vol.  i.  p.  566)  declares  it  to  be  a 
late  addition.  For  many  opinions  to  the  same  effect  see  Ameis-Hentze's 
Iliad  (Anhang,  Heft  i.  pp.  159-176).  The  Fourteenth  Iliad  (3),  so  far  as 
it  concerns  us,  as  also  the  Twentieth  (4),  and  the  Twenty-first  (5),  are 
similarly  divorced  by  the  consensus  of  competent  critics  from  the  un 
doubtedly  oldest  groundwork  of  the  Iliad.  These  facts  are  the  more 
worth  considering  because  none  of  these  exclusions  was  made  with  any 
reference  to  or  knowledge  of  the  question  here  under  discussion.  See  for 
(3)  the  Fourteenth  Iliad,  Bernhardy  (Litteraturgeschichte,  pp.  164  and 
165  of  the  first  part  of  the  History  of  Greek  Poetry}.  At  p.  169  (ibid. ) 
Bernhardy  begins  his  careful  presentation  of  the  various  views  of  various 
scholars  about  the  closing  books  of  the  Iliad,  and  makes  it  tolerably  plain 
that  (4)  and  (5)  are  to  be  classed  with  (i),  (2),  (3),  and  (6),  as  belonging 
to  a  comparatively  modern  phase  of  the  Homeric  era.  In  confirmation  of 
this  view  see  Bergk  in  the  work  above  cited,  pp.  609-613  for  (3),  pp. 
633-634  for  (4),  and  p.  636  for  (5).  If  still  further  confirmation  be 
desired,  it  may  be  found  in  Ameis  -  Hentze's  Iliad  (Erlduterungcn  zu 
xiii.-xv.  pp.  45-69,  and  ibid.,  Erlauterungen  zu  xix.-xxi.  pp.  45-63, 
and  p.  95). 


318  APPENDIX  VI  1 

If  Aphrodite  was  the  daughter  of  Zeus  and  Dione,  she  was 
a  native  goddess  of  the  Greeks  ;  and  her  name  was  not 
Cypris,  nor  was  her  home  at  Paphos  or  elsewhere  in  Cyprus. 
Logic  does  not,  however,  avail  when  you  are  dealing  with 
poetry  and  religion.  Homer  and  the  other  poets  are  so 
independent  of  "  arguments  about  it  and  about  "  that,  in  a  far 
later  day,  Theocritus,  when  he  wrote  his  seventeenth  Idyll  l 
in  praise  of  Ptolemy,  spoke  of  "  The  mistress  of  Cyprus, 
daughter  revered  of  Zeus  and  Dione." 

The  most  reasonable  way  of  explaining  this  poetical 
vagueness  of  Homer  points  to  a  confirmation  of  the  usual 
theory  that  Aphrodite  was  essentially  of  Paphos  and  the 
east.  The  earliest  Homeric  presentation  of  her  divinity 
by  giving  her  no  birthplace,  and  by  silence  as  to  her 
parents,  treats  her  as  a  comparatively  unknown  power. 
Later  on,  when  her  commanding  influence  secures  attention 
to  her  antecedents,  a  native  and  aboriginally  Greek  worship 
asserts  its  claims,  which  ran  parallel  to  hers.  This  Greek 
goddess  was  no  daughter  of  Zeus  and  Dione,  but  Dione 
herself.  Dione  was  apparently  for  the  early  Greeks  what 
Freya  was  to  the  Germans,  and  what  Venus  was  to  the 
early  Romans.  Dione2  was  worshipped  by  the  side  of 
Zeus  at  Dodona.  Euripides  —  fragment  177,  from  the  lost 
play  of  Antigone  —  apostrophises  Dionysus  as  the  son  of 
Dione,  and  thus  makes  her  the  same  with  Thyone-Semele, 
originally  an  earth-goddess.  All  this  goes  to  show  what 
were  the  most  ancient  Greek  associations,  surviving  vaguely 
in  all  pious  minds  of  Greece,  which  would  sometimes  lay 
hold  upon  the  smiling  goddess  of  the  east.  These  associa 
tions  have  an  undeniable  solemnity  and  a  real  dignity 
which  are  often  lacking  in  the  more  eastern  aspect  of  the 
goddess,  and  they  doubtless  played  a  part  in  the  ennobling 
transformation  which  that  goddess  underwent  at  the  hands 
of  Hellas  ;  but  the  fact  remains  that  Dione  of  the  early 
Greeks  was  very  completely  overshadowed  and  supplanted 


1  Line  36,  KvTrpov  exoiffa  Aiuwas  TTOTVIOL 
2  See  Plew's  revision  of  Preller's  Greek  Mythology,  p.  271. 


APHRODITE— GREEK,  HITTITE,  PHOENICIAN     319 

by  Cypris,  the  goddess  of  Paphos,  and  through  Paphos  of 
Phoenicia  and  Assyria. 

The  first  victory  won  by  Cypris  was  in  the  matter  of  the 
name.  Where  the  case  is  of  conflicting  divinities,  every 
thing  is  in  a  name,  and  the  name  of  Dione  was  practically 
driven  out  by  Cypris-Aphrodite.  Whatever  may  be  the 
derivation  of  Aphrodite,  it  is  not  Greek,  and  Cypris  speaks 
for  itself,  in  spite  of  the  forced  attempt  of  an  over-ingenious 
Homeric  commentator  to  derive  it  from  a  Greek  word l 
meaning  to  be  pregnant.  Then,  with  this  prevalence  of  the 
new  name  partly  as  the  cause  and  partly  as  the  effect  of  it, 
came  the  increasing  renown  of  the  great  places  of  the 
eastern  worship  of  the  goddess.  Paphos  and  Cythera, 
both  of  them,  played  a  remarkable  role  in  the  development 
of  early  commerce  on  the  Mediterranean.  No  wonder 
then  if  the  early  Greek  poets  fancied  that  the  real  divinity 
was  Aphrodite-Cypris,  and  not  Dione  of  Dodona.  No 
wonder  if  the  old  Greek  divinity  merely  retained  sufficient 
hold  upon  the  imaginations  of  Greek-speaking  men  to  live 
on  as  mother  of  the  goddess.  A  conflict  between  Do- 
donaean  Dione  and  Aphrodite-Cypris,  where  the  latter  was 
victorious,  is  the  sufficient  explanation  of  the  anomalies  in 
Homer's  account  of  Aphrodite. 

If  this  account  of  the  overshadowing  of  the  native  Greek 
figure  of  Dione  -  Aphrodite  by  the  Cypris  -  Aphrodite  of 
Cythera  and  of  Paphos  is  deemed  convincing,  and  if  it  be 
admitted  that,  so  far  as  her  Greek  worshippers  were  con 
cerned,  Cypris-Aphrodite  was  above  and  before  all  the 
Paphian  goddess,  then  the  question  of  her  origin  at  Paphos 
must  be  reopened.  Some  of  those  who  are  especially 
entitled  to  a  respectful  hearing  upon  any  question  where 
Cyprus  is  concerned  are  inclined  to  believe  that  the  pre 
ponderance  in  earliest  times,  at  Paphos  and  in  Cyprus,  of 
Phoenician  influence  has  been  so  seriously  overstated  as  to 
obscure  the  really  characteristic  and  original  strain  of  the 
Paphian  goddess,  who  will,  they  think,  be  eventually  recog- 


320  APPENDIX  VII 

nised  as  the  Paphian  and  Cyprian  modification  of  a  divinity 
of  Asia  Minor.  The  dissemination  of  the  cult  of  this 
goddess  they  attribute  to  a  conquering  race  known  hitherto 
only  by  name,  and  chiefly  from  the  annals  of  Canaan  and 
Egypt.  This  is  Aphrodite  of  the  "  Hittites." 

In  the  chapter  above  on  Paphian  Aphrodite  room  has 
been  made  in  the  process  of  reshaping  submitted  to  by  the 
Phoenician  Ashtaroth-Astarte  for  a  noteworthy  element 
from  Asia  Minor  added  at  Paphos  to  the  goddess's  character. 
This  Paphian  and  non-Phoenician  strain  may  conveniently 
be  called  by  the  name  of  the  "  Hittites."  It  certainly 
constitutes  an  important  difference  between  Cypris-Aphro- 
dite  and  Ashtaroth,  but  for  all  that  I  do  not  find  myself 
convinced  of  its  all-importance.  I  am  still  convinced  that 
Phoenician  influence  was  the  "tone-giving"  one  at  the  temple 
of  Aphrodite  at  Old  Paphos,  as  it  evidently  was  at  Cythera 
and  at  Eryx.  In  order,  therefore,  that  the  last  great  light 
which  shines  upon  the  Paphian  goddess  from  a  rising 
interest  now  felt  in  the  new-found  "  Hittite "  theory  may 
not  be  dimmed  by  prepossessions  of  my  own,  I  have  asked 
and  obtained  the  permission  of  my  friend,  Mr.  D.  G. 
Hogarth,  to  print  portions  of  a  private  letter  which  he  was 
good  enough  to  address  to  me.  Little  as  I  think  anything 
which  one  of  his  competent  knowledge  and  especial  research 
may  say  on  this  subject  requires  either  explanation  or 
apology,  I  am  perhaps  bound  to  quote  his  remark  that 
"  future  research  may  easily  upset "  various  confirmations 
of  the  view  which  he  so  acutely  defends,  and  to  repeat  his 
observation  that  the  knowledge  of  the  "  Hittites  "  is  only 
at  its  beginnings.  The  following,  then,  is  his  account  of 
Aphrodite  of  the  "  Hittites,"  communicated  to  me  in  con 
nection  with  my  attempt  to  criticise  a  note  in  his  Devia 
Cypria^  a  book  which  I  have  always  used  with  advantage. 

"  Now,  as  to  the  general  question  of  my  note  in  Devia 
Cypria,  p.  26,  the  Phoenician  origin  of  Aphrodite,  and  the 
part  played  by  Phoenicians  in  Cyprus,  I  may  as  well  state 

1  See  above,  note  i,  p.  282. 


APHRODITE— GREEK,  HITTITE,  PHOENICIAN     321 

my  own  theory  for  what  it  is  worth.  My  idea  in  writing 
that  note  was  that  Aphrodite  is  the  Asiatic  goddess,  derived 
from  Asia  Minor,  where  she  was  universally  worshipped  at 
an  early  period — not  from  Phoenicia,  though  I  know  that 
Ashtaroth  is  only  another  avatar  of  the  same  divinity.  It 
seems  to  me  a  very  significant  fact  that  the  Cypriote 
Syllabary  is  entirely  non-Phoenician  ;  it  is  also  so  clumsy  a 
vehicle  for  a  Greek  language  that  it  must  have  been  rooted 
before  Phoenician  was  brought  into  contact  with  the 
islanders.  If  not,  it  is  difficult  to  understand  either  how  it 
was  not  supplanted  by  the  more  convenient  script,  or  how, 
on  the  other  hand,  it  could  itself  have  supplanted  the  latter. 
This  postulates,  therefore,  an  ancient  pre-Phoenician  popula 
tion  in  Cyprus.  Whom  did  they  worship?  In  early  Cypriote 
inscriptions  '  Aphrodite '  does  not  occur,  but  r)  Fdvaa-a-a 
and  r)  Ilafaja  do  occur.  Is  she  the  original  goddess,  only 
called  Aphrodite  much  later  ? 

"If  the  date  of  Cypriote  texts  were  better  assured,  I 
believe  that  it  could  be  proved  that  17  fai/ao-o-a  is  pre- 
Phoenician.  It  is  not  yet  proved,  but  it  is  probable,  that 
the  Cypriote  script  comes  from  the  '  Hittites ' ;  it  was 
employed  by  a  people  speaking  a  language  in  which  there 
were  no  Semitic  elements.  Why,  then,  should  the  great 
goddess  of  that  people  be  Semitic  ? 

"You  say  there  is  no  positive  evidence  against  Phoenicians 
being  the  originators  of  the  goddess  and  the  main  influence 
in  the  island.  Certainly  I  can  adduce  nothing  more  con 
vincing  than  the  above  considerations.  I  find  it  hard  to 
believe  that  a  people  who  left  so  little  influence  on  the 
speech  of  an  island  originated  its  chief  worship,  and  I 
point,  rather  as  confirmation  than  anything  else,  to  the 
scantiness  and  local  distribution  of  Phoenician  remains  in 
Cyprus.  I  know  that  future  research  may  easily  upset 
that  confirmation  ;  but  at  present  the  fact  that  only  Citium, 
where  there  was  an  independent  Phoenician  kingdom  in 
late  times,  and  Idalium,  which  depended  on  it,  have 
yielded  anything  of  importance,  is  remarkable.  By  the  way, 

Y 


322  APPENDIX  VII 

have  you  noticed  that  at  the  latter  there  has  been  found 
no  dedication  to  or  mention  of  Aphrodite,  whereas  Athena 
is  common  ? 

"  Neither  Herodotus  nor  Homer  can  furnish  very  strong 
evidence  for  events  previous  to  their  time.  I  take  it  that 
'  Hittites '  never  came  in  contact  with  western  Greeks,  and 
were  always  unknown  to  the  literature  of  the  latter ;  what 
they  did  directly  or  indirectly  in  Cyprus  would  naturally  be 
ascribed  to  Phoenicians.  You  say  yourself,  and  most 
truly,  that  Cinyras  is  more  like  Attis  than  Tamuz. 

"  This  is  a  crude  statement  of  the  idea  which  led  me  to 
write  the  note  to  which  you  refer.  There  is,  I  admit, 
another  theory  which  would  account  for  some  of  the  facts, 
though  it  will  not  explain  the  absence  of  Phoenician  remains, 
namely,  that  the  immigration  of  a  people  from  Asia  Minor 
at  an  early  period  drove  out  Phoenicians  who  had  peopled 
the  island  previously.  It  is  not  hard  to  believe  that  such 
Asiatic  immigrants,  being  worshippers  of  the  great  Nature 
goddess,  would  identify  her  with  their  predecessor's  Ash- 
taroth,  who  would  accordingly  survive  in  Cyprus  while  the 
other  gods  returned  to  Sidon  and  Tyre.  Some  day  I  hope 
to  discuss  this  problem  more  fully,  but  do  not  see  how 
either  theory  is  to  be  established  unless  a  date  can  be  found 
for  the  earliest  Cypriote  inscriptions.  When  we  found  such 
complete  incised  '  Hittite  '  texts  this  past  summer  in  Cappa- 
docia,  I  thought  that  we  might  be  able  to  show  a  fairly 
close  relationship  between  their  conventionalised  symbols  and 
the  Cypriote ;  but  there  is  no  more  resemblance  than  in 
the  case  of  the  well-known  symbols  in  relief.  True  that 
the  number  of  incised  *  Hittite '  inscriptions  known  at 
present  is  very  small ;  still  it  cannot  be  said  that  the  pro 
spect  of  finding  a  terminus  a  quo  for  Cypriote  by  comparing 
these  (probably)  late  examples  of  '  Hittite '  is  particularly 
bright  at  present. 

"  By  the  way,  a  propos  of  what  you  say  about  a  Thracian 
wave  on  page  280,  Ramsay  has  shown  conclusively  now  that 
this  wave  was  that  which  introduced  a  father  god  ;  Cyprus 


APHRODITE— GREEK,  HITTITE,  PHOENICIAN      323 

owns  a  mother,  and  her  aboriginal  must  belong  to  the 
earlier  time." 

After  all,  the  above  account  of  the  aboriginal  mother 
goddess  of  Cyprus  differs  from  my  own  only  in  the  com 
parative  importance  attached  to  the  main  elements — of 
Phoenician  and  of  "  Hittite "  origin — in  the  Paphian 
goddess  as  she  conceivably  existed  at  Paphos  before  the 
first  advent  of  the  Greeks.  Chiefly  to  Mr.  Hogarth,  but 
somewhat  also  to  Engel,  I  owe  the  conviction  that  there 
was  an  aboriginal  pre-Phoenician  goddess  at  Paphos.  I 
would  like  to  persuade  him  that  the  belief  of  Herodotus 
and  others,  who  had  some  facts  before  them  which  are 
inaccessible  to  us,  that  the  Phoenicians  did  visit  and  abide 
at  Paphos,  was  well  founded,  and  cannot  possibly  be  due 
to  a  confusion  between  "  Hittites "  and  Phoenicians. 
Furthermore,  I  would  urge  upon  him  that  these  Phoenicians 
so  transformed  the  Mother  Queen  of  the  "  Hittites  "  that 
the  resulting  Paphian  divinity  found  at  Paphos,  and 
adopted  by  the  early  Greeks,  was  Phoenician  rather  than 
"  Hittite." 

Any  other  view  of  the  case  makes  it  difficult  to  under 
stand  the  legend  of  the  founding  from  Eryx,  i.e.  by  the 
Sicilian  Phoenicians,  of  a  temple  of  Aphrodite  at  Psophis, 
in  Arcadia.  It  is  also  difficult  to  explain  upon  any  jother 
theory  the  early  bestowal  upon  Aphrodite  of  the  name 
Cytherea  derived  from  a  worship  established  at  Cythera  by 
purple-fishing  Phoenicians.  And  yet  this  epithet  seems  to 
be  about  as  old  as  that  of  Cypris. 

In  fact,  if  Paphian  Aphrodite  was  of  the  "  Hittites," 
when  Greeks  first  found  her  at  Paphos,  then  she  "  suffered 
a  sea-change  "  on  her  way  to  Greece,  which  left  her  indis 
tinguishable,  so  far  as  Greek  worshippers  were  concerned, 
from  the  Greco -Phoenician  goddess  on  Cythera  and  at 
Eryx. 


APPENDIX   VIII 

THE  OLYMPUS  AND  THE  BOCARUS  IN  CYPRUS 
— HETTORE  PODOCATHARO  AND  JOHN 
MEURSIUS 

THE  most  minute  and  comparatively  unimportant  questions 
require  sometimes  an  unusual  expenditure  of  time,  and 
what  seems  a  vast  deal  of  talk  about  nothing  at  all.  This 
general  truth  will,  I  fear,  receive  illustration  in  what  follows. 
If  it  were  possible  without  controversy  to  call  Mount 
Troodos — the  highest  mountain  of  Cyprus — Olympus, 
the  "mons  mamillae  similis  "  of  certain  well-known  old1 
maps  of  Cyprus,  and  one  or  two  unknown  or  hardly  known 
modern 2  ones ;  and  if  I  could  ignore  the  trivial  fact  that 
a  blunder  of  the  elder  John  Meursius  has  imported  the 
river-name  Bocarus  from  the  Athenian  island  of  Salamis, 

1  See  the  finest  of  Cardinal  Bessarion's  MSS. ,  a  copy  of  the  work  of 
Ptolemy,  whose  original  goes  back,  they  say,  to  the  twelfth  or  thirteenth 
century.      However  that  may  be,   Bessarion's  copy  now  in  the  Marcian 
Library  has  a  beautiful  map  of  Cyprus, — somewhat  misshapen, — where 
Olympus  figures.     A.   Ortelius  published  in   1573  a  map  entitled  "  Cypri 
Insulae  Nova  Descript."     Here  Troodos  does  not  appear  at  all  (Trodisi 
occurs  as  the  name  of  a  village),  although  the  map  is  for  modern  purposes 
and  contains  only  modern  geographical  names,  i.e.  names  used  by  the 
Lusignans  and  Venetians.       Paphos  for  instance  is  marked  Baffo,   olim 
Paphos. 

2  Chypre :  Histoire  et  Geographic,  par  le  Marquis  de  Sassenay  :  Paris 
1878.     On  the  excellent  map  here  published  and  originally  made  for  the 
"  Revue  de  Ge"ographie, "  the  Troodos  is  rightly  identified  as  the  ancient 
Olympus. 


OLYMPUS  AND  BOCARUS  IN  CYPRUS  325 

where  it  belongs,  into  Cyprus  where  it  has  no  place, — then 
the  tiresome  minutiae  of  this  appendix  might  be  dispensed 
with.  But  I  have  already  been  asked  by  the  kindest  and 
most  competent  of  critics  what  is  my  authority  for  saying 
that  the  Troodos,  and  therefore  not  Mount  Santa  Croce, 
or  Delia  Croce  (Stavrovuni),  was  anciently  called  Olympus. 
The  question  is  justified;  for  the  best  accredited  maps — that 
of  Major  (then  Captain)  Kitchener  and  those  of  Kiepert — 
identify  the  ancient  Olympus  in  question  with  Mount 
Santa  Croce  (Stavrovuni),  and  leave  us  to  believe 
that  Strabo  had  never  heard  of  the  only  remarkable 
mountain  in  Cyprus.  To  answer  this  question  and  to 
protect  my  account  of  the  shrine  of  Aphrodite  at  Old 
Paphos  from  the  reproach  of  containing  no  mention  of  the 
wonderful  river  Bocarus  of  one  hundred  mouths,  the 
following  description  of  blunders  within  blunders,  of  con 
fusions  ancient  and  modern,  is  offered  to  such  as  may  be 
minded  to  read  it. 

The  question  whether  the  ancients  applied  the  name 
Olympus  to  Mount  Troodos  or  Mount  Santa  Croce  in 
Cyprus  is  eventually  settled,  I  think — first,  by  Strabo's  epithet 
of  /xao-roetSes,  which  is  absurd  when  applied  to  Stavrovuni 
and  exactly  suits  Troodos  ;  and  secondly,  by  a  good  MS.  of 
Strabo.1  The  restoration  of  a  period  there  found,  in  place 

1  The  ordinarily  accepted  text  (Teubner)  of  Meineke  for  this  crucial 
passage,  of  Strabo,  book  xiv.  p.  683,  is  as  follows, — he  has  just  mentioned 
KLTLOV  (Larnaca)  :  elr'  'A/u,a0oOs  7r6Xis  Kal  fj,era^v  iro\Lxvf]  llaXcua,  Ka\ov- 
fj.tvr),  Kal  6pos  /macTToeidts  "OXi'/iTros"  elra  Kouptds  (&Kpa)  ^ppovrja^rjs. 
.  .  .  Thus  Olympus  would  be  Stavrovuni  or  Mount  Santa  Croce,  which 
lies  just  inland  from  the  place  assigned  by  Strabo  to  Palaea  between 
Larnaca  and  Limassol.  The  punctuation  of  the  same  passage  which  I 
propose  to  adopt  is  that  of  the  Marcian  Strabo  catalogued  in  Venice  as 
"Cod.  378  (Arm.  Ixviii.  Th.  xc.)  in  4  minori  membr.  fol.  43;  Saeculi 
circiter  xi. "  elr'  'Apadovs  7r6\ts'  Kal  /xera^i)  iro\ixvn  HaXcua  KaXov^vrj. 
Kat  #pos  /xatrroeiSes  "QXv/juros'  eira  Koi>/>«x?  xeppovTjcrudrjs.  .  .  .  This 
dissociates  Palaia — which  is  mentioned,  as  it  were,  parenthetically — from 
Olympus,  and  puts  the  latter  between  Amathus  (Limassol)  and  Curias. 
This  description  exactly  applies  to  the  Troodos,  since  Strabo  thought  of 
the  coast  directly  after  Curias  as  trending  northward,  see  lower  down  on 
the  same  p.  683  :  apx^]  8'oiV  roO  dvff/JiLKov  TrapdirXov  TO  Kovpiov, — Curium 
is  the  town  just  beyond  Curias  from  Limassol  (Amathus), — TOV 


326  APPENDIX  VIII 

of  the  comma  substituted  for  it  in  the  generally  accepted 
text,  settles  the  question.     Still  the  present  punctuation  of 
Strabo's  text  is  of  very  ancient  date,  since,  as  we  shall  see, 
Eustathius    read    it   as    Meineke   does.       It   will    be  well 
therefore  to  go  into  many  details  and  fully  to  understand 
how  such  a  conflict  of  opinions  has   gathered  around   so 
simple    a    matter    as    the    identification    of    this    Cypriote 
Olympus.      Above  all  we  must  remember  that  the   same 
name  of  Olympus  is  also  applied  by  Strabo  to  a  headland- 
promontory  at   the   extremest   north-eastern   point   of  the 
island.1     About  the  situation  of  this,  the  lesser  Olympus  of 
Cyprus,  there  are  not  two  opinions.      Now  I  will  pass  in 
review  some  of  the  accounts  given  at  various  more  or  less 
modem   times  of  the  greater   Olympus.       No  more  com 
manding  authority  than  that  of  His   Holiness  Pope  Pius 
the  Second  can  be  appealed  to   in  the  fifteenth  century. 
In  the  chapter  on  Asia  Minor  of  his  Opera  Geographica  et 
Historical  he   translates   from   Strabo   the  passage   where 
Olympus  is  mentioned, — he  does  not  speak  of  Strabo,  but 
simply  adopts  the  passage  as  his  own.      Having  mentioned 
Citium   [Larnaca],   he  writes  :    "  Deinde   Amathus  civitas, 
et  intermedio  spatio  oppidum   nomine  Palae  et  Olympus 
mons    qui    mamillae    speciem   praebuit    et    Curias    penin- 
sulam,   et  Curium  urbs."     Thus  the  Pope  Aeneas  Sylvius 
de'   Piccolomini  plainly  ignored   the  highest   mountain  of 
Cyprus  and  applied  the  name  Olympus   to  Mount   Santa 
Croce ;    or  did   he  simply  follow  the   punctuation  of   the 

irpbs  'PoSov.  ...  It  is  impossible,  I  think,  to  suppose  with  the  ingenious 
d'Anville  (Mtmoires  de  I 'Institut,  Ancienne  Se"rie,  vol.  xxxii.  p.  259)  in  his 
Recherches  Gdographiques  sur  lile  de  Chypre,  that  Strabo  meant  both 
Stavrovuni  and  Troodos  by  his  Olympus. 

1  Thus  the  name  Olympus  rightfully  occurs  twice  on  maps  of  ancient 
Cyprus.      Ortelius  in  his  ancient  map  (1580)  brings  it  in  thrice,   and  so 
does  M.  le  Comte  de  Mas  Latrie  on  a  map  published  in  1862. 

2  Pius  II.  is  not  speaking  from   personal  observation,  as  in  his  Coin- 
mentarii,  where  he  so  graphically  describes  the  situation  of  Monte  Oliveto 
and  the  Alban  Hills.      Cyprus,  throughout  the  fifteenth  century,  was  little 
visited,  and  indeed  even  in  Venice  was  regarded  as  a  place  of  banishment, 
as  was  shown  by  the  banishment  of  two  inconvenient  nobles  to  Nicosia  in 
1492. 


OLYMPUS  AND  BOCARUS  IN  CYPRUS  327 

Strabo  which  lay  open  before  him  ? l  Pius  the  Second 
wrote  the  above  while  he  was  pope  (i458-i464).2  Let  us 
now  appeal  from  him  to  the  German  Ziegler  (1532). 
Ziegler  simply  places  Olympus  in  the  southern  part  of 
the  island :  "  Orientalia  .  .  .  Salaminia  comprehendunt, 
Occidentalia  Paphia,  Meridionalia  Amathusia  et  Olympus 
mons,  Septentrionalia  Lapathea."  Now  this  unfortunately 
is  very  non-committal,  but  the  difference  is  very  appreciable 
between  a  statement  that  Olympus  is  between  Citium 
(Larnaca)  and  Amathus  (Limassol),  and  that  it  lies  in  the 
district  of  Amathus,  with  the  district  of  Lapethus  to  the 
north.  The  latter  description  of  the  mountain's  where 
abouts  almost  inevitably  attaches  the  name  Olympus  to  the 
Troodos  which  would  to-day  be  described  as  in  the  district 
of  Limassol,  since  the  road  leading  to  its  summit  begins 
there.  But  after  all  Ziegler  is  only  translating  the  well- 
known  passage  in  Ptolemy,  as  Aeneas  Sylvius  translated 
Strabo.  Ptolemy,  as  explained  by  the  traditional  map  in 
the  best  MSS.,  is  decidedly  though  not  decisively  in  favour 
of  identifying  the  ancient  Mount  Olympus  with  the 
Troodos.  What  is  really  needed  is  not  an  accumulation 
of  slightly  varying  translations  of  two  well-known  passages 
in  Strabo  and  Ptolemy  respectively,  but  if  possible  some 
trace  of  the  local  survival  in  Cyprus  of  the  name  Olympus. 
So  far  as  ancient  testimony  leaves  the  matter  in  doubt, 
this  sort  of  evidence  has  a  decisive  value;  particularly  when 
the  great  mountain  of  all  Cyprus  is  involved.  Traditions 
about  smaller  things  are  easily  changed  and  shifted. 

If  tradition  of  this  kind  exists,  its  traces  will  be  found 
in  the  utterances  of  a  cloud  of  witnesses  belonging  to  the 
sixteenth  century.  These  men  knew  Cyprus  at  the  time 

1  The  same  mistake,   if  mistake  there  be,  was  made  by  Eustathius  on 
//.  i.   1 8  :   eiffl  5e  Kal  erepot  "O\i>/X7rot'  tv  re  yap  \\e\oirovvricrq,  ws  KOLL  tv 
rots  TOU  TrepL7)yr)Tov  eypd(prj,  Kal  ev  Ki^Trpy  5£  6pos  yuacrroetS^s  /J-era^v  Km'ou 
Kal  'A.[j,adouvTos  ''OXv/u.7ros    Xtyerai'    Kal    a\\-r)    5e  ns   aVpcipeia   Kttarpou 
e/caXe?ro  oiirus  ev  y  vaos  'Axpaias  'AffipodirTjs,  advros  yvvai^i. 

2  I  quote  from  the  Helmstadt  republication  by  Joh.  Melchior   Guster- 
mann,  1690. 


328  APPENDIX   VIII 

when    that    island    was    more    nearly    in    the    centre    of 
European  affairs  than  ever  before  or  since.      Many  of  them 
had  an  especial   hold   upon  Venice   and  Cyprus   as  well. 
Like  the  Athenians  of  old,  the  Venetians  in  the  sixteenth 
century,    having    failed    to    acquire   any   foothold    on    the 
mainland  which  was  at  all  commensurate  to  their  maritime 
and  commercial  importance,  were  forced  by  circumstances 
to  make  much  of  their  islands,  and  the  popular  interest  in 
these  possessions  gave  rise  to  a  new  sort  of  geographical 
literature  in  the  numerous  books  published  about  islands 
exclusively.      That  the   Venetian  public  laid   stress   upon 
accuracy  in  the  depiction  of  any  scene  where  their  island 
domain  was  involved  is  curiously  exemplified  by  the  dim 
outline  of  the  mediaeval  Acropolis  of  Naxia  (the  capital  of 
Venetian   Naxos)   which    adorns    the    middle    distance   of 
one    of  the    most    beautiful    of   all    Venetian    pictures — 
Tintoretto's    "Marriage  of   Ariadne   and  Bacchus."     Tin 
toretto  was   born,    it   will   be   remembered,   in    1518,   and 
lived   until    1594,    and    therefore    the    various    "Isolarii" 
of   the   century  were    the   new   books   of  his   day.       The 
earliest  of  these  which  I  have  consulted  is  that  of  Bene 
detto    Bordone    of     Padua.        This     curious    book    was 
published     under    a    quaint    and    apparently    irrelevant x 
authorisation  from  Leo  the  Tenth  (Giovanni  de'  Medici). 
"  Da.  Rome,  apud  Sanctum  Petrum  sub  annulo  piscatoris. 
die  V.  Junii   1521   Pont.  Nostri  Anno  Nono."     But  it  is 
evident    that    this    imprimatur    was    not    enough,    and    a 
petition  had  to  be  addressed  to  the  Venetian  Signoria  as 
follows :  "Benedetto  Bordone  miniator  compare  humilmente 
davanti  a  le  Signorie  vostre   narrando  cum   sit  che  molti 
anni  si  habbi  faticato  di  &  notte  in  coponere  uno  libro, 
nel  quale  si  tratta  de  tutte  1'  Isole  del  mondo  si  antiche 
come  etia  moderne  &c.   .   .   ."     This  narration  secured  a 
monopoly  of  the  sale  for  ten  years  from  the  6th  of  March 
1526.       The   date  of   Bordone's   Isolario  was   1508:    see 

1  The  title  of  Bordone's  book  does  not  appear  among  those  recited  in 
the  document,  and  his  book  is  not  apparently  authorised  by  it. 


OLYMPUS  AND  BOCARUS  IN  CYPRUS  329 

Horatio  Brown's  Venetian  Printing  Press  (1891)  p.  103. 
Bordone  thought  himself  learned  in  books,  and  gives  much 
space  to  a  remarkably  confusing  disquisition  on  La  Musica, 
the  "terza  sorella"  of  Astrologia,  Of  this  last  he  says  "e 
nell  numero  delle  Muse,  adonque  ell'  e  vera."  His  real 
claim  to  a  hearing,  however,  is  that  he  has  travelled 
himself,1  or  at  least  that  he  has  talked  with  others  who 
have  travelled,  and  can  be  depended  upon  to  give  a  fresh 
and  unbookish  account  of  Cyprus  or  any  one  of  his  "isole 
si  antiche  come  etia  moderne."  Of  the  art  of  drawing  an 
accurate  map  Bordone  was  not  a  master,  since  he  was 
almost  the  inventor  of  woodcutting  for  maps.2  For  this 
reason  no  doubt,  speaking  in  the  third  person,  he  tells  the 
Signoria  of  "  le  sue  tante  fatiche  .  .  .  volendo  quelle  far 
imprimere  di  molte  spese  si  nel  stampare,  come  anchor  nel 
far  tagliar  la  forma  di  ciaschuna  Isola  come  essa  sta.  ..." 
However,  for  the  purpose  in  hand,  Bordone's  sketch  map 
is  all-sufficient.  He  places  "la  croce,"  the  equiva 
lent  of  Mount  Santa  Croce  (Stavrovuni),  close  to  the 
southern  shore  of  the  island  with  "  il  chito "  (Larnaca- 
Citium)  close  at  hand.  He  finds  no  way  of  indicating 
Mount  Olympus  on  this  map,  but  that  is  readily  ex 
plained  when  we  read  what  he  says  of  Olympus  in  the 
text  of  his  work.  Speaking  of  Cyprus,  he  there  says : 
"&  ha  nel  mezzo  il  monte  olimpo."  Bordone  was  so 
completely  of  his  day  and  generation  that  his  idea  of  a 
map  did  not  contemplate  the  possibility  of  anything  more 
complete  than  are  those  rude  sailors'  charts  called  Portulani 
of  which  the  Marcian  Library  and  the  Museo  Correr  at 
Venice  are  full.3  Olympus,  then,  was  in  the  centre  of  the 

1  This  is  shown  by  the  letter  of  dedication  :  "  allo  eccellente  chirurg- 
ico  Meser.  Baldassaro  Bordone,  Nepote  Suo,  accetti  queste  nostre  fatiche 
.  .  .  cagion  potrano  esser,  che  alchuno  pellegrino  ingegno  la  strada 
dinanzi  fatta  vedendosi "  etc.,  etc.,  the  passage  becomes  hopelessly 
muddled,  but  seems  to  imply  some  travelling  on  Bordone's  part. 

2  See  Brown,   The  Venetian  Printing  Press,  p.  103. 

3  These  never  give  any  features  of  the  interior  ;  frequently  the  space 
where  such   details  would  come   is   taken  up  with  pictures   for   the  de 
lectation  of  the  jolly  sailor's  eye.     See  Portulan  No.  9  (in  the  Marcian 


330  APPENDIX  VIII 

island,  and  was  not  the  well-known  mountain  on  the  coast 
where  every  Venetian  sailor  knew  the  monastery  and 
church  of  the  Holy  Cross.  Bordone  emphasises  this 
familiar  fact  on  his  map  by  drawing  a  picture  of  the 
church  and  the  mountain,  making  the  church  just  a  little 
larger  than  the  mountain.  The  whole  impression  left  by 
the  excellent  Bordone  is  that  of  a  sailor  telling  of  his 
travels  after  they  are  over  in  language  so  artless  that  it  is 
often  absolutely  unintelligible.  An  Isolario  exists  from 
which  Bordone  undoubtedly  took  the  general  idea  of  his 
book.  It  is  that  of  the  sea  captain  Bartolommeo  who  names 
himself  "da  li  Sonetti"  in  his  book  of  geographical  sonnets 
on  the  islands  of  the  Aegean  Sea,  which  he  dedicated  to  the 
doge  Giovanni  Mocenigo,  and  published  shortly  after  isoo.1 
That  is  about  eight  years  before  Bordone's  book  of  the  same 
name. 

But  this  form  of  geographical-historical  literature,  the 
peculiar  invention  of  the  Venice  of  the  sixteenth  century, 
did  not  perish  with  these  rude  first  attempts.  Its 
perfection  was  achieved  just  while  the  Turk  was  out 
rageously  hurling  himself  on  the  island  empire  of  Venice. 
The  first  edition  of  L'lsole  piu  famose  del  mondo,  by 
Tommaso  Procacci  da  Cosiglione,  was  published  in  1571, 
and  the  author  prefaces  his  account  of  Cyprus  with  the 
following  allusion  to  a  work  on  that  island  which  he  had 
seen  in  MS. :  "  Nel  descriuer  la  nobilissima  e  famosissima 
Isola  di  Cipro ;  io  faro  piu  breve  che  la  grandezza  & 
gloria  sua  merita ;  non  perche  le  cose  non  siano  molte  in 
numero  ;  ma  perche  essendo  stata  fatta  questa  descrittione 
auanti  a  me  dall'  illustre  &  uirtuosissimo  Signore  Hettore 

Library)  by  Battista  Agnese.  There  a  figure  labelled  ' '  Philipus  Rex 
Hispaniae  Rex  Angliae "  is  throned  in  mid-Spain,  while,  somewhat 
inconsistently,  England  is  adorned  with  the  "  Regina  Angliae"  who  sits 
bolt  upright  and  holds  a  sword.  In  Italy  ' '  Pontifes  Ivlivs  Tercius " 
holds  up  the  cross,  while  over  against  him  in  Bosnia  is  the  terrible 
"  Suleymanssach  imperator  turcorum."  The  date  of  this  sailor's  chart 
and  picture-book  is  1553. 

i  The  date  of  publication  is  apparently  very  doubtful,  1477  and  1485 
are  both  given  by  competent  authorities. 


OLYMPUS  AND  BOCARUS  IN  CYPRUS  331 

Podocatharo,1  cauallier  di  quel  Regno,  &  non  essendo 
anchora  stata  data  in  luce,2  hauendola  io  per  cortesia  di 
quell'  honorato  &  cortese  signore  letta,  &  ueduta  tutta ; 
non  e  honor  mio,  ne  creanza  di  nobile  spirito  far  torto  a 
quel  magnanimo  gentil'  huomo  a  cui  son  grandemente 
obligato.  Perb  coloro  che  al  presente  legeranno  questa 
descrittion  da  me  fatta,  sappianno  che  io  toccherb  som- 
mariamente  alcuni  soli  passi  piu  important!,  &  del  resto 
aspettino  di  douer  da  quel  libro,  che  il  Signor  Hettore 
Podocatharo  chiama  Ritratte  del  Regno  di  Cipro,  ueder 
pienamente,  e  in  giudizioso  stile,  quanto  a  questo  proposito 
appartenga." 

Procacci's  first  edition  came,  as  above  remarked,  in 
1571  ;  one  of  the  publishers,  Simone  Galignani  by  name, 
dedicated  it  to  Don  John  of  Austria,  who  was  in  a  few 
short  months  to  win  the  glorious  victory  of  Lepanto,  for 
which  Venice  is  still  full  of  his  praises.  But  in  spite  of 
this  rebuff  the  unmentionable  Turk  pressed  forward  still,  and 
before  Procacci's  second  edition  appeared  Cyprus  had,  by 
process  of  fire  and  sword,  and  by  murderous  treachery,  been 
torn  violently  away  from  the  European  empire  of  Venice,  just 

1  Hettore  Podocatharo  was  a  man  of  some  wit.     Cicogna  speaks  of 
his  praises    in  Ludovico  Domenichi's   Facezia,   Venice,    1574.      Procacci 
dedicated  to  Hettore  the  first  volume  of  his  Cagione  delle  Guerre  Antiche, 
see  Cicogna' s  her.    Venet.   p.    142.     The  name  Podocatharo  (Lightfoot) 
is  variously  spelled  in  old  documents,  but  Procacci's  orthography  is  correct, 
for   Cardinal   Ludovico   Podocatharo   has   adopted   it  on  the  illuminated 
page  which  marks  one  of  the  treasures  of  the  Biiliotheque  Nationale  as 
having   been   originally  his.       He  bequeathed    the    book    to   the    Pope's 
Library.      How  it  came  to  be  in  Paris  is  shown  by  a  scrawled  inscription 
on  a  scrap  of  paper  pasted  on  the   inside  of  its  boards  :    "  Exempl.  du 
Vatican,  avec  fig.  2,  dont  la  iiire  est  couple  par  moitie".  le  25  Pluv.  an  9." 
It    is    the    1481    Dante,    with    Landino's    notes    and    two    of   Botticelli's 
illustrations.      Ludovico  Podocatharo  was  born  in  Cyprus  in  1430.      Most 
of  his  precious  collection  of  antiquities,  etc.  was  left  to  Livio  Podocatharo. 
Some  few  of  the  Podocathari  migrated  to  Venice  in  the  fifteenth  century, 
says  Cicogna.      Certainly  the  name   is   not   unfamiliar  on   tombs   in   the 
churches  of  Venice. 

2  Procacci  also  mentions  Hettore  Podocatharo's   "  Ritratte  del  Regmo 
di  Cipro"  in  his  work  de    Funerali  Antichi,   with   the  comment   "non 
credo  che  sia  stumpnta  questa  storia." 


332  APPENDIX  VIII 

at  the  time  when  the  chance  of  some  measure  of  justice  for 
her  immemorially  downtrodden  natives  was  at  last  arriving. 
The  Hettore  Podocatharo,  from  whose  forthcoming  book 
the  scrupulous  Procacci  had  feared  to  borrow  too  much, 
fought  under  Astorre  Baglioni  in  defence  of  Nicosia.  He 
was  one  of  the  eleven  captains  from  whom  the  eleven 
"  Beluardi  "  or  Bulwarks  of  the  town  were  named.  One  of 
these  last  was  on  the  point  of  succumbing,  but,  as  maybe 
read  in  the  life  of  Astorre  Baglioni,  published  at  Verona  in 
1591  under  the  name  of  Brenzone,1  "accorsero  appresso 
Gio.  Antonio  da  Soelle,  &  Eftore  Podocatsaro,  & 
ualorosamente  scacciarono  i  nemici.  Indi  il  Podocatsaro 
seguitb  il  nimico  fino  fuori  del  Parapetto,  se  ben  la  sua 
mala  sorte  volse  che  una  scheggia  di  pietra  spezzata  dall' 
artigliaria  lo  percuotesse  nel  capo  ove  bisognb  iacere  in 
letto.  Fu  feruto  anco  Ercole  Podocatsaro."  The  above 
mentioned  Podocathari  both  survived  these  wounds  and 
the  earlier  horrors  of  the  sack  of  Nicosia,  where  they  lost 
their  kinsmen  Giulio  and  Ludovico  of  the  same  name ; 
Brenzone  can  hardly  spell  it  twice  in  the  same  way.  The 
grievously  wounded  Hettore  only  escaped  for  a  time 
however.  With  various  fugitives  and  300  soldiers  he  was 
defending  as  well  as  he  might  the  house  of  his  sister  the 
Countess  of  Tripoli,  when  the  wily  Mustafa  urged  him  to 
accept  terms  which  were  instantly  violated.  Mustafa2 

1  There  are  many  reasons  for  thinking  that  this  is  a  mutilated  version  of 
Procacci's  life  of  Baglioni,  of  which  I  have  found  some  mention,  but  which 
was  never  published  unless  this  be  it.     The  curious  relation  in  which  many 
passages  of  the  life  of  Baglioni  stand  on  the  one  hand  to  Procacci's  book  on 
Islands,  and  on  the  other  to  Estienne  de  Lusignan's  Chorograffia,  which 
last  especially  represents  Hettore  Podocatharo's  MSS.,  makes  it  probable 
that,  if  Brenzone  wrote  the  book,  both  Procacci  and  the  Chorograffia  were 
freely  used.     If  Procacci  wrote  it,  he  may  have  used  the  details  (gathered 
from  Hettore's  MSS.)  which  delicacy  had  prevented  his  using  in  his  book 
on  Islands. 

2  The  intense  hatred  of  Turks,  common  in  those  days  as  in  these,  is 
exhibited  characteristically  by  Brenzone  in  his  account  of  the  names  of 
those  who  headed  the  invasion  of  Cyprus  :    ' '  che  nome  piu  stravagante  e 
bestiale  di   quello  che  trovano  i  Turchi  ?  Mustafa,   indegno,   imaginario, 
bestiale,  diabolico.     Pialy,  nome  da  mulatieri.      Giurerei  che  s'uno  fosse 


OLYMPUS  AND  BOCARUS  IN  CYPRUS  333 

ordered  his  prisoner's  head  to  be  cut  off,  pretending  to  be 
in  a  passion  with  him.1 

After  his  friend  Podocatharo's  lamentable  death,  Procacci 
might  well  have  put  all  that  he  could  of  the  missing  and 
unpublished  Ritratte  del  regno  di  Cipro  into  his  second 
edition,  which  was  published  shortly  after  his  edition  of  the 
Isolani  of  Bembo,  about  1590.  But  in  that  preface  he 
does  not  say  more  of  the  book  soon  to  come  than  that  he 
has  included  in  it  many  new  islands.  A  comparison  of  the 
two  editions  shows  that  in  fact  little  or  nothing  has  been 
altered  in  the  very  good  account  of  Cyprus  originally  com 
posed  with  Podocatharo's  helping  MS.  It  is  high  time  now 
to  give  without  further  explanation  Procacci's  account  of 
the  Cypriote  mountains — Santa  Croce  and  Troodos — which 
have  been  identified  with  Olympus  by  one  authority  or 
another.  After  giving  a  description  of  the  northern  range 
of  mountains,  Procacci  addresses  himself  to  the  southern 
group,  to  which  both  of  the  heights  in  question  belong, 
saying  :  "  L'  altra  parte  de'  monti  trauersa  1'  Isola,  comin- 
ciando  dalP  antica  citta  Solia  [Soli],  ch'  era  XVIII  miglia 
lontana  da  Cormachiti  [Strabo's  Cape  Crommyon],  & 
andando  per  mezzo  dell'  isola  fino  al  monte  della  Croce 
[Stavrovuni]  che  risponde  a  Capo  Masoto  [Ptolemy's  Cape 
Dades  near  Larnaca-Citium],  &:  uanno  fino  a  Baffo 
[Paphos]  a  marina ;  d'  onde  uoltano  dalF  altra  parte,  & 
pure  a  marina  uanno  fino  a  Solia.  In  mezo  a  questi  e  il 
monte  Olimpo,  chiamato  con  uoce  Greca  Trohodos^  che  e 
altissimo  &  pieno  d'  alberi  d'  ogni  sorte,  gira  di  circonfer- 

nel  mezzo  di  cinquanta  asini,  anzi  lupi  e  dicessi  Pialy,  Occhiaij,  o  Mustafa 
fuggirebbono  piu  che  lepri  i  cani,  o  le  colombe  lo  spaviero. " 

1  Such  is  Brenzone's  account,  differing  materially  from  that  given  in  the 
Chorograffia  above  mentioned  and  written  by  ' '  Padre  Maestro  Angelo,  della 
famiglia  de  Caletus  Vicario  generale  di  terra  Santa"  who  was  in  Nicosia 
during  the  siege  and  whose  qualifications  as  a  "  persona  iuditiosa,  dotta, 
&  sanza  passione  &  veridica  "  are  much  vaunted.  Angelo  says  that  the 
Countess  of  Tripoli  declared  herself  Mustafa's  prisoner  in  his  absence 
and  barricaded  herself  in  her  house  till  he  should  come.  Her  brother 
"  Hector  Podochataro  "  was  summoned  for  a  pretended  cure  of  his  wounds 
and  treacherously  slain  on  the  road. 


334  APPENDIX  VIII 

entia  LIIII  miglia,  che  son  XVIII  leghe,  &  ad  ogni  lega 
e  posto  un  monastero  di  monaci  di  San  Basilic,  Greci :  e, 
in  ciascuno  si  trovano  fontane  in  abbondanza,  &  frutti 
d'  ogni  qualita,  onde  la  state  soleuano  i  nobili  Cipriotti  venire 
a  questi  luoghi  per  lor  diporta." 

The  absolute  equivalent  of  this  description,  though  its 
pure  Italian  is  somewhat  defaced,  recurs  in  the  life  of 
Astorre  Baglioni  above  quoted,  where  it  runs  as  follows  : 
"...  principalmente  il  monte  Olimpo,  monte  altissimo, 
monte  d'  ogni  bene  monte  che  dal  piedi  gira  XVIII  leghe. 
Monte  che  in  ogni  lega  si  trova  un  monastero  de'  Calseri, 
Monaci  di  San  Basilio  Monasterij  copiosi  d'  ogni  gratia, 
per  igrani,  frutti  &  fonti  d'  acque  soauissime.  Indi  erano 
chiamati  le  delitie  della  nobilta,  perche  1'  estate  andauano  a 
prender  aria  per  sanarsi :  sanati  conseruarsi." 

It  would  be  evident,  if  we  did  not  know  it  from  the 
express  statement  of  the  scrupulous  Procacci,  that  such  a 
description  as  either  of  the  above  originated  with  one  who 
was  more  than  casually  familiar  with  the  facts  whereof  he 
spoke.  It  is  still  more  evident  that,  since  the  great  name 
of  Podocatharo  is  connected  with  it,  a  new  value  immedi 
ately  attaches  to  it.  The  Podocathari  form  a  link  between 
Cyprus  under  Byzantine  rule  and  the  Cyprus  ruled  and 
misruled  by  Venice  and  the  Lusignans.  One  of  the 
earliest  records  of  the  family  which  I  have  been  able  to 
discover  is  in  a  curious  book  which  I  have  often  alluded 
to  above,  and  which  was  prepared  for  publication — in 
the  monastery  of  Santa  Catherina  di  Formello  —  in  the 
month  of  November  1570  at  Naples  by  Frate  Stephano 
Lusignano  di  Cipro,  or  rather  Estienne  de  Lusignan,  whose 
name  appears  on  its  title-page.  He  fled  the  kingdom 
ruled  by  his  ancestors  in  the  flurry  caused  by  the  first 
arrival  of  the  Turks,  and  probably  reached  Naples  as  early 
as  September — Nicosia  capitulated  on  September  the  gth  : 
the  invaders  appeared  at  Paphos  on  the  ist  of  July. 
Considering  the  minute  nature  of  this  uncommonly  good 
account  of  Cyprus,  it  is  incredible  that  any  man,  let  alone 


OLYMPUS  AND  BOCARUS  IN  CYPRUS  335 

so  inefficient l  a  man  as  Estienne  de  Lusignan,  should  have 
written  it  in  so  short  a  time  as  that  which  elapsed  between 
his  flight  and  the  aforesaid  month  of  November  in  the  same 
year,  unless  it  is  a  compilation  from  various  documents, 
among  the  most  valuable  of  which  would  be  Hettore 
Podocatharo's  missing  MSS.2 

1  The  inefficiency  of  Estienne  appears  especially  in  his  foolish  additions 
to  the  simple  title  of  Podocatharo's  work,  in  his  helpless  but  somewhat 
pathetic  apology  about  the  misprints  in  the  Italian  version.      He  left  the 
book  a  whole  year  in  the  printer's  hands  while  he  was  travelling  up  and 
down   Italy  to  gather  money  for  the  ransom  of  his  friends  in  Turkish 
captivity,    "  il  correttore  hebbe  molti  dinari    non  hauendo   lo  la  lingua 
Toschana  ne  Italiana  naturale,"  then  it  was  brought  to  Bologna  and  kept 
another  year,    it   was    "quasi   dicono    veduto    &    riueduto,    nondimeno 
contiene  molti  errori,  &  qui  remediar  volesse,   necessario  sarebbe   rino- 
uarlo,  &  le  mie  forze  sono  debolissime,"  then  follow  quantities  of  errata. 
But  the  most  foolish  thing  about  this  Italian  edition  is  its  dedication, 
' '  Al  Christianiss.  et  glorioso  Carlo  nono,  re  di  Franza,  et  al  felicissimo,  et 
vittorioso  Henrico  novo  re  de  Polonia."     He  did  not  take  the  precaution 
to  prove  his  royal  descent,  and  therefore  the  book  passed  unnoticed  of  his 
brothers  of  France  and  Poland.     It  is  ludicrous  to  notice  how  elaborately 
Estienne   proves    his  Lusignan  genealogy  in  documents   prefixed  to  the 
French  version  (Paris,  1580),  which  is,  however,  dedicated  no  longer  to  two 
kings,   but    to  the  edification  of  one  boy,  Guy  de  Saint  Gelais,   son  of 
Estienne's  kinsman,  Loys,  of  the  same  name. 

2  In  the  dedication  of  his  French  version  Estienne  says,    "  Je  vous  ay 
voulu  done  faire  present  de  ceste  Chronique,  restant  du  sac  de  Cypre  & 
sauve"e  des  mains  cruelles  des  Barbares. "      In  the  Italian  Chorograffia  at  p. 
75  he  says  ' '  La  predetta  Cronica  cominciando  dal  Re  Giouanni  fino  a 
qui,  1'  ho  cauata  dalla  cronica  Greca  di  Giorgio  Bustrone  il  quale  era  com- 
pagno  del  Re  Giacomo  auanti  che  fusse  Re,  &  anchora  dipoi.     Vero  e  che 
noi  habbiamo  aggiunto  alcune  cose  di  altri  Auttori,  &  multe  altre  lasciate 
per  breuita."     Careful  comparison  of  the  parts  of  his  work  which  deal  with 
the  various  classes  (Parici,  Perpiari,   Lefteri,  Albanesi,  Venetian!  bianchi) 
and  describe  the  government  with  similar  passages  on  the  same  subject  in 
Procacci's  book  shows  that  the  two  undoubtedly  derive  this  information 
from  the  same  source,  since  neither  can  possibly  have  copied  the  other. 
A  comparison  of  the  brief  description  given  of  Cyprus  in  Florio  Bustrone's 
Istoria  di  Cipro  (the  Cronica  alluded  to)  on  sheets   1-12  of  the  MSS. 
shows  that  both   Procacci  and   Estienne  certainly  neglected  it,   and  as 
certainly  used  better  and  fuller  information.     Bustrone  does  not  mention 
Giovanni  Podocatharo's  name  in  his  account  of  the  ransom  of  King  lano 
(sheet  173).      Estienne  says  that  Giovanni  Podocatharo  sold  all  he  had  to 
free  the  king,  giving  a  version  not  unlikely  to  have  been  found  in  Hettore 
Podocatharo's  book.     Procacci  tells  us  that  he  used  Hettore  Podocatharo's 
Ritratte  del  Regno  di  Cipro.     The  inference  is  irresistible  that  Estienne 
de  Lusignan  derives  his  most   interesting  and  important  matter  (found 


336  APPENDIX  VIII 

However  that  may  be,  the  title  of  the  book  in  question 
is  Chorograffia  et  breve  Historia  universale  deW  Isola  di 
Cipro  principiando  al  tempo  di  Noe  per  insino  al  1572; 
and  in  it  we  find  the  most  ample  information  (from  a 
prejudiced  source,  be  it  said)  about  the  Podocathari,  their 
first  exhibition  of  the  devoted  and  self-sacrificing  patriotism 
which  characterises  the  family  being  recorded  as  follows  : 
"  Giouanni  Podochataro  gentilhuomo  Ciprioto,  vendette 
tutto  il  suo  mobile  &  immobile,  &  cio  che  haueua,  &  con 
quelli  danari  riscattb  il  Re  dal  Cairo  \  al  quale  da  indi  in 
poi  fu  imposto,  che,  pagasse  ogni  anno  il  tribute  al  Cairo  : 
il  qual  tribute  fu  cresciuto  al  tempo  del  Re  bastardo  : 
dipoi  1'  anno  1516,  il  Turco  estirpo  il  Soldano  :  &  quel 
tribute  che  si  pagava  al  Sultano,  si  pagaua  al  Turco,  come 
dominatore  del  Sultano  :  &  hora  non  vuole  piu  il  tribute  ; 
ma  come  si  dice,  ha  preso  tutta  1'  Isola,  &  cosi  e  finite  il 
tribute." 

This  King  lane,  so  loyally  ransomed  by  Giovanni  Podo- 
catharo,  was  the  father  of  Agnesa  who  married  Ludovico, 
Duke  of  Savoy ;  and  the  title  of  kings  of  Cyprus  and 
Jerusalem,  still  attaching  to  the  house  of  Savoy,  is 
through  her  derived  from  this  monarch,  of  whom  after  his 
ransom  we  hear  that  he  spent  the  rest  of  his  days  in 
continual  starvation  because  of  the  constant  raids  of  the 
Mamelukes.  Undiscouraged  by  all  these  reverses,  lano's 
epitaph  (given  by  Bustrone)  says  of  him  : 

Caesar  erat  bello,  superans  grauitate  Catonem. 

In  the  time  of  the  "re  bastardo,"  this  lano's  grandson, 
a  "  Pietro  Podochataro "  figures  very  creditably — so  far 
as  can  be  ascertained.  Indeed  it  is  satisfactory  to  see 

in  Procacci  and  not  in  Bustrone)  from  Podocatharo's  MSS. ,  "saved,"  as 
he  himself  says, — though  he  is  talking  vaguely,  and  appears  to  mean  his 
own  book  or  Bustrone's — "  from  the  sack  of  Cyprus  and  snatched  from 
the  cruel  barbarians'  hand."  This  view  of  his  debt  to  Podocatharo  is 
confirmed  by  the  unusually  familiar  knowledge  possessed  by  Estienne's 
book  of  the  doings  and  havings  of  that  great  family.  For  Bustrone's 
Chronicle,  see  MSS.,  British  Museum  (additional  MSS.,  Earl  of  Guilford), 
No.  8630.  Bustrone  (Florio),  Istoria  di  Cipro. 


OLYMPUS  AND  BOCARUS  IN  CYPRUS  337 

that  the  family  recuperated  itself  financially1  so  that  the 
Podocathari  were  in  a  position  to  buy  from  the  Venetian 
Signoria  (after  1489,  the  date  of  Caterina  Cornaro's  formal 
abdication)  one  of  the  most  valuable  and  the  most  ancient 
of  all  the  domains  of  Cyprus,  Chiti  or  Citium  (Larnaca) : 
"  Chitheon  era  citta  primieramente  edificata  auanti  d'  ogni 
altra,  <$:  fu  edificata  dal  primo  habitatore  dell'  Isola,  cioe 
da  Cethin  pronepote  di  Noe  :  II  che  testificano  li  sacri 
espositori,  Girolamo,  la  Glosa  ordinaria,  &  altri  sopra  al 
23.  capitolo  di  Isaia,  &  al  secodo  di  Gieremia.  Questa  citta 
e  posta  alia  marina,  discosta  dalla  citta  di  Marium  cinque 
leghe,  &  e  verso  mezo  giorno,  &  hauea  gia  vn  Porto 
bello,  &  serrato,  come  dice  Strabone :  il  quale  hora  e 
distrutto  affatto  ;  &  si  vede  bene  il  vestigio.  Questa  era 
citta  Regale  anticamente,  .  .  .  Hora  la  predetta  citta  si 
chiama  il  casale  Chiti  :  il  quale  e  grande  e  pieno  di 
giardini,  &  d'  ogni  frutto,  &  questo  fu  feudo  di  Chiarione, 
ouer  Gariu  Lusugnano  :  del  quale  fu  priuato  dall'  ultimo  Re 
bastardo,  &:  dipoi  fu  venduto  dalla  Signoria  di  Venetia 
alii  Podochatari."  2  There  is  great  probability  in  favour  of 
supposing  that  the  "Re  bastardo"  rewarded3  Pietro  Podo- 

1  See  the  Chorograffia,  p.  30  verso,  "  Ma  poi  al  tempo  del  Re  di  Cipro 
bastardo,   furono  molti  Nobili  di  Cipro  morti,   &  altri  fuggiuano,  &  altri 
furono  disnobilitati  per  le  priuationi  delle  loro  faculti  :  perche  non  volcano 
adherire  aesso  bastardo."     The  confiscated  estates  were  given  to  "  nobili 
&  ignobili  "   from  Italy,  and  Pietro  Podocatharo  (who  was  sent  as  am 
bassador  to  Cairo  in  the  interest  of  Carlotta,  but  changed  sides  on  reaching 
Egypt)  was  handsomely  rewarded,  and  retrieved  the  family  fortunes.     Bus- 
trone  (197-199)  gives  a  list  of  those  benefited.      Pietro  Podocatharo  (198) 
comes  in  for  enormous  estates,  Giovanni  and  Philippe,  his  kinsmen,  get 
handsome  presents. 

2  The  parallel  passage  in  Procacci  is  as  follows  :   "  ne  meno  d'  essa 
(Amathus)  fu  seggio  reale  la  citta  di  Chitheon,  prima  di    tutte  1'   altre 
edificata,  da  Cithin,    nipote  di  Noe,  ch'  e  posta  alia  Marina  verso  mezo 
giorno,    c'   haueua   un   bel  porto  ;    £   hora  ridotta  in   casalc,   si  chiama 
Chiti,  ch'  era  le  delitie  di  quel  regno,  posseduto  da  Hettore  Podocatharo, 
Cauallier    Cipriotto,  che  di  queste  cose  scrisse  :    il  qual  u'  aueua  giardini 
bellissimi  &   ripieni  di  preciosi  frutti."     The  differences  between  this  and 
the  corresponding  passage  of  the   Chorograffia  go  to  show  that  Estienne 
simply  published  the  original  text  of  Hettore  Podocatharo. 

3  Of  the  four  names  of  places  bestowed  upon  him  according  to  Rustrone's 
list  (p.  198  recto],  I  can  only  locate  two  which  were  not  near  Chiti. 

Z 


338  APPENDIX  VIII 

catharo  for  his  support  by  giving  him  certain  rights  at  or  near 
Chiti.  However  that  may  be,  the  family  came  into  regular 
possession  of  the  property  when  Venetian  rule  was  estab 
lished.  Procacci  gives  a  most  appreciative  account  of  the 
life  which  Hettore  Podocatharo  led  there,  and  of  his  generous 
hospitality.  Evidently  this  brave  man  lived  there  the  life 
which  he  loved,  and  the  splendid  courage  with  which  he 
defended  his  native  Cyprus  was  born  of  a  most  deep  and 
loyal  attachment  attested  at  the  last  by  the  sacrifice  of  his 
life.  Against  the  sombre  background  formed  by  the  savage 
onslaught  of  those  ruthless  Turks  in  the  last  half  of  the 
year  1570,  Hettore  Podocatharo  at  his  peaceful  home  of 
Chiti  stands  out  as  a  gracious  and  genial  representative  of 
all  the  good  that  might  have  come  to  miserable  Cyprus  if 
he  and  such  as  he  had  been  suffered  to  guide  the  counsels 
of  those  in  power.  That  Hettore  did  have  great  influence 
at  Venice  is  abundantly  proved  by  the  terms  in  which  his 
fugitive  brother  Zuanne  Podocatharo  apostrophises  him  in 
a  hitherto  unpublished  appeal  which  Zuanne  delivered  on 
the  1 7th  of  May  1573  "auanti  il  serenissimo  prencipe 
Aluise  Mocenigo,  doppo  la  perdita  del  regno  di  Cipro." 
Standing  forth  with  the  orphaned  children  of  the  heroic 
Hettore  by  his  side,  Zuanne  feels  for  the  first  time  to  the 
full  how  much  he  has  lost  in  Hettore,  and  how  completely 
alone  are  he  and  the  orphans  his  nephews.  There  is  a 
human  pathos  in  the  situation,  and  a  Homeric  directness 
in  his  words,  that  makes  it  well  to  quote  them  :  "  Dhe  ! 
Pietosissimo  Prencipe,  uengavi  oramai  pieta  di  noi,  risguardi 
hormai  Vostra  Serenita  con  la  serena  fronte  questi  infelici 
figliuoli.  Questi  son  quelli  che  non  hanno  esperienza 
alcuna  di  pecato,  si  pub  noi  con  qualche  nostro  difetto  siamo 
posti  nella  miseria,  che  si  trouamo.  .  .  .  Questi  sono  quelli 
c'  hanno  perduto  tanta  espetatione,  e  tale  che  poteua  farli 
uiuer  per  sempre  contenti,  e  felici.  Vostra  Serenita  risguardi 
quest'  innocenza,  risguardi  quella  purita,  e  considerando  in 
che  grado  poteuano  esser,  et  in  quale  si  trovino  al  presente, 
cerchi  con  la  sua  pietade  uincer  in  parte  1'  impeso  della 


OLYMPUS  AND  BOCARUS  IN  CYPRUS  339 

nostra  fortuna.  .  .  .  Dhe  !  Clementissimo  Principe,  aprite 
hormai  le  uostre  misericordiose  bracchia,  riceuete  noi  con 
quella  carita,  che  si  richiede  ad  un  tanto  Prencipe,  e  il  debito 
della  pieta  Christiana,  e  la  riputatione  del  uostro  benign- 
issimo  Impero,  e  per  edificatione,  et  essempio,  delli  uostri 
altri  sudditi  porgeteci  hormai  il  uostro  Clementissimo  agiuto, 
accio  possiamo  sostentar  questi  figli,  accio  possiamo 
recuperar  gl'  altri,  che  sono  dispersi  per  le  bande  degl' 
infedeli,  e  qua  li  potiemo  hauere,  qua  li  cauaremo  fuori 
delle  mani  de'  nostri  empij  nemici,  quh.  li  metteremo  nel 
consortio  d'  altri  Christiani,  qua  li  donaremo  gl'  amplessi 
materni : l  ci  para  assai  esser  ristoradi  di  cotanta  nostra 
perdita.  Ma  hoi  me  !  me  misero  !  o  me  infelice !  perche 
non  sono  io  atto  a  poter  con  quelle  efficace  parole,  che  si 
conuiene  persuader  questo  cosi  pietoso  offitio  !  Doue  sei 
sfortunato  fratello,  qual  crudele  et  improuisa  morte  mi  t'  ha 
tolto  ?  Perche  non  sei  qui  in  tanta  necessita  presente  ad 
aiutarmi?  Tu,  fratello,  hai  potuto  molte  altre  volte  in 
questo  medesimo  luogho  per  beneficio  della  nostra  Patria 
con  le  tue  parole  intenerir  i  cuori  di  questi  sapientissimi 
senatori  ad  impetrar  quanto  sapessi  dimandare,  ed  io  col 
tuo  medesimo  spirito  rappresentando  questi  tuoi  orfani 
figli,  e  questi  altri  delli  tuoi  amici  e  parenti,  non  potrb 
mouer  a  pieta  li  piu  pietosi  Christiani  del  mondo  !  Tu, 
fratello,  molte  fiate  col  tuo  ornato  parlar  hai  potuto  saluar 
la  uitta,  le  facolta,  et  honor  de  mold,  et  io  in  questo  nostro 
esterminio  non  potrb  impetrar  dalla  benignita  istessa  il 
uiuer  de  tuoi  figliuoli,  che  si  muorono  della  fame  !  Almeno, 
fratello,  poiche  qui  presente  esser  non  puoi,  poscia  che  di 

1  Several  of  these  captives  were  ransomed,  see  the  Rdatione  di  Ales- 
sandro  Podocatharo  de  successi  di  Famagosta,  Venice,  (?)  1656.  He  gives  a 
moving  account  of  the  horrible  death  of  Bragadino,  and  of  various  horrors. 
He  tells  in  detail  of  the  devastation  of  the  Messarea,  the  Carpass,  part  of 
the  Viscontado,  and  of  the  Baffo  district.  He  says  that  he  was  ransomed 
for  325  cecchini.  Cicogna  mentions  a  letter  from  Pietro  Podocatharo  of 
March  3rd,  1577,  to  Cardinal  Comendone,  where  he  says  that  Livio  Podo 
catharo  came  to  Venice  for  money  to  ransom  himself,  his  brother  Giovanni, 
and  his  son.  Probably  the  Giovanni  in  question  is  the  Zuanne  who  is 
speaking. 


340 


APPENDIX  VIII 


far  questo  officio  non  t'  e  concesso,  tu  insieme  con  1'  altre 
anime  beate  delli  nostri  Cittadini  li  quali  si  hanno  tanto 
volontariamente,  fedelmente,  e  prontamente  offerti  alia 
morte  per  la  Patria,  presentatevi  tutti  insieme  nell'  imagina 
tion  di  questi  Clementissimi  Signori,  scoprite  loro  in  questo 
ponto  le  uostre  crudelissime  ferite,  rappresentate  il  lago  del 
uostro  sangue  sparso,  mostrate  le  uostre  ardenti  ceneri  e 
con  quest'  Imagine  tutti  congionti  insieme,  come  se  fosse 
uiui,  aprite  le  uostre  supplicheuoli  braccia  alii  misericordosi 
piedi  di  questo  gran  Prencipe,  con  lui  vi  delete,  con  lui  ui 
lamentate,  con  lui  piangete,  da  lui  per  noi  e  per  uostri 
figliuoli  impetrate  qualche  pietade,  e  qualche  agiuto 
accio  noi  raconsolati  alquanto  possiamo  passar  questo 
pocco  di  uiuer  che  auanza,  sotto  la  santa  e  benigna 
protettione  di  questa  gloriosia  Republica,  la  qual  piacera 
alia  Maesta  di  Dio  di  conservar,  et  crescer  con  ogni  felice 
euento." l 

With  this  moving  apostrophe  made  in  his  name,  Hettore 
Podocatharo,  "  quel  magnanimo  gentil'huomo  "  disappears 
from  view.  Procacci's  description  of  Troodos- Olympus 
given  above  was  no  doubt  inspired  by  Hettore's  fuller 
account,  but  still  it  gives  little  of  his  "ornato  parlar,"  which 
was  evidently  as  well  remembered  by  his  contemporaries  as 
his  wit.  One  of  the  most  valuable  parts  of  the  precious 
MS.  which  Procacci  saw  and  praised  is,  however,  preserved 
in  the  description  of  Mount  Olympus-Troodos  given  in 
Estienne  de  Lusignan's  Chorograffia.  This  account  of  the 
great  mountain  of  Cyprus  has  only  to  be  compared  with 
Procacci's  to  appear  plainly  as  the  original.  "  L'  altra  parte 
delli  monti  comincia  da  Solia  citta  antica  discosta  da  Cor- 
machiti  6.  leghe  in  circa :  &  vanno  essi  monti  per  mezo 
dell'  Isola  insino  al  monte  della  Croce ;  il  qual  monte 
risponde  al  capo  Masotto,  &  li  monti  vengono  li  vicino,  & 
vanno  a  marina  per  insino  a  Baffo ;  &  poi  voltano  dalP 
altra  parte,  &  vanno  a  marina  a  marina  per  insino  a  Solia. 
In  mezo  de  questi  monti  e  il  monte  Olimpo ;  il  quale  in 
1  Marciana,  Classe  VII.,  Cod.  DCXLIX. 


OLYMPUS  AND  BOCARUS  IN  CYPRUS  341 

greco  si  adimanda  Trohodos ; l  il  quale  e  altissimo ;  & 
come  si  ha  salito  alcuni  monti,  come  si  e  al  piede  di  esso ; 
&  ancho  e  dibisogno  salire  vna  lega  buona,  che  sono  miglia 
3.  &  quando  si  e  giunto  alia  cima,  si  discopre  quasi  il  mare 
intorno  dell'  Isola ;  eccetto  che  da  Carpasso,  che  non  si  pb 
bene  conoscer  la  terra  :  perb  si  vede  bene  il  mare.  Vedesi 
anchora  li  monti  di  Cilicia,  &  quando  e  chiaro  nello 
spuntare  del  Sole,  si  vede  anchora  li  monti  della  Soria 
[Syria].  Questo  monte  e  pieno  di  alberi  di  ogni  sorte ;  & 
ha  tma  pianura  grande  in  cima.  II  piede  del  monte  cir- 
conda  18.  leghe  che  fanno  miglia  54.  &:  ad  ogni  lega  e 
posto  vn  monasterio  de'  Calloiri  [an  Italianised  form  of 
KaAoy//poi,  the  modern  Greek  for  monk  or  priest]  ouer 
Monaci  di  San  Basilio  :  quali  Monasterii  sono  pieni  d'  ogni 
frutto,  &  di  fontane  in  abondanza;  onde  questi,  &  altri, 
che  si  ritrouano  nell'  Isola,  sono  li  sollazzi  delli  Cipriotti  al 
tempo  della  estade.  ...  In  cima  del  monte  Olimpo  e  vna 
Chiesa  di  San  Michele,  &  li  di  fuora  e  un  sasso  grande 
simile  a  quelli,  che  si  ritrouano  nelli  torrenti :  &  intorno 
intorno  a  quel  monte  alto  vna  lega  per  insino  al  piede  non 
si  ritroua  vn  altro  simile  :  &  li  Greci  villani  dicono  una 
fauola,  che  quella  pietra  e 2  quando  che  1'  area  di  Noe 
riposb  di  sopra :  &  questa  e  grande,  perche  quattro 
huomini  apena  la  possono  eleuare  da  terra :  &  quando  che 
nell'  Isola  sta  assai  a  piouer ;  uanno  tutti  quelli  Casali 
vicini  del  monte  in  processione  in  cima  di  quell'  alto 
monte,  &  con  certi  legni  leuano  in  alto  quel  sasso,  & 

1  This  seems  to  mean  that  the  "  nobili  "  always  called  it   "Olympo," 
while   the  villani  used   the   name   Trohodos.       See  the  map  of  Ortelius 
(1573),    prepared  for  these  "nobili,"    where   Olympus  is  the  only  name 
given,   and  Trohodos  does  not   appear.       Bustrone  identifies  the   Monte 
della  Croce  as  the  Olympus  where  was  a  temple  of  Aphrodite  Acraea. 
This  of  course  is  a  blunder,  since  that  temple  was  in  the  district  of  the 
Carpass.      The  real   Olympus   Bustrone   calls  Lambadista  or   Chionodes, 
appealing  to  the  author  of  the  life  of  S.  Barnaba,  who  does  not  bear  him 
out.      The  modern  name  he  gives  as  Triodos,  MS.  p.  10,  verso. 

2  There  is   confusion  here,    as   often  happens,    because   Estienne    did 
not  understand  Italian,   and  has,  for  that  reason,  made  sad   work   of  his 
original. 


342  APPENDIX  VIII 

sernpre  cantando  :  &  cosi  finite,  dicono,  che  non  passa 
molto,  che  pioue,  &  assai ;  laqual  cosa  io  giudico  essere 
superstitione ;  perb  lasso  il  giudicio  a  chi  ne  ha  cura." l 

Here  then  is  a  description  of  Mount  Olympus  of  Cyprus, 
and  an  identification  of  it  with  the  Troodos,  which  might 
well  be  set  up  against  the  authority  of  Strabo,  if  Strabo, 
read  as  he  should  be,  were  not  wholly  in  agreement  with 
it.  It  is  not  rash  to  assume  that  Hettore  Podocatharo 
was  educated  in  one  of  the  great  universities  of  Italy.2 
He  was  acquainted  with  the  text  of  Strabo,  and  yet  never 
dreamed  of  suspecting  him  of  identifying  Olympus  with  the 
Monte  Delia  Croce.  But,  more  than  that,  his  long  life  in 
Cyprus,  where  he  was  born,  the  long  line  of  Cypriote 
ancestry  from  which  he  sprung,  the  traditional  sympathy 
with  the  down-trodden  Parici — descendants  of  the  mass  of 
people  found  in  the  island  by  King  Richard  Coeur  de  Lion, 
and  sold  by  him  first  to  the  Knights  Templars,  and  then, 
upon  their  speedy  expulsion,  to  Guy  (Guido  or  Guyes)  of 
Lusignan — which  characterised  the  Podocathari,  and  none 
more  than  Hettore  himself,  all  this  makes  of  him  the 
representative  of  any  continuous  traditions  which  may 
have  survived  in  Cyprus.  One  of  the  most  likely  things  to 
have  survived  from  days  even  before  Strabo's  time  would 
certainly  be  the  ancient  name  of  the  delight  of  all  the 
nobility  of  Cyprus,  their  refuge  from  the  infamous  heats  of 
lower  parts  of  the  island — Mount  Troodos -Olympus.  I 
therefore  have  no  hesitation  whatsoever  in  quoting  Hettore 
Podocatharo  in  proof  against  all  available  authorities, 
ancient  and  modern,  of  the  identity  in  ancient  Greek  and 
Roman  times  of  the  Troodos  and  Mount  Olympus. 

None  of  the  numerous  and  confusing  accounts  of  the 
mountains  in  question,  given  after  the  sixteenth  century, 

1  Chorograffia,  pp.  4  verso,  5  recto. 

2  Cicogna  mentions  a  letter  from  Hettore  to  his  brother  Pietro,  who 
was    at    college    under   Paolo   Manuzio.      This  justifies    the    assumption 
that  Hettore  had  been  to  college  himself.      He  could  hardly  have  taken 
the  position  among  learned  Venetians  which  he  plainly  occupied  unless  he 
had  been  at  a  university  in  Italy. 


OL  YMPUS  AND  BOCAR US  IN  C  YPR US  343 

need  now  detain  us.  With  the  occupation  of  Cyprus  by 
the  Turks  all  opportunity  of  understanding  anything  about 
it  at  first  hand  was  at  an  end.  Perhaps  it  is  worth  while 
to  note  that  the  much  admired  geographer,  Abraham 
Ortelius,  called  by  Cambden  "  eximius  veteris  geographiae 
restaurator,"  did  nothing  for  the  understanding  of  Cyprus 
beyond  repeating  what  Bordone  had  said,  appropriating 
Procacci's  work  without  acknowledgment,  and  giving  the 
testimony  of  his  map  of  1573  to  show  that  the  name 
Olympus  was  current  then  for  Mount  Troodos. 

Very  similar  to  the  unsystematised  and  ill-digested  treat 
ment  of  Ortelius  is  that  which  John  Blaeuw x  gives  to  Cyprus 
in  1662.  Blaeuw  takes  Ortelius'  map  of  1573,  in  flagrant 
disagreement  with  a  smaller  one  elsewhere  in  the  same 
book,  without  any  acknowledgment ;  then  he  accompanies 
it  with  Procacci's  commentary,  so  far  as  the  Olympus  is 
concerned.  Plainly  no  one  cared  or  knew  about  Cyprus 
in  the  seventeenth  century,  although  a  multitude  of  maps 
and  descriptions  of  the  island  continued  to  appear.  The 
one  exception  is  a  belated  Venetian  "  Isolario "  by  the 
enterprising  Geographer  in  Ordinary  to  the  republic,  Coron- 
elli.  His  Isolario  appeared  in  1696,  and  is  absolutely 
correct  in  giving  (according  to  Strabo)  a  promontory 
Olympus,  which  lies  close  to  the  north-eastern  extremity  of 
the  island,  and  then  a  Mount  Olympus,  which  is  the 
Troodos  half-way  between  Alesandreta  and  Piscopia. 
Coronelli  may  have  had  before  him  the  old  MS.  of  Strabo 
(still  at  Venice),  which  I  have  quoted  as  punctuating  the 
passage  about  Olympus  correctly,  but  he  certainly  did  have 
in  mind  the  traditions  about  Cyprus  and  its  Olympus, 
which  Venice  still  preserved  from  the  days  of  her  ^gean 
supremacy.  No  one  knowing  Cyprus  well,  and  going  to 
Strabo  and  Ptolemy,  would  differ  from  Coronelli,  I  am 
sure. 

1  It  was  really  the  work  of  his  father  William,  quite  as  much  as  his. 
See  its  dedication  as  "  Parentis  sui  suosque  labores  geographicos." 
Various  editors  and  compilers  were  employed. 


344  APPENDIX  VIII 

The  only  reason  that  existed  between  1675  and  1870 
for  connecting  with  Cyprus  the  name  Bocams  as  that  of  a 
river  or  of  anything  else,  was  a  curious  series  of  blunders 
made  in  a  posthumously  published  MS.  by  John  Meursius. 
But  then  Dr.  J.  P.  Six  discovered  one  more  letter  than  the 
Due  de  Luynes  or  Dr.  Deeke  could  see  on  a  coin  in  the 
British  Museum.  Dr.  Six  is  the  greatest  expert  in  the 
reading  of  Cypriote  characters,  and  while  he  was  in  doubt 
all  was  uncertain ;  but  he  has  finally  decided  that  that  in 
scription  on  that  coin  will  not  after  all  bear  his  first  inter 
pretation,  and  does  not  therefore  contain  any  name  even 
remotely  resembling  "  Bocarus  "  or  PO-KA-RO-SE.1  What 
now  remains  is  to  see  what  are  the  passages  in  ancient 
authorities  where  the  name  Bocarus  occurs.  The  lexico 
grapher  Hesychius  has  made  the  following  entry  : 
Trora/xo?  ev  SaAa/zin  IK  rov  'A/ca^ai/ros  opovs 
This  passage  gives  for  the  island  of  Salamis  two  names  not 
ordinarily  seen  upon  its  maps,  Acamas,  as  the  name  of  a 
mountain,  and  Bocarus,  as  the  name  of  a  river.  This  last 
name  does  appear  on  the  map  of  Salamis  given  by  K.  O. 
Miiller  in  Ersch  and  Gruber's  Allgemeine  Encyclopaedic 
(see  the  article  "Attica").  In  the  Etymologiciun  Magnum 
may  be  found  :  B12KAP02  :  To  eap  viro  T/xuj^viW'  Trapa  TO 
TO>  f3to)  -%apav  (freptiv,  /3co^apo^,  KOU  /3w/<apoS.  Kai  Trora/Jios 
Se  2aAa/zti/os  oirrw  KaAov/xei/os.  From  this  Hesychius  gains 
confirmation,  and  we  learn  moreover  that  the  name  Bocarus 
meant  spring  in  the  speech  of  the  men  of  Troezen,  just 
across  the  Saronic  gulf  from  the  Salaminian  river  Bocarus. 
From  Stephanus  and  Eustathius  3  no  additional  information 
but  further  confirmation  is  to  be  derived.  After  these 

1  According  to  a  note  shown  me  at  the  British  Museum  as  containing 
Dr.  Six's  amended  reading,  the  letters  are  not  what  he  first  thought  them, 
and  the  word  should  begin,  not  end,  with  the  sigma  syllable. 

'2  Meursius,  as  will  appear,  cites  this  one  passage  four  times.  Once  of 
the  river  Bocarus  on  the  island  of  Salamis,  thrice  as  referring  to  a  stream 
in  Cyprus. 

3  See  his  commentary  on  the  Iliad,  ii.  637,  and  on  Dionysius  Periegetcs, 


OLYMPUS  AND  BOCARUS  IN  CYPRUS  345 

commentators  and  lexicographers,  it  is  well  to  search  Strabo. 
In  his  ninth  book,  at  page  394,  you  may  read  :  Bw/capos 
8'  fcrrlv  kv  ^aAa/zivi  Trora/xd?,  6  vvv  B<i>JcaAia  KaAor/^cvos.  Then 
comes  to  hand  a  passage  in  that  dreariest  of  poems,  Lyco- 
phron's  Alexandra,  vv.  447-452,  where  Teucer,  Agapenor, 
Acamas,  and  two  others  are  spoken  of  as  five  early  colonists 
of  Cyprus,  which  is  called  by  two  of  its  most  obscure  and 
completely  forgotten  names,  Sphekia  and  Kerastia.  Lyco- 
phron,  not  content  with  these  hard  names,  alludes  to  Paphos 
in  Cyprus  by  the  forgotten  name  of  its  river  Satrachus,1  to 
Curium  on  the  coast  near  Paphos  by  the  epithet  vXaT^s, 
there  given  to  Apollo  (see  Devia  Cypria,  24  and  ff.),  to 
Paphian  Aphrodite  by  an  obscure  name  given  her  at  Sparta 
coupled  with  the  qualification  "  Zerynthian,"  alluding  to  a 
cave  in  Thrace  where  the  goddess  was  worshipped.  Finally, 
having  thus  done  every  ingenious  thing  to  prevent  our  know 
ing  that  the  heroes  went  to  Cyprus,  and  settled  near  the 
shrine  of  Aphrodite  at  Paphos  and  at  Curium,  Lycophron 
plays  a  final  trick  upon  our  wits  by  describing  Salamis 
(from  which  Teucer  departed  for  Cyprus)  as  the  "  Caves  of 
Cychreus,  and  the  dells  of  Bocarus."  Salaminian  Cychreus 
was,  like  Cecrops,  half  man  and  half  serpent,  and,  by  virtue 
of  the  lower  half  of  him,  a  denizen  of  caves.  Now  that  by 
a  tedious  process  of  excavation  their  meaning  has  been  laid 
bare,  these  lines  of  Lycophron  may  be  cited  in  full  : 

1  Satrachus  is  a  name  which  wanders  up  and  down  on  the  various 
maps  of  Cyprus.  Usually  it  is  bestowed  upon  a  river  rising  from  Troodos- 
Olympus.  Sometimes  a  city  is  improvised  and  called  Satrachos.  The 
passage  of  Lycophron  given  in  the  text,  together  with  the  following  four 
lines  of  Nonnus,  Dionysiac.  xiii.  458  ff.  ,  ought  to  make  it  certain  that  the 
Diarrhizo  at  Old  Paphos  bore  the  name  Satrachos  or  Sestrachos.  For 
the  most  learned  and  conclusive  paper  on  the  subject,  see  the  Philologus 
for  1874,  vol.  xxxiii.  Dr.  Robert  Unger  there  (pp.  419-430)  proves  that, 
wherever  the  Bocarus  may  have  been,  it  was  not  at  Paphos,  because  the 
Satrachos  was  there  as  Nonnus  says  : 


e'£  vddrwv 

^XL  BaXaffffiybvov  Ha-pirjs  vv^-prfiov  vdup 
^.drpaxov  i/j.ep6eis,  tidi  TroXXd/as  old  pa  \nrovcra 
KUTT/HS  av  e~xo.lv  tafff  \e\ov  /mtvov  vita 


346  APPENDIX  VIII 

01  Trerre  Se  2(/»y/ceiai/  eis  KepaariW 

KOU  ^Larpayov  /?Aw^avre?  'YAarov  re  y^v, 

Mop<£(b  TrapoiK'ijO'ova'i  TTJV 

o  fji^v  Trar/DO?  /AO/xc/xucri 

Kv>(/3eroS  avrpeov  Bw/capoi'  re  va^arcov,,    .    .    . 

"  But  five  there  are  who  shall  house  them  near  to  the  Zeryn- 
thian  Morpho  (Paphian  Aphrodite)  and  go  to  the  Horned 
Isle  Sphekeian  (Cyprus),  even  to  Satrachos  (Paphos  on  the 
Satrachos  l),  and  the  land  of  Hylates  (Curium,  where  Apollo 
Hylates  was  worshipped).  By  his  father's  blame  the  first 
(Teucer)  driven  otit  from  the  caves  of  Cychreus  and  the  dells 
of  the  Bocarus  (Salamis).  ..." 

These  are  the  only  places  in  ancient  literature  where  the 
Bocarus  is  mentioned.  And  yet  many  find  it  difficult  to 
give  up  the  idea  that  there  is  some  authority  to  show  that 
the  ancients  called  a  river  at  Old  Paphos  (the  river  Satra 
chos)  by  the  name  Bocarus.  This  is  not  the  case,  how 
ever,  with  Bursian,  who  gives,  in  his  description  of  Salamis, 
the  following  account  of  its  streams  (i.  p.  363  of  his 
large  work) :  "  Zahlreiche  .  .  .  Gussbache,  deren  ansehn- 
lichster  (wahrscheinlich  der  an  der  Siidwestseite  der  Insel) 
den  Namen  Bfa>jca/>o?,  spater  BuicaAta  fiihrte,  durchfahren 
die  Abhange  der  ziemlich  sparlich  mit  Strandkiefern  und 
Strauchwerk  bewachsenen  Berge."  Ancient  authorities 
should  be  the  only  warrant  for  ancient  usage,  but  still  let 
us  further  appeal  to  John  Blaeuw  and  his  Atlas  Major  of 
1662,  already  cited  above.  The  description  there  given  of 
Salamis  is  as  follows  :  "  Posita  est  ea  contra  Eleusin,  Atticae 
urbem,  6°  confinia  Megaridis  Atticaeque.  Longitudo  eius 
est  stadiorum  70.  Primarium  insulae  oppidnm  ei  cog- 

1  See  John  Meursius'  own  note  on  this  line  in  his  juvenile  commen 
tary  ;  see  also  Tzetzes,  who  speaks  of  a  town  and  a  river  named  Satrachus. 
Jacob  Geel  and  Emperius  seem  to  have  written  in  1773  saying  that  Paphos 
was  on  the  Satrachus,  but  I  have  failed  to  find  the  letters  in  question. 
See  also  Musgrave's  note  on  Eurip.  Bacchae,  404:  "  Audiamus  modo 
Sestrachi  apud  Nonnum  descriptionem  .  .  .  Sestrachus  enim  dicitur,  non 
Bocarus,  qui  Paphum  alluit  fluvius." 


OLYMPUS  AND  BOCARUS  IN  CYPRUS  347 

nomine  est ;  et  Bocarus  sive  Bucolius  amnis,  mine  If  alls 
Jlunnia  indigttatur"  Whoever,  among  the  numerous  con 
tributors  to  "  Blaevius' "  Atlas  Major,  may  have  written 
this  description  of  Salamis,  certainly  could  not  yet  have 
heard  of  the  transference  to  Cyprus  of  this  Salaminian  river 
Bocarus ;  for  it  did  not  take  place,  so  far  as  the  world  at 
large  was  concerned,  until  1675,  when  John  George  Graev 
or  Graevius  promulgated  at  Amsterdam  the  Cyprus,  Rhodes, 
and  Crete  of  Meursius.  The  writer  in  Blaeuw's  Atlas  did, 
however,  have  access  to  other  works  of  Meursius,  published 
during  the  author's  life,  which  ended  at  Soroe  in  1639. 
Meursius,  in  his  Pisistratus,  cites,  with  no  word  of  dissent 
ing  comment,  Strabo's  account  of  the  Bocarus  as  a  river  in 
Salamis  off  Attica.  This  proves  at  least  that  the  learned 
Meursius  was  not  always  of  his  later  opinion  about  the 
Bocarus. 

In  fact,  without  unduly  disparaging  the  merit  of  Meursius 
as  a  pioneer  in  that  laborious  research  which,  when  accom 
panied  by  a  sound  judgment,  is  the  bone  and  sinew  of 
scholarship,  I  may  say  of  his  work  in  general  that  he  accumu 
lates  quotations  for  the  sake  of  having  many  of  them,  and 
often  falls  into  the  error  of  using  the  same  authority  on 
both  sides  of  a  difficult  question.  Meursius  lacked  common 
sense  ;  his  work  has  not  the  luminous  and  life-giving  quality 
that  would  have  made  him  a  worthy  successor  to  the  tradi 
tions  of  Erasmus  or  have  won  the  unqualified  admiration  of 
men  of  learning  who  came  after  him.1  The  most  concise 
qualification  of  Meursius  and  his  scholarship  will  be  found 
in  Professor  Frederick  Allen's  address  on  the  University 
of  Leyden,  given  as  president  of  the  American  Philological 
Association  in  1882.  Professor  Allen  there  said  that 
Meursius,  the  antiquary,  had  great  diligence  and  some 

1  Of  Johann  Friedrich  Gronovius,  for  instance,  who  came  to  a  pro 
fessorship  at  Leyden  in  1658,  upwards  of  thirty  years  after  Meursius 
resigned.  It  is  possible  that  he  was  a  pupil  of  Meursius  ;  they  certainly 
carried  on  a  correspondence.  His  opinion  of  Meursius  is  perhaps  re 
flected  by  his  son  Jacob  Gronovius,  who  succeeded  him  in  the 
professorship. 


348  APPENDIX  VIII 

constructive  power,  and  that  his  monographs  laid  a  good 
foundation  for  subsequent  work.  Every  one  who  is  in 
terested  in  Cyprus  is  in  fact  interested  in  Meursius,  and 
owes  to  his  industry  a  debt  of  gratitude.  It  is  therefore 
a  thankless  task  to  talk  of  the  serious  errors  in  Cypriote 
geography  whose  origin  can  be  traced  to  Meursius.  But 
for  all  that  Meursius  may  fairly  be  held  responsible  for 
the  idea  current  in  his  day  among  scholars  that  ancient 
geography  was  independent  of  the  configuration  of  the 
earth.  He  certainly  has  treated  Cyprus  as  a  sort  of  fairy 
land,  where  rivers  may  run  over  mountain-tops  in  order 
to  enable  scholars  to  defend  the  most  random  and  un 
premeditated  emendations. 

The  fact  is  that  John  Meursius  is  one  of  those  who 
early  showed  a  fatal  precocity.  At  the  age  of  thirteen,  says 
an  appreciative  biographer  (Schramm),  he  was  preternatural ly 
fond  of  Greek  :  "  Quippe  iam  Anno  aetatis  XIII  carmen 
Graecum  suo  Auctori  hand  inficiandum  elaborabat."  This 
is  faint  praise,  not  at  all  like  that  bestowed  upon  the 
feat  by  which  Meursius  in  his  sixteenth  year  indicated  his 
preference  for  the  least  poetical  of  Greek  writers.  He 
actually  wrote  at  that  tender  age  a  commentary  on  Lyco- 
phron  !  It  is  for  this  that  D.  Heinsius  praises  him  and 
declares  him  an  Apollo  : — 

Talis  erat,  famam  cui  pristina  tempora  debent, 
Meursius,  et  debent  tempora  nostra  suam  ; 

Qui  Siculi  vatis,1  noctemque  Lycophronis  atri, 
Dispulit  et  Phoebus  post  tria  lustra  fuit. 

Meursius  taught  Greek  at  Leyden  as  professor  for  four 
teen  years,  and  then  was  glad  to  get  away  from  his  native 
country  and  accept  the  post  of  Professor  of  Danish  History 
at  Soroe,  on  the  island  of  Seeland.  Thither  he  went  in 
1625,  escaping  from  Calvinism  and  the  Synod  of  Dort,  and 
abandoning  for  the  most  part  his  Greek  studies.  If  there 
had  been  any  chance  of  his  further  intellectual  growth,  this 

1  This  alludes  to  Meursius'  juvenile  essays  on  Theocritus. 


OLYMPUS  AND  BOCARUS  IN  CYPRUS  349 

change  from  Leyden  to  Denmark  cut  it  off.  His  publi 
cations  on  classical  subjects  ceased,  and  his  study  of 
Danish  history  was  so  perfunctory  that  he  never  even 
learned  enough  Danish  to  master  the  most  necessary 
records.  Hence  when  Lami  asked  for  the  MS.  left  be 
hind  on  this  subject,  he  was  assured  that  there  was  nothing 
extant  which  would  not  if  published  mar  the  reputation 
already  achieved  by  Meursius.  As  a  boy  he  tried  to  do 
work  which  many  a  man  would  shirk ;  in  the  prime  of  his 
manhood  at  forty-five  he  was  subject  to  lapses  of  judgment, 
and  at  times  he  could  become  hopelessly  infantile.1  He 
did,  however,  preserve  to  the  last  that  command  over  many 
books  which  Graevius  praises,  saying  of  Meursius :  "  Nihil 
enim  hoc  viro  in  evolvendis  omnibus  omnium  aetatum 
scriptoribus  qui  ex  illetabile  illi  barbariei  nocte  salvi  in 
lucem  horum  temporum  emerserunt,  fieri  potuit  diligentius." 
It  is  this  wonderful  acquaintance  of  his  with  swarms  of 
writers  whose  dulness  few  ever  had  the  courage  to  penetrate 
that  gives  to  Meursius'  work  on  Cyprus  its  great  value. 
As  I  am  speaking  of  his  shortcomings  in  that  work,  let  me 
first  quote  competent  criticisms  of  defects  in  his  work  at 
large.  To  begin  with  his  mechanical  habit  of  always  trans 
lating  Greek  quotations  into  Latin,  it  would  seem  that  he 
sometimes  made  these  translations  while  he  was  half  asleep. 
In  some  remarks  appended  to  Meursius'  Ceramicus — see 
Lami's  complete  edition — Gronovius  2  objects,  as  well  he 

1  In  justice  to  his  later  work  it  should  be  remembered  that  he  long 
suffered  from  and  eventually  died  of  a  most  painful  disease. 

2  Lami  publishes  a  curious  letter  which  shows  that  Abraham  Gronovius 
had  not  forgotten  how  thoroughly  his  father,  Jacob — and  for  that  matter 
his  grandfather — had  found  out  the  weak  points  in  Meursius'  scholarship. 
Abraham,  though  not  of  much  learning  or  consequence,  was,  if  only  for 
the  name  he  bore,   deemed  of  sufficient  consequence  to  receive   Lami's 
circular  describing    the  projected   edition   of   Meursius.       In    ans\ver   he 
writes,  April  1736,  to  Valentio  Gonzaga,  complimenting  Italian  printing, 
saying  :    ' '  Xon  possum  non  recordari  felicitatis  qua  olirn  istic  usi  fuerunt 
avus  meus,  Parens  atque  Patntns."      He   then   feelingly  alludes   to  the 
odium  which   Meursius   incurred   by   devoting  time  to  the  study  of  the 
Christian    fathers  :     ' '  Quamvis    hac    sedula     opera    sua,    et    Christiano 
digna,  ingens  odium  Meursius  apud  nescio  quos  dum  inter  vivos  ageret, 


350  APPENDIX  VIII 

may,  to  Meursius'  translation  of  KaAAtVn/  ?}  G'/j/sa  TO 
TTporcpov — "  Pulcherrima  venatio prius ; "  he  also  finds  fault 
with  Meursius  for  confusing  Hermes  KaTcu/^a-n/s  with  Zeus 
of  the  same  epithet  and  a  clap  of  thunder:  "  Quae  sunt 
arenae  sine  calce,"  too  much  sand  and  no  mortar,  humorously 
adds  the  indulgent  Gronovius.  There  are  indeed  many 
of  these  trackless  waste  places  in  Meursius'  best  work. 
James  Brucker  truthfully  says  that  Meursius  was  much 
imposed  upon  "  in  Themide  Attica,"  where  he  quotes  as 
the  real  laws  of  Athens  fictions  composed  as  rhetorical 
exercises.  Also  Brucker  is  right  in  speaking  severely  of 
Meursius'  "  De  denario  Pythagorico."  * 

Let  us  turn  now  to  chapter  xxii.  of  the  first  of  Meursius' 
two  books  on  Cyprus,  as  printed  by  Graevius,  reprinted  by 
Lami,  and  revised  in  an  autograph  MS.  of  the  Marcian 
Library.  From  this  last2  in  Meursius'  own  handwriting, 
with  his  own  corrections  and  additions,  I  shall  make  my 

in  se  contraxerit.  Ego  vero  ab  ipso  iuventutis  meae  lubrico,  patcrnis  in- 
stitutis  ita  formatus,  eos  qui  scripta  vel  ad  ritus  vel  ad  historiam  veteris 
Ecclesiae  spectantia  primi  luce  donaverint  .  .  .  Meursium  .  .  .  Monte- 
falconem  .  .  .  venerari  nunquam  desinam." 

1  It  would  seem  that  some  rated  Meursius  so  low  that  they  threw  cold 
water  on  Lami's  idea  of  publishing  his  complete  works.      See  the  letters  in 
answer  to  Lami's  circular  printed  at  the  beginning  of  the  Florence  edition 
(Florence,  1741-63,  12  vols.  folio).     See  also  the  end  of  Lami's  dedication. 

2  Meursius,  while  in  Denmark,  did  no  classical  work  apparently  except 
this  one  on  Cyprus,  Rhodes,  and  Crete.      Of  this  he  made  two  copies,  both 
of  which  came  by  his  death  into  the  hands  of  his  son.      One  of  these 
passed  by  sale  into  the  library  of  George  Seefeld,  out  of  which  it  was  forcibly 
taken  in  1658  when  Charles  X.  of  Sweden  made  his  famous  march  across  the 
frozen  Belt.      Most  of  the  looted  books,  and  Meursius1  Cyprus,  etc.  among 
them,  were  added  to  the  royal  library  at  Stockholm.      Thence  Graevius  bor 
rowed  the  Meursian  MS.  in  1675,  and  byway  of  recognising  the  kindness  of 
the  loan  dedicated  the  printed  book  to  Charles  XI.  whose  father  stole  the 
MS.      All   this  happened  thirty-nine  years   after  Meursius  died  ;    twelve 
years  later  this  MS.  was  burned  with  the  rest  of  the  Stockholm  library. 
John  Meursius,  junior,  being  a  worldly-minded  person,  had  presented  the 
other  autograph  MS.  of  Cyprus,  Rhodes,  and  Crete  (one  which  had  received 
annotations  and  various  polishings  and  reshapings  in  the  last  years  of  his 
father's  life)  to  the  Senate  and  Doge  of  Venice  with  a  request  that  he  be 
recompensed  for  this  and  other  services  by  the  title  of  ' '  Cavaliere  di  San 
Marco."      This  title  was  conferred,  and  the  three  volumes — one  for  each 
island,  Cyprus,  Rhodes,  and  Crete — were  duly  consigned  to  entire  oblivion 


OL  YMPUS  AND  BO CA R US  IN  C YPR US  351 

quotations.  After  citing  a  line  from  a  Homeric  Hymn  to 
Aphrodite,  where  the  Cypriote  town  of  Salamis  is  men 
tioned,  Meursius  goes  on  :  "  Perluebat  autem  earn  l  (Salami- 
nem)  fluvius  Bocarus,  ex  Acamante  monte  profluens. 
Hesychius  :  Boticapos  Trora/xos  ev  2aA.a/xivi,  e/<  rov  'A/ca/zcu'Tos 
opous  fapofjievos,  Bocarus  fluvius  in  Salamine^  ex  Acamante 
monte  profluens?  The  writer  of  this  has  apparently  never 
dreamt  that  Cyprus  really  exists  to  check  by  its  shape 
any  suggestions  or  emendations.  Such  a  thing  as  a  map 
of  Cyprus — he  might  easily  have  had  that  of  Abraham 
Ortelius — he  has  never  looked  at.  In  chapter  xxvii.  he 
again  quotes  Hesychius,  whose  account  of  Acamas  he 

in  the  library  of  the  Council  of  Ten.  Thus,  thirty-three  years  before 
Graevius  published  the  unfinished  Stockholm  MS. ,  the  completed  one 
was  placed  beyond  the  reach  of  probable  publication.  When  the  library 
of  the  Council  of  Ten  was  dispersed  and  for  the  most  part  absorbed  into 
the  Marcian  library  (1795),  there  were  no  longer  three  but  two  parts 
of  the  MS. — Cyprus  and  Rhodes.  The  MS.  on  Crete  was  apparently 
stolen  or  lost  at  some  earlier  time,  and  it  was  finally  bought  by  Cicogna. 
With  his  books  and  MSS.  it  passed  into  the  Museo  Correr,  where  I 
have  examined  it  in  detail,  as  also  the  Rhodes  MS.  in  the  Marcian 
library.  To  the  Marcian  MS.  of  Cyprus  I  gave  even  more  time,  com 
paring  it  with  Graevius'  Amsterdam  edition,  after  satisfying  myself  that 
there  is  no  independent  value  attaching  to  Lami's  Florence  reprint  of 
Graevius.  (Lami  has  corrected  a  few  gross  misprints  of  Graevius,  but  did 
not  for  instance  use  the  table  of  errata  given  at  the  end  of  Graevius' 
Crete,  nor  did  he  rectify  cross  references,  omissions,  or  wrong  numbering 
of  chapters.)  As  reproductions  of  the  MS.  burnt  at  Stockholm,  both  of 
these  editions  of  Amsterdam  and  of  Florence  are  disgraceful.  The  Venice 
MS.  shows  enough  about  the  Stockholm  version  to  make  that  much 
absolutely  certain.  As  for  the  difference  between  the  Stockholm  MS.  and 
the  Venetian  revised  one,  I  challenge  any  one  to  find  a  single  error  of  judg 
ment  which  Meursius,  on  mature  second  thought,  saw  fit  to  correct.  He  has 
made  himself  doubly  answerable  for  all  the  worst  of  his  blunders  by  rewrit 
ing  them  in  smoother  Latin  and  by  a  careful  index,  where  he  passes  them 
all  and  assigns  to  each  its  place  in  alphabetical  succession.  But  it  must 
be  allowed  that  he  has  added  many  new  and  some  important  citations. 
These  give  a  slight  value  to  the  Venice  MS.,  but  they  need  never  be 
published. 

1  Curiously  enough,  no  one  has  ever  taken  this  blunder  of  Meursius 
seriously,  as  many  have  taken  the  blunder  about  Paphos  and  the  Boc 
arus.  The  Pediaeus  is  too  prominent  a  river  to  have  its  name  changed 
with  ease  by  any  one.  Perhaps  if  the  name  Satrachus  had  been  equally 
well  known  the  Bocarus  would  never  have  invaded  Cyprus. 


352  APPENDIX  VIII 

thus  translates  :  "  Acamanta  indefessum,  et  nomen  pro- 
prium  unius  filiorum  Antenoris."  He  quoted  just  above 
from  Sextus  Empiricus  to  show  that  Acamas  was  a  son 
of  Theseus,  from  whom  the  promontory  was  named, 
and  now  adds  from  Hesychius  :  "  Etiam  mons  in  Cypro, 
ita  dictus,  nominatus  vero  est  ab  Acamante,  Demophontis 
quidem  fratre  sed  These!  filio,  ex  hoc  Bocarus  amnis 
profluebat.  Idem  Hesychius  alio  loco.  ..."  Then 
he  repeats  the  quotation  and  translation  given  in  chapter 
xxii.  The  heading  of  chapter  xxx.  Meursius  has  written 
out  as  follows:  "  Fluuii.  Aous.  Bocarus,  Locus  Euripidis 
correctus.  Salaminem,  et  Paphum,  perluebat  in  plures  alueos 
distributus,  'etc"  Then  in  the  body  of  the  chapter,1  when 
he  has  disposed  of  the  Aous,2  he  again  attacks  the  Bocarus  : 
"  Bocarus.  Hesychius  .  .  ."  —  here  follows  the  same  quota 
tion  twicemade  before  —  "Scio  ab  aliis3  Bocarum  in  Salamine 
Atticae  commemorari  :  sed  hie  fuerit  sane  Cypri,  in  qua 
Acamas,  unde  ortus.  Ac  corruptum  esse  puto  eius  nomen 
apud  Euripidem  in  Bacchis  — 

I  Koi/JLav  TTOTI  rav  Kvirpov 
Nacrov  ras  A  </>/oo8mxs 
av 


poai 


Utinam  veniam  in  Ciprum 

Insulam  Veneris  : 

Et  Paphum,  guam  centum  habentis  ostia 

Bocari  fluuii  fluxus 

Foecundant  sine  imbribus. 

1  This   chapter  is   practically  identical  in   Graevius,    Lami,    and  the 
Marcian  MSS.  ,  probably  because  the  latter  has  by  accident  no  margin  for 
additions. 

2  The  Aous,  be  it  said,  is  the  name  anciently  given  to  Mount  Santa 
Croce,  so  constantly  miscalled  Olympus. 

3  These  "other"  people  include  himself,  in  the  passage  of  his  Pisistratzis, 
and  also  in  his  early  commentary  on  Lycophron,  line  450. 

4  I  take  it  that  Engel  is  responsible  for  the  adoption   of  this  absurd 
invention  of  Meursius,  Ross  following  him. 


OLYMPUS  AND  BOCARUS  IN  CYPRUS  353 

Hodie  editur,  Bapftdpov  Trorapov,  Barbari  fluuii^  Ubi 
etiam  obseruandus  :  Paphum  quoque  perluisse,  aut  alluisse  : 
et  in  plures  alueos  scissum  proluxisse." 

I  suppose  that  a  river  which  could  flow  from  the 
Acamas  to  the  sea  at  Salamis  could  also  empty  into  the 
sea  through  a  hundred  mouths  at  Paphos,  but  it  would 
certainly,  in  so  doing,  crowd  out  the  other  Paphian  river 
called  Barbarus,  which  Meursius  mentions  on  the  strength 
of  this  same  passage  in  Euripides  at  the  fourteenth  chapter 
of  this  same  book.  The  heading  of  this  chapter,  in  Meur 
sius'  own  hand,  runs  as  follows  :  "  Paphus  antiqua ;  quam 
Agieos  condidit ;  Typhonis  nlius ;  sive  Cinyras ;  sive 
Paphus  unde  dicta  &  quando  condita :  prius  Erythra 
appelata.  In  excelso  loco  sita  agro  pingui,  quern  sine 
pluuia  Barbarus  fluuius  irrigabat,"  etc.  Then  in  the  body 
of  the  chapter  comes,  after  an  account  of  the  excellent 
fertility  of  the  surroundings  of  Paphos :  "  Et  irrigabat 
fluuius  Barbarus,  etiam  sine  ulla  pluuia,  Euripides 
Bacchis — 

Tld(f>ov  0'  av  e/carocTTo/xot 

Bapfidpov1  TTora/jiov  poal 

KapTT i£ov(r iv  dvofM/Spot. 

Et  Paphum  quam  centum  ostia  Jiabentis 

Barbari  fluuii  fluxus 

Frugiferam  reddunt  sine  imbribiis" 

The  indices,  carefully  prepared  by  Meursius  for  this  book, 
and  written  by  his  own  hand  in  the  Venice  MS.,  show  both 
the  Bocarus  and  the  Barbarus,2  as  well  as  the  name  of 
Euripides  among  the  authors  "  qui  hie  illustrantur, 
emendantur  aut  errare  ostenduntur."  The  object  of  this 

1  Mannert  and  others  are  guilty  of  taking  the  capital  R  of  this  line  as 
seriously  as  Meursius  does.  It  was  zeal  in  correcting  this  mistake  that  led 
Engel  to  leap  from  the  frying-pan  into  the  fire  and  speak  of  a  "  vielarmiges 
Fliisschen,  der  von  den  alten  Bokarus,  nicht  Barbarus  .  .  .  genannt  war" 
(Cyprus,  i.  p.  126). 

-  These  are  far  complcter  than  the  indices  which  Graevius  published  in 
his  edition.  Probably  these  last  were  done  by  some  hard-pressed  scribe 
of  Amsterdam. 

2   A 


354  APPENDIX  VIII 

long  account  of  much  blundering  will  be  reached  if  a  com 
parison  of  the  above  passages  will  only  show  that  Meursius' 
Cyprus  is  an  unsteady  guide,  and  that  until  it  was  written 
the  Bocarus  was  never  connected  either  by  uttered  speech 
or  in  writing,  or  on  any  map,  coin,  or  inscription,  with 
Cyprus  where  it  has  no  place. 


VIII 
APOLLO   AT   DELOS 

THE  place  of  our  choice  when  first  we  set  foot  on  Delos  is 
the  altar  where  Odysseus  worshipped  in  the  spring-time  of 
our  world. 

"  At  Delos  once  by  Apollo's  altar  I  saw  thy  like,"  says 
Odysseus  the  castaway  to  Princess  Nausicaa,  going  on  to 
compare  that  maiden's  peerless  beauty  and  grace  to  "  a 
palm's  tender  shoot  growing  upward."1  Such  a  vision,  as 
of  a  delicate  tree  uplifting  its  graceful  branches  and  exquisite 
stem  against  the  pure  and  glowing  sky  of  some  well-loved 
Italian  picture,  is  a  fitting  symbol  of  the  worship  of  Delian 
Apollo  and  his  sister  Artemis,  guardians  of  purity  and  truth 
for  Hellas  and  the  ancient  world. 

Purification  and  purity,  these  two  words  must  always  be 
on  men's  lips  when  they  talk  of  Delos.  The  spiritual  needs 
and  the  moral  perfections  which  these  words  imply  were  in 
the  hearts  of  the  votaries  who  came  to  worship  and  observe 
the  rights  prescribed.  Unspeakable  feelings  and  inex 
plicable  associations  clustered  for  religious  Greeks  about 
the  story  of  Leto's  wanderings  and  her  final  deliverance 

1  Odyssey,  vi.  162  and  ff. 


356  APOLLO  AT  DELOS  vm 

upon  the  barren   Delian   rock — the  birth  of  Apollo  and 
Artemis.1 

All  the  efforts  of  poets,  philosophers,  geographers,  and 
historians  have  been  expended  in  the  most  various  utterance 
of  the  terror  and  the  pity,  the  awful  gladness  that  assailed 
the  mind  when  contemplating  Delos,  and  the  moment 
when  Apollo  "leaped  forth  to  the  light  of  day."2  Certain 
pious  representations  of  the  birth  of  the  Virgin  and  of 
various  saints  may  be  appealed  to  as  a  Christian  analogy. 
We  hear  from  Theognis 3  how  at  that  hour  "  The  universal 
shores  of  Delos  all  were  loaded  with  ambrosial  fragrance, 
and  the  immense  earth  was  moved  to  laughter,  while  re 
joicing  visited  the  depths  of  gray  ocean."  "Earth  smiled 
beneath  her,"  says  the  Homeric  bard4  of  Leto,  and  we 
hear  from  Euripides  in  two  of  his  most  pathetic  strains5 
how  the  "  first  parent  of  palm-trees  and  the  earliest  growth 
of  the  laurel "  were  called  into  being  for  the  comfort  of 
Leto,  the  mother.  Her  bed  was  near  the  Inopus,  a  stream 
specially  hallowed  in  the  minds  of  latter-day  believers  by  its 
fabulous  connection  with  the  far-off  river  Nile  whose  floods 
found  an  underground  way  to  rocky  Delos.6  Moreover 
Leto  had  close  at  hand  the  most  sacred  lake,  which  in 
some  versions  of  the  birth-legend  quite  takes  the  place  of 
the  Inopus,  and  in  almost  all  is  mentioned  with  a  peculiar 

1  According  to   the  Homeric   Hymn    to   Delian   Apollo    (16   and   ff. ) 
Artemis  was  born  on  Ortygia  and  Apollo  on  Delos.      Baumeister  seems 
justified  in  rejecting  these  lines.      Ortygia  was  an  older  name  for  Delos. 
Some  have  tried  to  make  out  that  it  means  Rhenea. 

2  Homeric  Hymn  to  Delian  Apollo,  119. 

3  Bergk's  ed. ,  vv.  7-10. 

4  Hymn  to  Delian  Apollo,  118. 

5  Hecuba,  459  ;  Iph.   Taur.  noo. 

6  Strabo,   271,   Callimachus's  Hymn  to  Delos,   206  ff.      See  also  the 
sonnets  on  "  Sdiles  "  in  the  curious  book  of  Bartolommeo  da  li  Sonnetti. 
The  story  of  to-day  substitutes  the  Jordan  for  the  Nile. 


vin        APOLLO  AND  ARTEMIS  BORN  AT  DELOS         357 

reverence.1  From  the  predestined  moment  of  fate  when 
Apollo  and  Artemis  were  born,  Delos  was  changed  :  "  Of  gold 
from  that  hour  were  all  her  foundations,  the  ripples  of  her 
wheel-shaped  lake  were  liquid  gold,  golden  was  the  sheltering 
palm-tree,  Inopus  rolled  a  flood  all  gold,  and  golden  was  the 
ground  from  which  the  mother  lifted  up  her  son  new-born."2 
"  A  flush  of  golden  flowers,  as  it  were  a  forest  flowering  on 
a  mountain-peak,  covered  the  chosen  island  where  no  flowers 
had  grown  till  then." 3  To-day  a  wealth  of  flowers,  gold  and 
red,  is  almost  the  only  remnant  of  past  glories  that  time  and 
man's  destructive  hand  have  left  on  Delos. 

Few  scenes  upon  which  the  religious  imagination  of  man 
has  loved  to  dwell  have  been  made  more  touching  or  filled 
more  full  of  the  pathos  which  consoles  than  the  stay  upon 
Delos  of  lonely  Leto,4  who  sealed  with  suffering  an  "  argu 
ment  of  never  ending  love."  Euripides  discoursing  upon  this 
theme  might  use  the  further  words  of  Shakspeare  saying — 

The  pretie  and  sweet  maner  of  it 

Forst  those  waters  from  me,  which  I  would  have  stopt, 
But  I  had  not  so  much  of  man  in  me 
But  all  my  mother  came  into  my  eyes, 
And  gave  me  up  to  tears. 

Truly  human  are  the  "tricklings  of  tears  down  dropping 
fast "  whereof  the  exiled  maidens  speak  in  the  play  just 
after  their  thoughts  have  dwelt  upon  the  touching  tale  of 

1  Herod,  ii.  170.  -  Callimachus'  Hymn  to  Delos,  260  ff. 

3  Homeric  Hymn  to  Delian  Apollo,  135  and   139,   Baumeister's   text, 
cf.  ibid.  53  and  ff. 

4  The   literary  record  of  Leto's   character  is   less  vivid  and  complete 
than  would  be  expected   from  the   prominence  given   to   her  in  various 
pictorial  and  sculptured  records.      Hesiod  gives,  however,  a  most  beautiful 
account  of  her,    ThcogoHv,  406  and   ff.      "She  was  a  comforter  always, 
and  gentle  to  mortal  men  as  to  the  deathless  gods,  —  a  comforter  from  the 
beginning,  the  most  soothing  presence  of  all  on  Olympus." 


35«  APOLLO  AT  DELOS  vm 

Leto,  into  which  they  plaintively  weave  the  sorrows  of  their 
loneliness  and  exile.1 

Bird,  that  adown  the  rock-ridged  main, 
Halcyon,  a  pitiful  refrain 

Singest,  to  all  thy  knowers  known, 
For  husband  lost  thy  tuneful  moan 
Shrilled  forth  !  be  other  lays  to  thine 
Likened,  a  wingless  songster's — mine, 
For  noon-tide  throngs  in  Greece  my  yearning  strain, 
Yearning  for  Artemis,  easer  of  pain, 

'Xeath  Cynthus'  heights  abiding  where 

Thrive  palms  with  delicate  leafage  clean, 
And  laurels'  tender  shoots  and  fair, 

Sprouts  of  the  olive  blest,  gray-green, 
That  Leto's  birth-pangs  dear  helped  bear ; 
There  swirls  the  lake's  disc  too  in  eddying  coils 

Whereon  the  sweet-voiced  swan  is  seen 
Who  for  the  muse's  sake  melodious  toils. 

The  religious -minded  Greek  might  well  attach  an  im 
portance  to  the  commonest  things  in  such  a  land  of  miracles 
as  Delos.  For  the  very  island  itself  had  been  originally  a 
wanderer  over  the  seas,  driven  forth  from  the  starry  heavens 
where  it  shone  in  the  beginning.2  When  the  twin  gods 

1  Iph.  Taur.  1089-1105. 

2  There  is  a  curious  confusion  of  geography,  astronomy,  and  myth 
ology  in  accounts  given — all  of  them  by  late  writers  such  as  Apollodorus, 
Callimachus  —  of  Delos  before  the  birth  of  Apollo  and  Artemis.     The 
learned  fabulist  Hyginus  has  gathered  them  together  ;  and  underlying  the 
whole  was  probably  some  genuine  religious  myth  at  which  we  dimly  guess. 
It  appears  that  according  to  one  current  version  Poseidon  hid  in  his  watery 
depths  the  floating  island  with  Leto  upon  it  while  the  grim  emissary  of 
Hera,  Pytho,  vainly  searched  for  her.      This  harmonises  with  the  tradition 
that    originally    Delos,    like    its    neighbour   Tenos,   was    a    possession  of 
Poseidon  (Strabo,  37,  Aeneid,  iii.   74).      But  now  comes  an  inexplicably 
barren   tale  which   seems  chiefly  to  exist  in  order  to  account  for  certain 
names  of  Delos,  and  which  most  effectually  hides  any  traces  which  may 
be    nvolved  in  it  of  genuine  popular  myth-making.     A  sister  of  Leto, 
Asteria,  in  order  to  escape  from  the  love  of  Zeus,  is  changed  into  a  quail 
and  finally  into  Ortygia  or  quail  island.     Asteria  was  at  the  beginning  a 


vin  DELOS  A  WANDERING  ISLAND  359 

were  born  the  holy  Cyclades,  motionless  till  then,  danced 
for  very  joy  around  Delos,  while  the  holy  island  itself— 
always  a  wanderer  over  seas  till  then — was  made  to  stand 
still.  "  For  in  the  foretime  it  was  a  wanderer,"  says  Pindar,1 
"  at  the  mercy  of  every  dashing  wave  and  of  the  whirling 
winds ;  but  the  daughter  of  Coeus  set  foot  upon  it  wild 
with  shooting  pains  the  forerunners  of  her  deliverance  ;  then 
it  was  that  on  earth's  stablished  foundations  four  upright 
pillars  arose,  and  upon  their  capitals  columns  that  were 
footed  in  adamant  held  firmly  aloft  the  rock2  where  she 
first  in  her  travail  had  sight  of  her  blessed  progeny." 
Meanwhile  the  islands  centered  round  about  Delos,  the 
chosen  Cyclades,3  went  through  their  dance,  and  swans 
from  Asia  went  singing  seven  times  around  the  holy  island, 
and  then  the  babes  were  born.  In  memory  of  these  joyous 
circlings  of  the  swans  before  his  birth,  Apollo  afterwards 
set  seven  strings  upon  his  lyre.4  Because,  moreover, 
Leto  had  respite  from  travail-pangs  on  the  seventh  day  of 
the  month,  that  day  was  made  holy.5 

star  in  the  sky,  and  her  two  names  were  merged  with  herself  in  the  floating 
island.  When  Apollo  and  Artemis  were  finally  born,  their  birth-place 
Asteria-Ortygia  became  Delos  for  insufficient  reasons  more  than  sufficiently 
dwelt  upon  by  Callimachus  in  his  Hymn  to  Delos. 

1  Quoted  by  Strabo,  p.  485. 

'2  Virgil,  Aeneid,  iii.  73  ff. ,  gives  an  alternative  account  of  the  early 
career  of  Delos  which  amounts  to  much  the  same  as  Pindar's,  only  in  place 
of  pillars  chains  fasten  the  island.  Its  anchors  are  Myconos  and  Gyaros. 
In  common  with  Myconos  the  two  sacred  isles  Delos  and  Rhenea  are  of 
granite.  Hence  the  notion  that  in  the  record  (i)  of  Delos  as  an  island  not 
originally  fixed  in  its  place  like  others,  but  left  a  wanderer  (Callimachus' 
Hymn  to  Delos,  30  ff.  and  92  f.),  (2)  of  Delos  as  being  hidden  under  water 
till  Zeus  or  Poseidon  drew  it  out  (see  Etym.  Mag.,  s.  v.  A^Xoj),  we  have 
a  mythological  account  of  the  comparatively  recent  and  volcanic  origin  of 
Delos.  If  this  way  of  treating  the  myth  could  only  be  deemed  reasonable 
it  would  have  a  certain  value  for  geologists  which  it  now  entirely  lacks. 

3  See  Appendix  IX.  on  the  Cyclades. 

4  Callim.  Hymn  to  Delos,  300  and  ff. 

5  Hesiod,  \Vorks  and  Days,  770  and  f. 


360  APOLLO  AT  DELOS  vin 

In  thinking  of  Delos,  the  first  importance  was  given  by 
religious  Greeks  to  a  feeling  that  the  birth  of  Apollo  and 
Artemis  made  the  island  especially  pure,  and  in  order  to 
preserve  that  quality  without  stain  it  was  in  time  enacted 
by  a  gradual  process  of  evolution  in  the  idea  of  what  purity 
required  (somewhat  accelerated  and  much  discredited  by 
political  circumstances)  that  there  should  be  no  birth,  no 
death,  and  no  burial  upon  the  holy  isle.1  Because  Leto 
and  her  children  had  communicated  to  Delos  by  contact 
the  inherent  virtues  indwelling  in  them,  it  was  meet  that  all 
occasions  of  contamination  from  mortals  at  the  supreme 
physical  moments  of  birth  and  death  should  be  removed.2 
Delos  itself  was  in  the  end  so  over-jealously  guarded  from 
the  contaminations  of  birth  and  death,  that  the  dying  had 
to  be  moved  across  the  sheltered  narrows  to  Rhenea  before 
they  quite  gave  up  the  ghost,  and  anticipated  births  were  at 
last  so  regulated  as  to  take  place  also  upon  that  adjacent 
island.  These  regulations  for  defence  against  contamina- 

1  There  was,  furthermore,  a  curious  provision  that  no  dogs  be  allowed 
on  Delos  (Strabo,  486). 

2  A  certain  help  in  gaining  the  Greek  point  of  view  may  be  derived 
from  the  fourth  chapter  of  Mr.  Frazer's  Golden  Bough,  where  many  curious 
customs  really  far  removed  from  the  Greeks  are  described.      In  some  of 
these  the  view  of  supreme  physical  moments  here  suggested  may  be  found. 
The  notion  that  there  is  a  taint  derived  from  the  presence  of  death  has 
not  survived  with  any  religious  sanction,  though  the  Old  Testament  is  full 
of  it.      Perhaps  though  a  touch  of  it  has  lingered  to  give  special  point  to 
Me"rime"e's  saying  that  he  came  near  dying  while  calling  at  the  house  of  a 
person  with  whom  he  was  not  sufficiently  intimate  to  justify  his  taking 
that  liberty.     The  taint  derived  from  proximity  to  a  birth,  however,  was 
quite  as   much   provided  against  by   the  priests   of  the  medieval  church 
as  by  those  of  Apollo  and  Aesculapius.      That  the  same  view  was  enter 
tained  under  the  Jewish  dispensation  is  noted  in  our  own  Milton's  touching 
lines, 

' '  Methought  I  saw  my  late  espoused  saint 
Brought  to  me  like  Alcestis  from  the  grave,    .    .   . 
Mine,  as  whom  washt  from  spot  of  child-bed  taint 
Purification  in  the  old  law  did  save." 


vin  THE  CONSECRATION  OF  RHENEA  361 

tion  seem  to  have  passed  from  Uelos  to  the  Epidaurian 
shrine  and  precinct  of  the  Apolline  god  Aesculapius.  In 
imperial  times  certainly  there  was  constructed  outside  of 
the  Epidaurian  precinct  a  place  of  refuge  called  the  house 
of  birth  and  of  death. 

But  before  Rhenea  could  become  a  sort  of  second 
Delos,  and  take  upon  itself  contaminations  for  the  sake  of 
leaving  the  Holy  Island  pure,  an  ideal  consecration  was 
required.  In  the  record  of  this  may  be  found  another  and 
real  though  not  logical  proof  that  in  Delos  resided  a  virtue 
of  inherent  purity,  greater  perhaps  than  that  attributed  by  the 
faithful  to  the  shrines  of  most  powerful  saints.  The  record 
in  question  is  of  a  gift  that  Polycrates  made  to  Delian  Apollo.1 
Polycrates,  among  his  many  proverbial  strokes  of  good  luck, 
hit  upon  the  idea  of  dedicating  to  Apollo  at  Delos  the 
island  of  Rhenea  which  he  had  conquered.  He  decided  to 
offer  the  island  in  just  the  same  way  that  he  would  have 
chosen  for  offering  a  statue  or  a  piece  of  furniture  to  the 
temple.  How  then  was  his  island, — a  small  island  it  was,  to 
be  sure,  but  still  at  least  four  times  the  size  of  Delos, — how 
was  Rhenea  to  be  offered  up  at  the  shrine  of  Apollo  ? 
The  universally  prevalent  religious  conception  of  Delos  as 
the  purest  of  all  spots  helped  him  to  the  way.  He  stretched 
across  the  not  very  deep  roadstead  between  Rhenea  and 
Delos  cables  or  chains  of  iron,  and  brought  about  through 
the  500  yards  of  water  that  intervened  a  contact  of  the 
impure  with  the  pure.  Thus  the  greater  was  contained  by 
the  less,  thus  the  island  now  called  Megale  Dili,  Great 
Delos,  was  made  for  certain  purposes  a  part  of  its  smaller 
neighbour.  It  was  a  very  fortunate  stroke  for  Polycrates, 
since  the  world  was  amazed  and  delighted,  and  at  the  same 
1  Thuc.  iii.  104. 


362  APOLLO  AT  DELOS  vin 

time  was  informed  that  the  new  master  of  the  island  of 
Samos  had  made  himself  the  best  friend  of  Apollo  at 
Delos. 

The  most  spiritual  bond,  however,  that  ever  attached 
a  whole  community  to  the  distant  shrine  of  a  god  was 
that  which  bound  Athens  to  Delos,  during  the  absence 
of  the  holy  ship  on  its  yearly  visit  to  the  island.  Dur 
ing  that  month  all  Athens  was  consecrate  to  Apollo, 
and  thus  a  month  was  added  to  the  life  of  Socrates. 
No  death  could  be  inflicted  by  the  state  while  it  was  in 
official  congress  with  the  Holy  Island,  for  this  would  blot, 
through  indirect  contact,  the  perfect  purity  of  Apollo's 
home.1 

The  early  history  of  Delos  is  a  chapter  of  lustrations  and 
purifications.  Pisistratus  of  Athens  devoted  his  efforts  in 
the  archipelago  to  a  removal  of  all  tombs  that  were  within 
sight  of  Apollo's  temple.2  A  century  later,  in  426  B.C.,  the 
Athenians,  self-governed  at  last  and  not  ruled  by  any  tyrant, 
renewed  and  improved  upon  the  policy  of  Pisistratus.  In 
this  year  they  began  to  take  possession  of  Delos  for 
avowedly  religious  but  also  for  commercial  reasons,  and 
instituted  the  great  Delian  festival  called  the  Delia.  At 
this  time  all  mortal  remains  were  dug  up  and  removed  from 
Delos.3  Thus  even  at  Delos  it  was  only  gradually  that  the 
stringent  regulations  of  ceremonial  purity  were  developed, 
and  generations  of  increasing  stringency  were  required  before 
the  taint  of  births  and  deaths  was  secluded  from  Delos  and 

1  A  similar  sanctity  by  spiritual  contact  is   constantly  implied  in  the 
terms  in  which  the  Cyclades  are  spoken  of  by  various  authors.      In  return 
for  this  their  homage  glorified  Delos.     As  Strabo  says,  485  :  ^vdo^ov  5' 
eTToi-rjaav   avr^v   cu   Trepiot/ctSes   vTjffoi    /caXoi^ei/ai   Ki</cAd5es,    Kara  ri^ty 
Tre/ATTOucrai  §rnj.oala  dewpovs  re  /cat  Ovffias,  /cat  %o/30i)s  irapdtvwv,  Travrjyupeis 
re  tv  aurrj  crvvayovffai  /j.eyd\as. 

2  Thuc.  iii.  104.  3  Ibid. 


vin  THE  EXILE  OF  THE  DELIANS  363 

confined  to  Rhenea.  Moreover  the  harsh  rules  set  up  by 
Athens  for  her  own  purposes  were  a  real  violation  of  the 
spirit  of  Apolline  religion.1  The  climax  was  reached  four 
years  after  this  second  purification  of  Delos  by  Athens.  In 
422  B.C.  the  Athenians  declared  that  all  the  Delians  were  a 
source  of  contamination  to  the  island,  and  thus,  having 
previously  removed  the  dead,  they  now  drove  out  the 
living.2 

The  conduct  of  Athens  at  Delos  from  426-422  B.C.  did 
not  deceive  religious  Greeks,  who  saw  in  it  a  determination 
on  her  part  to  suffer  no  commercial  rivalry,  and  resented 
the  outrage  offered  by  the  Athenians  elsewhere  to  a 
sanctuary  of  Delian  Apollo.3  Apollo  himself  from  his 
Delphian  oracle  interfered  and  secured  the  restoration  of 
the  Delians  after  a  most  unhappy  period  of  exile.4  An  echo 
of  the  public  interest  taken  by  religious  Greece  in  Delos 
and  the  Delians,  an  interest  which  would  resent  their  un 
merited  wrongs,  is  found  in  the  Hecuba,  which  was  brought 
out  between  426  B.C. — the  date  of  the  establishment  of  the 
Delia,  and  422  B.C. — the  date  of  the  expulsion  by  Athens  of 
the  Delians  from  Delos.  The  captive  Trojan  maidens  thus 
sing  of  Delos  and  the  Delians  : 

"  Where  the  first  of  palms  that  grew,  and  the  laurel-tree 
shot  upward  holy  branches  of  tender  green  to  give  comfort 
to  blessed  Leto  in  her  travail-pangs  for  Zeus.  With  Delian 
maids  shall  I  there  join  in  praise  of  the  fillet  and  bow  of 
heavenly  Artemis  ?  " 5 

1   Sec  M.  Homolle's  article  "  Delia"  in  Daremberg  and  Saglio. 
-  Thuc.  v.  i. 

3  At  Delium  in  Boeotia  (Thuc.  iv.  89-100).      After  the  battle  in  424,  the 
Boeotians  organised  a  Delian   festival  at    Delium.      This    was    a   protest 
against  the  policy  of  Athens  at  Delos. 

4  Thuc.  v.  32. 

5  Hecuba,  457  ff.  ;   and  cf.  I  ph.  Taur.  1089-1105,  quoted  above. 


364  APOLLO  AT  DELOS  vm 

Great  as  was  the  interest  of  Euripides  in  the  Delians  and 
their  traditional  observances,  he  yielded  to  none  in  a  sense 
that  Delos  must  be  perfectly  pure.  This  abundantly 
appears  in  the  Ion,  where  Creiisa  cries  out  in  the  temple  at 
Delphi,  making  appeal  from  injustice  to  the  holiness  of 
Delos.  She  thinks  that  Delphian  Apollo  has  grievously 
wronged  her,  and  believes  momentarily  that  he  has  no  pur 
pose  of  reparation :  therefore  she  cries  aloud  against  him. 
But  even  under  these  circumstances  the  genuine  piety  of 
her  heart  did  not  escape  the  poet.  What  more  true  and 
more  touching  proof  of  it  can  be  found  indeed  than  the 
closing  words  of  Creiisa,  whereby  the  harshness  and  horror 
of  the  tale  she  has  just  told  of  Apollo's  wrongdoing  are  made 
but  another  means  of  proclaiming  the  beauty  of  holiness  that 
guards  his  sacred  island  ? 

"Delos  abhors  thee,  she  cries;  even  the  laurel-shoot 
growing  close  to  the  palm-tree  of  delicate  leafage,  where  in 
the  pangs  of  her  holy  travail  Leto  brought  thee  to  birth  for 
Zeus."  l 

This  amounts  very  nearly  to  a  worship  of  the  island 
itself,  and  certainly  Euripides,  by  a  magnificently  religious 
exaggeration,  has  here  gone  farther  even  than  the  celebrated 
apostrophe  of  Delos  by  Pindar,  who  addressed  the  island 
as  follows  : 

"Hail!  thou  that  wert  stablished  by  a  god,  thy  upspringing 
was  most  longed  for  of  the  children  of  bright-vestured  Leto  ! 
Moveless  miracle  of  all  the  breadth  of  earth,  Delos  named 
by  mortals,  and  by  the  blessed  gods  called  the  far-gleaming 
star  of  darkling  earth." 

The  natural  beauty  which  flashes  from  Delos  in  these 
lines  of  Pindar  is  a  glittering  loveliness  shared  with  her  by 

1  Ion,  919  and  ff. 


N 


vin       ARCHILOCHVS  THE  POET  OF  THE  AEGEAN      365 

all  the  holy  Cyclades ;  and  like  the  holiness  which  brought 
worshippers  of  yore,  it  is  a  garment  bestowed  by  the  coming 
of  the  sun  god.  Untouched  of  sunlight  Delos  and  Rhenea 
with  all  the  twelve  that  circle  round  them,  white  Paros  and 
fertile  Naxos,  rugged  Myconos  and  Tenos  with  rocky  Syros 
— all  of  them,  untouched  of  sunlight,  seem  desolate  and 
drear  ;  but  let  Apollo  touch  them  with  the  arrows  of  his  day, 
and  they  are  then  like  flashing  prisms  that  he  has  set  upon 
the  sea,  or  blossom-like  they  glisten  white  on  blue,  and  the 
poet  .of  to-day  still  marvels  to  see  them  there — 

Lily  on  lily  that  o'erlace  the  sea. 

The  sea  and  the  sunlight  of  his  Aegean  home  entered 
long  since  into  the  heart  of  the  ancient  Aegean  poet  Archi- 
lochus  and  made  him,  even  more  than  Pindar,  capable  of 
worthily  setting  forth  the  strong  and  beautifully  awful  majesty 
of  Delian  Apollo.  Archilochus  chose  rather  for  his  theme, 
and  for  his  native  land,  the  whole  breadth  of  the  Aegean 
than  the  pent-up  Paros  where  he  came  to  birth. 

"  Away  with  Paros,  her  figs  and  fishy  life  !  " 1  he  cries,  and 
launches  his  barque  upon  the  sparkling  sea,  singing, 

Wood  makes  the  trough  to  knead  my  bread  withal, 
Wood  makes  the  cask  to  keep  Ismarian  wine, 
Wood  makes  the  deck  where  drinking  I  recline.2 

Although  the  poetry  of  great  Archilochus  has  disappeared 
with  the  exception  of  a  few  lines,  in  some  of  those  few  that 
remain  we  may  read  of  the  presence  in  his  heart  of  the 
glorious  Delian  sun -god  Apollo.  The  very  flame  of  swift 
power  in  the  death -dealing  sun,  pouring  ceaselessly  upon 
the  fervid  flanks  of  Delian  Cynthus,  or  making  the  while 

1  Bcrgk,  Frag.  51.  2  Ib'nL  3. 


366  APOLLO  AT  DELOS  vm 

sea-walls  of  Paros  glow  again  with  intolerable  heat,  was  felt 
by  Archilochus,  when  indignant  he  cries  : 

Sirius  will  flash,  I  hope,  most  fiercely  out, 
And  utterly  consume  them.1 

There  were  moments  in  the  life  of  this  man,  the  poet  of  the 
whole  Archipelago,  and  more  especially  of  that  greater 
Delos  formed  by  all  the  Cyclades, — moments,  I  say,  there 
were  in  this  poet's  life  when  the  intensity  of  his  scorn  swept 
him  away  as  completely  as  if  he  had  been  caught  in  the 
whirl  of  the  tides  of  his  native  seas.  At  such  times  he 
would  look  in  wrathful  and  untolerating  expectation  towards 
the  Holy  Isle  of  Delos  and  thence  invoke  Apollo,  the  dealer 
of  sudden  death  : 

Apollo,  take  the  guilty  ones  away, 

Unmask  them,  Lord,  and  slay  as  thou  canst  slay.2 

For  a  moment  the  largeness  of  such  noble  scorn  trans 
figures  Archilochus ;  the  inspiration  of  deep  utterance 
makes  him  for  one  instant  like  the  god  whose  power  he 
invokes,  but  it  lasts  but  the  space  of  a  moment.  Though 
our  poet's  anger  resembles  that  of  high  Apollo,  he  is  other 
wise  not  like  him.  Archilochus  never  dreamt  even  of  that 
chivalrous  regard  for  womankind  which  belonged  to  the 
ideal  godhead  of  Apollo. 

Let  us  now  examine  first  this  great  quality,  which  some 
times  tempts  one  to  call  Apollo  a  King  Arthur  of  the 
Greeks. 

Does  Apollo  the  giver  of  swift  and  painless  death  slay 
women  as  well  as  men  ?  Here  is  one  test  of  his  chivalry. 
Homer  may  be  called  in  for  the  first  witness,  he  knew 
1  Bergk,  Frag.  61.  2  Ibid.  27 


vin      APOLLO  A  KING  ARTHUR  OF  THE  CREEKS      367 

Apollo  first.  In  describing  Syrie,  commonly  identified  with 
Syros  one  of  the  twelve  Cyclades,  Homer  says  : 

When  it  cometh  to  pass  that  men  grow  old  in  that  country, 
The  god  of  the  silver  bow,  Apollo,  and  Artemis  with  him 
Suddenly  comes  upon  them  and  slays  with  shafts  that  are 
gentle.1 

Again  Apollo  and  Artemis  come  together  and  avenge  upon 
Niobe  and  her  children  the  blasphemous  outrage  uttered 
against  their  dear  mother  Leto.  The  two,  brother  and 
sister,  act  in  concert ;  Apollo's  shafts  strike  down  the  sons 
of  Niobe,  her  daughters  are  slain  by  the  unerring  aim  of 
Artemis. 

Sometimes,  we  hear,  these  twin  gods  dealt  destruction 
by  command  of  the  other  gods.  Sometimes  Artemis  acts 
alone,  as  in  the  case  so  bitterly  complained  of  by  Calypso 
when, 

Artemis  dread,  the  golden-throned,  on  the  island  Ortygia,2 
Suddenly  came  upon  him  3  and  slew  with  her  shafts  that  are 
gentle.4 

Apollo's  anger  against  the  guilt  of  his  love  Coronis,  soon 
to  be  mother  of  the  blameless  Aesculapius,  could  not  make 
him  raise  his  hands  against  her, — she  died,  "  slain,"  says 
Pindar,  "  by  golden  shafts  of  Artemis  shot  forth.  She  went 
from  her  chamber  down  to  Hades'  house  through  con 
trivance  of  Apollo."5  Thus  it  appears  that  Apollo  and 
Artemis  are  appointed  to  bring  a  sudden  death,  whether 
painless  and  peaceful — as  to  those  who  have  lived  right 
eously  through  a  long  life — or  a  destruction  whirled  on  the 
guilty  like  a  flash  from  the  death-dealing  anger  of  heaven. 

1  Odyssey,  xv.  409  and  ff.  2  Delos. 

3  Orion  the  lover  of  the  Dawn.  4  Odyssey,  v.  123  and  ff. 

5  Pyth. ,  iii.  8  and  ff. 


368  APOLLO  AT  DELOS  vm 

In  any  case  Apollo  gives  no  death  to  womankind,  in   all 
such  cases  his  sister  Artemis  intervenes. 

Other  Greek  divinities  are  not,  like  Apollo,  incapable  of 
bringing  death  to  women.  Furthermore,  Apollo's  conduct 
towards  his  various  loves  shows  often  in  other  ways  the 
true  aspect  of  chivalry.  The  love  of  Apollo  and  Daphne, 
the  story  of  Marpessa's  marriage,  both  represent  Apollo  the 
lover  submitting  to  be  scorned  and  rejected  with  no  thought 
of  revenge,  but  rather  (in  Daphne's  case  at  least)  with  a 
persistency  in  loving  which  foreshadows  the  more  complete 
ideal  of  devoted  chivalry.  As  a  lover  Apollo  is  not  angered 
by  refusal,  and  entire  rejection  does  not  end  his  suit. 
Though  Daphne  eluded  him  utterly,  he  takes  to  be  his  own 
the  laurel  tree,  which  was  peculiarly  hers,  since  it  was  she. 
This  laurel  became  one  of  the  symbols  of  purification  and 
purity  in  the  Apolline  ritual.  It  is  even  possible  that  the 
whole  myth  of  the  love  of  Apollo  for  Daphne  sprang  up  as 
a  link  to  join  a  form  of  native  and  immemorial  tree-worship 
to  the  later  and  higher  service  of  Apollo.  That  a  myth  of 
this  kind,  chiefly  intended  as  a  connecting-link,  should  come, 
by  the  way,  to  present  pictorially  so  high-minded  a  mood  of 
unsensual  and  unselfish  love  is  most  significant.  A  similar 
link  between  the  Thracian  Dionysus  and  the  native  Icarian 
tree-worship  was  that  of  the  suicide  of  Erigone  and  the 
mania  for  self-destruction  that  came  upon  all  the  maidens 
of  Icaria.  These  two  episodes  embody  the  popular  ap 
praisal,  so  to  speak,  of  Dionysus  and  Apollo.  Dionysus 
was  wrapped  in  a  mystery  of  cruel  horror,  he  drove  men  to 
nameless  destruction,  whereas  Apollo  was  the  noble-minded 
god,  shining  before  men  as  a  beacon  of  purity,  and  support 
ing  their  moral  weakness  by  his  own  sublime  and  undying 
strength. 


vni  APOLLO  AND  CYRENE  369 

Zeus  in  his  loves  forms  a  great  contrast  to  the  unsensual 
Apollo.  Zeus  transforms  himself  into  this  animal  or  that 
for  more  certain  and  secret  pursuit.  These  animals  are  an 
incarnation,  as  it  were,  of  the  carnal  impulse  in  the  god. 
Such  a  mark  of  sensuality  is  not  set  upon  the  love-pursuit 
of  Apollo.  Moreover  Apollo  is  represented  as  accepting 
the  duties  along  with  the  privileges  of  a  lover.1  The  re 
sponsibilities  of  fatherhood,  so  little  and  so  ineffectually 
heeded  by  Zeus,  come  home  in  all  their  fulness  to  Apollo. 

Many  stories  present  these  high  aspects  of  Apollo's 
commerce  with  men.  For  one  of  them,  his  appreciation 
of  those  charms  which  elude  perception  by  mere  sense, 
Pindar's  account  of  the  loves  of  Apollo  and  Cyrene  is 
perhaps  the  most  adequate.2  Wandering  in  the  woods, 
Apollo  comes  of  a  sudden  upon  Cyrene  engaged  in  a  hand 
to  hand  struggle  with  a  lion.  The  maiden  is  alone,  but 
unafraid.  The  god  calls  out  straightway  to  Chiron,  that 
gentlest  and  wisest  of  all  centaurs  :  "  Leave  thy  dread  cave, 
come  forth  and  marvel  at  this  great  prowess;  how  with 
mind  undaunted  the  girl  maintains  her  struggle,  showing  a 
spirit  that  towers  above  trials,  a  stedfast  soul  unstormed  by 
fear.  What  mortal  begat  her?  Plucked  from  what  tribe 
doth  she  dwell  in  the  hidden  nooks  of  the  shadow-flinging 
hills  ?  "  Truly  Apollo's  bearing  is  chivalrous  and  the  nobility 
of  passion  can  no  farther  go. 

In  the  answer  made  by  Chiron  to  Apollo's  questions 
about  Cyrene,  the  god  is  spoken  of  as  one  for  whom  it  is 
unlawful  to  have  part  in  a  lie,3  and  this  recalls  another 
place  where  Pindar  says  that  Apollo  sets  not  his  hand  to 
falsehood,4  and  also  various  passages  where  Plato  maintains 

1  See  the  Ion  of  Euripides,  passim.  2  Pyth.  ix. 

:i  Ibid.  42.  4  Pyth.  iii.  9,  \J/€v8tuv  oi'x  ^Tr 

2    B 


370  APOLLO  AT  DELOS  vin 

the  unerring  and  unswerving  truthfulness  of  the  god.1 
Truthfulness  is  also  among  the  cardinal  virtues  of  the 
Knights  of  the  Round  Table,  and  thus  again  is  the  chivalry 
of  Apollo  made  manifest.  Apollo  was  in  fact  bound  not 
to  come  near  to  falsehood  by  a  law,  a  Thesmos.  This  law 
was  an  obligation,  self-imposed  no  doubt,  but  yet  stronger 
than  any  other  impulse  in  the  god,  and  therefore  stronger 
than  himself.  He  would  not  speak  untruth  because  he 
could  not,  and  he  could  not  because  he  would  not.  This 
is  a  standard  which  suffers  nothing  by  comparison  with  the 
medieval  point  of  honour. 

Further  consideration  of  Apollo  as  the  infallible  speaker 
of  truth  by  his  oracles  would  lead  rather  to  his  Delphian 
than  his  Delian  shrine,  for  the  oracle  at  Delos  never  had 
great  influence  until  Alexandrine  times  2  when  oracles  began 
already  to  be  neglected.  Apollo,  we  are  told,  gave  oracles 
at  Delos  in  the  summer  season,  but  not  during  the  winter. 
Then  he  gave  answers  at  Patara  in  Lycia.  According  to 
still  another  account  Apollo  absented  himself  during  the 
winter  months  to  sojourn  among  the  Hyperboreans  in  the 
uttermost  parts  of  the  earth.  Kindred  stories  about  regular 
absences  of  the  god  from  Delphi  are  told  and  have  been 
dealt  with  in  the  opening  chapter  of  this  book.  Dismissing 
these  questions  without  going  more  deeply  into  them,  we 
may  briefly  add  that  the  two  great  Apolline  principles  were 
represented  by  his  two  great  shrines  at  Delphi  and  Delos. 
Apollo  was  everywhere  the  god  of  purity  and  truth,  but  his 
Delian  worshippers  had  first  in  their  minds  his  purity,  while 
men  flocked  to  Delphi  that  before  all  things  else  they  might 
hear  truth.  This  is  broadly  true  of  Apollo,  and  it  also 


/.  21  B,  Rep.  ii.  382  E  and  383  B. 
Callimachus,  Hymn  to  Delos,  vv.  1-5  ;  Virg.  Aen.  iii.  85-101. 


vin  SELF-DISCIPLINE  OF  APOLLO  371 

seems  true  to  say  that  in  many  local  stories,  notably  in 
those  of  Delphi,  the  god  is  represented  as  having  attained 
the  purity  of  his  heart  and  the  clearness  of  his  infallible 
mind  by  a  process  of  trial,  a  period  of  tribulation,  which  so 
regenerated  him  that  he  was  fitted  to  be  the  moral  pro 
tagonist  of  Olympus.1  There  was  a  commemoration  at 
Delphi  of  Apollo's  self-purification  after  he  slew  the  dread 
serpent  Pytho.  Apollo  was  condemned  in  another  story  to 
serve  a  mortal  for  nine  years  in  expiation  of  his  slaying 
the  Cyclops.  These  are  Apollo's  victories  over  himself, 
his  acts  of  submission  to  that  higher  and  self-imposed  thes- 
mos,  which  among  other  things  forbade  him — as  Pindar's 
Chiron  says — from  touching  falsehood.  Through  the  moral 
superiority  to  other  ideal  figures  in  Greek  mythology  thus 
achieved,  Apollo  was  enabled  to  possess  for  his  own  both 
Delphi  and  Delos,  where  other  gods  were  in  possession 
when  he  came.  For  instance,  Delos  apparently  belonged 
to  Poseidon  before  it  came  to  Apollo,  and  the  same  seems 
true  of  Delphi.  Actual  record  of  the  superiority  of 
strenuous  and  self- disciplining  Apollo  is  preserved,  first 
by  the  great  Delphian  motto  on  his  temple  "  Know  thyself," 
and  secondly,  by  legends  of  Apollo's  superiority  in  various 
strenuous  ways  to  other  gods.  The  contest  between  him 
and  Heracles  for  the  tripod  ended  in  a  compromise  where 
Apollo's  superiority  is  plainly  shown.  M.  Ronchaud 2  has 
indeed  most  admirably  said  that  "  Apollo,  when  he  has 
dealings  with  other  divinities,  always  shows  a  certain  moral 

1  See  chapter  iv.  of  C.  F.  Keary's  Outlines  of  Primitive  Belief.     Mr. 
Keary  says,  p.  191 :  "  The  history  of  the  development  of  Apollo's  character, 
then,  is  the  gradual  exaltation  of  his  nature  to  suit  the  growing  needs  of 
men.      In  the  Iliad,  though  Zeus  is  the  most  mighty  of  the  two,  Apollo's 
is  certainly  the  more  majestic  figure." 

2  Article  "  Apollo,"  in  Daremberg  and  Saglio. 


372  APOLLO  AT  DELOS  vm 

superiority.  His  standard  is  higher  than  theirs.  Poseidon 
was  his  fellow-labourer  in  the  building  of  the  walls  of  Troy, 
but  the  possession  of  Delphi  and  its  oracle,  originally  shared 
by  Poseidon  with  Earth,  passed  into  Apollo's  hands,  and 
Poseidon  was  dispossessed.  Apollo's  superiority  shows 
itself  also  in  the  Homeric  record  of  his  strife  with  Hermes. 
He  is  the  rival  of  Hermes  in  inventing  the  lyre,  and  wins 
the  day  over  him  in  a  race  at  Olympia.  Ares  himself  cannot 
withstand  Apollo  at  boxing,  and  as  for  the  insolent  Phorbas, 
Apollo  punishes  him  with  death.  Apollo  taught  Heracles 
the  use  of  the  bow^and  in  various  points  the  legends  of 
Heracles  run  parallel  to  those  of  Apollo.  But  Apollo 
stands  far  above  Heracles,  and  looks  down  from  the  heights 
of  his  divine  perfection." 

A  further  insight  into  the  nature  of  the  purifier  purified, 
of  Delian  Apollo  and  his  services  at  Delos,  will  be  gained 
by  looking  into  the  various  festivals  celebrated  on  the 
island  at  various  periods.  The  festival  in  honour  of  Apollo's 
birthday  was — like  the  flower  festivals  of  Dionysus — in  the 
spring.  Upon  the  sixth  of  the  month  Thargelion  began 
this  Apolline  Christmas  season,  of  which  a  glorified  record 
is  preserved  in  the  Homeric  Hymn  to  Delian  Apollo.  This 
may  be  summarised  as  follows  : — 

Apollo  has  many  sanctuaries,   "thick -growing  forests," 
"  high  points  of  far  outlook,"  1  "  high-standing  headlands,"- 
Delos,  though,  is  his  best  beloved  shrine. 

There  lonians,  trailing  long  robes,  are  wont  to  assemble. 
With  them  they  bring  the  wives  they  wedded,  and  children 

are  with  them. 
They  meanwhile,  on  combats  of  boxing,  singing,  and  dances, 

1  Such  is  the  temple  of  Apollo  at  Bassae. 


vin  THE  IONIANS  AT  DELOS  373 

Always  intent,  whenever  the  contest  begins,  are  contented, 
Seeming   predestined    to    live    thus    ageless    for    ever   and 
deathless. 

The  joy  of  the  occasion  lent  to  all  that  partook  of  it  a 
glamour  of  immortality  ;  the  stranger  standing  by  thinks  he 
is  seeing  the  "  Charm  of  the  World,"  and  rejoices  exceed 
ingly  at  sight  of  "goodly  men  and  women  with  beauteous 
girdles,"  of  swift  ships  and  abundance  of  treasures.  Finally, 
as  a  climax  to  his  description  of  the  festal  joy,  the  bard 
exclaims  in  wondering  delight : — 

Look  !  the  daughters  of  Delos,  handmaidens  of  him  the  far- 
darter, 

Singing  begin,  first  of  all  with  hymns  in  praise  of  Apollo, 

Of  Leto  next  they  sing,  and  of  Artemis  shooting  her  arrows. 

Heroes  of  old  they  praise,  and  the  glories  of  men  and  of 
women. 

The  tribes  of  mankind  are  enchanted  with  these  songs 
wherein 

All   men's   manner   of  speech,  the   Castanet's   rhythm   and 

motion, 
Well  they  counterfeit  all,  so  that  each  would  feel  sure  he 

had  spoken, 
Spoken  himself,  so  deftly  devised  are  their  songs  and  so 

sweetly. 

This  is  the  earliest  record  of  the  Ionian  festival  in  honour 
of  Delian  Apollo,  and  after  reading  this,  it  becomes  difficult 
to  believe  that  Apollo  was  of  purely  Dorian  origin,  as  some 
have  maintained.1 

A  later  testimony  to  this  earliest  festival  of  Apollo  at 

1  As  this  Homeric  Hymn  throws  a  welcome  light  upon  Delos,  it  is  satis 
factory  to  note,  before  turning  away  from  it,  that  inscriptions  found  on  the 
island  have  thrown  light  upon  the  questioned  authenticity  of  its  second 
part.  See  the  interesting  article  by  Gabriel  Daures  on  "  Excavations  at 


374  APOLLO  AT  DELOS  vm 

Delos  is  found  in  Thucydides  (iii.  104)  above  referred  to. 
From  him  we  learn  that  it  was  a  yearly  celebration,  and 
did  not  occur  every  fourth  year  like  the  festival  called  the 
Delia,  founded  at  Delos  by  Athens  in  426  B.C. 

In  fact  the  successful  attempt  begun  in  that  year  by  the 
Athenian  democracy  to  supplant  the  local  traditions  of 
Delos  by  closely  allied  Athenian  ones,  and  the  remodelling 
of  the  ancient  Apollonian  festival,  has  given  rise  to  much 
confusion  both  in  mythology  and  ritual.  It  would  be  useless 
to  attempt  now  a  disentanglement  of  such  confusion.  It  is 
enough  to  bear  in  mind  what  M.  Homolle  has  so  clearly 
pointed  out,  that  (i)  the  earliest  festivals  described  in  the 
Homeric  Hymn  were  quite  distinct  both  from  (2)  the 
Athenian  Delia  which  supplanted  them,  and  also  from  (3) 
the  festival  called  the  Apollonia,  which  was  an  attempt  to 
revive  something  like  them,  made  so  late  in  the  day  that 
all  the  traditions  that  surrounded  the  Homeric  bard  had 
died  away.  The  Apollonia  were  celebrated  during  the 
century  and  a  half  of  independence  enjoyed  by  Delos.  The 
last  stage  of  Delian  merry-making  begins  in  166  B.C.,  a  year 
memorable  for  the  final  reappearance  at  Delos,  under  Roman 
auspices,  of  Athenian  power.  Abundant  testimony  from  the 
end  of  the  second  century  B.C.  tells  of  the  second  revival  of 
the  Delia l  in  those  latter  days.  The  celebration  was  only 
in  part  religious  and,  as  had  been  the  case  with  every 
festival  ever  celebrated  at  Delos,  there  was  also  a  sort  of 

Delos,"  Nouvelle  Revue,  Sept.  1880.  M.  Homolle  found  an  inscription 
which  proves  that  the  last  half  of  the  hymn  was  called  Homeric  in  the 
third  century  B.C.,  not  only  popularly  but  by  a  competently  qualified 
judge. 

1  Thus,  to  summarise,  the  Delia  were  founded  in  426  B.C.,  suspended 
in  Lysander's  day  after  he  took  Athens,  until  386  B.C.  With  the  second 
sea-supremacy  of  Athens  they  were  revived  and  lasted  until  330  B.C. , 
then  came  the  revival  in  166  B.C. 


vin  VICISSITUDES  OF  DELIAN  FESTIVALS  375 

fair.1  But  apparently  the  commercial  spirit  finally  got  the 
upper  hand  in  these  latest  days  ;  and  Delos,  at  the  time  when 
Menophanes,  a  lieutenant  of  Mithridates,  brought  final 
destruction  upon  it,  was  largely  in  the  hands  of  enterprising 
Phoenicians,  and  Jews  who  knew  not  Apollo.  Delos 
became  more  famous  as  the  greatest  slave  market  in 
the  world  than  as  the  birth  -  place  of  Leto's  twin  chil 
dren.  Perhaps  it  was  some  sense  of  the  incongruity  of 
this  which  made  the  praises  of  Delos  so  irksome  a  theme 
to  Virgil. 

But  even  in  Virgil's  day  the  eye  of  faith  could  see  much 
that  was  beautiful  and  inspiring.  All  the  shrines  at  Delos 
had  been  deserted  for  years  before  Ovid  wrote  his  descrip 
tion  of  Candida  Delos  crowded  with  gifts  from  kings  and 
peopled  everywhere  with  votive  statues.  But  still  the 
splendour  of  Athenian  festivals  there  celebrated  irradiated 
the  memory  of  what  Delos  had  been.  And  religious-minded 
and  imaginative  men  remembered  more  vividly  the  glories 
of  the  remoter  past  when  the  sordid  and  commercial  thing 
that  Delos  had  become  was  swept  away.  Hence,  if  we  try 
to  gain  some  fragmentary  knowledge  of  the  manner  of 
Athenian  celebrations  at  Delos,  we  shall  but  do  (as  far  as 
may  be)  what  those  who  knew  the  island  in  the  day  of  her 
decay  and  destruction  were  prone  to  do. 

The  celebration  was  in  mid-spring,  before  May  was  one 
week  old,  lasting  at  least  two  days, — the  first  was  Apollo's 
birthday,  the  second  was  that  of  Artemis.  For  this  festival 2 

1  Tenos  has  reproduced  in  modern  Christian  times  many  of  the  features 
of  ancient  festivals  at  Delos.     Twice  a  year  the  Evangelistria  draws  crowds 
who  come  to  pray,   and  many  of  whom  stay  to  buy  and  sell.     These 
Tenian  festivals  were  begun  towards  the  end  of  the  first  quarter  of  this 
century. 

2  I  am  speaking  of  the  quadrennial  festival ;  there  were  minor  celebra 
tions  on  the  "off  years." 


376  APOLLO  AT  DELOS  vin 

extensive  preparations  were  made  at  Athens.  Choruses 
were  trained,1  deputations  were  chosen  and  qualified, 
victims  were  bought  and  put  into  a  good  condition  for 
sacrifice,  the  sacred  Delian  ship  was  made  in  every  way 
seaworthy,  and  the  signs  of  heaven  were  consulted  for  a 
favourable  time  of  departure. 

The  Delian  deputation  or  Theory — whose  members  were 
called  by  the  special  name  of  Deliastae  —  must  not  be 
confused  with  the  singing  choruses  of  youths  and  maidens 
that  accompanied  them.  The  whole  number  of  celebrants 
who  constituted  the  Athenian  contribution  to  the  quad 
rennial  festivities  at  Delos  would  therefore  be  very  large. 
When  Callias,  the  son  of  Hipponicus,  was  the  head  of  the 
deputation,  its  numbers  were  118.  Nicias  went  with  103. 
This  whole  array,  collectively  named  the  Delias,  had  to 
sail  from  the  Piraeus,  and  go  out  past  Sunium  into  the 
open  sea  where  began  the  Cyclades.  As  Attic  Sunium 
disappeared  in  the  distance,  they  passed  Ceos,  Cythnos, 
Seriphos,  and  Syros  on  the  right,  and  on  the  left  Andros 
and  Tenos.  When  they  came  in  sight  of  Myconos  to  the 
far  east,  they  stopped  short  of  it  and  ran  into  the  narrow 
and  sheltered  channel  between  Delos  and  Rhenea — the  gift 
of  Polycrates  to  Apollo.  In  the  beautiful  days  of  late  April 
and  early  May  they  would  have  a  glimpse  from  afar  of  Naxos 2 
and  Paros  in  the  southern  group  of  the  holy  Cyclades. 

1  The  choruses  from  Athens  were  renowned  for  beauty  and  for  artistic 
perfections,    Xen.   Mem.  III.   iii.   12.      The  Deliastae  —  a  committee  in 
special  charge  of  the  representation  of  Athens  at  the  Delian  festival — were 
chosen  from  the  Eumolpidae  and  the  Kerykes,  families  identified  with  the 
worship  at  Eleusis.     They  had  to  see  that  the  deputation  took  its  departure 
as  soon  as  favouring  signs  appeared  in  the  heavens.     The  Deliastae  were 
required   to    have  passed  a  year   of  probation,   say  in  the  Marathonian 
Delion.     For    references    and   further   details  see  M.    Homolle's  article 
"  Delia"  in  Daremberg  and  Saglio. 

2  For  views  of  Naxos,  see  Appendix  XI.  i.  84-86. 


vin  FROM  ATHENS  TO  DELOS  377 

Without  unfavourable  winds  the  journey  could  be  made 
in  four  days,  although  to  be  sure  the  ship  dedicated  to 
Apollo's  service  was  none  of  the  newest  or  swiftest.  This 
Uelian  boat  was  called  the  Theoris,  and  was  a  small  old 
fangled  craft,  with  thirty  oars,  a  triaconter,  kept  always  in 
the  very  best  repair.  The  very  faithful  tried  to  believe  it 
was  the  identical  boat  upon  which  Theseus  set  sail  for 
Crete,  vowing  to  found  a  festival  for  Apollo,1,  if  victory 
crowned  his  expedition.  Evidently  other  boats  were 
needed,  for,  apart  from  the  five  score  persons,  more  or  less, 
who  had  to  find  ship  room,  enough  victims  for  a  hecatomb 
had  to  be  transported  to  Delos.2  Often,  too,  the  high-born 
Athenians  who  wished  their  horses  to  compete  in  the  races 
at  the  Delian  festival 3  must  have  required  additional  trans 
port.  For  Delos  still  more  than  Ithaca  is  and  always  was 
"  a  place  for  goats "  rather  than  horses.  In  Delos  there 
were  no  wide  courses,  and  no  meadow  save  the  one  around 
the  sacred  lake.4  When  the  fleet  was  ready  in  the  harbour 
and  the  ministrant  who  made  sacrifice  daily  at  the  altar  of 
Apollo  gave  word  that  a  favouring  flash  of  lightning 
indicated  the  right  moment  for  weighing  anchor,  a  wreath 
was  set  upon  the  prow  of  the  sacred  Theoris  and  the 
pilgrims  sailed  away  with  appropriate  ceremonies  and  the 
singing  of  songs,  which  were  sung  all  the  way  to  Delos. 5 

1  Plutarch,  Theseus,  23,  and  cf.  Pausanias,  VIII.  xlviii.    3. 

2  The  accounts  for  377-374,  M.  Homolle  says,  cover  a  purchase  ot  109 
animals  for  sacrifice. 

3  The  Athenians  of  wealth  and  position  took  all  this  trouble,— at  the 
time  the  Delia  were  first  founded, — the  more  readily  if  the  Peloponnesian 
managers  of  the  Olympic  games,  being  Spartan  sympathisers,   are  sup 
posed  to  have  made  participation  in  the  races  at  Olympia  more  or  less 
disagreeable  for  them. 

4  See  Appendix  XI.  i.  71,  72. 

5  M.  Homolle  quotes  the  words  aSeis  uxrTrep  eis  A  77X01'  TrX/ajy,  Parocw- 
iogr.  (Gottingen),  p.  42. 


378  APOLLO  A  T  DELOS  vm 

Around  the  sacred  boats  sent  by  the  state  swarmed  many 
others  carrying  pilgrims  on  their  own  account,  and  some 
of  them  freighted  with  merchandise.  The  same  thing 
happens  to-day,  mutatis  mutandis^  when  the  faithful  sail 
to  Tenos  for  the  festival  and  fair  of  the  Holy  Evangel- 
istria. 

Perhaps  the  most  memorably  magnificent  of  all  these 
sacred  embassies  to  Delos  was  that  already  mentioned  de 
spatched  from  Athens  shortly  after  the  conclusion  of  the 
Peace  of  Nicias.  It  may  be  remembered  that  just  before 
this  time  the  repentant  Athenians  had  rescinded  their 
harsh  decree  of  banishment,  and  thus  the  Delians  had 
but  recently  returned  from  exile.  This  sacred  embassy 
was  headed  by  Nicias,  an  Athenian  whose  rather  mechani 
cal  piety  coupled  with  misfortunes  bravely  borne  made 
his  career  most  touching,  and  all  but  nobly  tragic.  In 
company  of  a  number  of  boats  bearing  the  youthful 
chorus,  and  of  other  craft  of  the  heavier  kind  for  freight, 
the  Theoris  brought  up  not,  as  in  former  years,  in  Delos 
proper,  but  across  the  way  on  Rhenea.  This  landing 
must  have  been  on  the  eve  of  the  great  Delian  festival. 
All  night  long  rumblings  and  voices  were  borne  over 
the  waters  to  the  listening  Delians.  In  the  morning  a 
gorgeous  pontoon  bridge  connected  Rhenea  with  the  Holy 
Island.  Across  it  came  the  procession  and  Nicias,  all  in 
festal  array,  and  wearing  golden  crowns.  After  the  bridge 
was  crossed  they  passed  northward  singing  all  the  way  to 
where  in  later  years  the  roadway  was  hemmed  in  by  the 
portico  of  Philip  (the  Fifth  of  Macedonia),  and  they  entered 
the  precinct  by  a  Ceremonial  Gate  (the  southward  gate 
built  by  the  Athenians).  After  this,  they  moved  northward 
still,  passing  to  the  westward  of  the  great  temple  of  Apollo, 


vin  THE  PROCESSION  OF  NICIAS  379 

and  also  of  the  smaller  temple  of  Leto  by  its  side.  The 
processional  way  here  as  at  Eleusis  was  lined  on  either  side 
with  various  statues  and  monuments,  and  these  were  more 
numerous  as  the  neighbourhood  of  the  temple  was  reached. 
After  joining  the  sacred  way  that  led  inward  from  the  larger 
North  Propylaea,  Nicias  came  to  the  ancient  temple  of 
Artemis  or,  as  some  think  of  the  Seven  Gods.1  Here  were 
deposited  the  laurel  crowns  of  gold  periodically  sent  to 
Apollo  of  Delos  from  Athens  during  Athenian  supremacy. 

After  the  procession  had  gone  the  rounds  prescribed, 
visiting  shrines  and  temples,  crowning  the  ancient  wooden 
statue  of  Aphrodite  left  at  Delos  by  Ariadne,2  who  had  it 
from  Daedalus,  its  fashioner,  the  second  stage  of  the  cele 
bration  was  reached,  the  sacrifice.3  The  hundred  kine 
decked  for  the  offering  and  with  their  horns  gilded  were 
sacrificed  on  all  the  altars  save  one  alone — the  bloodless 
altar  of  Apollo,  the  father  of  the  Ionian  race.  There  only 
first-fruits  were  offered,  and  no  doubt  thither  the  gifts  of  the 
mysterious  Hyperboreans4  found  their  way.  Then  came 
perhaps  certain  ceremonies  of  dedication  and  purification, 

1  The  uncertainty  about  this  temple,   the  existence   and  approximate 
position  of  which  are  known  through  inscriptions,   is  due   to  subsequent 
building,  which  in  the  Middle  Ages  swept  all  its  foundation  marks  from  the 
face  of  the  rock.      Here  stood  perhaps  the  only  Christian  building  of  any 
pretensions  that  Delos  has  ever  possessed — some  sort  of  a  chapel  built  by 
the  Knights  Hospitallers,  which  had  in  turn  almost  completely  disappeared 
when  excavations  began  in  earnest  under  M.   Homolle.     He  found  here 
one  of  the  most  remarkable  statues  of  antiquity — representing,  let  us  say, 
Artemis.     Take,  if  you  will,  a  very  tall  tombstone  and  round  off  its  angles 
and  corners  till  its  form  is  nearly  cylindrical.      Divide  the  result  into  three 
parts,  not  equal,  but  nearly  so,  one  for  the  legs,  one  for  the  body,  one  for 
the  head.     Something  should  be  done,  no  matter  what,   about  the  arms. 
Then  your  Delian  statue  of  Artemis,  not  beautiful,  but  most  ancient,  is 
complete.      Brunn  has  published  it  in  his  Denkmiiler. 

2  Pausan.  IX.  xl.  3  and  4  ;  Callimach.  Hymn  to  Delos,  308  ff. 

3  Plutarch,  Nicias,  3. 

4  Mannhardt    has  an  ingenious  theory  that  they  were   Thracians  or 
Macedonians. 


380  APOLLO  AT  DELOS  vin 

after  which  the  games  began — contests  of  physical  strength 
and  skill,  horse  racing,  and  musical  competition.  Every 
thing  which  Athens  could  do  was  done  here  in  Apollo's 
honour,  with  all  the  more  splendour  because  Athens  was 
solely  responsible  for  the  festival.  Nicias  saw,  no  doubt, 
the  older  temple  of  Apollo,  supplanted  later  on  by  the  one 
whose  ruins  we  contemplate  to-day.  That  temple,  like  its 
successor,  fronted  eastward  and  therefore  away  from  the 
channel  and  towards  Mount  Cynthus.  Between  its  back  or 
western  end  and  the  sea  stood  in  the  day  of  Nicias  the 
colossal  statue  of  Apollo  dedicated  by  the  Naxians.  This 
figure  towered  up  to  a  height  of  twenty-four  feet,  and  the 
ancient  Nicias  seems  to  have  been  deeply  moved  by  its 
god-like  proportions.  That  he  might  vie  with  the  ancient 
Naxian  worshippers  of  Apollo,  he  gave  money  before  he  left 
for  setting  up  a  colossal  palm-tree  of  bronze  by  the  side  of 
their  colossus.  This  was  done,  and  the  two  stood  there  for 
many  a  year  side  by  side.  So  close  were  they,  indeed,  that 
there  was  a  great  disaster  generations  afterwards  when  a 
windstorm  swept  across  from  Tenos,  where  Aeolus  himself 
was  housed  on  Mount  Cycnias.  The  too  blustering  Aeolus 
visited  Apollo's  precinct  rudely,  and  caused  the  brazen 
palm-tree  to  crash  down  against  the  statue  of  Apollo  and 
overturn  it  utterly.  Such  was  the  unlucky  gift  given  by  one 
who  at  the  time  considered  himself  no  doubt  the  luckiest  of 
men. 

But  the  pious  Nicias  did  more  than  this  :  he  gave  ground 
to  the  temple  worth  10,000  drachmae.  He  provided  that 
the  income  from  this  should  be  spent  for  celebrating  what 
sounds  very  much  like  a  mass  for  his  soul.  He  stipulated 
for  a  yearly  sacrifice  and  an  accompanying  entertainment  at 
which  the  Delians  were  to  pray  for  the  gods  to  grant  an 


vui  THE  EARLIEST  DEL1AN  FESTIVAL  381 

abundant  good  fortune  to  Nicias.  The  time  came  only  too 
soon  when  Nicias  —  overwhelmed  by  the  incessant  agonies 
of  the  most  painful  and  incurable  of  diseases  —  needed 
sorely  all  these  prayers  and  more.  A  hopeless  invalid,  he 
died  under  the  most  tragical  circumstances,  slain  by  decree 
of  the  Syracusan  mob,  and  at  Athens  his  pious  name  was 
execrated. 

Such,  in  outline,  was  the  manner  of  honouring  Apollo 
used  by  the  Athenians  during  the  various  periods  of  assured 
supremacy  when  Athenian  officers  took  from  the  Delians  all 
control  over  the  concerns  of  Delian  worship.  Such  was  the 
Athenian  festival  called  the  Delia  and  first  celebrated  in 
426  B.C. 

For  the  earliest  festival  we  are  constrained  to  fall  back 
chiefly  upon  the  pleasing  picture  of  the  Homeric  Hymn  to 
Delian  Apollo.  A  sufficient  account  of  this  has  been  given 
above.  There  are  also  some  verses  of  Theognis  l  more  or 
less  plainly  referring  to  it.  Aside  from  this  we  only  know 
a  detail  here  and  there  about  ancient  images,  and  remark 
able  practices  associated  with  a  fabulous  and  more  or 
less  prehistoric  past.  There  is  the  story  given  by  Aris- 


;,  avrbs  /j.ev  etrvpyiiMTas  irb\iv 
'A\Ka66(f}  n^XoTros  iraidl 
avrbs  5£  arparbv  vfipurrriv  M^ 
rrjade  TrbXevs,  iva  crot  Xaoi  i 


Kiddpyis  yd'  eparrj  daXtrj 
T€  ff 


Bergk,  773-779.     The  last  four  lines  are  thus  translated  by  Frere  : 

So  shall  thy  people  each  returning  spring 

Slay  fatted  hecatombs  ;  and  gladly  bring 

Fair  gifts  with  chaunted  hymns  and  lively  song, 

Dances  and  feasts,  and  happy  shouts  among  ; 

Before  thy  altar,  glorifying  thee, 

In  peace  and  health,  and  wealth,  cheerful  and  free. 


382  APOLLO  A7^  DELOS  vm 

totle x  of  Pythagoras'  visit  to  Delos,  wherein  we  hear  that, 
passing  others  by,  he  made  sacrifice  upon  the  altar  of  Apollo 
Genetor  or  Patroos,  where  the  shedding  of  no  blood  was 
tolerated.  We  cannot  know  what  may  have  been  the 
Delian  version  of  the  first  appearance  at  Delos  of  the 
ancient  wooden  image  of  Aphrodite,2  because  the  Athenian 
tale  has  alone  survived,  making  it  a  gift  from  Ariadne  when 
she  passed  that  way  with  Theseus.  That  the  ante-Athenian 
phase  of  worship  was  represented  by  many  curious  and 
clumsy  works  of  most  primitive  art  is  abundantly  shown  by 
the  seven  ancient  images  discovered  by  excavation  near  the 
Delian  shrine  of  Apollo.  Of  the  most  remarkable  of  these 
mention  has  been  made  above.3  Furthermore  it  appears 
that  the  mother  goddess  Leto  was  represented  in  her 
Delian  temple  by  a  wooden  idol  more  grotesquely  in 
adequate,  according  to  later  notions  of  Greek  art,  than  M. 
Homolle's  famous  Artemis.  That  the  art-critics  of  later 
times  had  much  to  say  on  this  score  appears  from  the 
following  anecdote  quoted  by  Athenaeus  4  from  Semus, — a 
Delian,  the  loss  of  whose  writings  has  no  doubt  deprived  us 
of  much  information  about  Delos  and  Delian  Apollo. 

Parmeniscus,  "the  man  who  never  laughed,"  consulted 
the  Boeotian  oracle  in  the  dark  cave  of  Trophonius  that  he 
might  in  some  way  break  the  spell.  The  god  gave  him 
answer  hexametrically  : 

Go  to  mother  at  home,  honour  her  with  exceeding  great 
kindness. 


1  Ar.   Fragm.   447,   quoted  by  Diog.    Laertius    and  lamblichus,   who 
gives  it  twice. 

2  For    the    curious    outline   of    Aphrodite's    history   at    Delos,    see 
Appendix  X. 

3  See  above,  note  i,  p.  379.  4  xiv.  614  B. 


vni  THE  TEMPLE  AND  IMAGE  OF  LETO  383 

The  obedient  Parmeniscus  sought  out  his  mother, 
and  amazed  her  by  his  unusual  attentions.  The  effect 
of  all  this  was  apparently  to  make  them  both  more 
hopelessly  solemn  than  ever.  Not  too  many  years  after 
wards,  our  solemn  friend  came  to  Delos,  —  as  Aegean 
voyagers  from  west  to  east  always  did  in  those  days. 
While  going  the  pious  rounds  on  the  island  Parmeniscus 
worshipped  in  the  Letoon  or  Temple  of  Leto,  next  that  of 
Apollo.  Accustomed  to  the  more  knowing  art  in  vogue  at 
his  native  Metapontum,  he  was  not  prepared  for  the 
wooden  idol  which  represented  Apollo's  mother  in  this 
ancient  shrine.  Therefore  on  catching  sight  of  the  idol  his 
devotions  were  interrupted  by  uncontrollable  fits  of  laughter. 
Leto,  Apollo's  mother,  was  the  mother  whom  he  had  been 
commanded  to  "  honour  with  exceeding  great  kindness." 

Aristotle  has  quoted  in  his  Ethics  l  the  curious  inscription 
which  met  the  eyes  of  worshippers  at  this  temple  of  Leto, 

Righteousness  the  noblest  is  ; 

Health  is  better  than  the  best  ; 
Sweetest  though,  of  all  that  is, 

Is  getting  what  you  love. 

These  words,  if  connected  with  Leto,  who  was  "a  com 
forter  from  the  beginning,  the  most  soothing  presence 
of  all  on  Olympus,"  must  be  expressive  of  her  typical 
attitude.  Like  an  indulgent  mother  with  her  children, 
she  chiefly  wished  that  men  should  have  their  heart's 
desire.  Attached  to  the  personality  of  the  goddess  of 
self- devoted  and  uncomplaining  love,  they  gain  a  new 

1  Eudemian  Ethics,  at  the  beginning  of  the  first  book,  and  also  toward 
the  end  of  the  eighth  chapter  of  the  first  book  of  the  Nicomachean  Ethics. 
In  the  former  place  it  is  described  as  on  the  Delian  Letoon,  in  the  latter 
simply  as  TO  A^Xia/cov  e7r£-ypa/z/ua. 


384  APOLLO  AT  DELOS  vin 

and  a  higher  meaning,  and  with  Leto's  life  to  point 
their  moral,  are  full  of  a  religious  significance.  Therefore, 
it  ceases  to  surprise  us  to  find  Aristotle  twice  quoting 
the  saying,  and  also  we  can  better  understand  why  this 
same  sentiment  without  substantial  variation  occurs  among 
the  maxims  of  Theognis,1  a  poet  to  whose  imagination 
the  Delian  myth  of  Leto  strongly  appealed.2  Let  the 
motto  of  the  Delian  Letoon  be  an  offset  to  the  Delphian 
motto,  "  Know  thyself,"  and  temper  its  too  exclusively  in 
tellectual  bias.  Here  too  is  the  needed  contradiction 
(always  given  by  the  heart  to  the  head)  of  the  other 
Delphian  motto  "  Nothing  too  much."  The  mother  goddess 
Leto  was  given  over  to  unmeasured  love,  and  the  affection 
which  her  love  inspired  in  her  children  Apollo  and  Artemis 
was  the  best  proof  that  you  can  never  go  too  far  in 
love. 

As  a  touching  mark  of  the  spirit  of  grateful  affection  and 
simple  trust  which  could  find  expression  at  Apollo's  Delian 
shrine,  the  inscribed  gift  picked  up  there  by  Professor 
Ulrichs  is  very  precious.  It  was  a  cheap  leaden  quiver, 
the  gift  of  shipwrecked  sailors  who  had  come  near  to 
starvation,  upon  which  was  stamped, 

For  by  these  we  were  rescued  from  starving. 

Such  indications  as  a  discovery  of  this  kind  gives  of  the 
living  spirit  of  religion  working  in  men's  hearts  and  guiding 
their  lives  are  of  far  more  real  importance  than  the  record 
preserved  by  Callimachus  and  others  of  the  building  of 
an  altar  by  Apollo  with  materials  sought  on  Mount  Cynthus 
by  his  sister  Artemis.3  It  was  made,  we  are  told,  entirely 

1  Bergk,  255  and  f.  -  See  above,  note  i,  p.  381. 

3  Callimachus,  Hymn  to  Apollo,  58-63. 


viii  THE  CYNTHIAN  CAVE  TEMPLE  385 

of  horns  such  as  the  goddess  could  gather  among  the 
granite  boulders  of  that  eminence,  too  irreverently  designated 
by  the  traveller  Tournefort  as  a  colline  desagreable.  Another 
curiosity  which  attaches  itself  also  to  the  very  earliest  forms 
of  worship  used  before  the  dawn  of  known  history  is  far 
away  from  the  town  of  Delos  and  the  precinct  of  Apollo 
visited  by  Nicias.  For  lack  of  a  better  name  it  may  be 
called  the  "  Cave  Temple  of  Apollo," x  though  it  has  been 
called  the  "  Grotto  of  the  Sun  "  by  Burnouf  and  the  "  Cave 
of  the  Dragon  "  by  another,  and  more  prosaically  still  by  the 
unenthusiastic  Tournefort  "  a  stone  sentry  box."  This  cave- 
temple  is  formed  by  placing  aslant  against  each  other,  on  the 
top  of  a  natural  rift  in  the  solid  rock,  five  huge  and  rough- 
hewn  granite  slabs,  two  on  the  north  and  three  on  the 
south.  These  two  rows  rest  upon  and  are  crowded  against 
uneven  ledges,  rudely  fashioned  at  their  lower  edges  in  the 
side  of  the  reshaped  rift  or  gully  in  the  Cynthian  mountain 
side.  Their  upper  edges  meet  together  forming  the  point 
to  the  most  inartificial  seeming  gable.  This  appearance 
of  inartificiality  is  enhanced  by  a  heap  of  granite  boulders 
accumulated  on  the  sides  of  this  rude  pediment,  which 
seems  the  chance  product  of  nature's  workmanship.  This 
impression  is  increased  by  the  circumstance  that  a  huge 
spot  at  the  back  of  this  cave  was  probably  never  roofed 
over.  Within  this  strange  place,  half  natural  and  half 
artificial,  half  cave  and  half  temple,  were  found  by  Lebegue,2 
who  excavated  here  for  Burnouf  early  in  the  seventies, 
the  feet  of  what  he  thinks  must  have  been  a  fine  statue  of 
Apollo.  A  curious  stone  of  unshapely  form  upon  which 

1  For  photographs  of  this  and  of  various  Delian  scenes,  see  Appendix 
XI.  i.  The  Cyclades,  71-83. 

2  See  for  more  details  his  Recherche*  sur  Dtlos,  where  M.  Burnouf 's 
astronomical  points  are  also  presented. 

2  C 


386  APOLLO  AT  DELOS  vm 

these  feet  rested  is  talked  of  by  M.  Lebegue  as  undoubtedly 
a  Baetylus  or  fetish  stone,  which  according  to  this  view 
must  have  been  worshipped  here,  as  a  cone  was  worshipped 
in  Aphrodite's  ancient  Paphian  precinct.  In  front  of  this 
sentry-box  cave-temple  is  a  terrace  built  up  by  means  of 
a  most  carefully  constructed  wall.  This,  though  of  very 
early  workmanship,  is  of  much  later  date  than  the  cave- 
temple  itself.  Upon  this  terrace  there  are  traces  of  some 
thing  like  a  tomb  and  of  the  charred  remains  of  sacrifice. 
Also  a  stone  footing  for  a  very  large  tripod  appears  there, 
and  furthermore,  in  front  of  and  close  to  the  cave-temple 
itself  is  a  row  of  stone  bases ;  perhaps  these  were  for  small 
tripods. 

And  now,  having  given  as  full  an  account  briefly  as 
might  be  of  the  Athenian  Delia,  and  of  the  more  ancient 
observances  and  objects  of  worship  at  Delos,  what  yet 
remains  to  be  spoken  of  is  the  festival  of  Apollo  as  it  was 
celebrated  during  the  day  of  Delian  independence.  When 
Athenian  interference  ceased,  the  intellectual  leadership  of 
Athens  was  still  felt,  and  therefore  the  new  Apollonia, 
which  should  have  been  a  revival  of  the  most  ancient 
observances,  were  in  fact  almost  the  same  thing  which 
the  Athenians  called  the  Delia,  only  no  Theoris  came 
from  Athens.  The  four  crowns  of  gold  sent  periodically 
by  Athens  while  she  ruled  the  festival  were  sent  no  more, 
and  no  more  votive  gifts  were  made  to  the  god  by 
Athenians.  Delos  was  no  longer  beautified  by  Athenian 
architects,  such  as  those  whose  ruined  work  is  still  seen 
in  the  scattered  fragments  of  the  most  important  temple 
of  Delos.1  It  was  to  the  munificence  of  a  Macedonian, 

1  The  temple  was  built  about  a  century  after  422  (when  Nicias  headed 
the  Athenian  Delias),  and  certainly  before  315  B.C.  It  was  of  about  the 


vin          LA17ER-DA  Y  WORSHIPPERS  AT  DELOS  ;S - 

Philip  the  Fifth,  that  Apollo  owed  the  chief  addition  to  his 
splendid  buildings  made  during  the  era  of  Delian  in 
dependence.1  During  these  years  Delos  welcomed  as  the 
votaries  of  its  god  not  Athenians  but  merchants  from 
Eastern  Tyre  and  Sidon,  traders  from  far-off  Panticapaeum 
on  the  sea  of  Azof,  and  also  from  the  far  West  The 
part  in  Delian  affairs  played  by  the  islands  round  about, 
the  Cyclades,  was  also  greater  than  in  the  day  when  Athens 
ruled  supreme.  Indeed,  the  sanctuary  became  a  centre  of 
Aegean  affairs  as  well  as  of  Aegean  religion.  It  was,  as 
M.  Homolle  puts  it,  not  a  place  for  worship  only,  but 
equally  a  "  Recorder's  office  for  the  safe-keeping  of  important 
decrees,  a  sort  of  treasury  department  or  bank  for  the 
whole  Archipelago,  and  also  a  central  museum  for  the 
islands  of  the  Aegean."  2 

Meanwhile  festivals  under  the  name  of  Apollonia  took 
place  with  but  few  alterations  from  the  Athenian  programme 
at  the  Delia  which  they  supplanted.  There  were  all  sorts 

size  and  dimensions  of  the  Theseum  at  Athens,  it  had  almost  precisely 
the  same  width,  but  was  appreciably  shorter.  For  all  manner  of  archaeo 
logical  details  concerning  Delos,  it  is  a  pleasure  to  refer  to  so  charming 
a  book  as  that  just  published  by  M.  Ch.  Diehl  under  the  title  of  Excursions 
arth&logiqves  en  Gr*ce  (Paris,  1890).  M.  Diehl  has  written  just  the 
entertaining  and  untechnical  sort  of  book  which  was  required,  and  has 
been  extremely  happy  in  what  he  says  about  Delos. 

1  The  porch  of  Philip,  whose  massive  blocks  form  one  of  the  most 
imposing  (and  to  the  hasty  sightseer  impeding)  of  the  present  Delian  ruin- 
heaps,  was  built  toward  the  dose  of  the  era  of  independence,  circa  iSo 
B.C.  It  is  interesting  to  the  student  of  architecture,  because  in  it  is 
represented  the  last  and  worst  extremity  of  die  Dork  style.  Its  columns 
have  lost  the  strength  and  the  upbearing  swing  that  belong  to  those  of  the 
Parthenon,  and  are  transformed  into  dull  and  unprofitable  posts.  See 
the  instructive  account  of  this  whole  matter  in  the  Antiq ttifies  of  Ionia, 
Pan  IV.  1 88 1,  published  by  the  Dilettanti  Society. 

5  See  in  the  Bulletin  de  la  sotitU  de  ftyn&M  4*  t'cst,  1881  (i« 
trimestre),  M.  Homolle's  lecture  on  Delos  given  at  Nancy.  To  under 
stand  what  the  temple  was,  he  tells  us  to  combine  the  Musee  de  Cluny, 
a  garde-meuble,  a  cour  de  comtes,  the  Bank  of  France,  the  Credit  Fonder, 
and  the  Madeleine, 


388  APOLLO  AT  DELOS  vni 

of  foot  races,  the  stadion,  the  'diaulos,  the  dolichos,  the 
hoplite  race, — and  there  was  the  pentathlon.  Furthermore 
torch  races  were  indulged  in.  It  may  be  difficult  to  be 
sure  how  many  of  these  features  were  new,  and  how  many 
were  simply  in  continuation  of  the  established  Athenian 
programme.  It  may  not  be  possible  to  know  whether  the 
far-fetched  practice  of  horse-racing  introduced  by  the 
Athenians  at  Delos  was  maintained  by  the  islanders  when 
freed  from  Athenian  supervision.  Of  one  thing  though  we 
may  be  certain :  there  was  abundant  dancing.  One  has 
but  to  visit  the  Archipelago  in  carnival  time  to-day  to  see 
a  tolerably  good  reproduction  in  modern  surroundings 
of  what  has  always  characterised  Aegean  merrymakings. 
"At  Delos,"  says  Lucian,  "they  could  not  so  much  as 
make  sacrifices  without  dancing  and  music  as  well."  After 
that  he  proceeds  to  describe  the  measures  trod  at  Delian 
merrymakings  by  especially  chosen  dancers,  and  by  choirs 
of  picked  youths  trained  for  the  delectation  of  those  who 
resorted  to  Apollo's  festival.1 

The  Greeks  of  the  Archipelago  have  at  all  times  been 
under  the  spell  of  the  swaying  surges  of  ocean,  which  was 
the  background  of  their  home  life  and  home  joy.  These 
islanders  live  now  as  of  old  face  to  face  with  the  strongest 
moods  of  the  great  sea,  not  as  in  the  far  recesses  of  Venetian 
lagoons,  where  the  whims  of  ocean  are  moderated. 
Accordingly,  while  the  graceful  swerve  of  the  moderated 

1  This  spirit  of  the  ancient  dances  took  shape  before  my  eyes  in  a 
band  of  Syriote  peasants  whom  I  saw  dancing  in  front  of  a  village  church 
on  the  last  day  of  the  carnival.  There  they  were,  old  men  and  young 
men,  maidens  and  women  of  maturer  years,  all  merrily  dancing  and 
singing,  while  the  kindly  priest  whom  the  refrain  of  their  song  was  chaffing 
looked  on  contentedly.  I  recollect  particularly  the  hilarious  conduct  of  a 
certain  genial  and  one-eyed  villager,  Socrates,  who  danced  until  he  lost 
both  shoes.  For  a  band  of  Syriote  butchers  caught  and  photographed  in 
the  act  of  dancing  the  Caramanian  dance,  see  Appendix  XI.  i.  64. 


vin  THE  INHERITANCE  OF  DELOS  389 

pulses  of  the  sea  has  shaped  the  motions  of  Venetians, 
and  lent  a  swaying  outline  to  the  houses  men  live  in  at 
Venice,  in  the  Archipelago  the  very  quicksilver  rhythm 
and  instant  sweep  of  ocean  have  passed  into  the  limbs  and 
hearts  of  the  Greek  islanders. 

By  the  mark  of  good-humoured  merriment  and  kindly 
spirit  of  comradeship  we  know  under  its  modern  disguise 
the  ancient  spirit  of  Apollo's  Delian  festivals,  which  may  be 
studied  with  especial  advantage  when  the  faithful  gather  at 
Tenos.  Three  neighbouring  islands  cast  lots  as  it  were  for 
the  rich  vestment  of  sanctity,  splendour,  and  power  which 
magnified  Delos  of  yore  and  exalted  Delian  Apollo.  The 
period  of  storm  and  stress  which  issued  in  Greek  independ 
ence  built  up  at  Syra  l  a  great  commercial  port,  instituted 
at  Tenos  the  sanctuary  of  the  Evangelistria,2  and  finally 
assured  to  the  inhabitants  of  Myconos  the  possession  of 
untenanted  Delos  and  Rhenea. 

To  this  same  establishment  of  Greek  independence  we 
owe  the  opportunity  for  studying  the  worship  of  Apollo  at 
Delos  unmolested  of  Turks  or  other  wild  beasts.  To  the 
indefatigable  labours  and  the  wonderful  resources  of  the  first 
established  school  at  Athens  we  owe  the  abundant  material 
now  open  to  the  student  of  Delian  secular  and  religious 
antiquities.  The  head  and  front  of  the  latest  and  most  im 
portant  discoveries  there  is  M.  Theophile  Homolle.  To 
that  distinguished  scholar  we  owe  the  deepest  gratitude  for 

1  Ross  tells  of  a  scheme  for  diverting  the  trade  of  Syra  to  Delos  which 
was  seriously  considered,    but   finally   abandoned  early  in    this    century. 
See  for  the  best  account  of  the  growth  of  Syra,  Loukis  Laras,  by  D. 
Bike"las  (the  English  translation  is  by  Mr.  Gennadius).     For  pictures  of  the 
modern  life  of  the  Cyclades,  see  Appendix  XI.  i.  58-70. 

2  Agitation   for   building  the  church  where  the  miraculous  picture  is 
housed  began  about   1820,  just  at  the  time  when  the  picture  itself  had 
been  unearthed  in  accordance  with  the  dream  which  persistently  visited  a 
pious  nun  of  Tenos.      See  Appendix  XI.  i.  87. 


390  APOLLO  AT  DELOS  vin 

all  the  self-denying  enthusiasm  and  undiscouraged  perse 
verance  which  have  led  to  his  brilliant  results.  The  work 
which  shall  embody  and  bring  to  a  climax  all  M.  Homolle's 
studies  is  as  yet  incomplete,  but  he  has  already  done  so 
much  with  his  accumulated  material  that  scholars  declare 
him,  and  rightly,  to  be  one  of  the  few  foremost  archaeolo 
gists  of  our  day. 


APPENDIX  IX 
THE   CYCLADES   AND   SPORADES 

IT  is  difficult  to  understand  the  relation  borne  by  the 
various  islands  of  the  Aegean  to  the  worship  of  Apollo  at 
Delos  and  to  each  other,  without  going  into  the  history  of 
the  two  terms  Cyclades  and  Sporades.  Neither  of  these 
terms  was  known  in  Homer's  or  Hesiod's  day,  and  of  the 
two  the  term  Cyclades  was  not  only  the  first  to  be  used 
later  on,  but  was  also  far  more  widely  known  throughout 
antiquity.  The  word  Sporades  was  used  apparently,  so  far 
as  it  came  into  a  general  usage,  by  a  sort  of  analogy  to 
the  use  of  the  word  Cyclades.  It  covers  somewhat  vaguely 
the  smaller  islands,  too  insignificant '  either  to  be  named 
singly  and  stand  alone  or  to  be  classed  among  the  Cyclades. 
In  most  cases,  also,  the  islands  so  named  are  too  far  away 
from  Delos  to  make  the  term  Cyclades  possible  for  them. 
The  Cyclades  were  certain  illustrious *  islands  more  or  less 
accurately  described  as  centred  around  Delos  and  Rhenea, 
and  were  especially  favoured  by  Apollo,  who  colonised  them 
through  his  son  Ion ; 2  the  Sporades  were  small  islands  to 
whose  population  might  be  more  especially  applied  the 
Euripidean  line  (Rhesus,  701),  v^a-norrjv  a-rropdSa  KCKT^TCU 
piov.  They  were  in  no  sense  a  group,  but  were  scattered 
broadcast,3  and  each  one  was  so  small  that  any  life  but  an 

1  Theocritus,  Id.  XVII.  90  and  f.  2  Euripides,  Ion,  1571-1600. 

3  Etym.  Magn.  s.v.  SiropaSes  vrjffoi:  ?}  dia  rb  ff-rropddrjv  Kc'iaOa.i,  T)  dirb 

TOU  cnrapTbv  KO.I 


392  APPENDIX  IX. 

unsettled  and  vagrant  one  was  more  or  less  difficult  upon 
it.  As  the  islands  answering  to  this  description  were 
chiefly  toward  the  western  coast  of  Asia  Minor,  there  was 
a  tendency  from  the  moment  the  term  Sporades  came  into 
use  to  divide  the  islands,  as  they  are  now  definitely  divided, 
into  two  categories,  an  eastern  one  of  Sporades,  and  a 
western  one — which  alone  could  be  called  a  group — of 
Cyclades.  But  this  was  never  strictly  done  in  antiquity. 

In  the  Homeric  Hymn  to  Delian  Apollo 1  there  still  sur 
vives  a  certain  confusion  noticeable  in  the  Odyssey  between 
islands  and  such  peninsulas  as  were  in  any  literal  way 
almost  islands ;  therefore  it  would  be  vain  to  expect  from 
this  source  any  classification  of  islands.  The  same  imper 
fect  knowledge  of  islands,  as  such,  characterises  Hesiod. 
Undoubtedly  the  notion  of  a  group  of  Cyclades  around 
Delos  grew  up  after  the  day  of  the  Homeric  poems  and 
hymns,  and  its  use  was  at  first  primarily  determined  by 
religious  and  mythological  considerations  which,  being  in  a 
sense  the  reverse  of  geographical,  did  not  require  anything 
so  definite  as  a  catalogue  of  islands  to  which  the  name  was 
applied.  Herodotus  may  have  been  more  definite  and 
strictly  geographical  in  his  understanding  of  how  the  word 
applied.  He  says  nothing,  however,  to  encourage  the  idea, 
for  he  declares  that  none  of  the  Cyclades  were  subject  to 
Darius  (v.  30),  and  means  evidently  to  distinguish  them 
from  islands  toward  Asia  Minor,  whose  Ionian  inhabitants 
voluntarily  submitted  to  Persia  (i.  169).  He  had  no 
general  name  for  these  latter,  and  the  only  phrase  ot 
general  import  which  he  uses  for  all  the  islands  in  question 
is  "the  islands  in  the  Aegean"  (ii.  97),  where  it  may  per 
haps  be  claimed  that  the  context  suggests  chiefly  the 
Cyclades.  Thucydides,  on  the  other  hand,  evidently 
thought  it  important  to  be  rather  more  clear  in  his  own 
mind  about  the  islands  of  the  Aegean.  He  speaks  (ii.  9) 
collectively  of  the  Cyclades  and  the  islands  later  known  as 
the  Sporades  as  "  the  islands  inside  of  the  Peloponnesus 

1  vv.  30-45. 


THE  CYCLADES  AND  SPORADES  393 

and  Crete."  That  he  had  a  definite  list  or  group  of  islands 
which  he  designated  as  the  Cyclades  is  made  very  probable 
by  this  same  passage,  where  he  uses  the  phrase,  "all  the 
Cyclades  except  Melos  and  Thera."  For  the  islands  after 
wards  called  the  Sporades,  Thucydides  had  no  name  whatso 
ever,  and  may  so  far  forth  be  classed  with  the  early  poets  and 
Herodotus.  In  the  speech  of  Athena  at  the  close  of  the 
Ion,  we  have  finally  the  most  important  mention  of  the 
Cyclades  from  a  religious  and  mythological  point  of  view. 
There  they  are  plainly  mentioned  as  chosen  islands,  in 
whose  colonisation  and  civilisation  Apollo  Genetor  or 
Patroos,  the  patron  and  father  of  all  lonians,  was  imme 
diately  concerned.1  This  Apolline  consecration  of  the 
Cyclades  is  often  made  prominent  in  later  writers,  such  as 
Callimachus,2  and  there  is  a  trace  of  it  in  the  general  tone 
of  Herodotus,  who  distinguishes  the  Cyclades  from  com 
mon  islands  by  saying  that  they  never  became  subject  to 
Darius. 

In  the  days  of  the  organised  Roman  empire  all  this  was 
changed ;  it  became  imperative  to  have  clearly  defined 
geographical  terms,  and  therefore  the  religious  mystery  of 
vagueness  in  the  use  of  the  term  Cyclades  disappeared.  A 
list  of  Cyclades  was  made  out,  and  the  newer  term  Sporades 
was  applied  to  the  excluded  islands,  most  of  them  east  of 
Delos.  It  must  be  remembered,  however,  that  none  of  these 
were  important  and  well  known.  Rhodes,  Chios,  Samos,  and 
islands  of  that  ilk,  were  not  counted  either  as  Cyclades  or 
as  Sporades.3  A  learned  Spaniard  Hyginus,  and  the  accom- 

1  Ion,  1583.     For  the  point  here  made  it  is  not  necessary  to  suppose 
that  Euripides  used  the  word  KVK\ddas  as  a  proper  name,  though  I  incline 
to  think  he  does,  as  do  also  most  editors  of  the  play. 

2  Hymn  to  Delos,  300  and  ff. 

3  The  only  exception  to  this  would  be  the  name  of  firapxla  v/jcruv 
KvKXdSwv,   sometimes  given  to  Diocletian's  Insularum  provincia.     This 
was   one   of  seven  subdivisions  of  the  province  Asia,   and   included  53 
islands,     among    which   were    Rhodes,    Cos,    Samos,    Chios,     Mytilene, 
Methymne,  Tenedos,   Porselene,   Andros,   Tenos,  Naxos,  Paros,  Siphnos, 
Melos,    los,    Thera,    Amorgos,    Astypalaea.       See    Marquardt,    Romische 
Staatsverwaltung,  vol.  i.  p.  348  ;   cf.  also  note  2,  p.  397  below. 


394  APPENDIX  IX 

plished  Greek  geographer  Strabo,  both  of  them,  give  some 
knowledge  of  the  definite  and  purely  geographical  meaning 
which  in  the  days  of  Augustus  and  later  attached  itself  to  the 
two  terms  in  question.  Under  the  title  of  "  Insulae  maxi- 
mae,"1  Hyginus  begins  a  long  enumeration,  where  may  be 
found  Sicily,  Sardinia,  Crete,  Cyprus,  Rhodes,  Euboea, 
Tenedos,  and  Corsica.  As  his  very  last  item  he  gives  a  list 
of  the  Cyclades  : — "  Cyclades  insulae  sunt  novem,  id  est 
Andros,  Myconos,  Delos,  Tenos,  Naxos,  Seriphus,  Gyarus, 
Paros,  Rhene  "  (Rhenea).  The  first  point  noticeable  here 
is  that  the  only  two  islands  specifically  named  by  Thucy- 
dides  as  Cyclades,  Melos  and  Thera,  are  not  included  in 
Hyginus'  list.  Another  point  against  the  list  is  that  Delos 
and  Rhenea  appear  in  it,  and  thus  themselves  figure 
among  their  own  surroundings.  This  is  enough  to  dis 
credit  Hyginus'  list,  even  if  we  had  not  Strabo's.2 

Strabo  has  evidently  considered  the  whole  question  in 
all  its  bearings,  for  he  not  only  gives  his  own  opinion  in 

1  Fab.  276.  Curiously  enough  Mauritania,  Egypt,  and  Sicyon  are 
classed  here  among  insulae.  This  has  something  to  do  with  an  ambiguity 
in  the  Latin  word  insula,  which  applied  to  buildings  and  precincts  which 
were  definitely  marked  off  from  their  surroundings.  That  there  was,  how 
ever,  some  real  confusion  in  Hyginus'  mind  is  shown  by  his  phrase  for 
Egypt:  "  quam  Nilus  circumlavat. " 

'z  See  Theocritus,  Id.  XVII.  58  and  ff.,  where  there  is  a  curious  reproduc 
tion  of  the  Delian  birth  legend  of  Apollo.  Ptolemy,  whose  praises  the 
poem  sings,  was  born  on  Cos,  which  comparatively  unwieldy  island  cried 
aloud  for  joy,  and  took  Ptolemy  new-born  into  her  arms,  exclaiming  in 
substance  :  ' '  Blessings  on  thee,  and  mayest  thou  honour  me  even  as 
Apollo  honoured  Delos,  and  love  me  as  Apollo  loved  Rhenea."  This 
shows  how  inseparable  were  Delos  and  Rhenea  in  the  poet's  mind,  and 
makes  evident  that  it  was  as  absurd  to  count  the  one  as  the  other  among 
the  Cyclades.  Delos  and  Rhenea  were  the  centre,  while  the  Cyclades 
were  the  circumference.  In  the  Encyclop&dia  Britannica,  be  it  said, 
only  ten  Cyclades  are  named,  for  two  of  the  twelve  there  given  are  Delos 
and  Rhenea.  Now  the  only  two  authorities  who  name  Delos  and  Rhenea 
among  the  islands  that  surrounded  them  are  Hyginus,  whose  list  is  of 
nine  only,  and  Stephanus  Byzantius,  who  names  twenty-three  :  twenty- 
one  beside  the  two  centre  islands.  It  seems  useless  to  attempt  thus  to 
tamper  with  the  authorities  at  this  late  day,  and  by  far  the  best  considered 
list  is  undoubtedly  Strabo's,  from  which  Delos  and  Rhenea  are  excluded. 
See  Pauly,  s.v.  "Sporades. " 


THE  CYC  LADES  AND  SPORADES  395 

the  matter  by  naming  the  Cyclades,  but  also  he  corrects 
Artemidorus.  He  enables  us,  furthermore,  to  get  a  fairly 
adequate  list  of  the  islands  which  he  considered  to  be 
Sporades,  and  he  separates  from  both  Cyclades  and 
Sporades  various  important  islands  and  groups  of  islands 
along  the  coast  of  Asia  Minor. 

To  begin  with,  Strabo  does  not  regard  the  Cyclades 
solely  as  a  collective  name  for  a  group  of  islands.  They 
are,  as  contrasted  with  the  Sporades,  famous  islands  singled 
out  from  among  less  noteworthy  ones  :  kv  8e  ravrats  (islands 
near  Crete)  at  re  Kv/cAaSes  curt  Kat  at  27ro/)tt8e5,  at  /xev  a£tai 
^Hi/Tip??  at  8'  da-rjfjLorepai,  p.  474,  book  x. ;  again  in  speaking 
of  islands  in  the  Aegean  sea — a  much  smaller  expanse, 
according  to  his  definition  of  it,  than  what  we  call  the 
Aegean — he  says  tv  8e  TO>  Atyatw  /xaAAoy  airny  re  rj 
/cat  at  TTf.pl  avrrjv  KvtfAaSes  /cat  at  ravrats 
2/7ro£>aS€9,  £>v  etcrt  Kat  at  Xt'^Oeio'a.i  Trepl  Ttjv  KpryTT^v,  p.  4^5' 
ibid. ;  and  again,  a  little  farther  down  the  same  page, 
Strabo  indicates  the  religious  nature  of  the  bond  between 
Delos  and  the  Cyclades,  not  without  a  confirmation  of  his 
previous  implication  that  the  Cyclades  are  islands  of 
especial  note,  and  therefore  set  apart  from  the  others. 
These  are  his  words  :  cvSogov  8'  eTrofyo-av  avrrjv  (Delos)  at 
s  vfjcroi,  KaAoryxevat  K/uKAaSes,  Kara  rt/x^v  Trc/xTrowai 
$ea)/)oi;s  re  Kat  0wrtas  Acat  ^opovs  Tra.p6f.vutv 
Ls  re  tv  avry  crvva,yov(ra,L  /xeyaAag.  The  sub- 
stance  of  this  important  passage  is  that  Delos  largely 
owed  its  glory  to  the  honours  paid  it  by  the  surrounding 
Cyclades,  whose  communities  as  such  constantly  deputed 
sacred  embassies,  provided  solemn  sacrifices,  and  sent 
choirs  of  maidens  to  add  beauty  and  solemnity  to  Delian 
festal  gatherings.  Such  in  general  terms  were  the  Cyclades, 
islands  far  more  noteworthy  than  the  Sporades.  A  detailed 
examination  of  Strabo's  account  and  list  first  of  the 
Sporades  and  then  of  the  Cyclades  is  now  necessary.  If  it 
be  desired  to  give  a  list  of  the  Sporades  according  to 
Strabo,  the  matter  will  be  a  difficult  one.  But  though  he 


396  APPENDIX  IX 

does  not  mention  all,  the  following  are  some  of  them. 
Thera  (p.  484,  and  cf.  p.  485  quoted  above),  Thucydides 
to  the  contrary  notwithstanding,  is  one  of  Strabo's  Aegean 
Sporades,  Amorgos  is  distinctly  classed  as  such  (p.  487 
end).  Anaphe  is  classed  along  with  Thera  (p.  485,  cf. 
Apollonius  Rhod.  iv.  1709),  Sicinos,  los,  Pholegandros, 
and  Gyaros,  as  of  the  Sporades,  so  that  all  insignificant 
islands  in  the  neighbourhood  of  the  Cyclades  are  added  to 
those  of  equally  small  pretensions  to  the  eastward — the 
Carpathos  group,  the  Calydnae  isles,  and  all  manner  of 
small  fry  in  the  neighbourhood  of  Rhodes,  Cos,  and  Samos 
— these  are  the  Sporades.1 

And  now  for  the  list  01  the  chosen  Cyclades  given  by 
Strabo  (p.  485,  book  x.)  He  says  somewhat  vaguely : 
"  The  number  given  to  begin  with  was  twelve,2  but  several 
have  been  added  to  the  list."  Then  he  quotes  Artemi- 
dorus — an  Ephesian  who  wrote  eleven  books  on  geography 

1  For  fuller  information  from  other  authors  as  well  as  Strabo,  see  the 
article  "Sporades"  in  Pape's  Worterbuch  der  Eigennamen,  where  a  list 
is  given  as  follows  : — Anaphe,  Astypalaea,  Amorgos- Patage,  Autoniate  or 
Hiera,     Ascania,     Azibinthia,     Atragia,     Aigilia,     Bouporthmos  -  Machia, 
Gyaros,   Gerus,   Donusa,    Dionysia,   Elaphonesos,   Helene   and  Eulimna, 
Thera,  Therasia,  Icaria  (Icaros),  los,  Hieracia,  Hippouris,  Casos,  Crapa- 
thos  (Carpathos),  Calydna  and  Calydna,  Calymna,  Cimolos,  Cos  (Coos), 
Corsia,  Cinaethos,  Corassiae,  Caminia,  Cinara,  Cythnosa.nd  Cothon,  Leros, 
Lebinthos,    Lea,   Melos,   Nisyros,   Nicasia,   Patmos,  Proconnesos,  Paros, 
Platea,    Sicinos,    Seriphos,    Scylos,    Sapyle,    Syrnos,     Schinussa,    Syme, 
Telos,    Tenos,  Tenedos,  Hypere,  Pholegandros,  Phacusia,  Chalcia,   Odia, 
Oletandros,  Olearos.      I  have  italicised  those  names  which  are  also  to  be 
found  on  Strabo's  list  of  twelve  Cyclades.     The  result  is  that  there  are 
sixty-three  islands  to  which,  by  some  one  or  another,  the  name  Sporades 
has   been   given,    and   that  only  six  islands    (Andros,    Myconos,    Naxos, 
Syros,  Ceos,  and  Siphnos)  on  the  usual  list  of  Cyclades  have  never,  so  far 
as  we  know,  been  classed  as  Sporades. 

2  Pape,  quoting  (s.v.  Ku/cXds)  Steph.  Byz.,  says  that  according  to  the 
ancients  there  were  more  than  twelve,  and  then  gives  the  list  of  Stephanus, 
as  follows: — (i)  Aegina  ;   (2)  Amorgos;    (3)  Andros;   (4)  Antissa  ;    (5) 
Aspis ;     (6)  Astypalaea;     (7)   Delos  ;    (8)   Icaros;    (9)   los;    (10)    Kos  ; 
( 1 1)  Casos  and  Nasion  ;   (12)  Cythnos  ;    (13)  Melos  ;    (14)  Myconos ;   (15) 
Naxos ;     (16)    Nisyros;    (17)   Paros ;     (18)    Peparethos  ;    (19)   Siphnos  ; 
(20)  Telos;    (21)   Tenos  ;   (22)  Tragiae  ;    (23)  Olearos.      It  will  be  seen 
that  except  Cimolos,  Syros,  Seriphos,  and  Ceos,  all  of  Strabo's  twelve  are 
on  this  list  in  italics. 


THE  CYC  LADES  AND  SPORADES  397 

early  in  the  first  century  B.C. — who  enumerates  fifteen 
Cyclades  as  follows  : — (i)  Ceos  ;  (2)  Sithnos  ;  (3)  Seriphos ; 
(4)  Melos;  (5)  Siphnos ;  (6)  Cimolos ;  (7)  Prepesinthos ; 
(8)  Oliaros;  (9)  Paros ;  (10)  Naxos ;  (n)  Syros ;  (12) 
Myconos;  (i3)Tenos;  (14)  Andros;1  (15)  Gyaros.  Out 
of  these  Strabo  takes,  without  comment,  (8)  Oliaros  and 
(7)  Prepesinthos,  and  with  a  reason  (15)  Gyaros,  leaving 
just  twelve,  which  number,  he  says,  figured  as  that  of  the 
Cyclades  at  the  very  first.  It  is  plain  that  Thucydides  had 
a  somewhat  different  list,  since  Thera  was  upon  it,  an  island 
not  thought  of  by  Artemidorus,  and  classed  by  Strabo  among 
the  Sporades. 

The  reason  which  Strabo  gives  for  excluding  Gyaros 
from  the  chosen  islands  is  evidently  the  fact  that  it  was 
bare  of  all  resources.  He  tells  an  anecdote  to  illustrate 
this,  and  finally  quotes  a  line  from  the  Elegant  Trifles  of 
Aratus,  where  Leto  is  reproachfully  apostrophised  for 
passing  her  votary  by,  even  as  she  passed  Pholegandros  or 
Gyaros  by.  Thus  we  are  brought  round  again  to  the 
divine  selection  of  certain  islands  to  be  the  holy  Cyclades. 

After  the  day  of  Strabo  the  number  twelve  was  appar 
ently  adhered  to,  for  the  phrase  Dodekanisia — Twelve- 
islands — survived  into  Byzantine  and  mediaeval  times,  and 
finally  seems  to  have  stood  for  many,  if  not  all,  islands  in 
the  Aegean.2  The  word  Archipelago,  which  arose  in  the 
later  days  of  Italian  supremacy,  seems  never  wholly  to 
have  lost  its  reference  to  the  sea.  Bursian  (quoting 
Forbiger)  is  my  authority  for  understanding  it  as  a  cor 
ruption  of  Aegaeon  pelagos? 

1  It  is  interesting  to  note  that  two  of  the  Bahamas  bear  the  name 
Andros,  not,  however,  taken  from  the  Aegean,  but  probably  from  Governor 
Andros  of  memory  unblessed  in  the  colonial  records  of  the  United  States 
of  America.     A  French  navigator  also  tried  to  fix  the  name   Grandes 
Cyclades  upon  a  group  lying  south  of  the  Caroline  Islands. 

2  It  would  seem  that  Diocletian's   e-rrapxia  vyvuv  l\.vK\d8wv  survived 
under  this  altered  name,  which  corrected  the  implication  that  all  its  fifty- 
three  islands  were  of  the  sacred  twelve. 

3  Bursian,  iii.  p.  351,  notes  ;   Forbiger,  Handbuch  der  alien  Geographic, 
ii.   p.    19  ff.  :    Alyatov  TrtXayos,    Aegeopelago,   Agiopelago,  Azopelago, 


398  APPENDIX  IX 

It  remains  now  to  consider  what  are  the  modern 
Cyclades  and  Sporades.  The  modern  Department  of  the 
Cyclades  covers  practically  all  islands  that  group  them 
selves  around  Delos,  Rhenea,  and  the  twelve  Cyclades  of 
old.  These  are  those  islands  lying  east  of  the  Pelopon 
nesus  and  north  of  Crete  which  are  not  misruled  by  Turkey, 
but  enjoy  freedom  under  the  kingdom  of  Greece.  East  of 
these  lie  the  Turkish  islands,  to  which  (large  and  small 
alike)  is  given  the  name  of  Sporades.  Certain  Greek 
islands,  north  and  east  of  Euboea,  are  now  sometimes 
called  the  Northern  Sporades,  a  convenient  use  of  the  term 
which  Strabo  would  not  have  found  it  easy  to  understand. 

Archipelago.     The  form  Arcipelago  occurs  first  in  a  treaty  of  June  3oth, 
1268,  between  Michel  Palaeologus  and  Venice. 


APPENDIX    X 

THE   WORSHIP   OF   APHRODITE  AND   OF 
STRANGE   GODS   AT   DELOS 

PAUSANIAS  (IX.  xl.  3),  in  speaking  of  ancient  wooden  idols 
(£oava)  traditionally  attributed  to  Daedalus,  says:  "The 
Delians  also  have  a  rather  small  wooden  image  of  Aphro 
dite,  the  right  arm  of  which  by  the  lapse  of  years  has 
suffered  grievous  disfigurement.  The  lower  part  of  it  is 
square,  and  there  are  no  feet.  I  am  convinced  that 
Ariadne  received  this  image  from  Daedalus,  and  that  when 
she  went  with  Theseus  she  took  it  with  her  from  home. 
Now  the  Delians  say  that  when  Theseus  had  been  parted 
from  her  he  dedicated  the  image  to  Delian  Apollo,  that  he 
might  not,  by  bringing  it  home  with  him,  have  the  remem 
brance  of  Ariadne  revived  and  be  constantly  renewing  his 
griefs l  on  account  of  the  love  of  her." 

The  form  in  which  Pausanias  gives  this  legend,  which  is 
no  doubt  its  latest  one,  throws  an  interesting  light  upon  the 
following  observation  2  of  M.  Homolle  :  "  II  n'est  pour  ainsi 
dire,  pas  une  legende  delienne  qui  n'ait  sa  contre-partie  dans 
une  legende  athenienne,3  destinee  a  prouver  la  primaute 
religieuse  d'Athenes  et  ses  droits  sur  De'los."  In  the  above 
passage  of  Pausanias  we  read  an  Athenian  legend  which  had 

1  Paus.  I.  xxii.  5.         -  Note  32  to  his  article  "  Delia"  above  quoted. 

,3  This  is  true  even  of  the  birth-legend  of  Apollo  and  Artemis^  partially 
transferred  by  Attic  legend  to  Cape  Zoster,  in  Attica.  Pausanias,  I.  xxxii. 
i.  Cf.  Baiter  and  Sauppe,  Orat.  Att.,  Hyperides,  fr.  286,  39,  and  286,  65. 


400  APPENDIX  X 

evidently  driven  the  original  Delian  account  of  the  ancient 
statue  of  Aphrodite  from  the  minds  even  of  the  native 
Delians.  If  Theseus  played  any  part  in  their  original  story, 
it  was  probably  not  the  beau  role  of  a  faithful  lover ;  it  is 
only  or  chiefly  in  Athenian  legends  that  Theseus  is  the 
plaintiff  (o  dfaipeOek).1 

This  story  of  Theseus  then  may  be  classed  with  the 
other  Attico-Delian  legend  to  the  effect  that  Theseus  taught 
the  islanders  their  characteristic  crane-dance.  They  danced 
this  around  the  altar  of  horns  in  a  strange  building 2  placed 
in  front  of  Apollo's  Delian  temple,  which  has  been  ingeni 
ously  described  under  the  name  of  the  Hall  of  Bulls.  This 
dance  is  reported  to  have  been  a  representation  of  the  way 
of  Theseus  through  the  mazes  of  the  Cretan  Labyrinth.  If 
the  name  crane-dance  implies  resemblance  to  the  lines  of 
flocking  cranes  that  move  across  Greek  skies,  then  the 
comparison  to  Theseus  in  the  Labyrinth  falls  of  its  own 
weight,  and  a  far  nearer  parallel  is  the  modern  peasant's 
dance  called  the  Syrtos.3 

To  reconstruct  the  forgotten  Delian  legend  of  the 
crane-dance  would  be  as  impossible  as  to  ascertain  what 
account  the  ancient  Delians  gave  of  their  wooden  idol  of 
Aphrodite.  It  is  probable  that  anciently  the  local  cults  of 
Delos  and  Naxos  were  most  closely  united ;  the  antiquity  of 
the  colossal  statue  set  up  by  the  Naxians  at  Delos  proves 
their  especial  devotion  to  Apollo,  as  do  also  abundant 
traces  of  his  early  worship  on  Naxos.  The  reasons  for 
identifying  Ariadne  with  Aphrodite,  as  one  phase  of  that 

1  Cf.  Pausanias,  X.  xxix.  4. 

2  It  appears  to  have  been  upwards  of  220  feet  long  and  only  40  feet 
wide.      See    M.    Homolle's  account   in    the   Bulletin  de  Correspondance 
Helllnique. 

3  This  dance  is  consecrated  in  the  minds  of  Greek  patriots  by  the  fact 
that  it  is  associated  with  the  heroic  defence  of  the  Khan  of  Gravia.     There, 
as  the  enemy  approached,  and  it  became  evident  that  a  picked  company 
must  stand  out  against  them,   the  gallant  Odysseus  led  off  the  Syrtos 
whereby  he  gathered  his  chosen  band  of  180  into  the  Khan  of  Gravia, 
that  modern  Greek  Thermopylae  whose  Leonidas  survived  with  nearly  all 
his  men. 


WORSHIP  OF  APHRODITE  AND  STRANGE  GODS     401 

elusive  godhead,  are  abundant,  and  therefore  it  seems  likely 
that  the  most  ancient  worship  of  Aphrodite  at  Delos  was 
the  same  in  origin  with  that  of  Ariadne  upon  Naxos.  There 
appear  in  fact  to  have  been  two  aspects  of  this  Aphrodite- 
Ariadne,  (i)  the  one  whom  Dionysus  espoused,  a  triumph 
ant  and  immortal  goddess,  (2)  she  who  was  forsaken  of 
Theseus  and  doomed  to  a  lonely  death.  As  the  spring 
time  bride  of  Dionysus,  Ariadne  was  the  gladsome  spirit  of 
love  and  vegetation.  As  the  forsaken  spouse  of  Theseus, 
she  was  that  same  spirit  doomed  to  a  wintry  eclipse.  Both 
of  these  phases  recur  in  Cypriote  as  in  Assyrian  legends. 
In  Cyprus  Aphrodite  was  entombed  as  Ariadne-Aphrodite. 
The  legend  of  Theseus'  abandonment  of  Ariadne  in  the 
form  less  creditable  to  him  was  also  current  in  Cyprus. 

The  other  links  between  Delos  and  Aphrodite,  not  the 
Paphian  goddess  in  particular,  but  the  goddess  at  large,  and 
more  especially  her  eastern  prototypes  and  parallels,  belong 
to  the  latter  days  of  Delian  independence  and  to  that  final 
period  when,  after  166  B.C.,  Delos  was  restored  to  control 
nominally  Athenian  but  really  Roman.  This  was  a  time 
when  distinctions  between  the  gods  of  one  people  and 
those  of  another  were  falling  away,  and  when  each  god  of 
Greece  and  Rome  tended  to  become  every  other  one.  By 
this  time  Apollo  certainly  may  well  have  begun  to  feel  that 
he  had  little  pre-eminence  at  Delos,  and  could  hardly 
recognise  in  the  great  emporium  for  buying  and  selling 
slaves  the  island  of  his  birth. 

Not  far  from  the  most  ancient  cave  temple  on  the  flanks 
of  Mount  Cynthus  was  set  apart  what  may  be  called  the 
precinct  of  the  foreign  gods,  and  there  an  inscription  has 
been  found  to  Eros  Harpocrates  Apollo}- 

Here  is  combination  and  to  spare.  In  the  Eleusinian 
rites  divinities  such  as  Rhea  Cybele  and  Demeter  were 
merged  into  one,  but  only  in  the  fulness  of  time,  when 
their  worships  and  their  stories,  after  running  parallel,  had 


1  See  upon  this  whole  subject  two  admirable  articles  by  M.  Hauvette 
Besnault  in  the  Bulletin  de  Correspondance  Helttnique  for  1882. 

2   D 


402  APPENDIX  X 

gradually  been  united.  This  gradual  fusion  was  impossible 
in  the  case  of  Apollo  and  Harpocrates  or  Horus,  two  names 
for  one  Egyptian  god  whose  resemblance  to  Apollo  was 
purely  superficial.  Through  this  Egyptian  interloper  Apollo 
on  his  own  native  soil  becomes  one  with  the  god  Eros,  and 
thus  enters  into  union  with  Aphrodite  the  mother  of  all 
loves.  But  perhaps  this  point  should  not  be  insisted  upon, 
and  we  should  rather  say  that  the  precinct  of  the  foreign 
gods  on  Mount  Cynthus  of  Delos  became  foreign  soil, 
where  Apollo  was  neither  really  himself  nor  even  first  among 
the  strange  gods  who  there  broke  down  the  reserve  of  his 
nature  and  made  themselves  identical  with  him.  Such  an 
exterritorial  character  in  this  precinct  seems  implied  by  an 
inscribed  enumeration  of  divinities  where  Apollo  is  neither 
first  nor  last  among  a  whole  procession  of  Egyptian  gods 
and  goddesses — Harpocrates,  Serapis,  Apollo,  Isis,  and 
Anubis.  This  precinct  was  certainly  not  in  existence  until 
the  days  of  the  breaking  up  of  pagan  divinities,  and  its 
Egyptian  gods  were  unknown  in  the  early  days  of  the  purer, 
nobler,  and  more  exclusively  Grecian  rite  of  Delian  Apollo. 
And  yet  its  nearest  neighbouring  shrines  are  the  oldest  on 
the  whole  island.  The  precinct  itself  is  on  the  western 
slope  of  Mount  Cynthus,  next  to  the  most  ancient  holy  way. 
This  holy  way  led  from  the  summit — where  was  the  old- 
time  temple  of  Zeus  Cynthius  and  Athena  Cynthia — to 
that  mysteriously  primeval  place  of  Delian  worship  called 
the  cave-temple  of  Apollo.  Near  by  and  a  little  below  on 
the  downward  journey  to  the  plain  and  city  of  Delos  lay 
this  precinct  of  the  foreign  gods.  In  it  were  two  diminutive 
temples  or  shrines,  one  of  Serapis  and  one  of  Isis.  Just 
below  and  northward  runs  the  bed  of  a  ravine  bordered  by 
what  is  believed  to  have  been  the  Cabirion,  a  temple  for 
the  worship  of  the  more  or  less  unclean  and  unmentionable 
Cabiri.1  These  gods  came  originally  from  Phoenicia,  and 

1  The  German  School  at  Athens  made  important  discoveries  in 
excavating  the  Cabirion  close  to  Thebes  in  Boeotia.  This  was  in  the 
winter  1888-89. 


WORSHIP  OF  APHRODITE  AND  STRANGE  GODS     403 

were  associated  with  the  mysteries  of  Samothrace,  a  north 
ward  island  where  were  cultivated  the  less  noble  and  more 
questionable  aspects  of  a  nature  worship  in  substance  not 
unlike  that  of  Eleusis. 

The  presence  here  of  Phoenician  Cabiri  may  prepare  us 
for  another  Phoenician  divinity  with  whom  we  have  been 
lately  occupied  :  I  mean  the  so-called  Syrian  goddess.  The 
ready  confusions  and  hastily  made  conglomerations  of  latter- 
day  pagan  worship  are  nowhere  more  conspicuous  than  in 
the  latest,  the  cosmopolitan  era  of  Delos.  No  worship 
more  fully  illustrates  it  than  this  of  Syrian  Aphrodite.  We 
must  not,  though  we  rightly  call  her  Aphrodite,  connect  her 
too  closely  with  the  Paphian  goddess.  On  the  other  hand 
she  was  in  the  latter  days  not  purely  Phoenician,  but  was 
associated  in  worship  with  Egyptian  Isis ;  and  her  cult, 
with  that  of  the  other  strange  gods  in  this  precinct,  was 
supervised  and  administered  by  Greek  officials. 

This  had  a  curious  result,  i.e.  the  establishment  by 
Phoenicians  under  native  management  of  a  second  worship 
of  the  Syrian  goddess  Atargatis.  The  reason  for  this  striking 
duplication  of  sanctuaries  and  observances  in  honour  of 
one  and  the  same  goddess  is  that  newly  arrived  Syrians, 
merchants  fresh  from  Beyrut,  from  Antioch  and  from  Sidon, 
found  at  Delos,  in  the  Syrian  goddess  worshipped  in  Greek 
fashion  on  the  north-western  spur  of  Mount  Cynthus, 
nothing  which  they  recognised  as  their  own. 

Three  several  times,  then,  and  under  three  guises,  did 
Aphrodite  visit  Delos.  Once  as  Ariadne  forsaken  of  the 
hero  Theseus,  once  as  a  sort  of  Isis  in  the  company  of  the 
Cabiri  and  the  chief  gods  of  Egypt,  and  once  as  the 
Atargatis  worshipped  by  the  Phoenician  and  Syrian  colony 
established  for  commerce  in  later  days. 


APPENDIX   XI 

PHOTOGRAPHS    REFERRED   TO   FOR 
ILLUSTRATIONS 

N.B. — By  the  kindness  of  Mr.  Leaf  I  am  able  to  place  certain 
numbers  on  his  list^  as  well  as  many  upon  that  for  which 
I  am  now  chiefly  responsible^  at  the  disposal  of  those  who 
might  wish  to  procure  illustrations.  The  lists  in  question 
are  issued  by  the  Society  for  the  Promotion  of  Hellenic 
Studies.  Those  of  my  readers  who  belong  to  that  society 
can  procure  all  these  and  many  other  Greek  pictures  at 
cost  price.  Those  who  are  not  members  of  the  Hellenic 
Society  will  Jind  below  the  prices  at  which  they  can  pro 
cure  various  illustrations  as  enumerated. — L.  D. 

I. —  LIST  OF  PHOTOGRAPHS  TAKEN  IN  GREECE  AND 
CYPRUS,  JANUARY  TO  JUNE  1888,  BY  MR.  MALCOLM 
MACMILLAN  AND  MR.  Louis  DYER. 

Notice. — These  photographs  measure  8x6  inches  and 
may  be  procured  at  a  cost,  including  postage,  of  123.  per 
dozen  from  Messrs.  Walker  &  Boutall,  16  Clifford's  Inn, 
Fleet  Street,  London,  E.G.  Single  prints  may  be  obtained 
at  is.  each. 


PHOTOGRAPHS  FOR  ILLUSTRATIONS  405 

(a)  ATHENS 
THE  ACROPOLIS 

No. 

1.  The  Theatre  of  Dionysus. 

2.  A  Cyclopean  wall  (near  the  Asclepieum). 

3.  A  Cyclopean  pit  (near  the  same). 

7.  The  Bastion  of  Odysseus,  the  defender  of  Gravia  (since 

destroyed). 

8.  The  Propylaea  (from  Cimon's  Wall). 

9.  ,,  ,,         (from  the  top  of  the  Parthenon). 

14.  The  Parthenon,  N.W.  corner  through  a  Byzantine  arch 
of  the  Erechtheum,  showing  in  the  middle  dis 
tance  foundations  attributed  to  the  age  of  Pisis- 
tratus. 

1 8.  The  same  (foundations  seen  from  the  roof  of  the 
Acropolis  Museum). 

ACROPOLIS  MUSEUM 
2 1 .  An  ephebus  (of  the  Apolline  type). 

LOWER  TOWN 

23.  The  Temple  of  Zeus  (Olympieum) — connected  by  its 
final  dedication  with  the  deification  of  the  Roman 
emperors — and  the  Bed  of  Illissus. 

THE  CARAPANOS  MUSEUM 
25.  Aphrodite  and  Eros  (terra  cotta). 

CENTRAL  MUSEUM 

30.  A  satyr  found  near  Lamia.     ;.« 

32.  A  sleeping  maenad  found  S.  of  the  Acropolis. 


406  APPENDIX  XI 

(b)  ATTICA 
SUBURBS  OF  ATHENS 

No 

34.  Colonos,  the  Cephissus   (crossed  lower  down  by  the 

Eleusinian  procession). 

35.  „        Hill  of  Demeter  Euchloos. 

THE  PIRAEUS 

40.  The  babe  Plutus  found  in  the  water  near  Eetionea 
(Central  Museum). 

ELEUSIS 

42.  General  view  including  the  "Secos." 

43.  Lower  gateway,  medallion. 

44.  Appius  Pulcher's  gateway,  debris. 

ICARIA  (STO  DIONYSO) 

46.  Rapendosa  valley  and  cave. 

47.  „         cave. 

48.  View  from  the  brow  of  Rapendosa  cliff,  forming  the 

Sto  Dionyso  valley  towards  Marathon  and  Styra 
in  Euboea. 

49.  View  from  the  same  toward  the  Pentelic  range. 

THE  AMERICAN  FIND  AT  ICARIA 

50.  The  ruined  church  untouched. 

51.  The  ruined  church  pulled  down. 

52.  Replica  of  "  Marathonian  Soldier,"   a   head  (archaic) 

and  a  Bas-relief  (stele). 

53.  The  replica  and  a  muleteer. 

54.  A  crown  of  Icarian  ivy  devoted  to  Dionysus. 

55.  A  bas-relief  from  the  Icarian  Pythion. 


PHOTOGRAPHS  FOR  ILLUSTRATIONS  407 

SUNIUM 

No. 

56.  The  temple  of  Athena,  near  view. 

57.  The  same,  far  view,  being  "Cap  Colonnas." 

(f)  THE  CYCLADES 

SYRA 

58.  Old  Syra — The  Roman  Catholic  Upper  Town. 

59.  The  same  (distant  view). 

60.  A  glimpse  down  a  street  of  Old  Syra. 

6 1.  Hermupolis — Shipping,  a  Chiote  Bombarda. 

62.  ,,  ,,  a  Perama. 

63.  „                  „  a  Trechanderi    from    Siphnos, 
a    Goelette    and  a  Trechanderi    from    Santorin 
(Thera). 

64.  "  La  Caramanienne  " — performed  by  Syriote  butchers. 

65.  The  Psariana. 

66.  The  Potamos  :  a  street  in  Hermupolis. 

67.  Episcopio,  view  from  the  church  terrace  (inland). 

68.  The  same,  towards  Rhenea. 

THE  SYRIOTE  MUSEUM 

69.  Stele  from  Paros,  a  poor  man's  gravestone. 

70.  Inscription  from  los. 

DELOS 

71.  The  Lake  of  Leto. 

7  2.   Mount  Cynthus,  from  the  lake. 

73.  ,,  from  Apollo's  temple. 

74.  Mount  Cynthus,  Cave  temple  from  a  Roman  house. 

75.  ,,  Cave  temple  (foundations). 


408  APPENDIX  XI 

No. 

77.  Temple  (on  the  slope   of  Mount  Cynthus).       Draped 

female  statue  of  Isis. 

78.  Portrait-statue    of  Caius   Ofellius    by  Dionysius   and 

Timarchides  of  Athens. 

79.  Ruins  of  Apollo's  temple. 

80.  Acroterion  from  same  (Central  Museum). 

81. 

82.  The  Naxian  Colossus. 

83.  Rhenea   from    Mt.    Cynthus   and    the    lesser  of  the 

Rheumatiari  Reefs. 

NAXOS 

84.  The  Gateway  of  Dionysus. 

85.  Mt.  Coronis. 

86.  The  Valley  of  Paratrecho,  and  Mt.  Zia  or  Ozia. 

TENOS 

87.  Mt.  Burgo  and  the  Sanctuary  of  the  Evangelistria. 


(^—CYPRUS 

LlMASSOL 

88.  St.  Nicholas  monastery,  a  ruin  near  Cape  Gatto. 

89.  The  castle  of  the  Knights  Templars  at  Colossi. 

BAFFO 

90.  The  ruins  at   Old   Paphos  (Couclia),  after  the  British 

excavations. 

91.  Inscription  from  Old  Paphos  (elaeochristion). 

92.  The  Eros  of  Paphos,  from  temple  of  Aphrodite.     Ibid. 

93.  Same,  profile  view. 

94.  A  terra  cotta  head  from  Old  Paphos. 

95.  The  Coucliote  Diggers  at  Old  Paphos. 

96.  The  Bleeding  Column,  New  Paphos. 


PHOTOGRAPHS  FOR  ILL  US  TRA  7  'IONS  409 

LAPETHIA — CERYNIA 

No. 

97.  A  tomb  and  a  monastery  at  Lapethus. 

98.  A  Byzantine  fort  in  Cerynia. 

99.  The  cloister  at  Bello  Pais. 

100.  The  castle  "  Dieu  d'Amour." 

SALAMIS 

10 1.  The  rampart  and  moat  of  Famagosta. 

102.  St.  George  and  the  cathedral-mosque  of  Famagosta. 

103.  Famagosta  cathedral,  from  the  rampart. 

104.  The  same,  nearer  view. 

105.  The  same,  chantry  door. 

1 06.  The  same,  minaret. 

107.  Gateway  of  the  Lusignan  palace  at  Famagosta. 

THE  VISCONTA 

1 08.  St.  Sophia,  the  cathedral-mosque  of  Nicosia. 

(£)_WESTERN  GREECE 

AEGINA 

109.  The  temple  of  Athena,  from  below, 
no.  The  same  from  nearer  N.E. 

in.  „  „       S.E. 

ARGOLIS 

112.  Tiryns  citadel  from  the  west. 

113.  The  same,  gallery  toward  Nauplia. 

114.  Argos  museum,  a  Medusa. 

115.  Mycenae,  the  Lions'  gate. 

1 1 6.  A  Cyclopean  bridge  near  Epidaurus 

117.  Theatre  seats  at  Epidaurus. 

1 1 8.  „     (orchestra)        ,, 

119.  „         (stage) 


410  APPENDIX  XI 

ARCADIA 

No. 

120.  The  battlefield  of  Tegea. 

121.  Heads  from  a  Tegean  temple  (of  Athena  Alea),  C.M. 

122.  Bassae-Phigalia,  temple  from  N.E. 

123.  The  same,  interior. 

LACONIA 

124.  Sellasia  valley  (Skiritis). 

125.  The  valley  of  the  Eurotas  from  Vrylias  (Skiritis). 

126.  Spartan  museum,  the  Omphalos  relief. 

127.  An  Amazon,  etc.     Ibid. 

129.  The  Langgada.  Pass  (Taygetus)  and  distant  Parnon. 

130.  The  summit  of  Langgada  Pass,  Mt.  Rindomo  (Biscuit- 

mountain). 

131.  The  same  in  another  direction,  southerly  Mt.  Pigadia. 

OLYMPIA 

133.  Ruins  of  the  Heraeum  from  the  gymnasium. 

134.  The  Hermes  (bearing  the  babe  Dionysus)  from  the 

Heraeum,  now  in  the  Syngro  Museum. 

135.  Temple  of  Zeus,  from  Pelopion,  E.J 

136.  „  „  „  W.| 

137.  The   same,    Metope,    Nemean   labour   of  Heracles, 

Syngro  Museum. 

138.  The  same,  Metope,  Athena,  nearer  and  front  view. 

139.  „  ,,        Augean  labour,  Syngro  Museum. 

140.  .,  „        the  fetching  of  Cerberus,  Syngro 

Museum. 

141.  „         Western  pediment,  Syngro  Museum. 

142.  „          Eastern          „  „  „ 

143.  ,,         Apollo    from    western    pediment,    Syngro 

Museum. 

144.  The  Kladeos,  from  the  eastern  pediment. 

145.  The  same,  debris  on  the  south  side. 

146.  A  well  near  the  same. 

147.  N.W.  entrance  of  the  Stadium. 

148.  A  pugilist  of  note,  bronze  in  the  Syngro  Museum. 


PHOTOGRAPHS  FOR  ILLUSTRATIONS  411 

CEPHALLENIA 

No. 

149.  A  view  toward  Ithaca  from  the  road  near  Same. 

ITHACA 

150.  View  from  Mount  Aetos  northward. 

151.  View  on  Mount  Aetos  (cyclopean  wall  of  "Odysseus' 

castle  "). 

152.  The  Grotto  of  the  Nymphs  (so  called). 


II. — CATALOGUE  OF  PHOTOGRAPHS  OF  GREECE  BY 
MR.  WALTER  LEAF. 

These  Photographs  measure  about  7x5  inches,  and  may 
be  procured  at  a  cost  of  is.  each  in  silver,  or  is.  6d.  in 
platinum,  from  Mr.  CASSTINE,  Photographic  Studio,  Swanley, 
Kent.  Platinum  is  recommended  only  for  those  marked 
with  an  asterisk.  The  profits  on  the  sale  will  go  to  the 
Homes  for  Working  Boys,  Swanley. 

ATHENS 

2.   *The  British  School :  Lycabettus  in  background. 

4.  *The  Acropolis  from  monument  of  Philopappus. 

5.  *The  same  :  larger  scale. 

7.   *  Acropolis  from  Areopagus. 

ELEUSIS 

33.  * Sekos  from  S.W.  angle. 

34.  -fView  towards  S.E. 

35.  36,  37-    Sekos  from  N.W.  angle.     (These  three  form 

a  panoramic  view.) 

38.  *Precinct  of  Pluto  from  S. 

39.  *Precinct  of  Pluto  from  N. 

40.  *  Substructures  of  Sekos. 


412  APPENDIX  XI 

SUNIUM 

No. 

42.   *From  S.E.  43.   *From  N.E. 

44.   *From  E.  45.  *From  N. 

46.   *From  W.  47.   *  Interior,  looking  W. 

OROPOS 

51.  Theatre  at  Amphiareion. 

52.  *The  same,  showing  Proscenium. 

53.  *The  same,  from  N.W. 

56.   *The  same,  showing  Seat  of  Priest. 

CORINTH 

65.     Temple  from  N.E.  66.  Temple  from  E. 

DELPHI 

73.  *  General  View. 

74.  ^Substructure  of  Peribolos  and  Athenian  Stoa. 

76.  *Castalian  Spring. 

77.  ^Relief  in  Museum. 

MONASTERY  OF  ST.   LUKE,  STIRIS. 

80.  Church,  West  Front. 

8 1.  Church,  South  Side. 

82.  83.     East  End  (these  two  form  a  single  view). 

86.  General  View  from  S.E.  :  Parnassus  in  background. 


HELICON   (HIERON   OF  THE   MUSES) 

87.  General  View  :  Hill  of.Ascra  to  right. 

88,  89.  Proscenium  of  Theatre. 


PHOTOGRAPHS  FOR  ILLUSTRATIONS  413 


III. — LIST  OF  ENLARGEMENTS  FROM  MR.  WALTER 
LEAF'S  PHOTOGRAPHS. 

*#*  These  are  selected  from  the  list  of  eighty-nine  small 
photographs  which  are  already  accessible  to  members  of  the 
Society  for  the  Promotion  of  Hellenic  Studies.  They  are 
also  to  be  had  by  the  public  at  large  through  the  Autotype 
Company,  at  No.  74  New  Oxford  Street,  London,  W.C. 

No.  SUBJECTS  DIMENSIONS 

1.  Athens  from  the   Monument  of  Philo- 

pappus      .  .  .  .  17^x1 1 1  in. 

2.  Temple  of  Sunium,  from  N.E. .  ,..«  17^  x  nj  ,, 

3.  Temple  of  Sunium,  East  end    .  .  17  J  x  12     ,, 

4.  Temple  of  Corinth       .  .  \~  .  17^x11     „ 

5.  Delphi:   General  View  .  .  17!  x  n     „ 

6.  Delphi :  Peribolos  Wall  and  Stoa  of  the 

Athenians  .  .  .17^x11,, 

7.  Eleusis :    Remains  of  the  Hall  of  the 

Mysteries.  .-  .  .      17^  x  n|  ,, 

8.  Eleusis:  Precinct  of  Pluto        .  I7ix  ITi  » 

9.  View  of  St.  Luke,  Stiris :    Parnassus  in 

the  background     .  .  .      i7jxn|  „ 


INDICES 


I. — INDEX  OF  AUTHORS  AND  SOURCES 


ACTS,  chap,  xiii.,  290 

Adams,  Henry,  History  of  United 
States,  267 

Aelian,  story  of  the  cure  of  a  horse 
by  Serapis  from,  254 

Aeneas  Sylvius  de'  Piccolomini,  326 

Aeschylus,  Edoni,  fr.  55,  158 ; 
Eumen.  22  ff. ,  23 ;  Prom.  806, 177 

Agnese,  Battista,  Portulano  of,  de 
scribed,  329  f. 

Allen,  Professor  F.  D. ,  on  the  Uni 
versity  of  Leyden,  347 

Ameis-Hentze's  Iliad,  317 

Andocides,   de  Mysteriis,   110-112, 

^  75 

Anthology,  ix.  75,  108 

Anthropological  Institute,  Journal  of, 
i69i. 

d'Anville,  Recherches  Gtographiques 
sur  I'ile  de  Chypre,  326 

Apollinaris  Sidonius,  witness  for 
Apollonius,  265  f. 

Apollodorus,  Bibliotheca,  I.  iii.  5,  49; 
I.  v.  2,  64  ;  I.  v.  4,  63 ;  III.  x.  i, 
1 1 9  ;  III.  xiii.  6,  6j  ;  III.  xiv.  7, 
sos,  107 

Apollonius  Rhodius,  iv.  1709,  396 

Apollonius  of  Tyana,  letters  attri 
buted  to,  xxxiii. ,  22 3 

Apuleius,  261  f. 

Arabantinos,  work  on  Epirus,  89 


Aratus,    Elegant    Trifles   by,    397  ; 

Phaen.  91,  138,  in 
Archaeological  Society,  the  Greek,  at 

Eleusis,  186 
Archaeology  (American],  Journal  of , 

2,   104 

Archtologique,  Revue,  227,  229 
Archilochus,  Frag.  3,  Frag.  51,  j6j 
Architects,  Royal  Society  of  British, 
Proceedings   of,   232,   314  ;    plan 
reproduced  from,  398 
Aristides,    prayer    of,     to     inspired 

dreams,  236,  242 

Aristophanes,  Birds,  28,  schol.  on, 
874,  8^,  96.  Frogs,  jibes  at 
Dionysus,  78 ;  312,  154  ff. ,  316- 
459,  212  ;  372-459,  185  ;  schol. 
on  158,  181.  Schol.  on  Knights, 
697,  707.  Peace,  374,  181. 
Plutus,  727,  177.  Thesmophoria- 
zousae,  68,  Wasps,  9,  85 
Aristotle,  Const.  Ath.  onEpimenides, 
123  f.  ;  3,  j-jo;  13,  129;  15, 
126,  137;  16,  126.  Eth.Eud.\.\ 
init.  Eth.  Nicom.  i.  8  fin. ,  383. 
Fr.  447,  382.  Politics,  v.  9, 
129  ;  v.  12,  126.  Rhetoric,  iii. 
1 6,  cites  Antigone,  911  f.,  166. 
irepl  6av^,a.ff'nt3v  aKova/jidruv,  cxxii. 

133-  93- 
Artemidorus,  on  the  Cyclades,  395  f. 


INDEX 


415 


Athenaeus,  Deipnos.  ii.  40  A,  ///  ; 
iii.  78,  82  ;  416  B,  57  ;  533  C, 
125  ;  609,  126  ;  xiv.  614  B,  382 

Attar,  Leonida,  unpublished  map  of 
Cyprus  by,  287 

BACCHYLIDES,  fr.  64,  62 

Bachut,  E.,  Histoire  de  la  Mtdecine, 

etc.,  no 

Backofen's  Mutterrecht,  168 
Baglione,  life  of  Astorre,  332 
Bahr,  on  Orpheus,  in  Pauly,  127 
Barry,  F.   W. ,   ' '  Report  on  Census 

of  Cyprus,  1884,"  272 
Bartolommeo      "da     li      Sonetti," 

rhymed     account     of     Venetian 

islands  by,  330 ;  on  Cythera,  286  ; 

sonnets  on  "  Sdiles, "  356 
Bayle's    Dictionary,     translated    in 

1735,  266 

Bede,  Commentary  on  the  Acts,  290 
Benndorf,  795 
Bentley,  775,  262 
Bergeat,  Herr  Alfred,  researches  of, 

on  Cypriote  geology,  275  f. 
Bergk,         Griechische       Litteratur- 

geschichte,  112,  317 
Bernhardy,    G. ,   Griechische  Litter- 

aturgeschichte,     115,     116,     123, 

127,  736,  317.     On  Philostratus, 

260 
Bibliotheque  Nationale,  Dante  (148 1 ) 

from  the  Vatican  in  the,  331 
Bie,  Oscar,  Die  Musen  in  derAntikcn 

Kunst,  104 
Bikelas,  D.,  on  Greek  doctors,  267 

ff. ;   Loukis  Laras  by,  389 
Birdwood,    Sir    George,    K.C.I.E. , 

85  ;  on  Dionysus,  164  f. 
Birmingham  Speculative  Club,  essay 

read  before  the,  226 
Blaeuw,  Atlas  Major  of  John  and 

William,  343,  346  f. 
Boccaccio,  Teseide,  VII.  stanza  43, 

286 
Boissier,   Gaston,  ' '  La  religion  Ro- 

maine,"  15 
Bordone,     Benedetto,     Isolario    of, 

328  i. 


Braun,  Kmil,  Dr. ,  Kimstvorstellungen 

des  gefliigelten  Dionysus,  779 
Brenzone,    life  of  Astorre   Baglioni 

by.  332 

Hritannica,  Encyclopaedia,  article 
"Cyclades,"  criticised,  394 

British  Museum,  additional  MSS., 
Earl  of  Guilford,  No.  8630. 

3J6 
Brown,    Horatio   F. ,     The    Venetian 

Printing  Press,  329 
Brucker,     James,      on       Meursius, 

350 
Brunn,    Denkmdler     der     Antiken 

Kunst,    by    Dr.,    795,  249,    270, 

379 
Buck,  Carl,  on  the  American  find  at 

Icaria,  104 
Bursian,  346,  397 
Bustrone's   Istoria   di  Cipro  MSS., 

JSjf- 
Byzantius,  Stephanus,  394 

CALLIMACHUS,  Hymn  to  Apollo,  58- 
63,  384.  Hymn  to  Delos,  1-5, 
370  \  30  ff.  and  92  f. ,  359  ;  82 
ff.,  78;  206  ff.,  356;  260  ff., 
357  ;  300  ff.,  359,  393  \  308  ff., 

379 
Carr,    Lucien,  Women   among    the 

Huron-Iroquois,  777 
Casaubon,     Meric,   on    Apollonius, 

266 
Cassel,  Verein  deutscher  Philologen 

und  Schulmaenner  at,  104 
Cervantes,  "  Journey  to  Parnassus," 

24 
Chaucer,    "  Knight's  Tale,"    1363, 

286 

Christus  Fattens,  4,  139 
Cicero,    Ad  Atticum,    vi.    i,    i8S\ 

v.  15,  vi.  i,  258.     Phil.   II.  xliii. 

no,  XIII.  xix.  41,  37 
Cicogna,  331 
Claudian,  on  Cyprus,  274 
Clement,  of  Alexandria,  Protrept.  iii. 

ad  fin.,  309  ;  p.  3°.  777 
Clement  the  Roman,  Homil.  v.  23, 

309 


416 


INDEX 


Colvin,  Mr.  Sidney,  on  the  Homeric 
Hymn  to  Demeter,  54  f. ,  37  f. , 
ISO 

Comparetti,  on  date  of  Orphic  frag 
ments,  181 

Corlieu,  "  Atude  me"dicale  sur  la 
retraite  des  10,000,"  225 

Cornhill,    vol.    xxxiii.    June,    1876, 

54 
Coronelli,   I  solaria  of,  343  ;  Morea 

of,  288 
Correr  Museo,  unpublished  map  in, 

287 
Cudvvorth,  265  f. 

DANTE,  Par.  xix.  22,  74 ;  xxi.  71, 
42 

Daremberg,  Charles,  Etat  de  la 
Mtdecine  entre  Homere  et  Hippo- 
crate,  226  ;  on  Hippocrates,  irepi 
ei}<r%77/iocn;i>?7S,  2jS ;  Histoire  des 
Sciences  Mtdicales,  223  ;  ' '  etudes 
d '  arche"ologie  me"dicale  sur 
Homere,"  227  ;  list  of  Alexand 
rine  doctors,  2ig 

Daremberg  et  Saglio,  Dictionnaire 
des  Antiquit^s,  66,  i8j  f . ,  363, 
371,  376  f. ,  J99 

Daures,  Gabriel,  on  "  Delos,"  373  f. 

Dawkins,  Mr.  Clinton,  164 

Diehl,Ch. ,  Excursions  archtologiques 
en  Grece,  387 

Dilettanti,  Society  of,  Unedited 
Antiquities  of  Attica,  i86\  Anti 
quities  of  Ionia,  387 

Dindorf,  rejects  Bacch.  286-297,  767 

Dio  Cassius,  37  f. ,  261 

Diogenes  Laertius,-  125,  382  ;  a 
quotation  from  Aristotle  by,  382 

Domenichi,  Ludovico,  Facezia,  331 

Dorpfeld,  Dr.  William,  i,  186,  194, 
197,  202 

Dryden,  Palamon  and  Arcite,  286 

Duemmler,  account  of  an  early 
Semitic  occupation  of  Cyprus,  280 

Duncker,  272 

Dyer,  L. ,  779 

'Epdoft&s,  Athenian  periodical,  89 


Eckermann,  Conversations  with 
Gothe,  162,  166 

Edinburgh,  Philosophical  Institute 
of,  3 

Egger,  A.  E. ,  /j 

Emperius  places  the  Satrachus  at 
Paphos,  346 

Engel,  "Cyprus,"  j/j  f.,  323, 
352  f. 

Epiphanies  in  Cyprus,  278  f. 

'E7ri0ew/)tcrts,  'Ai>aToAt/c77,  Athenian 
periodical,  89 

Erasmus,  237,  260 

Ersch  and  Gruber,  Allgemeine  En 
cyclopaedic,  ' '  Attica, ' '  344 

Etymologicum  Magmim,  107,  344, 

J59>  391 

Eunapius,  188 

Euripides,  Alcestis,  360,  777.  Bac 
chanals,  136-162  ;  lines  274-276, 

f^5  :  3°3  ft.  99  I  385-  4°6  ff-, 
287  ;  404  ff. ,  348,  352  f.  ;  877- 
881,  897-901,  184 ;  1344,  4. 
Electra,  71.  Frag.  177,  318 ; 
394.  30  I  752,  23.  Hecuba,  457 
ff.,  363 ;  459,  jj-6 ;  1267,  96. 
Helena,  1098,  304.  Heraclidae, 
66.  Ion,  23,  3s;  713,  j- ;  919 
ff.,  364  ;  1571-1600,  391  ;  1583, 
393.  Iph.  in  Tauris,  34  f.  ; 
264-274,  146  ',  1089-1105,  358, 
363;  1193,  210;  1244,  25  > 
1401  f.,  jjo.  Rhesus,  36,  99  ; 
701,  391.  Phoen.  205  ff.  and 
226  ff.,  25  ;  687,  48 
Eusebius  of  Caesarea,  263,  290 
Eustathius,  on  Dion.  Per.,  344. 
On  //.  i.  18,  327  ;  ii.  637,  z6j, 
344  \  vi.  132,  132 
Evenus,  of  Ascalon,  Anthol.  ix.  75, 
108 

FABRICIUS,  note  on  Dio,  44,  6,  37  f. 
Felix,  Quintus  Minucius,  the  Octav- 

ius  of,  8 

Fitzgerald's  Rubaiyat,  91 
Foerster,  R. ,  Raub  der  Kara,  70 
Folk-Lore,    J.    G.    Frazer   on    May 

festival,  8 


INDEX 


417 


Forbiger,  quoted  by  Bursian,  j>p7  f. 
Fortnightly,  Mr.    Pater's   articles  in 

the,  27 
Foster,    B.   W.,    M.D.,  Essay  read 

before     Birmingham     Speculative 

Club,  226 

Foucart,  M. ,  work  at  Delphi,  33 
Fourcade,  unpublished  report  of,  on 

Cypriote  agriculture,  281 
Frazer,  ].  G. ,  the  Golden  Bough,  27, 

50,  /J9  ;  on  May  festivals,  8g 
Friedlaender,  Darstellungen  aus  der 

Sittengeschichte  Rom  s,  etc. ,  8 
Furtwaengler,  on  Alcamenes'  Aphro 
dite,  270  ;  in  Roscher's  Lexicon, 

271 

GAILLARD'S  Medical  Journal,  241 

Galen,  23 4  f. 

Gardner,    Mr.   Ernest,    Excavations 

in  Cyprus,  305 
Gardner,  Professor,  on  sculptures  of 

Dionysus  and  Icarius,  105 
Gaudry,      M.      Albert,      Recherches 

scientifiques   dans   V  Orient,   275- 

277,  280 
Geel  places    Satrachus    at    Paphos, 

34^ 

Gellius,  xv.  10,  no 

Gennadius,  389 

Geographical  Societies,  publications 
of,  Berlin,  275  ;  de  I'Est,  387  ; 
Munich,  275  ;  France,  275 

Gdographie,  Revue  de,  324 

Geological  map  of  Cyprus,  275 

Ge"ologique,  Socie'te',  275 

Gildersleeve,  Professor  B.  L. ,  Essays 
and  Studies,  238  f. ,  297,  303 

Girard,  Paul,  L ' Ascltpieion,  etc.,  2 

Gladstone,  the  Right  Hon.  W.   E. 
on  Eumaeus,  47 

Gloucester,  Warburton,  bishop  of, 
180 

Goethe,  on  Aberglaube,  7  ;  wished 
Antigone  904  ff.  proved  spurious, 
166.  The  Bacchanals,  translation 
from,  162  ;  Conversations  with 
Eckermann,  162,  166  ;  on  th 
sublime  in  Euripides,  162 


Grecques,     Association    des    Etudes 

(Monuments),  70 

regory,  St.,  of  Narianzum,  4 

rohs,  Hugo,  on  Dio  Cassius,  j8 
Sronovius,  A.,  349  f.  ;  ].  ¥.,347 

uardia,  La  Mldecine  a  travers  Us 
Siccles,  226 

HARRISON,  Miss  Jane,  and  Mrs. 
Verrall,  Mythology  and  Monu 
ments  of  Athens,  27 ,  106,  no, 
no,  124,  176,  foj 

Hauvette,  M. ,  on  strange  gods  at 
Delos,  401 

Heffter,  j/j  f. 

Hellenic  Studies,  Journal  of,  2,  33, 
inf.,  181,  joj 

HelUnique,  Bulletin  de  Corres- 
pondance,  2,  400  f. 

Hermesianax,  of  Cyprus,  25- 

Herodotus,  i.  14,  02  ;  28,  85  ;  30, 
122 ;  35,  02;  59,  120;  105, 
273  ;  131,  208;  169,  J02  ;  199, 
208.  ii.  48  f.,  164  ;  52,  77  ;  81, 
164  I  97.  392'<  146,  v6j  ;  170, 
357  '  I7L  jr.  iii.  90,  85  ;  97, 
6j;  119,  166.  iv.  79,  77  ;  94, 
86.  v.  4  f. ,  84  ;  30,  J02  ;  67, 
IZ5  '•  74'  *9S-  vi.  64,  75,  198. 
vii.  6,  127  ;  75,  85.  viii.  32, 
sj  ;  65,  209  f.  ;  138,  02 

Herophilus,  237  f. 

Hesiod,  Theogony,  347,  355,  08 ; 
406  ff.,  J57;  455.  9$  \  502, 
112  \  769,  08;  912,  jo; 
913,  08 ;  969  ff. ,  62,  07. 
Works  and  Days,  298,  j/  ;  463- 
466,  jrj  ;  466,  32 ;  502,  132 ;  504 
ff. ,  Ji2  ;  700  f. ,  JJ9 

Hesychius,  82,  120,  132,  344 

Heuzey,  M. ,  on  Demeter  Graia, 
70 

Hieronymus,  note  on  Eusebius,  200 

Hippocrates,  220,  223,  225,  234, 
238  ;  irepl  fixrx'rllJLOffvv'n*i  23& 

Hogarth,  Mr.  D.  G. ,  on  Aphrodite 
of  the  Hittites,  320-322  ;  Dei'ia 
Cypria,  282,  200,  343 

Holwerda,  272 


2   E 


4i8 


INDEX 


Homer,  Iliad,  ii.  695,  47.     iii.  374, 

316;   413  ft.,  27 j.      v.    125,  47  ; 

131,    312,    371,382,   316 ;    500, 

^7;     820,    316.       vi.     135,    77. 

xiii.  322,  46,  32.      xiv.  193,  224, 

316 ;    325,    97  ;    326,    48.       xvi. 

514  ff.,  29.      xx.  105,  j/A      xxi. 

76,  ^r6  ;  416,  316.     xxiii.  220,  9^. 

Odyssey,    i.    93   f . ,   285,    <lr.      ii. 

125,    48  \     214,     359,     61.       iv. 

83,  2c£/ ;    702,    61,      v.    123  ft"., 

367  ;    125   ff. ,    6.2.      vi.    162  ff. , 

SSf.     viii.  288,  317  ;  308,  j/<5  ; 

363,  287.     ix.  196-213,    91.     xi. 

97-  ;   217,  48.     xiii.  250,  67.      xv. 

409.  367-     xviii.  193,  j/7.     xix. 

171  ff.,  61 
Homeric  Hymns,    i.    372 ;    16   ff. , 

jj6;    30-45,  392;    53  ff.,  jj7; 

118    f.,   Jj6;     135,     139,    357. 

iv.   59,    2c?7  ;    247  ff. ,   271,  273. 

v.  54-68 ;  4,  46  ;  272,  2/5.     vii. 

28,  284.     xvi.  .243-.     xxxiv.  163  f. 
Homolle,  M.  The"ophile,  2, 33,  389  f. , 

400  ;  work  at  Delphi,  33  ;  lecture 

on  Delos,  387  ;  article  ' '  Delia  "  in 

Daremberg  and  Saglio,  363,  376 

f-.  399 
Horace,  Od.   iv.  5,  39,  38  ;    iv.  14, 

25.  95 

Hutchinson,  W.,  on  Northumber 
land,  49 

Hyginus,  Fab.,  59,  65  ;  130,  /-jj- ; 
147,  64  ;  191,  92  ;  225,  119  ; 
276,  JW 

Hyperides,  fr.  286  (39,  65),  J99 

IAMBLICHUS,  a  quotation  from 
Aristotle  by,  382 

Imola,  Benvenuto  da,  Commentary 
on  Dante,  25 

Im  Thurm,  testimony  on  couvade, 
170 

Inscriptions,  43  ;  C.I.G.  3525,  14  • 
C.LL.  iii.  i,  685,  87  ;  found  at 
Delos,  274,  373 ;  found  at  Epi- 
daurus,  M.  Kabbadias  and  the, 
237  •  ]•  P-  Six  on  Cypriote  coin 
bearing  an,  344 


Institut,  Memoires  de  V ,  326 
Isaiah,  281 

JACOB,  on  Ant.  904  ff. ,  166 
Jebb,  Professor,  3,  ii,  133,  166  f. 
Jerusalem,   letter  to  the  bishop  of, 

278 

Jowett,  Dr.,  127 
Judaeus,   Philo,  Legatio  ad  Caium, 

14 

Justin   Martyr,  judgment  on  Apol- 
lonius  attributed  to,  266 

KABBADIAS,  M.,  2,  237 

Kalogeropoulos,  P.  D. ,  88  f. 

Kampouroglous,  89 

Keary,  C.  F.,  6,  9,  16,  52,  371 

Kenyon,  Mr.  F.  G.,  126,  132,  /J7 

Kern,  Dr. ,  194  f. 

Kiel,  Schriften   der  Universitat  zu, 

104 
Kitchener,   Major,  map  of  Cyprus, 

325 

Kock,  Dr.  Theodor,  215 
Krause,  44 

LAMI,  edition  of  Meursius  by,  349  f. 

Landor,  /jo 

Lang,  R.  Hamilton,  280,  288 

Largus,  Scribonius,  237 

La  Roche,  edition  of  Odyssey,  61  '.' 

Lebegue,  Recherches  sur  Ddlos,  385 

Lehrs,  Populare  Aufsatze,  10,  20  f. 

Lenormant,    M.    F. ,   66,    70,    fSj, 

217 

Leunclavius,  37 

Lloyd,  Bishop,  letter  to  Bentley,  262 
Lobeck's  Aglaophamus,  127 
Lucan,  24  f. 
Lucian,    107,    239,    261-263,    270, 

388 

Lucretius,  300,  304 
Lumbroso,  G.,  13 
Lusignan,  Estienne  de,  275,  285  f. , 

334  ff.,  342 
Lyall,  Sir  Alfred,  12 
Lycophron's  Alexandra,    447-452, 

345*- 

Lyon,  Professor  D.  G.,  301 


INDEX 


419 


Macmillaris  Magazine,  27,  101 

Macrobius,  30,  93,  96 

Malgaigne,  223 

Mannert,  353 

Mannhardt,  William,  27,  53,  97, 
182,  208,  379 

Marcellinus,  Ammianus,  308 

Marcian  Library,  unpublished  MSS. 
of  Meursius,  274  ;  unpublished 
MS.  of  a  speech  by  Zuanne  Po- 
docatharo,  340 ;  Portulano  No. 
9,  329  f.  ;  Bessarion's  MSS.  of 
Strabo,  325 

Marianus,  Scotus,  290 

Marquardt,  38,  44,  393 

Martial,  276 

Mas  Latrie,  Comte  de,  278,  326 

Maximus  Tyrius,  293 

Medecine,  Gazette  de,  223 

Medicine,  Academy  of  (N.Y.),  241 ; 
Gaillard's  Journal  of,  241 

Menander,  151 

Merriam,  Dr.  A.  C.,  13-15,  32,  107 
f.,  in,  115-117,  133,  237,  241 

Meursius,  John,  346-354  ;  Cerami- 
cus,  349  ;  (MSS.)  Cyprus,  274  f., 
290,  350-352  ;  (Graev's  edition) 
Cyprus,  347  ;  "  De  denario 
Pythagorico, "  350  ;  on  Lycoph- 
ron,  346 ',  Pisistratus,  347  ; 
"  Themis  Attica,"  350  ;  on  Theo 
critus,  348 

Middleton,  Professor  J.  H.,  33 

Milton,  79,  360  ;  Comus,  146,  148, 

150 

Minucius  Felix,  20 
Moellendorf,  Wilamovitz,  238 
Mommsen,  13,  83 
More,  Henry,  the  Platonist,  19,  264 
Miiller,  E.,  258 
Miiller,  K.  O. ,  map  of  Salamis  by, 

344 

Munro,  J.  Arthur  R.,  280 
Musgrave,  346 

Nation,  The,   of  New  York,  2,   23, 

305 

Nazianzum,  Gregory  of,  707 
Newton,  Sir  Charles,  i,  69  f. ,  795 


Nipperdey,  on  Tacitus,  43 
Nonnus,  92,  109,  139  ;  Dionysiaca, 

i.    5,  142  \    x.    init.  99  ;    xiii.  458 

ff-.  345  I    xiv-    I2o  ff-i  99;    xix. 

261  ff. ,  95;   xxvii.  283-307,  707; 

xlv.   99,   142  ;    xlvi.   passim  707- 

77O 

Northumberland,  A   view   of,  etc., 
49  f. 

Nouvelle  Revue,  374 

OBERHUMMER,  Dr.,  275  f.,  278 

Olympiodorus,  87 

Omar,  the  Persian,  97 

Origen,  258,  264,  278 

Orphic    .Hymns   on    Dionysus,    22, 

162 

Orphic  fragments,  7<?7 
Ortelius,  Abraham,  maps  of  Cyprus, 

324,  326,  343,  135 
Osann,    on    the    Lesser    Mysteries, 

104,  124 
Ovid,  64  f. ,  375,   Fasti,  iii.  717, 142 ; 

v.  145,  38.     Her.  ii.  65 

PAMPHOS,  122,  175 

Pandora,  Athenian  periodical,  89 

Pape,  Diet,  of  Proper  Names,  132, 

39<> 
Pater,  Mr.,  26  f.,  70,  101,  136, 169, 

235  i.,  242 
Pathe,  C.  von  Wulfften,  on  Philos- 

tratus,  259 
Patzig,  H.,  27 
Pauly's      Real  -  Encyclopaedic      dcr 

Alterthumswissenschaft,  44,  118, 

127 
Pausanias,  32,  218,  247,  285  ;  I.  ii. 

5,  104  ;  ii.  7,  135  ;  iv.  5,  92  ;  xiv. 

1-4,  776  ;  ibid.  3  f.,  124  \  xxii.  i, 

5,  j>99  ;    xxxi.   4,  31,  82  ;    xxxii. 
i,  J99  ;  xxxviii.  3,  64,  122,  175; 
ibid.   7,   64.      II.   ii.   5,   6,  7<?9  ; 
ibid.  14,  65 ;    xxiv.  6,  99  ;    xxxv. 
4,  776.      III.  xix.  6,  779  ;    xxiii. 
i,  j>77.      IV.  iii.    10,  65.      V.  x. 
8,  9.       VIII.  v.  2,  286  ;    xxiv.  i, 

6,  286 ;    xxv.   4,   176 ;    xlviii.  3, 
J77 ;    liii.   3,    286.      IX.   xii.   j, 


420 


INDEX 


140  ;    xx.    i,   1 1 9  ;    xl.    3,  399 
ibid.  3,  4,  379.     X.  iv.   3,  132 
xxix.  4,  ^oo ;  xxxii.  6,  25- 
Peabody  Museum,  Report  of  Trus 

tees  of,  171 

Penrose,  Mr.  F.  C. ,  232,  314 
Perrot  and  Chipiez,  310 
Perry,  Mr.  T.  S.,  if 2 
Petelia,  Orphic  inscriptions  found  at 

181 

Pfander  on  Euripides,  157 
Pherecydes,  64 
Philios,  Dr.,  i,  196  f. 
Philo,  Judaeus,  14 
Philological    Association,     Transac 

tions  of  the  American,  JTJ,  347 
Philologus,  the,  j6,  345 
Philostratus,  258-261*   263  \    Epist 

39.  ^07 
Phurnutus    (Cornutus),    quoted     b; 

Meursius,  274 
Piccolomini,    Aeneas     Sylvius    de' 

3*61 

Pindar,  71,  130.    Olympian  Odes,  i 
102  ;   ii,  71  \  12,  102.     Pythian 
Odes,  3,  <?ff.,  369;   9,  369 
Pius  II.  (de1  Piccolomini),  326  f. 
Plato,  19.     Apol.  21  B,  370.     Epi- 
nomis,  487  E,   297.      Laws,  viii. 
init. ,  177.     Rep.  ii.   364  B,  181 ; 
382  £-383  B,  370 
Plautus,  Mercator,  init.,  272 
Pliny,  41  f.,   253.      Nat.  Hist.  II. 
vi.  12,  vii.  18,  19,  42  •    VII.  viii. 
45.  42  ;  59.  107  ;  195,  294 
Plutarch,   Alexander,  97  ;    Antony, 
24,    79  ;      Aristides,     27,     210  ; 
Demetrius,  26,  209  ;  Flamininus, 
16,  40  ;  Nicias,  3,  J79  ;  Pericles, 
13,  7-97;  Solon,  12,  .T2?;  29,  z/d, 
7.2<5  ;   Theseus,  23,  .577 
Plutarch,  Pseudo-,  De  Anima,  no  ; 
tract  on  Isis  and  Osiris,  22  ;    De 
Fluviis,   25  ;    De  primo  frigido, 
113 ;    on   Hesiod,   132 ;    Quaest. 
Rom. ,  310  ;    Sympos. ,   103,  106, 
258 

Podocatharo,  the  family,  330-343  ; 
Alessandro  P.,  339  ;   Hettore  P., 


Ritratte  del  Regno  di  Cipro,  330- 
33S>  340-342 ;  Zuanne  P. ,  his 
appeal  to  Alvise  Mocenigo 
(Marcian  MSS. ),  338-340 

Polemon,  218 

Porphyry,  72 

Potli,  88 

Praktika,  i 

Preller's  Greek  Mythology,  48,  98, 
177  ;  Plew's  revision,  318  \  Ro 
bert's  revision,  272,  298 ;  Ritter 
and,  Hist.  Phil.,  177 

Procacci,  Tommaso,  L 'hole  piu 
famose  del  mondo,  330-333,  337 
f.  ;  de'  Funerali  Antichi,  331  \ 
Cagione  delle  Guerre  Antiche,  331 

Propertius,  14 

Ptolemy,   Bessarion's  MS.  of,  324 

Pullan,  R.  P.,  z 

QUINTILIAN,  42 

RAGOZIN,  Mme.  Zenai'de,  302 
Ralli  and  Potli,  Church  History,  88 
Rapp,    Beziehungen    des    Dionysos- 

kultes  zu  Thracien,  83 
Reber,    Professor,   History  of  Art, 

308 
Ribbeck,  Otto ,  A  nfaenge  des  Dionysos- 

kults  in  Attica,   104,    125,  129, 

132 

fitter  and  Preller,  777 
Robert,  Carl,  interpretation  of  vase- 
pictures  by,  94 
^onchaud,  M. ,  371 
Roscher,    Dr.    W.    H. ,  402 ;    Aus- 
fiihrliches  Lexicon  der  Griechis- 

chen  und  Roemischen  Mythologie, 

120,  139,  270,  292 
Ross,  Dr.  L. ,  290,  352,  389 
oyal  Institute,  lecture  before,  39 
luhnken,  54 

IALIGNACIUS,      BARTHOLOMAEUS, 

Itinerarii,  290 
andys,  Dr.,  142,  146,  148 
appho,  274,  304 
assenay,  le  Marquis  de,  324 
cherer,  Chr.,  176  f.,  194 


INDEX 


421 


Schliemann,  Dr.  H.,  106,  286 

Schone  on  Euripides,  757 

Schramm's  Life  of  Meursius,  348 

Schwalbe,  166 

Seeley,  Professor,  on  Roman  Im 
perialism,  39 

Semus  of  Delos,  382 

Seneca,  290 

Servius,  confusions  about  Parnassus, 
24,  8s,  107 

Sibylline  Oracles,  290 

Sidonius  Apollinaris,  265 

Six,  Dr.  J.  P. ,  344 

Smith,  Mr.  Cecil,  181 

Smith,  Mr.  R.  Elsey,  252,  305-308 

Sophocles,  133.  Antigone,  904-915, 
discussion  of  interpolation  of, 
166  f. ;  1115-1152,  133  f.  ;  1119, 
the  reading  Icaria  justified,  133  ; 
1126,  1144  f.,  25;  1200,  777. 
Philoctetes,  177,  237.  Triptole- 
mus,  77 

Stephanus,  Byzantinus,  105,  181, 
274,  344 

Strabo,  183  f. ,  325,  327>  34f>  395- 
397  \  37.  358'>  271,  356  \  371, 
103  ;  375.  237  ;  383,  218 ;  394, 
345  \  395.  l89'>  412.  **8\  4J7f-. 
25 ",  463-474.  94 I  467.  i&4\ 
468,  132,  159  ;  474,  395  ;  479. 
123  f.  ;  484,  306  ;  480,  360  ; 

485.  359>  Jte.  395  f-  J  564. 
^5  ;  633,  122 ;  683,  the  tradi 
tional  punctuation  changed  accord 
ing  to  Bessarion's  MSS.,  325 

Suetonius,  14,  37  f. 

Sybaris,  Orphic  fragments  found  at, 
181 

Syncellus,  7/9 

TACITUS,  Ann.  i.  72,  iv.  42,  xii. 
4,  xiii.  ii,  43  ;  iv.  20,  42;  iv. 
55,  44.  Hist.  iv.  81,  42 


Themistius,  ir6 

Theocritus,  318,  394 

Theognis,  356,  381,  384 

Theopompus,  126 

Thiersch,  130 

Thraemer,  779 

Thucydides,  7J7,  361-363 

Tomaschek,  86 

Tournefort's  description  of  Cyprus, 

385 

Tylor,  Dr.,  769  f.,  289-207 
Tyrius,  Maximus,  2OJ 
Tyrrell,  Professor,  757,  162,  166  f. 

UNGER,  Dr.  Robert,  345 
Uspergensis,  Abbas,  290 

VERCOUTRE,  Dr.,  229 

Verrall,  Mrs.,  27,  124 

Vienna,  Reports  of  Academy,  86 

Virgil,  Aeneid,  iii.  15,  85  \  73,  359  \ 

74,jj<?;  85-101, 370.     Georgics, 

i.  21-31,  14 
Vitruvius,  797 
Voigt,  F.  A.,  97,  120,  159, 

176  f. 

WARBURTON,  The  Divine  Legation 

of  Moses,  1 80 
Wecklein,    rejects   Bacch.    286-297, 

167 
Wegener,    Dr.,    on    the    Hymn    to 

Demeter,  56 
Wilmans,  Dr. ,  on  Dio  Cassius,  38 

XENOPHON,  his  record  of  military 
medicine,  225.  Anab.  I.  ii.  13, 
92.  Mem.  III.  iii.  12,  376  ; 
IV.  ii.  8  ff. ,  225 

ZENODOTUS,  on  Odyssey  text,  61 
Ziegler,  on  Cyprus,  308,  327 
Zonaras,  37 


422 


INDEX 


II. — GENERAL  INDEX 


ABONOTICHUS,  serpent  impostures  of 

Alexander  of,  254 
Abraham,  statue  of,  in  Lararium  of 

Severus,  265 
Academy,    Cicero  proposes   a   gate 

for  the,  1 88 
Acamantis,       Philonides         named 

Cyprus,  274 
Acamas,  a  mountain  of  Cyprus  and 

of  Salamis,  187,  344,  JS2  f- 
Acanthus-leaf,  on  the  capitals  of  the 

Epidaurian  Tholos,  252 
Accad,   Aphrodite    came    from  un 
known,  303 
Accounts,    inscriptions    concerning, 

found  at  Eleusis  and  Delos,  195, 

207,  387 
Achilles,  the  Apolline  ideal  and,  9  ; 

fire-baptism  of,  64 ;  Chiron's  pupil, 

232 
Acropolis  (Athenian),    chamber  on, 

devoted  to  remains  of  Aesculapian 

temple,  249 
Acropolis    (Eleusinian),    187,     199, 

203  f. 
Acta   publica,    consulted     by    Dio 

Cassius,  38 
Actor,   inspiration  of,  by  Dionysus, 

117 

Adonis,  291  f. 
Aee'tes,  231 
Aegae,  Apollonius  and  Maximus  at 

the  temple  of  Aesculapius  at,  255, 

257*. 
Aegean,    Archilochus    the   poet    of 

the,  365 ;    festivals  of  the,  389  ; 

Venetians  and  Greeks  in  the,  328, 

389  ;    Delian  centre  of  the,  387  ; 

Archipelago,  a  name  of  the,  397 
Aegina,  187,  223 
Aegipan,  98-100 

Aegyptos,  river,  Nysa  close  to,  163  f. 
Aelian,  ridiculous  credulity  of,  230 


Aeneas,  virtues  of,  39 ;  Chiron's 
pupil,  232 

Aeolus,  housed  on  Tenos,  380 

Aeschylus,  the  record  of  the  cou- 
vade  and,  169-171  \  Dionysus  the 
tutelary  god  of,  81 

Aesculapius,  15,  220,  222,  231  f., 
235  f->  239  f.,  242,  245,  255; 
medicine  and,  15,  220,  226,  234, 
f. ,  237  f.  ;  myth  of,  15,  220  f., 
230  f.,  232  f.,  234  f.,  244-246, 
254,  256  ;  other  gods  and,  15, 
240,  242  f. ;  Amphiaraus  and,  232 ; 
Aphrodite  and,  270  f.  ;  Apollo 
and,  15,  219  f. ,  231,  239-242 ; 
Apollonius  of  Tyana  and,  220, 
255<  257  '•  attendant  divinities 
and,  240  ]  Demeter  and,  194, 
219  f.,  270  f.  ;  Dionysus  and, 
194,  220  f. ,  247 ;  Persephone 
and,  219  f.  ;  Serapis  and,  254  ; 
Zeus  and,  is,  21,  219  f. ,  239  ; 
worship  of,  229  f. ,  236,  238  i. ,  243, 
247,  249,  254,361 ;  at  Aegae,  255- 
258  ;  at  Athens,  2,  5,  242,  248  f.  ; 
at  Cos,  194  f.,  221,  225  ;  at  Epi- 
daurus,  2,  3,  219  f. ,  220,  243-245, 
248-254  •  among  the  Lapithae 
and  Phlegyae,  194  f. ,  221,  233  f. ; 
at  and  near  Rome,  221,  234$, ,  243, 
254 ;  in  Thessaly,  194  f. ,  221, 

233*. 

Agamemnon,  285,  293 
Aganippe,  24 
Agapenor,  285  f. 
Agave,  137,  147,  150,  160-162 
Agelastos  petra,  the,  at  Eleusis,  61 , 

207 

Aglaurus,  mother  of  Ceryx,  122 
ayvr],  used  of  Persephone,  48 
Agriculture,  in  Cyprus,  281  ;  know 
ledge    of,    came    from    Demeter 
through    Triptolemus,     71 ;    im- 


INDEX 


423 


plements    of,     Cinyras    invented, 
294 

Agrionia,  103 

Aidoneus,  49  f. ,  55'  57,  Z78>  '94 
Alyafov     Trt\ayos,     corrupted    into 

Archipelago,  J97  f. 
Aiorai,  706 

Aithousa,      mother      of     Eleuther, 
daughter  of  Alcyone  and  Poseidon, 
etc.,  779 
Ajax,  46,  78 
Alaric,  188,  265 
Alban  hills,  326 

Albanians,    May   festivals    of,    89 
wasting  of  Eleusis  by  the,  188  f. , 
796 

Alcaeus,  702 
Alcamenes,  statue  of  Aphrodite  by, 

270,  273 
Alcibiades,  215 
Alcmaeon,  dissection  of  animals  by 

223 

Alcmaeonidae,  arraignment  of,  124 
Aletis,  song  in  memory  of  Erigone 

106,  115 

Alexander,  Bacchanals  of  the  hous 

of,  138  ;    Dionysus  and,  79  ;   th 

empire  of,  44  f.  ;    the  Emathia 

conqueror,  79 

Alexander,     of    Abonotichus,     254 

263 

Alexandria,  Caesareum  at,  13  f. 

Alexandrine    doctors,    Daremberg 

list  of,    279  ;    sailors,   worship  o 

Caesar  eTri/Sar^ptos  by,  14 

Ali  Pasha,  of  Janina,  268 

Alliance,  the  Eleusinian,  of  divinitie 

77,    776,    277,    276 

Alpheius,  163 

Amadis  de  Gaule,  anatomy  of,  com 

pared  to  Homer's,  228 
Amathus,  a  Phoenician  foundatio 

281,  291  f. 
Amathusia,  Xenagoras  called  Cypru 

274 

Amazons,  38,  249 
American  excavation  at  Icaria,  10 

114 
Amphiaraus,  a  Boeotian  Aesculap 


and  a  pupil  of  Chiron,  shrine  at 
Oropus  of,  232 

mphictyon,    in    Dionysus -legend, 
7/9 

mphipolis,    Phyllis    on   the   shore 
near,  65 

mphipolitans,     deified      Brasidas, 
40  f. 

nacreon,  sang  under  inspiration  of 
wine,  102 

natomy,  Homer's  clear  notions  of, 
227-230 ;    origin  of,    234,     237 ; 
heroic  interest  in,  227  ;   influence 
of  Homer  in,  229 
naxagoras,  loftier  teachings  of,  in 
the  Bacchanals,  138 
ndania,    competition    with   Celeae 
and  Eleusis  of,  65 
Androclus,  son  of  Codrus,  122 
Andros,  on  the  way  to  Delos,  376  ; 

among  the  Bahamas,  J97 
Anemones,  on  the  Tholos  of  Poly- 

cletus,  251 
\nimism,  27 

Anthesteria,  at  Athens,  130  f. 
Anthesterion,  month  of  the  Lesser 

Mysteries,  208 
\nthropology,  scientific  method  in, 

169  f. 

-\nthroporraistes,    dvdpuTroppa.l<TTT]S, 
or  man  wrecker,    Dionysus'   sur 
name,  100 
Antigone,     the,     interpolation     in, 

i66f. 
Antonines,  the  age  of  the,  an  age  of 

valetudinarians,  235 
Antony,  Mark,  as  Caesar's  flamen, 
38 ;  masquerade  of,  as  Dionysus, 
79 

Anubis  and  Apollo  at  Delos,  402 
Apamea,  14 

Aphrodite,  2,  5  ;  nature  of,  273  ', 
powers  of,  277  ;  came  from  every 
where,  272 ;  a  nature  goddess, 
271,  273;  a  jealous  power,  273;  a 
restraining  power,  300  ;  bore  a 
sword,  46,  270,  273  ;  a  consoling 
power,  277,  298,  303  ;  a  goddess 
of  the  dead,  292,  309  f.  ;  sur- 


424 


INDEX 


named  Sosandra,  271,  298,  303 
ancient  influence  of,  277,  303  f. 
modern  views  about  the  origin  of 
315 ;  had  a  strain  neither  Greek 
nor  Phoenician,  282,  324  ;    cam< 
from  Accad,  303  ;   worship  of,  in 
the    East,    273  f.,    293,    297   f. 
Greek  influences  never  quite  pre 
vailed  with,  271-273,  297,  317 
Aphrodite,  of  the  Greeks,  270-273 
284,  297  f. ,  303  f. ,  315  f. ,  319 
of  the  Hittites,  315,  320  f.  ;    o 
the  Phoenicians,  273  f. ,  282,  284 
286,   297  f. ,  303,  31 5,  318,  320, 

323 

of   the  poets. — Euripides  on, 

298-300 ;     Homer    on,    298    f. , 
304,  317  f.  ;  Pindar  on,  299  f. 

Aphrodite's  associates. — Adonis  and, 
291  ;  Aesculapius  and,  270  f. 
Apollo  and,  271,  379,  382,  399- 
403;  Ariadne  and,  400  f.  ;  Cybele 
and,  21,  303  ;  Demeter  and, 
21,  69,  303  ;  Dionysus  and,  148, 
273,  3io 

sanctuaries        (Greek).  — At 

Athens,    270     f.  ;      at    Cythera, 
285  f. ,  323  ;   at  Delos,  ^79,  382, 
399,  403  ;  towards  Eleusis,  218, 
271,  296  f. ,  304,  314  ;  at  Psophis 
and  Tegea,  285  f. ,  323 

sanctuaries         (Cypriote). — 

'53.    274>    2$S<    287,    296,  317, 
321  f.  ;  at  Old  Paphos,  272,  279, 
292 

Zerynthian  sanctuary,  345  f. 

Apolline   perfection,    Apollonius    of 

Tyana  and,  220 

Apollo,  5,  8,  9,  14  f.,  17  f. ,  24,  28- 
30,  33  f.,  169-171,  219,  242,  264, 
366-369,  371  f. 

Apollo's  associates.  —  Aesculapius, 
220,  240-242 ;  Aphrodite,  271, 
300  ;  Ares,  372  ;  Artemis,  355, 
367 ;  Athena,  218  ;  Augustus, 
14  ;  Bacchus,  25  ;  Coronis,  245  ; 
Daphne,  368 ;  Demeter,  75-9, 
218 ;  Diana,  150  ;  Dionysus,  21, 
30  f. ,  34,  36,  Sf,  102,  104,  159, 


162, 368  ;  Eleuther,  119  f.  ;  Har- 
pocrates,  401  f.  ;  Hecate,  159  ; 
Heracles,  371  f.  ;  Hermes,  372  ; 
Horus,  401  f.  ;  Marpessa,  368  ; 
the  Muses,  102,  104,  159  ;  Nero, 
114  ;  Poseidon,  372  ;  Proserpina, 
r59  '•  Pythagoras,  219  ;  Zagreus, 
128 ;  Zeus,  26,  369 

Apollo's  sanctuaries. — At  Bassae, 
372  ;  at  Curium,  345  f.  ;  on  Mt. 
Cynortion,  246  ;  at  Daphne, 
218 ';  at  Delos,  2,  5,  35,  201, 
360-363, 370,  372,  375,  37^-380, 
382,  384-389,  394,  401  f.  ;  at 
Delphi,  17,  26,  29,  33-36,  370, 
372 ;  at  Icaria,  102,  104 ;  at 
Marathon,  104 ;  on  Parnassus, 
104 

story,    no,     128,    219,    231, 

239,    294,    356-350,    369,     372, 
379.  382,  391,  393,  399 

Apollonia,  celebrated  during  Delian 
independence,  374,  386-389 

Apollonius  means  one  under  Apollo's 
guidance,  219 ;  seven  Alexandrine 
doctors  named,  219 

Apollonius  of  Tyana,  reasons  for 
the  discredit  of,  258  f. ,  261  f.  ; 
important  facts  concerning,  220, 
255,  257,  261,  263-266 ;  relation 
to  Christians  and  Christianity  of, 
258,  260  f.,  263  f. 

Appius,  Claudius  Pulcher,  gate  of, 
at  Eleusis,  188,  190 

Arabian  strain  of  Dionysus,  168 

Araby  the  blest  worships  Dionysos, 
140 

Arcadia,  temples  of  Aphrodite  in 
285  f. 

Arcadian  forms  in  Cypriote,  280 ; 
legends  of  Demeter  Erinys  and 
her  daughter  Persephone  A^o"- 
Troiva,  48 

Arcadians,  the,  when  and  whence 
they  entered  Cyprus,  280 

Archaeological  Society,  Greek,  i  f. 

Archaeology,  7 

Archedemus,  his  Athenian  grinders, 
214 


INDEX 


425 


Archelaus,  wild  religion  at  the  cour 

of,  137 

Archilochus,  the  poet  of  the  Aegean 
365 ;  and  Apollo,  j66  ;  and 
Paros,  365 

Archipelago,  a  corruption  of  Aiyaiov 
irtXayos,  397  f. 

Archipelago,  Athens  in  close  com 
munion  with  the  islands  of,  81 
the  carnival  of,  and  the  Delia  and 
Apollonia,  388  f.  ;  tales  of  Diony 
sus  in,  ii f  ;  Thracians  in  earliest 
days  controlled  the,  80  ;  seat  of 
Thracian  power  in,  was  Naxos, 
80 

Archon,  marriage  of  Dionysus  with 
the  wife  of  king,  130  f. 

Arcturus-Icarius,  no 

Ares,  Apollo  boxed  with,  372 

Aresthanos  found  Aesculapius  on 
Mt.  Titthion,  244  f. 

Argive  account  of  Eubouleus-Diony- 
sus,  176 

Argolis,  Polycletus  of,  his  two  Epi- 
daurian  masterpieces,  246 

Ariadne,  130,  379,  401 ;  Aphrodite 
and,  400  f. ;  Daedalus  and,  399  ; 
Demeter  and,  62;  Dionysus  and, 
130  ;  Persephone  and,  62  \  Thes 
eus  and,  at  Delos,  382 

Arician  Grove,  picturesque  but  com 
paratively  unimportant  rites  of, 

27 

Aristaeus,  the  giver  of  honey,  reared 
Dionysus,  143 

Aristophanes,  68,  212-216  ;  Apollo 
and,  28  ;  Dionysus  and,  81 ; 
Eleusinia  and,  212-216 

Aristotle  on  Dithyrambs  and 
Tragedy,  116  ;  corroborates  the 
Parian  Marble,  132 

Art,  Delian  and  Metapontine,  383 

Artemis,  245,  jjj  f. ,  367,  399  ;  at 
Delos,  336,  379,  385  ;  at  Eleusis, 
189  f. ;  at  Epidaurus,  230 

Arts,  Pygmalion  and  Cinyras  origin 
ated  the  useful,  294 

Ascalon,  Evenus  of,  108 

Asclepieion  at  Athens,  2 


Ascoliasmos,  at  Dionysiac  festivals, 
origin  of,  108 

Ashdod,  Dagon  at,  and  Eastern 
Aphrodite,  297  f. 

Ashtaroth  and  Cypris  -  Aphrodite 
differentiated,  320  f. 

Asia,  Central,  12 

Asia  Minor,  i,  2  ;  attempt  to  derive 
Aphrodite  from,  271  f.,  320, 
322  ;  Athenians  in,  122,  132  ; 
contributions  from,  to  the  myth 
of  Cinyras,  293,  295 ;  Cnidian 
sanctuary  in,  70  ;  north  Cyprus, 
an  extract  from,  277  ;  Maenads 
followed  Dionysus  from,  141  ; 
Lenaeon,  a  month  name  of,  132 ; 
land  of  Phrygian  flute  and  Lydian 
airs,  294 ;  Sporades  lie  near, 
392 ;  Taurus  range  and  north 
range  of  Cyprus,  273  ;  Thrace  of, 
84 

,  ten  in  Asia,  44 
,  apxiepevs  rrjs,  43 

Aspelia,   Xenagoras  called   Cyprus, 

274 

Assumption,  the,  and  the  bringing  of 
Semele  to  Olympus,  777 

Assurbanipal,  Sardanapalus,  poem 
on  Ishtar  found  in  library  of,  300 

Assyria,  Aphrodite  ultimately  from, 
272  ;  Cinyras,  king  of,  291 

Assyrian  and  Greek  strains  in  Aphro 
dite,  298 ;  idea  of  Ishtar,  273, 
303  ;  stranger,  Phoenician  wine 
trade  and  Dionysus  the,  165 ', 
rosettes  an  inheritance  from,  2jf 

Astarte  -  Aschera,  and  Aphrodite, 
277 

Asteria  and  Ortygia,  barren  legend 
of,  358 

Astronomy,  of  Icarian  story,  no  f. ; 
in  a  Delian  legend,  358 

Astynomus,  called  Cyprus  Crypton 
and  Colinia,  274 

Atargatis,  worship  of,  added  to  that 
of  the  Syrian  goddess  at  Delos, 

403 

Athena,  appeared  to  Alaric  and 
saved  Athens,  263 


426 


INDEX 


Athena,  frequent  mention  of,  at 
Citium,  322 ;  Cynthia,  Delian 
temple  of,  402 ;  and  Demeter, 
shared  Apollo's  temple  at  Daphne, 
218 

Athene  Sotera,  Pisistratus  and,  126  ; 
and  the  heart  of  Zagreus,  181 

Athenian  shrine  of  Rome  and 
Augustus  on  the  Acropolis,  40  ; 
temple  of  Aesculapius  next  the 
Acropolis  (inscription),  241 ; 
and  Argive  account  of  Eubouleus- 
Dionysus,  176  ;  Confederacy,  12  ; 
religion,  became  the  same  with 
Attic,  122  f . 

Athenian  worship,  at  Delos  (a)  archi 
tecture,  386,  (b]  festivals,  374-381, 
(c)  legends,  382, 399  i. ;  at  Eleusis, 

122,  124,  207-218 

Athenian  worship  of  Dionysus,  123, 
134  ;  derivation  of,  81,  130, 
137,  162  \  festivals  of,  128-132, 

!34 

Athenian  theatre  of  Dionysus,  114, 
ng 

Athenians  in  Asia  Minor,  122,  132  ; 
Venetians  and,  131,  328 

Athens  typically  free,  n  ;  the  Archi 
pelago  and,  81 ;  Cecrops  at,  64  ; 
commercial  jealousy  of,  363 ; 
Democedes  at,  224  \  Eleusinion 
at,  124,  207  ;  Eleusis  and,  122, 
208 ;  Eleutherae  and,  118  f.  ; 
Epimenides  at,  124  •  Icarian 
influences  at,  114,  117,  129,  132 ; 
religious  innovations  at,  129  f. ; 
worship  of  Aphrodite  at,  270 ; 
worship  of  Apollo  (Delian)  at, 
362  f. ,  376  f. ,  386  f.  ;  worship  of 
Apollo  (Delphian)  at,  34;  worship 
of  Aesculapius  at,  219,  222,  241, 
248  f. ,  256  ;  worship  of  Demeter 
and  the  Eleusinian  divinities  at, 
77,  118,  122,  124,  207-211,  215- 
217  ',  worship  of  Dionysus  at,  36, 

77,  80   f.,   89  f.,    II7-IIO,,   121  f. , 

123,  129-131,  fjj-sjj 

Athens  and  Venice,  religions  con 
trasted,  131 


Atlas,  father  of  Aithousa's  mother 
Alcyone,  119 

Atthis,  Adonis  closely  connected 
with,  291,  322 

Attic  Demeter  legend,  60  ;  Diony 
sus  legend  and  worship,  34,  104 
f. ,  113,  143  ;  confederation,  60  ; 
religion,  122  f. ,  129  f. ,  218 ;  sense 
of  measure,  34,  104  f. ,  113 ; 
tragedy,  117 

Attica,  the  Bacchanals  called  the 
Passion  Play  of,  137  ;  song  of  Bac 
chanals  and  winter  festivals  of, 
156  ',  country  demes  of,  and 
Dionysus,  34  \  Demeter  and 
Dionysus  came  to,  105 ;  Euripides 
wrote  for  Greece  as  well  as  for, 
143  ;  exclusiveness  of  local  reli 
gions  in,  122  f.  ;  Icaria  the  Pieria 
of,  in ;  belief  in  Pisistratus  of 
countrymen  of,  126 ;  political 
and  religious  fusion  of,  122  ;  early 
influence  of  Thracians  on,  in  \ 
beautiful  Triptolemus  myth  of, 

71 
Atticus,  lukewarm  about  building  in 

Academy,  188,  190 

Aufidus,  bull-shaped,  95 

Augustales,  in  Italy,  Sicily,  Gaul, 
Spain,  etc.,  43 

Augustalia,  local  celebrations  analo 
gous  to,  43  f. 

Augustan  poets,  imperial  religion 
and  the,  38 

Augusti,  oaths  In  acta  divi,  43 

Augustin,  St.,  commends  self-con 
trol  of  Apollonius,  264 

Augustus,  deification  of  genius  of, 
38  ;  worshipped  as  'AXe^/ca/cos, 
Zeus,  Apollo,  Poseidon,  14 ; 
Egyptian  style  of,  14 ;  Julius 
Caesar  and,  37  ;  masquerading 
scheme  of,  37  f .  ;  Rome  and  wor 
ship  of,  40 ;  wished  to  govern  not 
to  reign,  37  ;  worship  of,  Rome 
and,  45;  worshipped  by  "circum 
locution,"  40 ;  and  temples  for 
imperial  worship,  38  ',  temple  at 
Jerusalem  and,  8 


INDEX 


427 


Aurelian,  vision  by,  of  Apollonius, 

26S 

Autonoe,  on  Cithaeron,  75-0 
Autopsy   of  the  Greater  Mysteries, 

209 
Axius,  Dionysus  crosses  the,  757 

BABYLON,  Aphrodite-Mylitta  at,  303 
Bacchanal,  Apollo  a  prophet,  30 
Bacchanalian      revels,       Euripides' 
understanding  of,  fjo  f.  ;  victory, 
184 1 

Bacchanals  of  everyday  life,  7/2, 
138  ;  full  of  Dionysus,  136  ;  com 
parison  of,  with  Pans,  99  f.  ; 
in  the  Thiasos,  100 ;  Thebes 
the  mother  of,  133 ;  Thracian 
originals  of,  97 

Bacchanals,  The,  of  Euripides,  pre 
figures  future  and  sums  up  the 
past,  138  ;  music  of,  144  ;  called 
the  Passion  Play  of  Attica,  7J7  ; 
many  aspects  of,  136  f. ,  7jj ; 
relation  of,  to  Dionysus,  (a)  its 
scene  is  full  of  Dionysus,  7J9  f., 
(b}  it  presents  in  full  the  Bacchic 
cult,  757,  (c)  a  gospel  of  Dionysus, 
136,  (d)  fullest  presentation  of 
Dionysus,  727,  (e)  contains  proto- 
Thracian  Dionysus  and  teachings 
of  philosophy,  138  ;  all  characters 
in,  are  prophets  of  Dionysus,  156 
f.  ;  stand  for  Dionysus,  7J9 ; 
Dionysus  not  the  sole  representa 
tive  of  the  god  in,  7j>9,  i^S  • 
analysis  of  the  action  of,  140- 
143,  145-160,  162  ;  details  of, 

(a)  written    in    Macedonia,    137, 

(b)  lines  of,  preserved  in  Christus 
Pattens,  139  ;    (c)  Milton's  emen 
dation  of,   146,  (d)  interpolation 
in,  766  f.  ;    (e)  Cadmus  in,  140, 
145-147  f. ,  (/)  Tiresias  in,  145 

Baccheios,  Dionysus,  at  Corinth,  759 
Bacchic  power,  wine  represented  the 
sterner  as  well  as  the  more  charm 
ing  side  of,  97 

—  cult,  presented   in  full  by  the 
Bacchanals,  757 


Bacchic  worship,  stage  of,  investi 
gated  by  Americans,  104',  re 
verence  for  Cybele  in,  7^7;  the 
wanton  ferule  in,  142  ;  ivy  and 
oak  for,  142 

dance,  its  graceful  undulations 

and  fitful  variations,  707,  146 

images  at  Corinth  of  Pentheus' 

tree,  150 

instruments,  mention  of,  143 

madness,  the  method  of,  145 

Bacchus,  Homeric  notion  of,  77 ; 
dance  of,  142  ;  lover  of  laurel, 
30 ;  torches  of,  247 ;  religious 
consolation  of,  135 

Bactrian  forts,  Dionysus  worshipped 
at,  140 

Baetylus,  on  Mt.  Cynthus,  386 

Daffo  olim  Paphos,  324 

Balkan  Peninsula,  85,  89 

Bank,  the  Delian  shrine  a,  387 

Baptism,  analogy  of,  to  Lesser  Mys 
teries,  209 

/3dp/3apoj,  Meursius  makes  a  river  of 
adjective,  353  f. 

Barnabas,  St. ,  in  Cyprus,  290 

fo,  the  dpxuv,  with  four  ^TTI- 
administered      cult     of 
Athens  and  Eleusis,  208 

Basilissa,  marriage  of  Dionysus  with, 
130  f. 

Basque  country,  persistence  of  cou- 
vade  in  the,  770 

Bas-relief  of  Dionysus  and  Icarius, 
106 

Bassae,  temple  'of  Apollo  at,  204, 
206,  372 

Bassarids  of  Dionysus,  79,  97.  See 
Maenads. 

Bavarian  killing  of  Pfingstl,  159 

Beans,  Cyamites  the  giver  of,  and 
the  Mystae,  277  ;  excluded  from 
Demeter's  sanctuary  at  Eleusis, 
277  f. ;  horror  of,  felt  by  Pytha 
goreans,  218 

Bentley,  abandoned  edition  of  Apol 
lonius,  262 

Bermius,  rose  gardens  on  flanks  of, 
92 


428 


INDEX 


Birth,  forbidden  on  Delos,  360  \ 
principle  of  second,  justifies 
Orestes,  169  f. 

Birth-legend  of  Aesculapius,  75-, 
233,  245 ;  of  Aphrodite,  271, 
284,  296,3T7  f.,Js6-3S9>  399  > 
of  Dionysus,  128,  149,  165,  168- 

173 

Birthday -festival  of  Apollo  and 
Dionysus'  flower  feast,  372 

Bishop,  functions  of,  analogous  to 
those  of  'Ao-tdpx^s,  44 

Bi0wid/>x?7S,  44 

Blizzard,  the  most  ancient  record  of 
a,  113 

Bocalia,  an  alternative  name  for 
Bocarus,  345  f. 

Bocarus,  a  Salaminian  river -name 
wrongly  connected  with  Cyprus, 
324  f. ,  344'347>  352-354 

Body,  mind  cured  through,  235 

Boedromion,  month  of  the  Greater 
Mysteries,  208 

Boeotia,  Amphiaraus  the  Aescula 
pius  of,  232  ;  Eleutherae  on  Attic 
borders  of,  118 ;  Lenaeon  in,  112, 
132  ;  early  Thracians  in  various 
parts  of,  80 

Boeotian  festival  of  Agrionia,  con 
nection  of  Dionysus  and  Muses 
in,  103 

Boeotians  organised  a  Delian  festi 
val,  363 

Bondholders,  the  Turkish,  and 
modern  Cyprus,  278 

Bootes-Icarius  as  vindemiator,  ITT  ; 
Icarius  and  his  wain  become, 
no 

Bordone,  identifies  Troodos  and 
Olympus,  329 

Botticelli  and  spirit  of  mystery  at 
Eleusis,  182  ;  illustrations  of 
Dante  by,  331 

Brasidas,  deified  by  Amphipolitans, 
40  f. 

Brenzone,  account  of  sack  of  Nicosia 
by,  332  i. ;  reproduces  Procacci  on 
Troodos-Olympus,  334 

Bricks,    devised   by   Cinyras,    2()4 ; 


transport    of,     from     Athens    to 

Eleusis,  207 
Bridge,    jibes  at,    in   procession   of 

Mystae,  214,  217 
Brightness,   elements    of    Dionysus 

coupled  with  swift,  102 
British  Museum,  Demotic  stelae  in,  14 
Bromius,  wine  gift  of,  109  ;    mystic 

maids  sealed  for  service  of,  87 
Brotherhood  of  Dionysus  and  Apollo, 

3i 
Brumalia,    Roman     and    Thracian 

festival   of,    86-88,     112.       See 

Rosalia 
Bucolion,     scene    of    marriage    of 

Dionysus,  13 T 
Bull,  represents  water,  95 ;  Dionysus 

shaped  as  a,  95 
Bulls,  the  Delian  Hall  of,  400 
Bumna,    a   modern    name   for    the 

Bocarus,  347 
Bustrone,  calls  Troodos  Lambadista 

or  Chionodes,  341 
Butes,  leader  of  the  early  Thracian 

migration  to  the  Aegean,  80 
Byblus,    Aphrodite   worshipped   at, 

as  a  cone,  293 ;  Cinyras,  a  founder 

priest  and  king  at,  293 
Byzantium,    Cyprus   under  rule   of, 

278  f. 

CABIRION,  the  Delian,  and  the  The- 
ban,  402 

Cadmus,  the  Bacchanals  of  Euri 
pides  and,  140,  145-147 ;  the 
Dionysus-legend  and, y<5z  f.,  164 

Caesar,  Julius,  14,  37 

Caesarea-by-the  sea,  14 

Caesareum  at  Alexandria,  the,  13-15 

Calamis,  statue  of  Aphrodite  Sosan- 
dra  by,  299 

Calendar,  juggle  with,  at  Athens  for 
initiation  of  Demetrius,  209 

Callias  at  the  Delia,  376 

Callichorus,  the  well  where  Demeter 
sorrowed,  207 

Canicula-Maera,  no 

Capitoline  Museum,  bust  of  Virgil 
in,  K)5 


INDEX 


429 


Cappadocian,  Apollonius  the,  sneered 

at  by  Dio  Cassius,  261 
Caracalla,     coins    of,    44  ;    son    of 

Julia  Domna,   264  ;  partiality  for 

Apollonius  of  Tyana,  261,  263 
Caramania,    anciently    invaded    by 

Thracians,    280  ;    mountains    of, 

376 
Caratheodori,     two     cousins,     both 

fathers  of,  were  doctors,  268 
Carinondas    and  Apollonius,    magi 

cians,  262 
Carnival  -  time   in    the   Archipelago 

and    the    Delia    and    Apollonia, 


Carthage,    Aphrodite-Ashtaroth    at, 

303 

Castalia,  streams  of,  23,  35,  133 
Caste,  ii 

Castor,  a  pupil  of  Chiron,  232 
Cavafy,   Dr.,  at  St.   George's  Hos 

pital,  269 
Cavaliere   di    San    Marco,     gift    of 

Meursius'  MS.  Cyprus  gained  his 

son  the  title  of,  jjo 
Cave  at  Icaria,  709 
Cave-like  arches  at  Eleusis,  797 
Cave  -temple   of  Apollo   on    Delos, 

sssi 

Caves,  frequency  of,  in  Cyprus,  289 
Cecrops,  Cychreus  as,  345  ;  Eleusis 

as,  64  \  Pandion,   the  fifth  since, 

105 
Celeae,    Eleusinian    Demeter-legend 

and,  64  f.,  7  05 
Celeus  and  Eleusin,    61,    64  f.  ;    a 

hero  at  Celeae,   63  ;  an  interloper 

at  Eleusis,  6j  ;   Demeter's  visit  to, 

6j,  68,  105  ;  discrowned  by  Ovid, 

77  ;  Icarius  contrasted  with,  loj  ; 

the  sons  of,  62  f.  ;  the  daughters 

of,  <5j,  122,  175 
Centaurs,  138,  231,  249 
Central  Museum  at  Athens,  Epidaur- 

ian  victories  in,  249 
Ceos,    on  the  way  from  Athens  to 

Delos,  376 
Cephissus,  jibes  at  bridge  over,  214, 

217 


Cerastis,  Xenagoras,  called  Cyprus, 

274 

Ceres,  the  Roman,  jo 
Cerynia,   the  only  centre  of  life  in 

North  Cyprus,  277 
Ceryx,  son  of  Hermes,  successor  to 

Eumolpus,  122 
Cestus  of  Aphrodite,   borrowed   by 

Hera,  .299 
Chalcidice,    Greeks  of,   deified   Fla- 

mininus,  40 

Chastity  and  Cypris,  149-153 
Children,  growth  of,  linked  to  growth 

in  the  field,  jj  ;  surviving  Europ 
ean  customs  about,  jj 
Chinese,   use  of  play-actors'   masks 

by,  172 
Chios,  not  of  Cyclades  or  Sporades, 

393 

Chiotes,  for 

Chiron,  the  character  of,  231 ;  Aes 
culapius,  reared  by,  246 ;  loves 
of  Apollo  and,  369  ;  pupils  of, 
232 

Chiti  (Citium),  Podocatharo  estate 
of,  337  f.  ;  Procacci's  account  of, 

337 

Chittim,  Cyprus  named,  274  ;  Tar- 
shish  and,  in  Isaiah,  281 

Chivalry,  of  Apollo  in  love,  368  f.  ; 
Homeric,  scorned  Dionysus,  78 

Cholargia,  Xenocles  of,  at  Eleusis, 
797 

Choruses,  tragic,  at  Sicyon,  7/j 

Christ,  deification  of  emperors  led  to 
the  imitation  of,  265  ;  statue  of, 
in  the  Lararium  of  Severus,  263 

Christian  art,  spirit  of  ancient  mys 
teries  in,  182 

Christian  birth-legends,  compared 
with  the  Delian  story,  jj6 

building  at  Delos,  379 

opponents    took     trouble     to 


understand  Apollonius,  263 

—  ritual,  4 
Christian,  mimic  fight  between  Turk 

and,  90 
Christianity,    j,    5,    6,    fj    f. ,    79  ; 

Paganism  and,   44  f. ,  2jS,  260, 


43° 


INDEX 


266 ;  in  Thrace,  84,  88 ;  the 
Dionysus-worship  of  the  Baccha 
nals  and,  138 

Christians,  Apollonius  respected  by, 
261 ;  ransomed  from  Turks,  339 

Christopoulos,  a  lyric  poet,  studied 
medicine  at  Buda,  268 

Chrysopator,  epithet  of  Dionysus,  92 

Chrysopolitissa,  the  holy,  and  Aphro 
dite  Paphia,  304 

Church,  the  hall  at  Eleusis  a,  189 

Church  universal,  n,  15 

Churches,  ruins  of,  in  Cyprus,  283 

Cicero,  his  mocking  allusion  to 
Caesar's  divine  honours,  37  ;  pro 
poses  a  building  like  the  gate  of 
Appius,  188-190 

Cilicia,  mountains  of,  276 

Cimon,  Hall  of,  at  Eleusis,  199,  204 

Cinnamon  tree,  the,  and  Nysa,  164  f. 

Cinyradae.the,  at  Paphos,  291 

Cinyras,  legend  of,  291-295  ;  the 
Paphian  south  wing  is  perhaps  the 
tomb  of,  309,  314 

Circe  had  knowledge  of  miraculous 
drugs,  23T 

Cithaeron,  Mount,  118 ;  Dionysus- 
legend  and,  141,  147,  -fjo,  zjj  f. , 
158  \  false  connection  with 
Cythera  of,  286  ;  never  connected 
with  Aphrodite,  286 

Citium,  the  Roman  form  for  Chit- 
tim,  281 ;  Athena,  not  Aphrodite, 
mentioned  at,  322  ;  Phoenician 
remains  at,  282,  32 1\  Podocathari 
estate  at,  337 

Citizenship,  political,  invented  by 
Greece,  12 

Clairvoyance  of  Apollonius  in  Dio 
and  Philostratus,  261 

Clan  and  commonwealth,  n 

Classicists,  romanticists  compared 
with,  yjo 

Clemency  of  Augustus,  worship  of, 

40 
Cleomenes,  devastation  of  Eleusinian 

sanctuary  by,  198,  216 
Clisthenes,      effect      of     Sicyonian 
choruses  on,  115 


Clodones,  the,  are  Thracian  originals 

of  Bacchants,  97 

Clytemnestra,    murder   of,   and  the 
couvade,  169, 170  ;  struggle  of,  for 

Iphigenia  and  the  couvade,  171 
Cnidus,  i,  j,  59  f.,  70,  194,  195 
Cnossus,    Epimenides   from    district 

of,  124 
oan  practice  of  medicine,  work  on, 

attributed  to    Hippocrates,    225  ; 

school  and  early  medicine,  234 
Cock,  debt  of  Socrates  and  Crito  to 

Aesculapius,  239 
Codrus,  father  of  Androclus,  122 
Coins,  head  of  Apollonius  on,  265 
Colettis,  the  Prime  Minister,  was  a 

doctor,  268 

Colontas,  Demeter  and,  776 
Colos,  Garin  du,  at  Colossi,  288 
Colossi,  name  and  history  of,  287  f. 
Colossus  of  Naxians  at  Delos,  380  ; 

of    Rhodes     and     the     Cypriote 

Colossi,  287  f. 
Columbus  of  a  new  world,  Thespis 

the,  116 

Comedies,  at  Athenian  festivals,  134 
Comedy,  Icaria  the  cradle  of,  in, 

113,    117  ;    Susarion   invited    to 

Icaria  with  the  first,  114  f.  ;  tra 
gedy  and,  began  together,  117 
Commanderia   and  Madeira  wines, 

288 
Commanders,     Knights,      the,     at 

Colossi,  287 

Commandery  at  Colossi,  288 
Commandment,  Demeter's,  72 
Commercial    spirit    of    Delia    after 

second  revival,  375 
Commonwealth  and  clan,  u 
Concord,  of  Dionysus  and  Apollo, 

104 
Cone,  worship  of  Aphrodite  as  a,  at 

Byblus  and  Paphos,  293,  386 
Confirmation,    analogy   of,    to    the 

Greater  Mysteries,  209 
Conon  and  Evagoras  used  forests  of 

Cyprus,  308 
Contradictions,  in  the  pose  and  look 

of  Demeter  and  Persephone,  73  ; 


INDEX 


inherent  in  Dionysus,  78  f.,  113, 

rj6 

Coray,  a  doctor  at  Montpellier,  268 
Corfu,  May  festival  at,  89 
Corinth,  Aphrodite  at,  27  f  ;   Bacchic 

images  of  Pentheus'  wood  at,  759 
Corinthian      capital,     the      earliest 

known,  252 
Corn,  contrasted  with  wine  and  the 

vine,  707  ;  lady,  the,  of,  jo  ;  mother 

of,  or  Kornmutter,  470 
Coroebus,  a  builder  of  the  temple  at 

Eleusis,  797 
Coronelli  calls  the  Troodos  Olympus, 

343 

Coronis,  daughter  of  Thessalian 
Phlegyas,  2jj,  244  f. ,  367 

Corybantes,  at  birth  of  Zeus-Diony 
sus,  99,  142  f. 

Corycian,  cave,  133 ;  Dionysus  on 
the  heights,  757 

Cos,  inscriptions  at,  237  ;  coming  of 
Aesculapius  to,  221  ;  date  of 
Aesculapian  foundation  of,  243  ; 
Ptolemy  and,  likened  to  Apollo 
and  Delos,  394 

—  temple  of  Aesculapius  at,  the 
school  of  Hippocrates,  225 

Cothonea,  given  for  Metanira,  6j 

Couclia,  modern  name  of  Old 
Paphos,  289  ;  description  of  the 
village  of,  296 ;  Tschiflik  at, 
290  ;  further  excavations  at,  de 
sirable,  314 

Courage,  Homeric  conception  of,  78 

Couvade,  Apollo  in  the  Eumenides 
and,  769  ^  '•  l^e'  anc^  Clytem- 
nestra's  death,  769  f.  ;  Dionysus' 
second  birth  and,  168-170  ;  the, 
and  Dionysus'  love  for  Semele, 
769  ;  Greek  transformation  of 
problem  of,  to  beauty,  777  ;  sur 
vival  of,  to-day  in  Spain  and 
France,  770 ;  struggle  for  Iphi- 
genia  and  the,  777  ;  Zamacola  on 
the,  770 

Couvocles,  old  French  for  Couclia, 
modern  name  of  Old  Paphos,  296 

Cowardice,   the,   of   Dionysus,    per 


sonified  by  Satyrs,  99  ;  grotesque- 
ness  as  of,  attached  to  Dionysus, 
78 

Crane  dance,  legend  of,  at  Athens 
and  Delos,  400 

Creed,  6  f.,  16,  18,  fjj 
retan  birth  of  Epimenides,   effect 
of,   7^7  ;    legends,    Dionysus   in, 
81,  7J9,  142 

rete,  Aphrodite  in,  277  ;  Ariadne 
came  from,  62  ;  Curetes  from,  97  ; 
Demeter-worship  in,  61  f.  ;  lasion 
in,  62  ;  Rhea  and  Cybele  in,  143  ; 
Zeus  of,  a  Dionysus,  142 

Crissaean  plain,  23 

Crito,  debt  of  a  cock  to  Aescula 
pius,  2J9 

Croce,  Mount  St.,  wrongly  identi 
fied  with  Olympus  of  Cyprus,  J2j  ; 
is  probably  the  ancient  Aous, 

352 

Cronos,  Aphrodite  on  the  Olympian 
hill  of,  277  ;  Demeter's  father,  jo 

Cross  of  the  penitent  thief,  discovery 
of,  in  Cyprus,  275- ;  tempests 
stilled  by  a  nail  of  the  true,  275 

Croton,  Aesculapius  probably  not 
worshipped  at,  225  f.  ;  Alcmaeon 
the  Pythagorean  of,  223  ;  Demo- 
cedes  came  from,  223  \  renown  of 
its  school  of  medicine,  223  ;  the 
centre  of  Pythagoreanism,  223 

Cruelty,  in  Thracian  Dionysus-wor 
ship,  700 

Cruz,  Santa,  Cal. ,  and  Cape  Curias, 
288 

Crypton,  Astynomus,  called  Cyprus, 
274 

Cuneiform,  version  from  the,  JO7 

Curetes,  are  the  Satyrs  of  Crete,  9^, 
97,  142  f. 

Curias,  cape,  and  Santa  Cruz,  Cal. , 
288 ;  Strabo's  account  of  coast 
west  of  cape,  325 

Curium,  Apollo  i/XdrTjs  at,  345  f. 

Customs,  justice  done  by  Euripides 
to,  757 ;  variety  of  local,  con 
nected  with  Eleusinian  procession, 
218 


432 


INDEX 


Cyamites,  the  Mystae  and,  2/7 

Cybele,  Bacchic  worship  of,  141  ; 
Corybantes  of,  97  ;  Demeter  and 
Rhea  combined  with,  401  f.  ;  one 
of  eight  Eleusinian  gods,  178 ; 
Rhea  acts  in  Crete  as,  143 

Cychreus  (Salaminian),  Cecrops 
Eleusis  and,  64,  345  f. 

Cyclades,  influence  on  Athens  of, 
8 1 ;  Delos  and  the,  359,  j62, 
364  f. ,  387  ;  history  of  the  name, 
391-398  \  the  Holy  Islands  called, 
370 

Cyclopean  sanctuary  at  Eleusis, 
200  f. 

Cyclops,  Apollo's  expiation  for  slay 
ing  the,  37 i 

Cycnias,  a  mountain  of  Tenos,  380 

Cyllene,  seen  from  Eleusis,  187 

Cylon  avenged,  124 

Cy Ion's  failure  the  people's  victory, 

122 

Cynortion,  mountain  sacred  to 
Apollo,  246 

Cynthian  Cabirion  at  Delos,  402 ; 
cave-temple  of  Apollo,  385  f.  ; 
precinct  of  foreign  gods  became 
foreign  soil  to  Apollo,  402 

Cynthus,  j6j ;  temple  to  Zeus 
Cynthius  and  Athena  Cynthia  on 
Mount,  402  ;  altar  of  horns 
brought  by  Artemis  from,  384  f. 

Cyprians  knew  that  Aphrodite  was 
from  the  East,  273 

Cypriote  agriculture,  Lang  and 
Fourcade  on,  281  ;  characters, 
who  used  them  ?  280,  321 ;  pea 
santry,  Lang's  sympathy  for, 
280  i. 

Cypris-Aphrodite  and  Ashtaroth  dif 
ferentiated,  320 ;  not  of  Greek 
birth,  318 

Cypris  and  chastity,  ijo  f. ,  153 ; 
Homer  mentions  Aphrodite  as, 
284,  316 ;  Pontia,  Phaedra's 
name  for  Aphrodite,  297 

Cyprogeneia,  Aphrodite's  epithet,  296 

Cyprus,  2  ;  Aphrodite's  isle  is,  153, 
274,  287  ;  in  ancient  times,  272, 


274,  278,  280,  284,  292,  294, 
32r  f.  ;  description  of,  275-277, 
288  f.,  308,  328  ;  general  history 
of,  -276-280  ;  in  medieval  times, 
27S>  278>  326-328, sjs-jjs,  342 ', 
in  modern  times,  272,  27^,  278 'f., 
288 

Cyrene  and  Apollo,  300,  369 
Cythera,  Aphrodite  on,  277,  286  f., 
296,  317  ;  confused  withCithaeron 
and   Cythrea,    286  ;    Phoenicians 
on,  286,  323 

Cythnos  on  the  way  to  Delos,  376 
Cythrea   of   Cyprus,    confusion    of, 
with  Cythera,  286 

DAEDALUS,  Delian  image  of  Aphro 
dite  by,  J79,  J99 

Daeira,  surname  of  Persephone  at 
Eleusis,  48 

Dagon  at  Ashdod  and  Eastern  Aph 
rodite,  297  f. 

Daimon,  the  mystical,  of  early  Eleusis, 
and  Dionysus  -  Zagreus-Iacchos, 

*74 

Daisy,  the  Greek,  on  capitals  of  the 
Tholos,  252 

Dalin  (Idalium),  earthquake  at,  290 

Damaschino,  Dr.,  professor  in  the 
Faculty  of  Paris,  269 

Damasias,  Mr.  Kenyon's  note  on, 
132 

Damigeron  and  Apollonius,  sor 
cerers,  262 

Damis  the  Ninivite,  travellers'  tales 
of,  257-259  ;  notes  of,  given  to 
Philostratus  by  Julia,  260 

Dance,  Apollo  and  the,  29 ;  the 
Archipelago  and  the,  388  ;  Apollo 
and  Dionysus,  two  gods  of  the, 
104  ;  Dionysus  and  the,  06,  loj, 
133,  142,  146  f. ,  213  ;  of  Diony 
sus  on  Parnassus,  133,  149  ;  of 
Silenus,  95 ;  represented  Eleu 
sinian  and  Icarian  legends, 
107 

Dante,  Apollo  invoked,  24  f.  ;  Botti 
celli's  illustrations  of,  331  ;  idea 
of  Parnassus  in,  /// ;  Orphic 


INDEX 


433 


ideas  in,  25  ;    remembers  Lucan, 

24 

Daphnae  on  the  Sacred  Way,  218 

Daphne  and  Erigone,  myths  of,  368 

Dardanus  and  Apollonius  as  magi 
cians,  262 

Darius  cured  by  Democedes,  224 

Daulis,  80,  in 

Dead,  Dionysus  translated  from  the 
world  of  the,  102  ;  streams  poured 
by  Odysseus  for  the,  97 

Death,  Aesculapius  the  awakener 
from,  239,  245  ;  Apollo  the  dealer 
of,  366  ',  of  Dionysus,  112  ;  re 
presented  by  Dionysus,  28 ;  taint 
of,  mentioned  in  the  Old  Testa 
ment,  j6o  ;  Thracian  idea  of,  84 

Decelea,  fortification  of,  by  Spartans, 

215 

Deification  of  emperors  led  to  imita 
tion  of  Christ,  265 

Delian  festivals  of  Apollo,  104,  363, 
372-381,  386-389 

Delians,    exile   and   restoration    of, 

Jtj,  378 

Deliastae,  376 

Delion,  the  Marathonian,  lo.f,  376 

Delium,  festival  organised  at,  and 
battle  of,  363 

Delos,  2,  9  ;  description  and  general 
history  of,  361,  363,372-377, 379, 
383,  Sty,  J$9.  394<  4<>r  I  monu 
ments  at,  355, 371,  378-380, 384, 
386  f. ,  400  ;  mythology  of,  35, 
350,  356-359,  37  f>  379'  3^2-384, 
394,  399  f.  i  403  ;  ritual  and  wor 
ship  in  general  at,  255,  355,  360- 
364,  370,  376,  378-382,  387 

Delphi,  description  of,  22-24;  known 
details  of  monuments  and  worship 
at,  21  f.,  31-33)  jfo;  history  and 
mythology  of,  22,  31-35,  119, 
37 f ;  oracle  at,  26,  34-36,  81, 
122,363,  370  f. ;  revellers  blocked 
by  snow  above,  113 ;  worship 
and  ritual  at,  22,  29-33,  35,  364, 

370,  384 

Demaratus  inquires  about  the  Eleu- 
sinia,  209  f. 


Demeter,  i,  4,  j,  8,  17  ;  association 
and  combination  of,  with  other 
gods,  21,  69,  73  f.,  77,  107,  159, 
174,  176,  178  f.,  182-185,  208, 
218-220,  273,  303,  401  f.  ;  char 
acter  and  meaning  of  the  divinity 
of,  79  f..,  47,  51  L,  67-69,  71-74, 
133,  180,  182,  185,  215,  303; 
leading  points  in  the  myth  of,  50, 
55,  57 >59'  61-63,  66-68,  71  f., 
105,  in,  119,  175  f.,  185,  206  f. , 
215,  217,  303 ;  development  of 
the  worship  and  myth  of,  46,  48  f. , 
51  f.,54,56,  60  f.,  75,  77.  104, 
in,  i22i.,  125  f.,  163,  175  f., 
194,  221,  233  ;  monuments  of  the 
worship  of,  70,  72  f.,  189-196, 
124,  271  ;  processional  song  in 
honour  of,  213  ;  Dionysus  of  the 
Lesser  Mysteries  with,  125 

Demetrius  the  Phalerean  at  Eleusis, 
797 

Demetrius  Poliorcetes,  violated  the 
degrees  of  Eleusinian  initiation, 
209 

Democedes,  223  f. ,  226,  234 

Democracy,  triumph  of  Dionysus 
with,  122 

Democratic  reform  of  the  greater 
Dionysia,  134 

Demophoon,  legends  of,  62-66,  94, 

175'  229 

Demotic  stelae  from  Memphis,  14 
Dendrites,  Dionysus  as,  106, 159, 176 
Dendrophoria,       Thraco  -  Phrygian 

annual  custom,  759 
"Denis,   vive  notre  bon   pere,"   old 

French  song,  779 
Denmark,  Meursius  in,  348  i. 
Deo,  the  Eleusinian,  133 
Diana  and  Apollo,  150 
Diaulos,  at  the  Delia  and  Apollonia, 

388 
3td£w/xa      of     Eleusinian       temple, 

meaning  of,  797,  203 
Dicaeus,  describes  Eleusinia,  209  f. 
Dido-Anna  and  Aphrodite,  272 
Dili,   Megale,  the  modern  name  of 

Delos,  361 


2   F 


434 


INDEX 


Dio  Cassius,  trustworthiness  of  his 
account  of  honours  voted  to  Julius 
Caesar,  37  f.  ;  inconsistently  en 
dorses  clairvoyance  of  Apollonius, 
261  f. 

Diocles,  summoned  by  departing 
Demeter,  63 

Diocletian's  Insularum  provincia  in 
cluded  53  islands  called  Cyclades, 
393  !  Cyclades  came  to  be  called 
Dodecanisia,  ^97 

Diogenia,  daughter  of  Celeus  (Pam- 
phos),  63 

Dione  and  Aphrodite,  272,  284, 
jyj-f.,  318  L 

Dionysia,  the  City,  are  the  Greater, 
*35  !  foundation  of  the  Greater, 
f32>  I34  ^  I  the  Anthesteria  con 
trasted  with  the  Greater,  131  ; 
Thucydides  calls  the  Anthesteria 
the  older,  i  31  ;  the  Lesser  are  the 
Rural,  131  ;  the  Rural  took  shape 
in  Icaria,  113 

Dionysiac  customs,  persistence  in 
Thrace  of,  88 

Dionysiaca,  the,  of  Nonnus,  JTJQ 

Dionysodotos,  Apollo,  31 

Dionysus,  i,  4,  5,  9,  15,  17  f.  ;  the 
god  of  song  and  dance  and  tra 
gedy,  29,  31,  75,  79,  86,  97,  102- 
104,  us,  i  iji  *33>  sj6,  143  ;  the 
god  of  the  nether-world,  29,  31, 
34,  79,  97,  06  f.,  112,  127,  132, 


185,310  ;  the  god  of  the  real  world, 
—  (a)  in  general,  oo,  02,  102  f.  , 
130,  140,  144,  217,  —  (b]  of  trees 
and  vegetation,  82  f.  ,  106,  150,  — 
(c)  of  various  elements,  parts,  and 
events  in  nature,  24,  02-06,  100, 
102,  107,  133,  139  f.  ,  143,  145, 

140,  152,   156,   158,    178  f.  ,   212  ; 

attendants  and  others  variously 
possessed  by,  (a)  inspiration  and 
possession  by  (generally  con 
sidered),  04,  97  f.,  133,  136,  j-jo, 

141,  144,  155,   158  i.,—(b)   Bac 
chanals  possessed  by,  11  2,  156  f  .  ,  — 
(c)  Bassarids  possessed  by,  79,  — 


(d)  Corybantes  and  Curetes  at  the 
birth  of,  142, — (e)  Graces  led  by, 
104, 135, — (/)  Maenads  possessed 
by,  112,  141,  184  f. , — (g}  Midas 
as,  79,  03, — (h)  the  Muses  and, 
31,  06,  103  f.,  135, — (i)  nymphs, 
naiads,  and  waterfolk  of,  04, — (j) 
Pans  of,  99, — (k]  seasons  of,  06, 
— (/)  Satyrs  and,  79,  99, — (m)  the 
Thyades  of,  133  ;  the  Saviour  God 
(of  Eleutherae),  22,  34,  79,  118- 
120,  133,  145,  150,  161  f.  ;  festi 
vals  of,  and  ritual  in  general,  (a) 
characteristics  of  festivals  of,  4, 
85,  89  f. ,  106,  112  f. ,  132,  134, 
140,  372, — (b)  Thracian  festivals 
of,  86-00,  —  (c)  Athenian  festivals 
of,  113,  128-132, — (d)  Boeotian 
festivals  of,  103, — (e)  Icarian 
country  festivals  of,  108, 112-114,  — 
(f)  Eleusinian  rites  of,  214, — (g) 
observances  in  honour  of,  36,  140, 
155  ;  Thracian  origin  of,  22,  77, 
79,  82,  84,  86-00,  02,  04,  06, 
99  f.,  103,  106,  113,  126,  138, 
162,  164,  177 ,  180,  104  ;  Orphic 
features  in  the  myth  of,  112,  125- 
128,  156  L,  168-173,  177,  181  ; 
other  gods  and,  (a)  Adonis  as, 
201, — (b)  Aesculapius  and,  220  f., 
247, — (c)  Aphrodite  and,  273,- 
(d)  Apollo  and,  21  f. ,  29,  31,  102, 
104,  159,  162,  185,  372,—  (e) 
Ariadne  and,  130,  328, — (/)  Eleu 
sinian  Demeter  and,  77,  107,  125, 
174-176,  178  f. ,  182-185,  208, 
210,  212, — (g)  Hades  of  Eleusis 
and,  176-178, — [k]  Persephone 
and,  144,  178, — (*)  Plutus  and, 
08, — (j)  Zeus  and,  22,  142  ;  lead 
ing  points  in  the  myth  of,  (a] 
general,  34,  77,  101  f.,  143,  158, 
318, — (b)  birth-legends  of,  81,  in, 
114,  142,  148,  163-165,  318, 
— (c)  Attic  legend  of  (i. )  early 
Icarian  legends  of,  80,  105,  in, 
(ii. )  Semachidae  legends  of,  105, 
(iii. )  Eleutheraean  legends  of, 
118  f.,  (iv. )  later  Attic  story  of, 


INDEX 


435 


soj,  rn,  n 7,  779,  128,  ij2, — 
(d)  Eastern  myth  of,  02,  140  f. , 
W.  '57.  J6j-j6j,  i68,—(e) 
Euboean  legend  of,  143, — (/) 
Theban  legend  of,  140  f. ,  148, 
777  ;  development  of  the  myth  of, 
8 r  f.,  (a)  from  points  outside 
Greece,  34,  79,  81  f. ,  83,  140  f. , 
163-165,  1 68  f.,  233, — (<J)  from 
places  outside  of  Athens,  34^  104, 
IT i,  u8i.,  121, 123,  139,  773-, — 
(c)  progress  to  and  in  Athens,  34, 
joi,  1 21 -i 23,  125  f.,  130-132, 
154,  175,— (d]  to  its  latter-day 
shape,  2r  f.,  36,  79,  106, 
143,  186  ',  characteristics  of, 
(A)  violence,  extremes,  and  con 
tradictions  in,  (i.)  contradictions 
in  general,  30,  78,  83,  85,  93, 
10 1,  113,  115,  136,  144,  149, 
156,  (ii.)  the  pitiless  and  out 
rageous  god,  22,  79,  99-707-,  770, 

128,  757,  759  f.,  (iii.)  the 
cowardly  and  crazed  god,  77  f. , 
99,  779  ;  (B)  higher  religious  and 
moral  aspects  of,  (i. )  the  elusive 
mystery  and  truth  of  his  being, 
22,  20,  80,  05,  101-104,  f33  f-. 
136,  148,  779  f.,  183-185,  368, 
(ii. )  Christian  aspects  of,  34,  79, 
777,  134,  138,  145,  153-157, 
(iii. )  Athenian  perfections  of,  80  f. , 

702,     70J,     77J,     777,     162,     2l8, 

(iv. )  power  of  faith  in,  20,   76  f., 

97,     1 08,    7JO   f. ,     133    f.,    140    f., 

I52-i54>  i5<>  f- 
Diophantus  of  Sphettus,   his  prayer 

to  Aesculapius,  241 
Dirce,  Cadmus  sowed  dragon's  teeth 

near,    7./7  ;     Achelous'    daughter, 

invoked  for  Dionysus,  756 
Dissection  practised  by  Pythagorean 

Alcmaeon  at  Croton,  223 
Dithyrambic         contests,         tripod 

awarded  for  prize  in,  36 
Dithyrambus,       contradictions       of 

Dionysus  mirrored  in,  115  ;  influ- 

I^nce  of,  on  Thespis  and  tragedy, 
r/,-  f. 


Divine  man,  ideal  of,  evolved  among 
Greek  gods,  20 

Divinity  of  Aesculapius,  Demeter, 
and  Dionysus,  hardly  recognised 
by  Homer,  233 

Divus,  implications  of  the  use  of,  43 

Doctors  (Greek),  Homeric  status  of, 
222,  226,  230  ;  deference  for 
Apollo  and  Aesculapius  of  ancient, 
220,  230,  234,  238 ;  career  of 
Democedes  as  a  public,  223  f.  ; 
modern  status  of,  222,  267-269 

Dodecanisia,  a  name  for  many 
Aegean  islands,  ^97 

Dodona,  70  ;  Aphrodite's  origin  at, 
315  f.  ;  Dione  was  the  Aphrodite 
of,  318 

Dog-star,  Maera  became  the,  777 

Doge,  marriage  of,  with  the  Sea,  7J7 

Dogs,  the  nocturnal  touch  of, 
brought  healing,  235  f.  ;  forbidden 
at  Delos,  360  ;  frieze  of,  on  Arte 
mis'  temple  at  Epidaurus,  250 

Dolichos,  the,  at  the  Delia  and 
Apollonia,  388 

Domitian,  Apollonius  saw  from 
Ephesus  the  murder  of,  261 

Doric  style,  bad  example  of,  in 
Hadrian's  gate  at  Eleusis,  790, 
792  ;  last  extremity  of,  at  Delos, 
387  ;  of  the  Tholos  at  Epidaurus, 
250 

Dotian  plain,  Demeter  came  south 
from,  to  Attica  and  Cnidus,  793- 

Draconus,  promontory  of  Nicaria, 
where  Dionysus  was  born,  163 

Drama,  contrast  of  Greek  with 
Chinese  and  Japanese,  772 

Dreams,  circumstances  attending, 
and  results  gained  by,  in  Aescula- 
pian  temples,  235  f.  ;  prayed  to, 
as  the  children  of  Apollo,  242  ; 
porches  in  Aesculapian  sanctuaries 
for  awaiting,  248 

Drugs,  Homeric  knowledge  of, 
228  f.  ;  miraculous  power  of  cer 
tain,  in  Homer,  231 ;  knowledge 
possessed  by  Medea  and  Circe 
from  Aeetes  of  miraculous,  231 ; 


436 


INDEX 


Chiron  in  relation  to  knowledge  of, 
232 ;  Aesculapius  and  his  sons 
associated  with  miraculous,  231 ; 
Herophilus  called  the  hands  of 
the  gods,  237  ;  the  sun -god  is  the 
source  of  miraculous  knowledge 
of,  231,  237 
Drunkards,  Dionysus  and  brawls  of 

leering,  79 

Dryads,  the,  and  Dionysus,  06 
Dyalos,  Dionysus  surnamed,  97 


EAGLES,  golden,  at  the  Delphian 
navel-stone,  31 

Earth,  Agave  and  Maenads  are 
angry  powers  of,  160  ;  clung  to 
the  Telesterion  at  Eleusis,  206  ; 
Delphi  the  centre  of,  32  ;  king 
dom  of,  in  the  heavens,  13  ;  oracles 
of,  supplanted  at  Delphi,  jj- ;  Pen- 
theus'  father's  mother,  756,  158, 
160 ;  and  Poseidon  at  Delphi, 
372  ;  stricken  for  woes  of  Dem- 
eter-Cybele  and  Ishtar,  302  f. 

Earthquake,  Dionysus  directs  the, 
100,  144,  158 ;  Maenads  as  the 
many  -  handed,  143,  159  f-  ', 
earthquakes  in  Cyprus,  at  Dalin 
(Idalium)  and  Old  Paphos,  290 

East,  gods  from  the,  77  ;  Aphrodite 
from  the,  273  ;  Dionysus  en 
tangled  with,  in  Euboea,  Nicaria, 
and  Naxos,  163 ;  second  birth  of 
Dionysus  connected  with,  164 

Eastern  pedigree  of  Aphrodite  and 
Cinyras,  2O,r ;  uncleanliness  of 
Aphrodite,  297  ;  affinities  of 
Adonis  can  be  overstated,  201 

Echion  the  Earth-sprung,  Pentheus' 
father,  147,  156,  158 

Egypt,  one  of  Hyginus1  insulae,  394 ; 
the  gods  of,  77 — unknown  to 
early  Delos,  402 — invasion  of 
Delos  by,  401  i. — influence  on 
Aesculapius  worship  through 
Serapis  of,  254 

Egyptians,  their  knowledge  of  herbs 
in  Homer,  231 


Egyptian  poet  of  Dionysiaca,  92  ; 
doctors  and  Democedes,  224 ; 
strain  of  Dionysus,  168  ;  use  of 
rosettes,  251 

ei/)a0tu>T77S,  Dionysus  called,  166 

Elagabalus,  coins  of,  41 ;  his  reign 
a  Nightmare  of  Religious  Refor 
mation,  260 

Elements,  Dionysus  becomes  the  in 
carnation  of  the,  90  ;  wine,  fire, 
water,  gold,  and  unusual  bright 
ness  are  the  Dionysiac,  97,  92, 
102  ;  fusion  through  song  and  the 
dance  of  the  Dionysiac,  103 

Eleusinion  at  Athens,  the,  124,  207  f. 

Eleusis,  165  ',  landscape  and  posi 
tion  of,  SS>  57,  187  f->  218; 
Athens  and,  60,  122,  208,  376  ; 
account  of  monuments  of,  7,  j, 
44,  60  f. ,  186,  188-193,  797-207, 
218,  379 ;  chief  points  in  the 
legend  of,  jj,  57,  59-67,  63-63, 
in,  779,  793- ;  legend  of,  com 
pared  with  others,  33,  50,  243  ; 
compared  (i.  )with  Icarian  legends, 
104,  107,  133  ;  (ii. )  with  Pelepon- 
nesian  legends,  65,  105  ;  the 
mysteries  of,  4  f.,  68,  76,  122, 
180-182,  200  f. ,  2ij  f. ,  376,  403 ; 
the  Holy  Alliance  of,  77,  776- 
178,  211,  401  f.  ;  Aesculapius  at, 
2/9  ;  Aphrodite  near,  277,  273  \ 
Dionysus  at,  97,  727,  125,  174- 
777,  202  ;  Eubouleus  at,  776  ; 
Hades  at,  77^,  776  ;  Persephone 
Daeira  at,  48 

Eleutherae,  Dionysus  of,  34,  104, 
in,  118-120,  143,  162  ;  image 
of  Dionysus  at  Athens  from  the 
temple  at,  779 

Elis,  Aphrodite  in,  277 

Elysium,  75  f. ,  87  f. 

Emanations  from  Dionysus,  Muses, 
Hours,  etc.,  as,  06 

Emathia,  the  cradle  of  Philip's 
power,  79  ;  north  of  Olympus, 
where  was  Mount  Bermius  and 
the  gardens  of  Midas,  79  ;  har 
boured  myth  of  Dionysus,  79 


INDEX 


437 


Emathia  and  Pieria,  87 

Emathian  conqueror,  the  great,  79 

Emesa,  Julia  Domna,  daughter  of 
priest  at,  260 

Emperius,  places  Satrachos  at  Pa- 
phos,  346 

Emperors  of  Rome,  allegiance  to 
Venus-Aphrodite  of,  joo  ;  sanct 
ity  of  Apollonius  compared  with 
divinity  of,  265 ;  deification  of, 
rooted  in  previous  habits  of  mind, 
40  ;  not  work  of  clever  men,  37, 
40 

Engel  argues  that  Aphrodite  was 
Greek,  272,  315  f.  ;  is  responsible 
for  adoption  of  Paphian  Bocarus, 

352 
English  occupation  of  Cyprus,  277, 

279 

Epacria, — Plothea,  Semachidae,  and 
Icaria, — confederation  of  mount 
aineers,  105,  ng 

Epacria,  Icarian  legend  of  the  visit 
of  Dionysus  to,  105,  in 

iTTOuvri,  as  an  epithet  of  Persephone 
in  Iliad  and  Odyssey,  48 

Ephesus,  place  for  yearly  assembly 
of  'Acndpxcu,  44 ;  founded  by 
Androcles,  122 ;  privileges  of 
/3a<nXets  at,  122 ;  Domitian's 
death  witnessed  from,  by  Apollon 
ius,  261 

^TTi/SaTTjpia  and  eTrt/Scrnjpios,  13  f. 

Epicureanism,  a  refuge,  41  ;  first 
master  of  Apollonius  believed  in, 
251  I  Apollonius  rejected,  255 ; 
Lucian  biassed  against  Apollonius 
by,  263 

Epicurios,  Apollo,  father  of  Aescu 
lapius,  2/9 

Epidaurus,  2,  5,  243 ;  journey  to 
Hieron  of,  246  ;  description  of 
Hieron  at,  247  ;  precinct  of  Aes 
culapius  near,  248-254 ;  Tholos 
of  Polycletus  at,  250-254 ;  theatre 
of  Polycletus  at,  246  f.  ;  in 
scriptions  describing  miracles  at, 
237,  254  ;  statue  of  Aphrodite 
found  at,  270,  273 ;  legendary 


claims  of,  243  f.  ;  legend  of,  2/9, 

221,  233,   244  f.  ;  importance  in 
history  of  surgery  and  medicine  of, 

222,  237 

Epimenides  prepared  Athens  for 
Solon's  laws,  123, 131,  175  ;  from 
Phaestus  near  Cnossus,  124  ;  the 
Eleusinion  and  Lesser  Mysteries 
and,  124,  129  ;  statue  at  Athens 
of,  124,  208  ;  Dionysus  and,  127, 

*75 

Epione,  Gentleheart,  wife  of  Aescula 
pius,  243,  249 

Epirus,  theatrical  features  in  May 
festivals  of,  £9 

Episcopia  of  the  Venetian  Cornari, 

2c?9 

€TTiTV/ji.pla,  epithet  of  Aphrodite,  310 
Epoptical    Mysteries,   the,  time  of, 

208  f. 
Eprius    Marcellus,    the    career    of, 

43 

Erechtheus  killed  in  war  with  Eleusis, 
122 

Erichthonius, succeeded  Amphictyon, 
7/9 

Erigone,  legends  and  ritual  concern 
ing,  105,  107,  709  f.,  if 4  f.,  /JJ, 
368 

Erinys,  epithet  of  Demeter,  176 

Eros,  painting  of,  in  the  Tholos,  253  \ 
Harpocrates,  Apollo  and,  401 

Eryx,  foundation  of  Aphrodite-temple 
from,  286,  323 

Euboean  legends  of  Dionysus,  143, 

i<>3 

Eubouleus  at  Eleusis,  174,  170,  194 
Eucharis,  a  purely  Greek  epithet  of 

Aphrodite,  298 
Euhemerus,    his    views    applied    to 

traditions  of  Mt.  Meroe,  166 
Eumaeus,  character  of,  47 
I'.umcnides,     The,     protest    against 

matriarchy  in,  171 
Eumolpidae,   the    Deliastae   chosen 

from  the,  174,  176,  376 
Eumolpus  from  Thrace,  174  ;  father 

of  Immarados,    122 ;    station   of 

the  Mystae  at  the  tomb  of,  218  \ 


438 


INDEX 


cultivation  of  the  vine  and  trees 
by,  107  ;  a  son  of  Celeus,  63 

Eunapius,  testimony  in  favour  of 
Apollonius,  264 

Euripides,  general  account  of,  as  a 
thinker  and  poet,  18  f. ,  29,  71, 
136,  138,  143,  149,  157,  159, 
161  f. ,  166  i. ,  169-171,  178 ; 
Aphrodite  as  represented  by,  298- 
300,  303  ;  Apolline  legends  in, 
35>  35$>  3^4  ;  Dionysus-worship 
and  legends  of,  81,  136,  142  f. , 
150  f. ,  i  Si ',  the  Tiresias  of, 
143  f.  ;  debt  of  Milton  and  Goethe 
to,  146,  148,  150,  162 

Eusebius,  tribute  to  Apollonius, 
263  f. 

Euthydemus,  Socrates  thinks  his 
books  are  on  medicine,  224 

Evagoras  and  Conon,  used  forests 
of  Cyprus,  308 

Evangel  of  Apollonius,  Philostratus 
called  to  write  the,  259 

Evangelistria,  Tenian  feast  of  the 
Holy,  378,  389 

Evariges,  letter  of  Sidonius  Apollin- 
aris  to,  265  f. 

Evian  god,  the,  Evoe  to,  144 

Evoe,  97,  144 

Evolution,  natural,  of  politics  and 
religion,  16,  19-21 ,  53 

Examples  of  Odysseus  and  Pandarus 
an  incentive  to  know  medicine, 
228 

^oxcirare,  "your  excellence,"  was 
the  style  of  Greek  doctors,  268 

FAITH,    required    by    Aesculapius, 

*SS 

Falsehood,  divinities  free  from,  18; 

Apollo  could  not  touch,  369  f. 
Famagosta-Salamis,  history  of  ruins 

at,  283 
Family,   change   in,    the    Dionysus' 

birth-legend  and,  163 
Family-life,  Demeter  who  sanctified 

the  bonds  of,  72 
Farmer's   god,    Dionysus  the,   105, 

217 


Fast-day,   Icarian  observances  like, 

112 

Fates,  Aphrodite  the  eldest  of  the, 

272  ;  measure  led  by  the  glorious, 

in  yearly  Eleusinia,  215 
Father-god  of  Thracians  in  Cyprus, 

322 
Fatherhood,  the  duties  of  Apollo  and 

Zeus,  369 

Fathers  of  the  Christian  Church,  3 
Fawn  skins,  symbolise  for  Dionysus 

the  starry  sky,  140,  142 
Ferule,  the,  in  Bacchic  worship,  142 
Festivals,     Aegean     and    Apolline, 

372-381,    386-388;     Dionysiac, 

106,  112-114,  128-130,  134 
Fetichism  and  primitive  religion,  far 

below  Greek  religion,  27 
Fetich -stone  or    Baetylus,    on    Mt. 

Cynthos,  386 
Feuilletonist,  Philostratus  a  Parisian, 

260 
Fig-man,   Phytalus  the,  at  Lakkia- 

dae,  217 
Fig-tree,   in  bas-relief  of  Dionysus 

at  the  house  of  Icarius,  106 
Filiiis  dei,  Aesculapius  as,  240  f. 
Finances   the,    of  modern    Cyprus, 

278 
Fir-tree,  the,  of  Pentheus  at  Corinth, 

Thraco  -  Phrygian    dendrophoria, 

*59 

Fire,  association  of  Dionysus  with, 
1 6,  62,  64,  78,  92  f. ,  103,  133, 
145,  149,  156,  158  f.,  179  f., 

212,    245 

Flames,   personified  by  Agave  and 

Maenads,  137 

Flamininus,  deification  of,  40 
Flash  of  the  elements  of  Dionysus, 

92 

Flattery,  attributed  to  Greeks,  14 
Fleet,  the  clay  fleet  of  Cinyras,  293 
Flesh,  Dionysus  an  eater  of  raw,  96 
Flowers,  festival  of,   at  Athens,  130 

f.  ;    flowers    of    Greece    inspired 

Polycletus,  251,  253 
Forests  abounded  in  Cyprus,  308 ; 

forests,  Scandinavian,  6 


INDEX 


439 


Fortification,     Eleusis    a,    in    early 

days,  790,  198 
Frari  church  in  Venice,   picture  in, 

of  Cornari  of  Episcopia,  289 
Freedmen,     part    assigned     to,     in 

imperial  services,  J9 
Frenzy,  inspired  by  Dionysus,  100 
Frogs,  immense  success  of,   due   to 

religious  causes,  216 

GAIA,  j<5,  64 

Galignani,  Simone,  a  Venetian  pub 
lisher,  331 

Gamelion  and  Lenaeon,  132 

Gardens,  statue  of  Aphrodite  in  the, 
at  Athens,  277 

Garin,  Nicholas  du  Colos,  at  Colossi, 
28S 

Gebeleizis,  a  name  for  the  primitive 
Dionysus,  86 

Genetor,  the  Cyclades  are  chosen 
islands  of  Apollo,  393  ;  Delian 
altar  of  Apollo,  j>79,  382 

Genii,  of  later  Emperors,  38 

Genius  of  Roman  people,  time- 
honoured  worship  of,  40 

Genoese,  the,  in  Cyprus,  278 

Geographical  use  of  term  Cyclades 
under  Rome,  jgj 

Geography,  astronomy  and  mytho 
logy  and,  confused  in  a  Delian 
legend,  358;  Meursius'  idea  of 
ancient,  348 

Geologists,  on  Cyprus,  275  f. 

Gephyrismoi,  gibes  at  Cephissus- 
bridge,  218 

Germans,  the  Great  God  of,  6 

Gibes,  used  by  lambe  to  diver 
Demeter,  68 

Glaucus,  prayer  of,  29 

Goat,  vine  -  destroying,  slain  by 
Icarius,  108 

Gold,  significance  of,  in  Diony 
sus  legends,  92,  97,  102,  i.fo 
178 

Golgoi,  perplexing  connection  with 
Paphos,  28s 

Gorgon's  head,  sent  forth  by  drea 
Persephone,  48 


Gospels,  Godhead  of,  79  ;  the  life  of 
Apollonius  and  the,  25$ 

joths,  overran  Thrace  and  Illyria, 
87  ;  letter  to  the  King  of  the, 
26^  f- 

otterineinanderleben,  of  the  Olym 
pians,  20 

out,  appeal  of  Diophantus  for  cure 
of,  to  Aesculapius,  2.ff  f. 

}ozzoli,  Benozzo,  Pausias  and,  253 
Jraces,  the,  Dionysus  and,  96,  104, 

IJ5,  303 

'pcua,  Demeter  surnamed,  70 
irasshoppers,    invasion    of    Cyprus 

by,  280  f. 
Jravia,  Odysseus  danced  the  Syrtos 

at,  400 
recia   (Magna),   no  proof  of  early 

worship  of  Aesculapius  in,   225  ; 

Dionysus  in,  186 

reek  religion,    21,    26,   29,   75  f., 

779,  210 
ryaros,  not  of  the  Cyclades,  392  ;  an 

anchor  of  Delos,  361 

HADEPHAGIA,  Demeter,  57 
Hades,    the    Eleusinian   legend    of, 
58  f. ,  69,  73,   7S,  98  I  the  Eleu' 

sinian  worship  of,    77,    176-178, 

797,  218 
Hadrian's    gate    at    Eleusis,    a   bad 

imitation,  790,  792 
Hagnon  deposed,  and  Brasidas  put 

in  his  place  at  Amphipolis,  41 
Hair  worn  for  Dionysus,  7JJ 
Halicarnassus,  7 

Hall  of  Cimon,  the,  at  Eleusis,  799 
Harpocrates,  Horus,  and  Apollo  on 

Delos,  402 
Harvest  Queen,  the,  jo 

home,  Demeter  and,  69 

usages,  Dionysus  and,  217 

Health   (Hygieia),    Panacea,    Teles- 

phorus  (Convalescence),  attend  on 

Aesculapius,  240 
Heaven,  kingdom  on  earth  of,  13 
Hecate  sitting  in  her  cave,  j6,  759, 

iSl 

Helena,  St. ,  legend  of,  in  Cyprus,  275 


440 


INDEX 


Helicon,    placed    on    Parnassus    by 

Cervantes,  24  ;  Thracian  worship 

of  the  Muses  on,  103 
Hellas,    Eleusinian  divinities   at  the 

great  crisis  of,  211 
Hellenic     genius     in    religion    met 

Semitic    in    Cyprus,    273  ;    ideal, 
•    Apollo,  Dionysus,  and  the,  162 
Hellenism,   effect   on  Aphrodite  of, 

273 

Henry  the  Eighth,  preface  of  Eras 
mus  addressed  to,  260 
Hephaestus  fought  for  Greeks,  46 
Hera  set  Titans  on  Zagreus,   128, 

149,  167,  181 ;  sent  Pytho  against 

Leto  on  Delos,   j>j<?  ;    borrowed 

Aphrodite's  cestus,  299 
Heracles,  87,  371  f. 
Heraclitus  and  Dionysus,  777,  779  f. , 

184 

Hermes,  30,  114,  122,  372 
Hermione,  Demeter-legend  of,  176 
Hero  physician,  Athenian  shrine  of 

the,  2j>9  f. 
Herod,  14 
Heroes,   the  Homeric,  fought  hard 

that  we  might  think  clearly,  229  ; 

misprised  Demeter's  gifts,  46 
Herophilus  on  the  miraculous  nature 

of  drugs,  2j>7 
Hierocles  set  Apollonius  up  against 

Christ,  264 

Hieron  of  Aesculapius,  the,  244,  246 
Hierophant   in  the  Eleusinia,   209, 

212,   2IJ 

Hieroskipou    is    Strabo's  iepoKrjiria, 

290 

Hill,  holy,  of  Dionysus,  94 
Hipparchus  and  Onomacritus,  727 
Hippocrene  on  Parnassus,  24 
History   of   the  worship    of  Aphro 
dite,  303  f. 

Hittite  strain  discoverable  in  Aphro 
dite,  282,  320-323 
Hittites,  the,   in  Cyprus,   280,  321- 

323 

Home,  Demeter  the  goddess  of,  69 
Honey  in  legends  of  Dionysus,  108, 
143 


Hope,  Demeter  still  had,  77 
Horned  Isle  (Kerastia),  poetic  name 

of  Cyprus,  274 

Horns,  Delian  altar  of,  384  f. 
Horse,  cure  at  Delos  of  a,  254 
Horse-racing  at  Delos,  j>77,  380 
Horus,     Apollo    and    Harpocrates, 

402 
Hospitallers,    Knights,    at    Colossi, 

288  ;  on  Delos,  j>79 
Hospitals,    the   temples   of  Aescul 
apius  were  not,  229  f. 
Hostanes  and  Apollonius,  262 
Hours,  the,  and  Dionysus,  96 
Humanity  of  Aesculapius,  239 
Humility  of  Aesculapius,  242 
Huns  overran  Thrace  and  Illyria,  87 
Hunter,  Dionysus  a,  96,   101,   128, 

160,  185 
Huntress,  Artemis  the,  at  Epidaurus, 

250 
Huntsman,  Dionysus  the  wild,   96, 

707,  128,  160,  185 
Hygieia,  220,  240 
Hyperboreans  at  Delos,  J79 

IACCHAGOGOS,  another  name  for  the 

Hierophant,  2/7 
lacchos,  727  f. ,  ij 4  f. ,  ij8  f. ,  210, 

212-214 
lambe,  68 
lano,  King  of  Cyprus,  his  ransom, 

330 

lasion,  father  of  Plutus,  97 
Icaria,  (i)  the  Attic  deme  of,  5,  32, 

36,    80,  101,    104-115,    117-119, 

121,  129,  133,  368;    (2)   (Icaros 

or  Nicaria)  the  Island,  in,  163 

(see  Italia) 
Icarian  legends  and  customs,   104- 

115,  117-119,  123,  131  f. ,   164, 

208 
Icarius,    Si,   105,    107-11 i,  114  f. , 

135 
Icaros  (Icarus,   Icaria,   or  Nicaria), 

in,  163 
Ictinus  at  Bassae,  204  ;  at  Eleusis 

(Hall  of  Initiation),  191,   199  f., 

202-206 


INDEX 


441 


Idalian  fields,  death  of  Adonis  on, 

291 
Idalium,    dependent   on  Chittim    of 

the    Phoenicians,    at,    282,    291, 

3*1 

Illyrian  tribes  and  Thracians,  86  f. 
Immarados  of  Eleusis,  killed  in  war 

with  Athens,  122 
Immigration,      early     Greek,      into 

Cyprus,  280 
Immortality,  ideas  in  the  Mysteries 

about,  176,  181,  211 
Imperialism,  Roman,  fj-fj,  37-39, 

43  t 
Incubation  in  temples  of  Aesculapius, 

236 

India,  political  state  of,  12 
Indian     Dionysus,    a   deification    of 

Alexander's  conquests,  29 
Indo-Caucasus,  wine  first  cultivated 

on,  i6j 
Initiation  and  the  initiated  at  Eleusis, 

176,  181,  189,  208  i.,  218  i. 
Ino,  a  reveller  ofCithaeron,  147,  150 
Inopus,  Leto's  travail  on  the  banks 

of  the,  356 

Inquisition,  the  Spanish,  8 
Inscriptions,   early  Cypriote,  Aphro 
dite  does  not  occur  on,  j2r  f. 
Inspiration  of  Dionysus,   29,  96  f. , 

102,  117,  zjj,  220 
Instruments  used  by  doctors  in  days 

of  Democedes,  223 
Insula,    Hyginus1   idea  of,    includes 

Mauritania  and  Egypt,  394 

Insularum  provincia   of   Diocletian 

called  tirapxLa-  vrjffuv  KvK\d5uv 

393 

Intensity    of    Dionysus    lacking    in 

Aesculapius,  220 
Intolerance,        Mohammedan,        in 

Thrace,  85 
Ion  colonised  Cyclades  as  Apollo's 

son,  391 

Ionian  festival  at   Delos,  372-374 
Islands,  May  festivals  of,  89 

British  rule  of,  277 
race,  Apollo  the  father  of,  J79 

382,  393 


ophon,  Antigone,  904  ff.,  interpo 
lated  by,  166 

phigenia,  love  of,  for  Orestes,  /jo  ; 
struggle  for,  and  the  couvade, 
iji 

ris,  her  fruitless  message  to  Demeter, 

57 
shtar,    Aphrodite  and,    273,  joo  ; 

and    Demeter,    303  ;    descent    to 

Urugal  of,  30 f  ;  lament  for  Tam- 

muz  of,  joo  f. 

[sis  and  Apollo  at  Delos,  402 
Islands  and  peninsulas,  confusion  of, 

in  Homer,  392 

tsmenus,  the  softly  gliding,  133 
Isolarii,  Venetian  books  bearing  title 

of,  328-331 
[srael,  j,  18,  274 
ttalia,  read  Icaria  for,  in  Ant.  1119, 

133 

Ithaca,  no  wheat  land,  47 ,  377 
Ivy-god,  Apollo,  30  f. 
Ivy    and    parsley,    Mystae    crowned 

with,   217  ;  branches  for  Bacchic 

worship,  142 

JACK-IN-THE-GREEN,  Dionysus  at 
tached  to  himself  the  attributes 
of  a,  83 

Japanese  and  Chinese  drama,  masks 
in,  172 

Jason,  a  pupil  of  Chiron,  232 

jealousy  of  Aphrodite's  Assyrian 
nature,  298  ;  prayer  to  Aphrodite 
for  deliverance  from,  joo 

Jerusalem,  Augustus'  offerings  at,  S  ; 
buildings  at  Paphos  and  buildings 
at,  jo6 

Jews,  religion  of,  19 ;  Moeragenes 
maintains  Dionysus  is  the  god  of 
the,  106  ;  Delos  and  the,  J7J 

Jordan,  modern  Delian  legend  of 
the,  jj6 

Journalism,  Apollonius  the  victim  of, 
261 

Joy,  mixture  of,  with  sorrow,  in 
Demeter,  73 

Julii,  oaths  In  acta  divi,  43 

Julius,  Jupiter,  37 


442 


INDEX 


JuliaDomna,  Philostratus,  Romancer- 

in-ordinary  to,  258-260 
Jupiter,  philosophical  supreme  god, 

15,  20,  26  ;  Julius,  37 


,  44 
Kadapot,    on    Sybarite    inscription, 

176 

Kerastia,  a  name  of  Cyprus,  345  f. 
Kerykes,  the  Deliastae  chosen  from, 

J7* 
Kittim,    the    Phoenician    Larnaca- 

Citium,  28r 

Kolossi,  Garin  de,  at  Colossi,  288 
Kore  in  Orphic  story,  181 
Ki/JcXddwi  tirapxla  vyvuv,   Diocle 

tian's  Insularum  provincia,  called 

the,  303 

LABYRINTH   below   the  Epidaurian 

Tholos,  253  f.  ;    Theseus  in  the 

Cretan,  400 
Lakkiadae,  station  of  the  Mystae  at, 

217 
Lapethus,  importance  of,  in   North 

Cyprus,  277 
Lapithae,  Aesculapius  the  tribal  god 

of,  221,  233  ;   on  the   Epidaurian 

temple  of  Aesculapius,  249 
Lararium  of  Severus,  Christ,  Abra 

ham,    Orpheus,   -Apollonius,    and 

emperors  in  the,  265 
Lares       Compitales       and      genius 

Augusti,  38 
Larnaca,     early    fcmndation    at,    by 

Phoenicians,     281  ;     fortified    by 

Knights  Templars,  288 
Lasus  of  Hermione  detected  Onoma- 

critus,  727 

Laughless  stone  at  Eleusis,  60  f. 
Laurel  tree,  Lord  Bacchus,  lover  of 

the,  jo  f.  ;    Leto  and  the,  jj6, 

j6j  ;  connection  of  Apollo  through 

Daphne  with,  368 
Layard,  Sir  Henry,  de  Mas  Latrie's 

dedication  to,  278 
Lazarus  found  a  cross    in  Cyprus, 

275 
Lemnos,  destruction  of,  predicted,  127 


Lenae,  biennial  procession  of,  132 
Lenaea,  the,  and  mid-winter  festival 

of  Icaria,   132  ;    first   tragedy  at 

Athenian,  132 
Lenaean     Dionysus       earlier     than 

Dionysus  Eleuthereus,  118,  132 
Lenaeon    in     Boeotia    and     Icaria, 

112  f. ,  132  ;  Gamelion  and,  132  ; 

Brumalia  in  the  season  of,  112 
Lenides,  or  Lenae,  132 
Lepanto,  the  battle  of,  331 
Lepidus  and  the  favour  of  Tiberius, 

42 

Lesbos,  14 
Leto,   traditions  and  character    of, 

355-358,  363,  370,  383  f. 
Lightning,  Apollo  sent  the,   for  sig 
nal  to  Theoris,  j>77 
Ligyon,  Achilles,   so  named  for  not 

taking  mother's  milk,  64 
Liknites,  lacchos  surnamed,  128 
Limassol,  land  at,  for  old  Paphos, 

2?5>  2$7 
Limestone  in  Cyprus  is  friable,  288 

f. 
Lions,    frieze    of,     on    Aesculapius' 

Temple   at    Epidaurus,   248 ;    on 

cornice  of  the  Epidaurian  Tholos, 

95* 

Little-in-the-fields,  Mr.  Browning's 
name  for  the  rural  Dionysia,  131 

Lotus  blossom,  miraculous  effect 
of,  231 

Lusignan,  Hugues  I.  of,  gave  Col 
ossi  to  Hospitallers,  288 

Lusignans,  feudalism  of,  in  Cyprus, 
278 

Lusios,  Dionysus  called,  f^o,  162 

Lustrations  of  Delos,  362  f. 

Lycian  Patara,  oracles  of  Apollo  at, 
370 

Lycians  in  Cyprus,  279  f. 

Lycorea,  peak  of  Parnassus,  25 

Lycurgus  and  Dionysus,  story  of, 
77  ;  a  Thracian,  77  ;  his  hatchet, 

77 

Lydian  airs,  and  the  musical  es 
capade  of  Cinyras,  294 

Lydias,  Dionysus  and  the,  757 


INDEX 


443 


Lyre,  Apollo's,  lent  to  Dionysus,  31  ; 
Apollo  and  Hermes  invented  the, 
372  ;  seven  strings  of  Apollo's, 

359 
Lysander,  deification  of,  41 

MACEDONIA,  79,  86  f. ,  737,  22r, 

379,  386  f. 
Machaon,  son  of  Aesculapius,  222, 

230 
Macris,   Mysa,    daughter    of    Aris- 

taeus,  nurse  of  Dionysus,  143 
Madeira,   restocked  with  vines  from 

Cyprus,  288 
Madness,     Dionysus  -  worship   and, 

no,  120,  141,  145,  147,  i6r  f. 
Maenads,  in  legends  and  worship  of 

Dionysus,  31 ,  79,  100,  112,  128, 

I37<  f39<  *4f>  fJ5>  T50<  138-160 

181,  184  f. 
Maera,    in    the    Old    Attic     Icarian 

legend,  709-7/7,  77,/ 
Magdalene,    St.    Mary,   in  Cyprus, 

2J5 

Magicians,  Apollonius  among  the, 
262 

Magnes  of  Icaria,  777 

Maidenswell  at  Eleusis,  60  f. 

Malta,  ruins  on,  and  Paphian  ruins, 
306 

Mandra,  a  sheepfold  in  Cyprus, 
often  a  cave,  289 

Manwrecker,  Dionysus,  surnamed, 
100 

Maps  of  Cyprus,  modern,  274 

Maps,  neglect  by  Meursius  of  avail 
able,  Jj-7 

Marathon,  mountains  near,  early 
Thracians  settled  on,  77  f. ,  80, 
soj  ;  journey  from,  to  Icaria  over 
comes  Dionysus,  106  ;  the  wilder 
ness  of,  if  4  ;  the  Deliastae  at, 
376  ;  Delion  at,  376  ;  daisies  grow 
on  the  field  of,  252 

Marathonian  tetrapolis,  the  worship 
of  Apollo  from,  104 

Marathus  of  Phoenicia,  the  Snail's 
Tower  at,  and  the  south  wing  at 
Paphos,  308,  314 


Marco  (San),   title  of  Cavaliere  di, 

350 
Mardonius,   devastated  the  Hall  of 

Pisistratus,  79^ 
Maron,  wine  given  by,  97 
Marpessa  and  Apollo,  368 
Marriage  of  Dionysus  and  Ariadne, 

130  ;  of  Dionysus  and  Basilissa, 

130  f.  ;    of   Dionysus  and  Perse 
phone,  ij8  ;  of  the  doge  and  the 

sea,  7j7. 
Masks,  Japanese  and  Chinese,  use  of, 

in  plays,  172  ;  use  of,  in  tragedy, 

a  barbaric  survival,  172 
MacrroeiS^s.Strabo's  epithet  of,  proves 

his  Olympus  was  Troodos,  j2j 
Plater  dolorosa,  Demeter  as,  77 
Matriarchy,      protest       of      Apollo 

against,  777 
Mauritania,   one  of  Hyginus"   insu- 

lae,  394 
Mavrocordato,   Alex.,  was  a  doctor 

of  Padua,  268 
Maximus   of  Aegae   wrote  of,    and 

shared,  ascetic  life  of  Apollonius, 

*S7  I 

May-day  festival,  ancient,  at  Icaria, 
772  ;  at  modern  Corfu,  89 

Mayor  of  Thebes,  a  doctor,  267 

Measure,  Attic  sense  of,  in  Diony 
sus-worship,  /Of,  77j> 

Medea  had  knowledge  of  miracu 
lous  drugs,  231  ;  Aphrodite  as 
represented  in  the,  300 

Media  worships  Dionysus,  140 

Mediaeval  church,  laws  of  purifica 
tion  in,  360  ;  churches,  ruins  of, 
in  Cyprus,  283 

Mediation,  final  requirement  of,  by 
Aesculapius,  254,  256 

Medicine  (Greek)  early  (i.)  begin 
nings  of,  223-225— (ii.)  practical 
bearings  of  Homeric,  228  f.—  (Hi. ) 
positive  and  mythical  aspects  of, 
232-234;  later  perfection  of,  (i.) 
sacred,  222,  226,  230,  232,  237 
f. — (ii.)  secular,  223-22^,  229  f., 
234  f. ,  238  f.  ;  modern  Greek, 
study  of,  222,  268 


444 


INDEX 


Meeting-house   at    Eleusis,    and    in 

Thrace,     for     Dionysus -worship, 

i8g 
Megara,  and  Icaria,  relations  of,  to 

early  comedy,  115 
Meilichius,  Dionysus  surnamed,  101 
Melanaigis,      Dionysus,      seen      by 

daughters     of     Eleuther,      120  ; 

Dionysus,  madness  sent  by,  120, 

133 

Meleager,  a  pupil  of  Chiron,  232 

Mell  supper,  5 

Melpomene,  an  emanation  from 
Dionysus  Melpomenos  of  Eleu- 
therae,  104 

Melpomenos,  surname  of  Dionysus, 
the  god  of  song  and  dance,  104 

Memphis,  14 

Meroe,  thigh  mountain,  and  second 
birth  of  Dionysus,  765" 

/j.r)poTpa(pris,  Dionysus  called,  166 

Mesorea,  or  mid-mountain  of  Cy 
prus,  275-277  ;  devastation  of, 

1  339 

Messianic    vision    of    Euripides    in 

The  Bacchanals,  138 
Metagenes    of    Xypeta,    builder    of 

Eleusinian  temple,  191,  204 
Metanira,   legends  of,    62,    65,    68, 

107 
Metapontum,   story  of  Parmeniscus 

of,  383 
Methe,    painted   by   Pausias  in  the 

Tholos,  253 
Meursius,  John,  his  life  and  work, 

324,  344,  353,  347-351 
Midas,    Dionysus,    and  the  legends 

of,  79,  87,  92  f.,  204 
Milton's    debt    to    Euripides,    148, 

150 
Mimallones,   the  Thracian  originals 

of  Bacchanals,  97 
Mining  invented  by  Cinyras,  294 
Miracles  inscribed  at  Epidaurus,  254; 

wrought  by  Aesculapius,  234,  237 

f. ,  245  ;  of  Apollonius,  sanctioned 

by  Aesculapius,  .257 

—  wine  a  perpetual  source  of,  97 
of  Dionysus,  143,  149,  153 


Miraculous  drugs  in  Homer,  moly, 
nepenthe,  lotus-blossom,  231 

Mnesicles,  Propylaea  of,  imitated  at 
Eleusis,  790 

Mocenigo,  Isolario  of  Benedetto 
dedicated  to,  330 

Moeragenes,  maintained  that  Diony 
sus  was  the  god  of  the  Jews,  186 ; 
a  bandit  of  the  Taurus,  258  ;  the 
Athenian  ?  told  of  Apollonius' 
life,  258 

Moesa,  and  her  two  daughters,  of 
Julia  Domna's  clique,  259 

Moliere,  farce  made  of  Democedes' 
story,  224 

Moly,  Herophilus  on  drugs  com 
pared  with  Homer's  account  of, 
237 :  238 

Monastery  of  St.  Nicholas,  on  Cape 
Curias,  288 

Monks,  the  destroyers  of  Eleusis 
were,  188 

Monotheism,  of  Greek  religion,  10 

Morality,  the  Bacchanals,  7J7 

Mortality  of  Aesculapius,  his  mortal 
schooling,  231,  232 

Moscow  MSS.  of  hymns  to  Demeter 
and  Dionysus,  163 

Moses  and  Apollonius  as  magicians, 
262 

Mother  Rye,  Demeter  as,  51 

Motion,  of  the  elements  of  Diony 
sus,  92 

Mountaineer-festivals,    two    Icarian, 

772  f. 

Mourning    for   Dionysus,    especially 

in  Icaria,  96  f. ,  772  t. 
Musaeus,  Onomacritus  falsified,  727 
Musagetes,   surname  of  Apollo,  the 

god  of  song  and  dance,  104 
Muses,  the  worship  of,  with  Apollo 

and    Dionysus,    25,   31,    78,    96, 

102-104,  SJ5'  *59 
Music,  relation  of  instruments  of,  to 

Dionysus,  143  f. 
at  the  Delia,  380 


Musical  contest  of  Apollo  and  Ciny 
ras,  29^ 
Mustafa,  at  the  sack  of  Nicosia,  332 


INDEX 


445 


Mycenae  vases,  rosettes  on,  231 
Myconos  and  Delos,  330,  363,  376, 

Mylitta  and  Aphrodite  -  Urania, 
208 

Myron,  arraignment  of  Alcmaeonidae 
by,  124 

Mystae,  yearly  procession  from 
Athens  to  Eleusis  of,  212-218 

Mysteries,  the  Eleusinian  or  Greater, 
122,  123,  14 f,  173  f.,  180-180, 
207  -  210,  218  f.  ;  the  Lesser  at 
the  Athenian  Eleusinion,  124  f., 
127  ;  the  Samothracian,  403 

Mystery,  ancient  and  modern  mean 
ing  of,  178 

Mystes   and    Epoptes,    degrees   of, 

Mythology,  j,  4,  7;  era  of  conscious 
analysis  ;  in,  78;  (Greek)  con 
trasted  with  philosophy  and 
theology,  32  ;  the  critic  of,  must 
not  offend  the  poet,  28 

Myths  (Greek),  their  relation  to 
fetichism,  27  ;  lives  of  Christian 
saints  and  late  pagan,  263  ;  the 
beautiful,  of  early  Attica,  71,  104- 
112 

NAIADS,  give  not  wine  but  water, 
108 ;  torch-led  dance  of,  88  (see 
Fire  and  Maenads) 

Narcissus,  the,  in  myth  of  Perse 
phone,  36 

Nature  -  worship  in  Greece,  73, 
136  f.  ;  personifications  of,  7^', 
97  ;  theory  of,  at  the  bottom  of 
Eleusinian  and  Dionysiac  worship, 
140,  178-180,  182-183  ;  Aescula 
pius  and,  220 ;  Aphrodite  and, 
271,  273 

Nauplia,  road  from,  to  the  Hieron  of 
Aesculapius,  244 

Nausicaa,  compared  to  the  Delian 
palm-tree,  333 

Naxia,  Tintoretto  painted  Acropolis 
of,  in  his  Bacchus  and  Ariadne, 
328 

Naxos,  Delian  Apollo  and,  363,  376, 


380,  400  ;    Dionysus,  legends  of, 

Si,  in,  ijo,  163 
veuKopia,  competition  for,  .// 
Neocorion  at  Eleusis,  the,  797 
vfUKopos,     finally      removed     from 

meaning  temple-sweeper,  44 
Nepenthe,  Helen's,  stands  for  curing 

mind  through  body,  234  f. 
Nero,    the    foe    of    mankind,    42  \ 

manner  of  his  birth,  42  ;  probably 

gave  the  Icarian  sculptures  to  the 

Athenian  stage,   //./  ;    as  Apollo 

in  the  flesh,  114 
Nestor,  high  esteem  of  doctors  felt 

by,  222  ;  a  pupil  of  Chiron,  232  ; 

wounding   and   fall   of  horse  of, 

Netherworld-god,  Aesculapius  like 
Dionysus  a,  220 

New  York,  Varoschia,  Famagosta, 
Salamis,  and,  283 

Nicaria  (Icaros  or  Icaria),  an 
Aegean  island  .where  Dionysus 
was  born,  163  \  Draconus,  a  pro 
montory  on,  163 

Nicholas,  St..  a  monastery  on  Cape 
Curias,  288 

Nicias  at  Delos,.  378-381  ;  endow 
ment  of  prayers  by,  380 

Nicosia,  cathedral  of,  built  by 
Knights  Templars,  288;  Venetians 
banished  to,  326  ;  siege  and  sack 

of,  332-334 
Nile,    connected    with    the    Delian 

Inopus,  336  ;  the.  Cypriote  Pedi- 

aeus  and  the,  276 
I  Niniveh,  Aphrodite-Ishtar  at,  303  ; 

poem  on  Ishtar  found  at,  301  ; 

Adonis  of  Amathus  and  the  Tam- 

muz-Adonis  of,  20.2 
Ninivite,  Damis  the,  told  credulous 

tales      of      Apollonius'      travels, 

237  f- 
Niobids,  Apollo  and  Artemis  pursue, 

North   Cyprus,    an    elegant    extract 

from  Asia  Minor,  277 
North    range    of    Cyprus,    age  of, 


446 


INDEX 


Northern  origin  of  Aesculapius  and 
gods  at  Eleusis,  220  f. 

Northumberland,  Demeter's  coun 
terpart  in,  49  f. 

Notables,  connection  with  imperial 
worship  of  provincial  meetings  of, 

43*. 

Notre  Dame  de  Bon  Secours,  Aphro 
dite  Sosandra  as,  298 
Novices,  the  Eleusinian,  182,  208 
Nymphs  and  naiads,  78,  94 
Nysa,    curious  legends  of,  78,  134, 
143,  163-166,  168 

OAK,  branches  for  Bacchic  worship, 
142 

Oath,  doctors',  by  Apollo,  Aescula 
pius,  etc.,  220 

Oaths,  "  In  acta  divi  Augusti "  and 
"  divi  Julii,"  43 

Oceanus,    Persephone   seized   near, 

5* 

Oceanus,  Eleusis  son  of,  64 
Octavius,  an,  of  the  Emperor's  Gens, 

14 

Odysseus,  the  domain  of,  no  wheat- 
land,  47  ;  streams  poured  for  the 
dead  by,  97,  94  ;  anatomical 
knowledge  of,  228  ;  Chiron's  in 
struction  of,  232 ;  the  ship 
wrecked,  273 

Odysseus  of  Gravia,  a  modern 
Leonidas,  400 

Oil  contrasted  with  wine,  108 

Oliveto,  Pius  II.  describes  Monte, 
326 

Olympia,  9  f.  ;   271,  372 

Olympian  Zeus,  capitals  of  Athenian 
temple  of,  252  ;  mildness  of,  in 
Aesculapius,  220 

Olympus  of  the  gods,  jj,  20,  26,  28, 
57,  777,  240,371  ;  of  Thrace  and 
Macedonia,  20,  87 ,  ijj ;  of 
Cyprus,  an  investigation  of  its 
whereabouts,  324-343 

Omadios,  Dionysus  surnamed,  101 

Omestes,  Dionysus  surnamed,  101 

Omnipotence,  not  chiefly  repre 
hended  by  Greeks  in  Christian 


ideal,  20  ;  Greek  gods  had  local, 

10 
Onomacritus,     125-127,     174     f., 

181 

oiraiov  of  Eleusinian  temple,  797 
Opposites,  Demeter  a  curious  ming 
ling  of,  73 

Oracles,  34-36,  06,  238,  255 
Oreads,  Dionysus  and  the,  06 
Orestes,   Iphigenia's  love  for,   750  ; 

justified  by  the  principle  of  Diony 
sus'  second  birth,  169  f. 
Orgies    of   Dionysus,    34,    154  ;    of 

Demeter,  63 
Orientation  of  temples,  facts  about 

the,  192 
Origen,  read  four  books  of  Moera- 

genes  on  Apollonius,  238,  264 
Oropus,    discoveries   about   Amphi- 

araus  at,  232 
Orpheotelestae,  the,  iSr 
Orpheus,    Thracian   tradition   of,  9, 

IOJ>   *57>  2^2  ;  in  the  Lararium 

of  Severus,  263 
Orphic  doctrines  and  myths,  25,  87, 

125,  127,  178,  181 
Ortygia  and  Asteria,   barren  legend 

°f.  350,  35S 

Ostrogoths  at  Eleusis,  188 
Otherness,    equivalent    to    reality  in 

Thracian  conception  of  Dionysus 

and  his  world,  90 
Otherworlds  of   Iliad  and   Odyssey, 

49 

Ovid,  his  lowly  setting  of  the  stay  of 
Demeter  with  Celeus,  71 

PACTOLUS,  Dionysus  bade  Midas 
wash  in  floods  of,  92,  178 

Paean  Apollo,  the  sun-god,  father 
of  Aesculapius,  231,  241  f. 

Pagans,  it  is  best  to  make  tolerable 
sense  of  the  affairs  of,  19  ;  criti 
cism  of  the  solitary  Christian 
god  by,  20  ;  testimony  for  Apol 
lonius  of,  264  f. 

Paganism,  3,  7  ;  last  days  of,  21, 
45,  258,  261,  266 

Palaia,  a  town  between  Larnaca  and 


INDEX 


447 


Limassol    not  next    to  Olympus, 

325 
Palm    tree   and   laurel   of    Leto   at 

Delos,  356,  363,  380 
Palm  tree  in  bas-relief  of  Dionysus 

at  the  house  of  Icarius,  106 
Pammerope,  daughter  of  Celeus,  63 
Pan,  distinguished  from  Satyr,  99  ; 

personifies  cruelty  of  Dionysus,  99 
Panacea   named   in    doctors'    oath, 

220  ;  attends  on  Aesculapius,  240 
Panas,    Dr.,    of  the    Paris  faculty, 

269 
Pandarus,     ignorance    of    anatomy 

punished  in,  228 
Pandemos,    Aphrodite,     shrine    of, 

near  Asclepieium  at  Athens,  270  ; 

Mylitta  represented  by,  298 
Pandion,  dealings  with  Thracians  of 

king,   in  ;      the  fifth  king  since 

Cecrops  is,   705  ;    Demeter  came 

to  Eleusis  under  king,  7/9  ;  Attic 

Dionysus-legends  and  king,  779 
Panic    terrors  inspired  of  Dionysus 

and  his  Pans,  99 
Pans,  or  Aegipans,  98-100 
Papticapaeum,   men  of,     at    Delos, 

JS? 

IId0i/a,  i],  later  Aphrodite,  repre 
sented  by,  321 

Paphian  Aphrodite,  distinguished 
from  the  Syrian  Goddess  on  Delos, 
403 ',  cone  of,  and  Delian  Bae- 
tylus,  386  ;  Aphrodite  at  Tegea 
and,  285 

Paphos  (new),  283 

(old),  2,  5  ;  history  of  the 

temple  and  worship  of  Aphrodite 
at,  2J2,  279,  281  f. ,  284  f. , 
289-291,  293,  295-297.  304-308, 
ji 2  f. ,  324,  345  f. 

Parian  Marble,  Susarion  and  the, 
777  ;  Aristotle's  Constitution  of 
Athens  and  the,  132  ;  fixes  the 
date  of  the  first  tragedy  at  Athens, 

I32 
Parisian  feuilletonist,  Philostratus  a, 

260 
Parmeniscus  at  Delos,  story  of,  382 


Parnassus,  description  of,  23-25, 
33<  3°.  f33  '.  Dionysus  and 
winter  festivals  on  the,  77.2  f. , 
132  f. ,  149,  156  f.  ;  Apollo  and 
Dionysus  on  the,  104  ;  Daulis  on 
the,  777 

Paros,  Archilochus  and,  365  ;  Delos 
and,  365,  376 

Parsley  and  ivy,  Mystae  crowned 
with,  .277 

Paspati,  Dr.,  a  well-known  author, 
269 

Passion,  the,  4 ;  in  the  Dionysus- 
legend,  153-157,  181 

Passion- Play  of  Attica,  The  Baccha 
nals  called,  7J7 

Patara,  winter  oracles  of  Apollo  at, 

37? 

Patriarchy,  in  the  primitive  family, 

770 
Patrician  worship  of  Eleusis,  Icarian 

Dionysus  added  to  the,  208 
Patroos,  Apollo,  altar  at  Delos  of, 

379>     3& 2   I     tne    Cyclacles    are 

chosen  isles  of  Apollo,  393 
Paul,  St.,  Simon  Magus,  Apollonius 

and,  258,  260 
Pausias,  paintings  of,  in  the  Tholos, 

252  L 
Pauson,  compared  unfavourably  with 

Polygnotus,  252 
Peasant,  the  Greek,  of  to-day  and 

his    taste     in    wine,     706   ;    the 

Cypriote,   335,  341  ;    admiration 

for  doctors  of  the  Greek,  222 
Pediaeus,  the,  a  Cypriote  river  Nile, 

275.  35 f 
Pegasus,  of  Eleutherae,  j>./,  i2O\  on 

Parnassus,  24 
Pelasgians,       Thracians      identified 

with,  220 
Peleus,  the  fire-baptism  of  Achilles 

and,     64;     a    pupil    of  Chiron, 

232 
Peloponnesian  \V;ir,   Greek  religion 

stood    intact  until  time  of,   136  ; 

state  of  Greece  after  the,  ir 
Peninsulas  and  islands,  confusion  of, 

in  Homer's  day,  392 


448 


INDEX 


Pentelicus,  home  of  Dionysus  near, 
104  f. 

Pentheus,  Dionysus  flies  to  Icarius 
from,  107  ;  the  frenzy  of  Zagreus- 
Dionysus  and,  139  ;  in  The  Bac 
chanals  of  Euripides,  145-160 

Pericles,  and  the  temple  at  Eleusis, 
187 

Persephone,  earliest  and  forbidding 
conception  of,  47-49,  128  ;  later 
myth  of  (post-Homeric),  50,  54 
f-,  57-59 >  t>9<  75  *"•  ;  general 
religious  and  moral  aspects  of  the 
divinity  of,  48,  69,  73-75,  213  ; 
significance  at  Eleusis  of,  73  f., 
77,  144,  176,  178,  218  ;  lacchos 
and,  174  ;  Kore,  a  name  of 
(Orphic),  181 ;  Adonis  plays  the 
part  of,  297  ;  tie  between  Aescu 
lapius  and,  219  f.  ;  Erigone  con 
fused  with,  107  ;  Zagreus  the  son 
of,  128  ;  monuments  and  rites 
concerning,  59  f. ,  70,  194-196, 
213 

Persia,  Democedes  led  captive  to, 
224 

Persian  order,  the,  12,  17 

Persians,  destruction  of  the  Eleu- 
sinian  Hall  by  the,  198  ;  inter 
ference  with  Eleusinian  procession 
from  Athens  by  the,  211 

Personification,  implications  in  pri 
mitive  mind  of,  78 

Pfingstl,  killing  of  the,  compared  to 
Pentheus'  death,  759 

Phaedra  calls  Aphrodite  Cypris 
Pontia,  297 

Phaedrus,  stage  built  by,  at  Athens, 
114 

Phaedryades,  rocks  at  Delphi,  23 

Phaestus,  near  Cnossus,  Epimenides 
of,  124 

Pharsalia,  watchword  of,  300 ; 
Dante  remembered  Lucan's,  24 

Phidias,  Plutarch  would  leave  super 
vision  of  Eleusis-buildings  to,  191 

Philip,  Dionysus  and  the  house  of, 

79.  J3$ 
Philip,  portico  at  Delos  of,  379 


Philippi,  inscription  found  near,  87  \ 
Dionysus'  hill  near,  92 

Philo,  porch  of,  at  Eleusis,  790  f. 

Philonides  named  CyprusAcamantis, 
274 

Philosophers,  Apollonius  among  the 
illustrious,  263  f. 

Philosophy,  mythology  and  theology 
contrasted  with,  32 

Philostratus,  Apollonius  not  the  play 
acting  personage  of,  261  f. ,  264 

Phlegyae  and  Lapithae,  Aesculapius 
tribal  god  of,  221 ,  233 

Phlegyas,  visit  of,  to  Epidaurus,  243 

Phoebus,  35  f.  ;  leader  in  the  dance, 
29 

Phoenicians,  name  of  Dionysus  from, 
85  ;  Dionysus  as  a  child  of  the 
thigh  came  with  the,  163-165, 
168 ;  Moesa,  Julia  Domna,  and 
Ulpian  were  by  descent,  25-9 ; 
Aphrodite  a  goddess  of  the,  277, 
286,  323  ;  Atargatis  on  Delos 
worshipped  by  the,  403 ;  Cabiri 
brought  by  the,  402  ;  Delos  and 
the,  374  ;  Cyprus  and  the,  279, 
281-284,  319,  321  ;  Old  Paphos 
of  Cyprus  and  the,  281,  284  f., 
306,  308 ;  Hittites  confused  with 
the,  322  f.  ;  Selinus  and  the,  306 

Phrygians,  cousins  of  Thracians,  85, 
90,  103  ;  Dionysus  and  the,  77 , 
90,  95,  103,  140,  164,  178; 
Adonis  and  the,  291  ;  Cinyras  and 
the,  294 ;  Cybele  and  the,  97, 
143 ;  Rhea  of  Crete  and  Cybele 
of  the,  143 

Phye,  Thracian  girl  named,  126 

Phyllis  betrayed  by  Demophoon,  65 

Phytalus  entertained  Demeter  at 
Lakkiadae,  217 

Pieria,  a  centre  of  Dionysus -wor 
ship,  87,  137,  153 ;  spirits  of 
waters  the  nurses  of  Dionysus  in, 
94  ;  Icaria  an  Attic,  in 

Pieris,  a  centre  of  Dionysus-worship, 
87,  92,  94 

Pindar  remoulds  story  of  Tantalus, 
18  ;  transcends  idea  of  a  resistless 


INDEX 


and  unrelenting  God,  79  ; 
in  nobility  of  religious  thought, 
29  ;  solemn  aspect  of  Aphrodite's 
charm  seen  by,  299  f.  ;  his  ac 
count  of  Cinyras  as  Apollo's  and 
Aphrodite's  friend,  292,  29.;, 
299  f.  ;  account  of  Parnassus  and 
deluge  by,  25  ;  birth  legend  of 
Aesculapius  keeps  him  in  Thes- 
saly,  245  f.  ;  his  pious  preludes, 
102  ;  praises  of  Delos  by,  364 

Pine-tree  dedicated  to  Dionysus, 
rob  ;  Pentheus,  perched  in  a,  150, 
138  f. 

Pirates  carried  off  Dionysus,  168 

Pirithous  or  Apollo,  in  the  Olym 
pian  pediment,  9 

Piscopia  of  the  Venetian  Cornari, 
2c?9 

Pisistratidae,  727 

Pisistratus,  Solon's  censure  and 
acting  of,  if 6,  126;  Dionysus- 
legend  and  worship  developed 
by,  125-132,  137,  175  ;  second 
exile  of,  126,  137  ;  lustration  of 
Delos  by,  362  ;  tillers  of  the  soil 
befriended  by,  126  ;  Eleusinian 
Hall  of,  igS  f. ,  204 

Pissouri,    brighter   landscape    near, 

2<?9 

Pius  II.  translates  Strabo  on 
Cyprus,  326 

Planets,  Dionysus  and  the,  133 

Plataea,  77 

Plato,  arguments  about  Gods  of,  iS 
f .  ;  on  the  "Streamers,"  779; 
Aphrodite  Urania  of,  298 ;  pro 
test  against  the  poets  of,  26 ; 
leader  in  nobility  of  religious 
thought,  29 

Pleiades,  Alcyone  of,  779 

Plothea  of  Epacria,  105,  779 

Pluto,  Dionysus  brought  the  epithet 
to  Hades  at  Eleusis,  174,  777  f.  ; 
precinct  of,  in  Eleusinian  sanc 
tuary,  103-105 ;  of  eight  Eleu 
sinian  divinities,  178 

Pluto,  daughter  of  Oceanus  and 
Tethys,  08 


Pluto,  or  Plutus,  97 

Plutus,  nature  of  his  divinity  like 
Dionysus',  97  f. 

Plutus,  Demeter's  son,  97 

Podalirius,  son  of  Aesculapius,  in 
fallible  in  Homer,  230 

Podocathari,  fortunes  of  the  family 
of,  331  f. ,  336-340 

Podocatharo,  Giouanni,  his  loyalty 
to  king  lano,  336  \  Hettore,  330- 

333<  33S-342 
Cardinal  Ludovico,  collections 

of,  jji 

Pietro,    under    the    ' '  re   bas- 

tardo,"  reward  of,  336  f. 

Poetic  inspiration,  702 

Poet,  Plato's  prohibition  to  the,  18; 
Apollo  as  conceived  by  the,  25, 
33  ;  inconsistency  demanded  in 
any  treatment  of  Dionysus  by  the, 
136  ;  critics  of  Greek  mythology 
must  not  offend  the,  28  ;  Dante 
the  poet's,  33  ;  Eleusis  and  the, 
182  ;  religion  and  the,  38,  136 

Poetry,  Greek  mythology  and  tilt- 
abstract  spirit  of,  28  ;  Greek  re 
ligion  is  the  poetry  of,  33  ;  Icaria 
influenced  the  history  of,  ///  ; 
Apollo  and  Dionysus,  the  two 
gods  of  dancing,  song,  and,  104 

Politics,  ancient  and  modern,  10, 
15 

Pollux,  a  pupil  of  Chiron,  2j2 

Polycletus,  the  elder  and  the 
younger,  246  f.  ;  the  Tholos  of, 
at  Epidaurus,  250-254  ;  theatre 
built  by,  at  Epidaurus,  276 

Polycrates,  called  Democedes,  from 
Athens,  22^  ;  gave  Rhenea  to 
Apollo,  361  f. ,  376 

Polygnotus,  Pauson,  compared  un 
favourably  with,  252 

Polyphemus,  clever  wounding  of, 
228 

Polytheism,  3,  70,  20 

Polyxenus,  a  son  of  Celeus,  63 

Pomegranate-seed,  in  myth  of  Per 
sephone,  58 

\\ovr apxn^'  •// 


2    G 


450 


INDEX 


Pontia,  Cypris,  Phaedra's  name  for 
Aphrodite,  207 

Porch,  used  for  stoa  or  porticus, 
2j6,  248,  310 

Porches,  for  the  sick  in  Aesculapian 
sanctuaries,  247  f.  ;  need  of,  at 
Paphos,  jio 

Portulani,  in  the  Marcian  library, 
and  the  Museo  Correr,  j2q 

Poseidon,  Delos  and  Delphi  be 
longed  tO,  J2,  37  f.  ,  JjS 

—  father  of  Aithousa,  the  mother 
of  Eleuther,  ng 
and     Apollo      walled     Troy, 

372 

—  father  of  Persephone,  48 

Possession,  by  Dionysus,  o,  no, 
133  ;  the  whole  play  of  The  Bac 
chanals  a  case  of,  ij6 

Prayer  to  Dionysus,  133 

Prayers,  endowment  of,  at  Delos, 
j8o  ;  to  Aesculapius,  Apollo 
named  first  in,  220 

Preceptorery,  older  name  for  com- 
mandery,  288 

Prehistoric  man  in  Greek  art,  173  ; 
tribe  changes,  and  second  birth 
of  Dionysus,  163  (see  Primitive) 

Prescriptions  of  Aesculapius  given 
in  dreams,  2jj 

Priests,  of  Aesculapius,  2jo,  234  f. , 
241  ;  cleverness  of  Delphian,  30  ; 
Delphian,  sanctioned  brotherhood 
of  Apollo  and  Dionysus,  31  ;  of 
old  Paphos  descend  from  Cinyras, 
201 

Primitive  belief,  in  personification, 
78 ;  custom,  survives  in  tragic 
masks,  172  ;  Dionysus,  confusion 
about,  oo  ;  family,  160-171 ;  man, 
12  ;  medicine,  Chiron  and,  232  ; 
worship,  106,  13?,  105,  203 

Probation,  Dionysus  before  he  came 
to  Athens  underwent  a  triple, 
Si 

Procession  to  Eleusis,  many  local 
customs  connected  with,  218  ;  of 
the  Mystae  from  Aristophanes' 
Frogs,  212-216 


Processional  ways  at  Athens,  Delos, 

and  Eleusis,  103,  jyg 
Proclus,  an  Athenian  friend  of  Aes 
culapius,  256 

Prophecy,  coupled  with  wine,  03 
Prophet,  divine,  Dionysus  a,  78 
Prophets  of  Dionysus,   the  Baccha 
nals  are,  ij6  f. 
Propitiation  of  Dionysus,  134 

of  Eleusinian  gods,  desire  for, 


at  Athens  in  405  B.C.,  215  f. 
Propylaea,  Athenian,  307  ;    Eleusin 
ian,  186 
Proserpina,     rape    of,    localised     in 

Sicily,  62  (see  Persephone) 
Provindemiator,  star  in  the  Icarian 

legend,  in 
Prudentilla,     the     heiress,     wife     of 

Apuleius,  262 
Prussia,  East  and  West,  6 
Psophis,  Phoenician  foundations  at, 

285  i.t  323 
Ptolemy,  Cos  and,  likened  to  Apollo 

and      Delos,     304  ;      Theocritus 

praises,  318 

—   (the     geographer),    place    of 

Cypriote  Olympus   according  to, 

3*7 

Public  doctors  and  ancient  hospitals, 
229 

Punch  and  Judy  show,  Greek 
counterpart  of,  go 

Purification,  days  of,  at  Athens  before 
the  Eleusinia,  209  f.  ;  at  Delos, 
jjj  ;  Milton's  lines  upon,  360 

Purifying  powers  of  Dionysus,  133 

Purity,  requirement  of,  by  Aescu 
lapius,  23 j,  2jj  ;  ideal  of  Apollo 
and  Delos,  jjj,  360-363,  370 

Pygmalion  and  Cinyras,  kinship  of, 
204  ;  sculpture  invented  by, 
204 

Pyrasus  of  Demeter,  47,  221 

Pythagoras  and  the  Pythagoreans, 
72,  21  Si.,  223,  257,  382 

Pythian  monster  slain,  j>j  ;  Apollo's 
expiation  for  slaying  the,  371  ; 
sent  by  Hera  against  Leto,  358 

Pythoness,  description  of  the,  36 


INDEX 


QUAKER-MEETING,  and  first  tragic 
actor  of  Thespis,  7/7 

Quintilian,  Tacitus,  and  Pliny,  rever 
ence  for  the  imperial  idea  felt  by, 
4ri. 

Quiver,  votive  gift  at  Delos  of  a 
leaden,  384 

RANSOM  of  Christians  from  Turkish 

captivity,  jjjf.,  339 
Rapendosa,  the  valley  of,  at  Icaria, 

705,  109,  114 
Rarian   plain,   the,  55  f.,  66,    187, 

207 
Reality  of  Thracian  Dionysus-world, 

9° 

Reason,  inapplicable  to  the  gods, 
146  ;  difficulties  of,  in  Dionysus- 
legend,  148 

Religion,  primitive,  relation  to  Greek 
myths  of,  27  ;  Cyprus  a  meeting- 
ground  of  many  a,  27 7 

(Greek)  relentlessness  of  early 

phases  of,  35  ;  is  the  poetry  of 
poetry,  jj  ;  was  unshaken  until 
the  Peloponnesian  war,  136 ; 
Athenian  innovations  in,  729  f.  ; 
harmony  of  Greek  medicine  and, 
226,  230 

Resurrection,  and  the  offering  of  a 
cock  to  Aesculapius,  239 

Retzinato,  Dionysus  overcome  by, 
106 

Revels,  the  night-long,  of  Dionysus, 
140 

Rhea,  Demeter's  mother,  jo,  57  ;  of 
the  eight  Eleusinian  divinities, 
/7<?,  401 ;  a  Cretan  Cybele,  143 

Rhenea,   Delos  and,  360-362,  376, 

394 
Rhodes,  Telchines  of,  97  ;  Colossus 

of,    287  f.  ;    neither  of  Cyclades 

nor  of  Sporades,  303 
Ribaldry    of    Eleusinia,     connected 

with  lacchos-worship,  214 
Right  living,    Pythagorean  rules  of, 

72  ;  Demeter's  rules  of,  71  f. 
Roman  Brumalia  and  Rosalia,  86- 

8S  (see  Rosalia) 


Roman  Church,  15,  21 

—  hall  at  Eleusis,  20  f,  203 

-    Aesculapia,     distinction     be 
tween  Greek  and,  221,  254 

Romanticists  compared  with  classi 
cists,  /jo 

Rome,  3,  j,  7  f.  ,  10,  13  ;  and 
Augustus,  worship  of,  40,  45  ; 
temples  to  the  emperor  discouraged 
at,  38  ;  Cyprus  under,  278 

Rosalia,  Roman  and  Thracian  festi 
val  of,  86-88  ;  centres  of,  in 
Macedonia  and  Thrace,  87  ;  sur 
vival  of,  87-90  ;  and  Brumalia 
compared  with  Icarian  festivals, 
112 

Rose-gardens  of  Midas  in  Thrace, 
02 

Rosettes,  history  of,  as  ornaments,  25-7 

Rotunda  (Tholos)  of  Polycletus  at 
Epidaurus,  246  f.  ,  250-254 

Rustica  work  at  Old  Paphos,  308 

SABAE,  of  Thrace  and  Phrygia,  06 
Sabazius,   a  name  for  the  primitive 

Dionysus,    86  ;    and   the   Sabae, 

06  ;  Thracian  and  Phrygian  idea 

of,  77,  103 
Sacred    lake,    the,     in   the    Delian 

legend  of  Leto,  356 
Sacred    and    secular    medicine,   dis 

tinction  between,  232,  237 
Sahara,  mountains  of  Cyprus  and  of 

the,  276 
Sailors,   votive  inscription  to  Apollo 

of,  384 
Saints,   lives  of  Christian,   and  late 

Pagan  myths,  21,  265 
Saisara,   daughter  of  Celeus  (Pam- 

phos),  6j 
Salamis,    12  ;    Mount   Acamas   on, 

187,344;  Blaeuw's  description  of, 

346   f.  ;    Cychreus  and,    64,  j./t 

f.  ;     Bocarus,    a    river   on,    324, 

j/./;    of  Cyprus,   283,  J.//-J-/6. 


Samos,  not  of  Cyclades  or  Sporades, 
303  ;  Democedes  called  by  Poly- 
crates  to,  224  ;  sack  of,  by  IVrs- 


452 


INDEX 


ians,    224 ;    Prince    Carathe'odori 
of,  268 

Samosata,  Lucian  of,  a  Voltaire,  262 
Sanctuaries,  Greek,  2  f. ,  j",  7  ;  Chris 
tian,  2,    4   i.  ;    Aphrodite   at   all 
Greek,  271 
Sappho,  invocation  of,  to  Aphrodite, 

29S 
Sardanapalus   (Assurbanipal),   poem 

on  Ishtar  found  in  palace  of,  joo 
Sarmatians,  the,  overran  Thrace  and 

Illyria,  87 

Sathalia,  tempestuous  gulf  of,  275 
Satrachus  and  Bocarus,  345  f. 
Satrae,  the  indomitable,  84 
Satyr -friend    of  mystic  maids,    88 

(see  Maenads) 
Satyr- play,  the  Icarian,  132 
Satyrs,  half  beasts  and  half  men,  78 
—    and    Sileni,    companions    of 

Dionysus,  94,  99,  106,  128 
Saviour-god,  Dionysus  the,  134 
Savoy,  Democedes'  career  paralleled 

at  the  court  of,  224 
Science  has  its  place  in  nature  of 

Chiron,  232 
compatible  with  superstition  in 

worship  of  Aesculapius,  234 
and  religion,   harmony  of,    in 

Greek  medicine,  226,  238 
—  debt  to  Homeric  fighting  of, 

229 

Sculpture,  skill  of  Pygmalion  in,  294 
Scythian   superstition,    Hippocrates 

on  a  gross,  238 
Sea,  the  sound  of,  in  forests,  6  ;  part 

in  Greek  worship  played  by,  210  ; 

power  of,  to  purify,  210  ;  Diony 
sus  leaps  into,  77 
Seasons  attend  Dionysus,  78,  96 
Secular   and   sacred   medicine,    dis 
tinction  between,  232,  237 
Seefeld,  library  of,  looted  by  Charles 

X.  of  Sweden,  350 
o"/7/cos,  6  /j.vffTiK6s,  fSg 
Selinus,     pierced    stones    at,     and 

piercings  at  Old  Paphos,  306 
Semachidae,    Dionysus  -  legend     of, 

105,  119 


Semachus,      legend      of     Dionysus 

coming  to  visit,  105,  119 
Semele,  Theban  place  of  death  of, 

140  ;     Dionysus    and,     140-142, 

163,    7-77,    181  ;    Pentheus   and, 

148,  154  ;  Dione  and,  318 
Semitic,    Aphrodite's  decisive  traits 

are,  272  f. ,  280 

Senate  decrees   honours    to    Julius 
£  Caesar,  37,  40 
Serapis,   3  ;    priests  of  Aesculapius 

and,    254 ;    Delian    Apollo    and, 

402 

Seriphos  on  the  way  to  Delos,  376 
Serpent  impostures  of  Alexander  of 

Abonotichus,   254 ;  Zeus  in  form 

of,  181 
Serpents,  part  played  by,  in  miracles 

of  Aesculapius,  235,  253  f. 
Sestrachus  or  Satrachus,  a  river  at 

Old  Paphos,  346 
Severus,  Alexander,  Julia  Domna's 

great-nephew,  265 
Sheep-folds   are   often    in    Cypriote 

caves,  289 
Sicilian  life,  Aphrodite  at  centres  of, 

271* 

Sicily,    Greeks  in,  worshipped    De- 
meter,    51  ;    myth  of  Proserpina 

localised  in,  62 
Sicyon,  connection  of,  with  tragedy 

overstated,  115 
Sidon,  Cadmus  of,   and   Pentheus, 

147 ;     Aphrodite    Ashtaroth    at, 

303  ;  Cadmus  of,  brings  eastern 

tinge      into     Theban      Dionysus 

story,  164  ;  sent  gifts  to  Delian 

Apollo,  387  ;   early  intercourse  of, 

with  Cyprus,  281 
Sight,  use  of  terms  of,  to  describe 

the  Mysteries,  209 
Silence,     at      Eleusis,      180  -  182 ; 

Pausanias  warned  to  keep,  about 

Eleusinia,  218 
Sileni,  innumerable,  and  Satyrs  the 

mates  of  Dionysus,  94,  97 
Silenus,  old,  the  type  of  things  that 

flow,  78,  93-93>  98 
Silvius,  Aeneas,  326  f. 


INDEX 


453 


Simon  Magus,  Apollonius  ami   Paul 

from  the  same  quarter,  260 
Sisyphus,   shadowy  punishment  of, 

76 

Slave-market,  Delos  a,  375,  401 
Snow,  revellers  above  Delphi  blocked 

by,  nj 

Socrates,  loftier  teachings  of,  in  The 
Bacchanals,  138  ;  meaning  of  his 
dying  words,  239  ;  might  not  die 
while  Athens  was  consecrate  to 
Delian  Apollo,  362 
Solomon's  temple  and  the  Paphian 

ruin,  306,  313  f. 

Solon  and  Epimenides,  122  f.,  175 
—  and   the  tragedy  of  Thespis, 
116,  123,  126 
Son  of  God,  Aesculapius  described 

as  the,  240 

Song  and  dance  in  Dionysus-wor 
ship,  102-104 

Sophism,  the,  of  Sophocles,  166  f.  ; 
the,  of  Tiresias  in  The  Baccha 
nals,  149 

Sophocles,  the  Tiresias  of,  145, 
148*. 

Dionysus     the    tutelary    god 

of,  81 

—  sophism  of,  166  f.  ;  Icarian 
notion  of  Dionysus  simpler 
than  his,  in 

Sorcery,    Apuleius    in   his  own  de 
fence  on  the  charge  of,  262 
Soroe,  Meursius  at,  348  f. 
Sosandra     Aphrodite      at     Athens, 

271,  208,  300 
Sotera,    Athene,     and      Pisistratus, 

126 
Soudan,    Nysa   in   the   Troglodytic 

country  beyond  the,  165 
Soumali-country,   is   the  country  of 

the  cinnamon,  165 
Sparta,  interference  at  Eleusis  under 

Cleomenes,  216 
Spenser,     anatomy    of,     compared 

with  Homer's,  228 
Sphekia,  a  name  of  Cyprus,  j./j  f. 
Sphettus,    prayer  of  Diophantus  of 
Sphettus,  241 


Spintharus,  of  Corinth,  built  Ivl- 
phian  temple,  31 

Sporades,  the,  30 1 -398 

Spring,  celebration  of,  return  of 
Dionysus  in,  112  f.  ;  gibes  in 
celebration  of,  213 

Stadion  at  the  Apollonia  and  Delia, 
388 

Stage,  Athenian,  built  by  Nero  and 
by  Phaedrus,  //./ 

Stars,  fawn  -  skins  symbolise  the 
heaven  flecked  with,  140 

Statius,  linked  Icaria  and  Eleusis, 
707  ;  tells  of  Erigone's  wander 
ings,  TI4 

Stavrovuni,  Mount,  Santa  Croce,  or 
Delia  Croce,  325 

Stockholm,  MSS.  of  Meursius  used 
by  Graevius,  j>j/ 

Stoicism,  a  refuge,  41 

Stoics,  practically  believed  in  many 
gods,  20 

Strabo,  antiquity  of  Meineke's  mis- 
punctuation  of,  xiv.  p.  683,  325- 

327 

Strymon,  of  Orphic  fame,  and 
1'ieris,  87  ;  Pisistratus  in  exile 
near  the,  13  j 

Styx,  mother  of  Persephone,  79 

Sublime,  in  Euripides,  Gothe  on  the, 
162 

Suicide,  epidemic  -  mania  for,  of 
Icarian  maidens,  no  ;  epidemics 
of,  in  modern  times,  no 

Sultan,  Colossi,  appropriated  by  the, 
288 

Sun-god,  the  source  of  knowledge  of 
miraculous  drugs  is  the,  231 

Sunium,  on  way  to  Delos  from 
Athens,  376 

Surgeon,  Democedes  qualified  at 
Croton  as  a,  225 

Surgery,  Homeric  skill  in,  came 
down  to  professional  doctors, 
224 ;  necessity  of  warfare  tin- 
mother  of  invention  of,  226  ;  at 
Epidaurus,  237  ;  knowing-  of. 
upon  which  Hippocrates  dn-w. 
•  223  ;  sprang  from  positive  practi- 


2   G    2 


454 


INDEX 


cal  tendency  in  early  medicine, 
2JS,  237 

Surgical  operations  inspired  by 
Aesculapius,  226,  237 

Susarion,  comedy,  his  great  in 
vention,  ii 4  f. ,  7/7 

Suttee,  Thracian,  custom  analogous 
to,  84 

Swans,  song  of,    at  Apollo's  birth, 

3S9, 

Sybaris,  inscription  from  a  tomb  at, 

176 
Syra  has  one  part  of  ancient  Delian 

glory,  j><?9 

—  Aphrodite  drifted    to    Paphos 

from,  296 

Syrian  traits  in  Aphrodite,  297 
goddess,    worship    on    Delos, 

of  the,  403 
Syriote  peasant  dances  in  carnival, 

388 

Syros  and  Delos,  365,  376 
Syrtos,   the  modern   peasant -dance 

called  the,  400 

TABLE  mountains,  of  Cyprus   and 

the  Sahara,  276 
Taboos,  account  of,   in  the  Golden 

Bough,  360 
Tacitus,  reverence  of,  for  the  imperial 

idea,  41  f. 
Tammuz-  Adonis,     plays     part     of 

Dionysus    and  Persephone,   291  ; 

lament    for,    300    f.  ;     temple    at 

Amathus  of,  292 
Tantalus,  Pindar  remoulds  the  storv 

of,  r8 ;   punishment  of,  76 
Tar  in  Cypriote  wine,  288 
Tarsus,    Apollonius    removed    from 

Aegae  to,  257 
Taurus   mountains   of  Asia    Minor, 

northern  range  of  Cyprus  parallel 

to,  258,  275 
Taxes,  the,  levied  to-day  in  Cyprus, 

277 
Tegea,«  temple  of  Paphian  goddess 

at,  28s  f- 

Telamon,  a  pupil  of  Chiron,  232 
Telchines,  97 


Telesphorus  (Convalescence)  attends 

on  Aesculapius,  240 
Telesterion,    Hall    of    Initiation    at 

Eleusis,  iSg 
Tellus    the    Athenian,     death     and 

burial  of,  at  Eleusis,  122 
Tempests,   stilled  by  a   nail   of  the 

True  Cross,  275 
Templars,  Knights,  in  Cyprus,  288, 

341 

Tenedos,  101 
Teniote  festivals  and  modern  shrine, 

375,  389 

Ten os,  Aeolus  housed  on,  380  ;  of 
Poseidon,  358 ;  Delos  and,  j6j, 

376 

Tetrapolis,  Marathonian,  worship  of 
Apollo  from,  104 

Theatres,  Aesculapian  shrines  built 
near,  247 

Theban  Maenads,  ministers  of  earth 
quake,  i4S 

—  legend  of  Dionysus  and  Eastern 
stories,  164 

Thebes  to  be  decked  as  a  Maenad, 
142  ;  the  mother  of  Bacchanals, 
133 ;  Dionysus-driven  women  of,  in 
The  Bacchanals,  139  ;  Dionysus' 
connection  with,  76,  139,  141,  144 

Thelpusa,  Demeter  legend  of,  176 

Themistocles,  // 

Theocracy  of  Roman  Empire,  8,  26 

Theologians,  Homer  and  Hesiod  the 
first  of  Greece,  26 

Theology,  mythology  and  philo 
sophy  contrasted  with,  20,  52 

Theoris,  the  Delian  boat  of  Athens, 

376  f. 

Theory     sent      by     Deliastae    from 

Athens,  376 

Thera  and  the  Cyclades,  397 
Theseum     and     temple     of    Delian 

Apollo,  387 
Theseus,    Demophoon    the    son    of, 

6j  ;  the  altar  of  Zeus  and,  217  ; 

Chiron's  pupil,  232  ;    Theoris  and, 

377  ;  Ariadne  at  Delos  and,  382  ; 
Athenian  and  Delian  legends  of, 
400 


INDEX 


455 


'I'hcsinoi  of  Deineter,  51 
Thesmophoria,  the,  51 ,  6S 
Thesmos,      the       self-imposed,      of 

Apollo,  370  f. 
Thespis,    115-117,    126,    ij2,    137, 

f39 

Thcssaly,  myth  of  Persephone  wan 
dering  from,  J9  ;  Demeter  from, 
194,  221  ;  Aesculapian  legends  of, 

22 f,   234,    2.fS  f- 

Thetis,  her  attempted  fire-baptism  of 
Achilles,  64  ;  proteeted  the  fleeing 
Dionysus,  77 

Thiasos,  100  ;  blessings  of  belonging 
to,  141 

Thief,  discovery  of  the  cross  of  the 
penitent,  275 

Thigh-mountain,  second  birth  of 
Dionysus  and,  165 

Tholos,  the,  of  Polycletus  at  Epi- 
daurus,  250-254 

Thornton,  Dr.,  competed  for  build 
ing  the  Capitol,  267 

Thrace,  Dionysus  from,  77,  86,  90, 
103,  119,  121,  174,  189,  218 ; 
Dionysus  in,  85  f. ,  88,  go,  92-04, 
06,  174  ;  Phrygia  included  in  the 
larger,  84,  oo  ;  chronic  disturb 
ances  of,  <$j  ;  Valerius  found  three 
silver  statues  and  strange  rites  in, 
87 ;  Eumolpus  from,  174,  218 ; 
Pisistratus  exiled  to,  126,  137 ; 
Zerynthian  Aphrodite  in,  345  f. 

Thracian  elements  in  the  legend 
and  worship  of  Dionysus,  So,  82- 

85,  94,  96  f.,  / 00-104,  106,  135, 
138,    162,    177 ;     Brumalia    and 
Rosalia,    86-88 ;    oracle    on    Mt. 
Zilmissus,  93  ;  places  of  assembly, 
94 ;     history    in    early    days,    So, 

III  f. ,    221 

Thracians,  history  of  the,    So,   84, 

86,  103,   Iff,     220    f.,     280,    322  ; 

character  of  the,   84-90  ;   religion 
of  the,  So  f.,  83-87,  oo,   os,  103, 
106,  322  ;    the  Delian  Hyperbor 
eans  were,  379 
Thraco-Eleusinian,    Eumolpus  the, 

122 


Thraco  -  Phrygian  features  in  all 
Dionysus-legends  from  the  East, 
169 

Thrasymedes,  statue  of  Aesculapius 
by,  249 

Thriasian  plain,  55  f. ,  187 

Thucydides,  unknown  sense  of  the 
term  Cyclades  \n,  392,394  ;  semi- 
Thracian  parentage  of,  84 

Thyades,  with  Dionysus  on  Delphian 
pediment,  31,  133  ;  Thyone- 
Semele  and  Dione,  318 

Thyone-Semele  and  Dione,  318 

Tiber,  temple  of  Aesculapius  on  the 
island  in,  2.// 

Tiberius,  42  f. 

Tintoretto  painted  Naxia-acropolis, 
328 

Tiresias,  145-148,  166  f. 

Titans,  Zagreus  and  the,  128,  131, 
i8r 

Tithorea,  peak  of  Parnassus,  25 

Titian,  a  masterpiece  of,  289 

Titthion,  Aesculapius  exposed  on 
mount,  243,  245  f. 

Tityi,  94 

Tmolus,  Dionysus  born  on,  154 

Tolerance,  Apollo's  sense  of,  8,  29, 33 

Tragedy,  Icarian  legends  and,  ///, 
115-117  ;  rise  of  Athenian,  132, 
134-136,  159,  172;  Dionysus  and, 
7S 

Tree-worship,  78,  82  f.,  106  f.,  no, 
143,  159,  164 

Triaconter,  the  Theoris  a,  ^77 

Tricca,  Inscriptions  at,  237 

Trinity,  mystery  of,  compared  with 
Eleusinian  mystery,  178 

Triopian  promontory,  Cnidian  sanct 
uary  on,  70 

Tripod,  371 ;  a  symbol  of  unison  of 
Apollo  and  Dionysus,  35  f. 

Tripoli,  the  country  of,  in  the  sack 
of  Nicosia,  332 

Triptolemus,  a  son  of  Celeus,  63  ;  a 
son  of  Icarius,  707  ;  Icarius  and, 
707  ;  an  EU-usiniun  demigod,  6j 
f.  ;  suppression  of,  in  Eleusinian 
legend,  64-66,  175,  194  ',  Rarian 


456 


INDEX 


plain  and,  66,  71,  201  ;  Demeter's 
representative,  67,  124,  206  ;  one 
of  the  gods  at  Eleusis,  178  f. 

Troezen,  meaning  of  Bocarus  in 
dialect  of,  344 

Troglodytic    country,   Nysa   in  the, 

it>5 

Trohodos,  used  by  the  villani  for 
Olympus,  341 

Troodos,   276,  324  ;  tradition  iden 
tifying  Olympus  with,  327-343 
—  Strabo's  epithet    for  Olympus 
exactly  suits,  323 

Trophonius,  Parmeniscus  and  the 
oracle  of,  382 

Trullo,  council  of,  order  against  the 
Rosalia  at  the,  89 

Truthfulness  of  Apollo  conspicuous 
at  Delphi,  370 

Trygaeus,  had  to  be  initiated  before 
he  died,  181 

TVfjij3opvxos,  epithet  of  Aphrodite,  310 

Turk,  mimic  fight  of,  with  a  Chris 
tian,  90 

Turkish  bondholders  and  modern 
Cyprus,  278 

Turks  besiege  and  sack  Nicosia, 
332-334 ;  Brenzone's  hatred  of, 
332  ;  ransoms  of  Christians  from, 
jjj  f. ,  339  ;  their  laws  and  taxes 
in  Cyprus,  277  ;  population  of, 
in  modern  Cyprus,  272  ;  special 
causes  for  prominence  of  Greek 
doctors  under  the,  267 

Tyana,  saved  from  destruction  by 
Apollonius,  265 

Tyre,  Aphrodite-Ashtaroth  at,  303 

Tyre  and  Sidon,  early  intercourse 
of,  with  Cyprus,  281  ;  sent  gifts 
to  Delos,  381 

uXdrr;?,  epithet  of  Apollo  at  Curium, 

345  *• 

Ulpian,  of  Phoenician  descent  and 
of  Julia  Domna's  clique,  259 

Ulrichs,  Professor,  votive  inscription 
found  at  Delos  by,  384 

Unction  (extreme),  initiation  into 
the  Mysteries  compared  to,  181 


Unity,  of  Demeter  and  Persephone 
is  unity  of  growth  at  large,  74  ; 
of  god -doctrine  of  Xenophanes 
and  of  the  Mysteries,  180 

Urania,  transformation  in  meaning 
of  epithet,  297  f. 

Urugal,  Ishtar's  descent  to,  301 

VALERIUS,  in  Thrace,  87 
Varoschia,  Salamis-Famagosta  and, 

283 

Vegetation,  Dionysus  god  of  abun 
dant,  78 

Venetian  rule  in  Cyprus,  278,  289 
Venetians    and    marriage    of    doge 

with  the  sea,  131 
Venice   and    Greeks,    religions    of, 

131,  135,  328,  389 
Venus,  87,  300,  304 
Vespasian,  reported  miracle  of,  41  f. ; 

the  greatest  latter-day  guide,  42 
Viaticum,  181 
Victory,   statues   of,   on   Epidaurian 

temple,  249 
Vilaras   of  Janina,    a   poet   and    a 

doctor,  268 

Vindemiator,  Bootes-Icarius  as,  in 
Vine,    the   he-goat   and   the,    108  ; 

cultivation  of  the,  and  Dionysus, 

107 
Vintage,  autumn  Dionysus  festivals 

of  the,  129  f. 
Virgil,    Messianic    vision    of,    138  ; 

supposed  bust  of,  193  ;  the  mother 

Venus  of  his  song,  304 
Virgin  Mary,  the,  4,  71 
Virgo,  bright  star  e  near  wrist  of,  is 

frovindemiator,      in ;      Erigone 

and  the,  no 
Volcanic  origin  of  Delos,  jjg 

WARFARE  in  Homer,  debt  of 
modern  science  to,  228  f. 

Water,  given  with  barley  to  Deme 
ter,  68 ;  wine  tempered  with,  at 
Eleutherae,  34,  120  ;  in  Hera- 
clitus'  doctrine  and  Dionysus  - 
worship,  179  f.  ;  in  legend  and 
worship  of  Dionysus  and  his 


INDEX 


457 


creatures,   78,  94  f.  ,  102  f.,  108, 

f37'  '56>  JJ9 

\\'ine,  regarded  as  an  element,  97 
f.  ,  103  ;  first  culture  of,  /6jr  ; 
Phoenician  trade  in,  and  Diony 
sus,  j6j  ;  brought  by  Dionysus 
to  Icaria,  78,  107-109,  152  f.  ; 
power  over  the  dead  of,  97  ;  pro 
phetic  power  of,  oj  ;  represents 
the  power  of  Dionysus,  97  ; 
Eleutherae  and  the  use  of,  34, 
1  20  ;  power  over  poets  of,  102  ; 
the  story  of  Midas  and  the,  oj  ; 
Cypriote,  288  ;  Pramnian  given 
by  Maron,  97 

Winged  Dionysus,  the,  779 

Winter,  death  of  Dionysus,  grief 
of  Demeter  in,  iSj  ;  Icarian 
observances  in,  112  f.  ;  other 
Dionysiac  festivals  in,  132,  156 

Winter-oracles  of  Apollo  at  Patara, 

370 

Wodin,  6 
Wood,  Pentheus,  like  a  king  of  the, 


XANTHIAS   and   Dionysus,    witness 

the  march  of  the  Mystae,  212 
Xenagoras,  named  Cyprus  Cerastis, 

Aspelia,  and  Amathusia,  274 
Xenocles    of  Cholargia,    builder  of 

Eleusinian  oiralov,  797 
Xenophanes,  Demeter  stands  for  the 

idea  of  divinity  of,  779  f. 
Xenophon       nearly     contemporary 

with  Hippocrates,   225  ;    military 


medicine  and,    225  ;    Philostratus 

imitated  passages  from,  259 
Xerxes,   Onomacritus   and    Pisistra- 

tidae  at  the  court  of,  127 
£6ava,  attributed  to  Daedalus,  J99 
Xypeta,    Metagenes   of,   at  Eleusis, 

797 

ZAGREUS,  the  myth  of,  jo,  127  f.  ; 
at  the  Anthesteria,  7j7  ;  the 
mystical  Sal/uaov  of  Eleusis  and, 
77^  ;  the  doctrine  of  immortality 
and,  181  ;  the  pitiless  huntsman, 
185 

Zamacola,  on  the  couvade,  770 

Zamolxis,  a  name  for  the  primitive 
Dionysus,  86 

Zerynthian,  epithet  of  Aphrodite 
from  Thrace,  j/j  f. 

Zeus,  character  of,  12-14,  2°>  2#> 
2f<?  f.,  2j6,  264,  277,  j6g ; 
monuments  connected  with  the 
worship  of,  9  f. ,  14-17,  277,  252, 
402  ;  Aphrodite-legend  and,  277 
f. ,  284,  J7J  f.  ;  Asterinand,  358; 
Aesculapius  and,  2fo,  239  ; 
Demeter-legend  and,  48,  jo,  36 
f.  ;  Delphi  and  the  eagles  of,  31  ; 
Dionysus  legend  and,  22,  142  f. , 
163  ',  Zagreus-Dionysus  and,  128 
181 

Zilmissus,  oracle  on  Mount,  oj 

Zoroaster  and  Apollonius  as  magi 
cians,  262 

Zoster,  birth-legend  of  Apollo  and 
Artemis  transferred  to,  399 


THE  END 


Printed  by  R.  &  R.  CI.ARK, 


MACMILLAN  AND  CO.'S  PUBLICATIONS. 

WITH  ILLUSTRATIONS  AND  PLANS.       CROWN  SVO.      j6s. 

MYTHOLOGY  AND  MONUMENTS  OF  ANCIENT  A  Till 

Being  a  Translation  of  a  portion  of  the  "  Attica  "  of  Pausanias.      P.y 
MARGARET    DE    G.    VKRKALL.       With    Introductory    Kssay   and 
Archaeological  Commentary    by  JANE     E.    HARRISON,    Author    of 
"  Myths  of  the  Odyssey,"  "  Introductory  Studies  in  Greek  Art." 
***  A   translation,  by   Mrs.    Verrall,  of  the  account   of  Athens   and    Attira    in 
Pausanias'  description  of  Greece.     In  a  certain  sense  Miss  Harrison  supplies  a  Com 
mentary  on   Pausanias ;    but  her  primary  object  is  to  elucidate  the  mythology  of 
Athens,  and  with  this  intent  she  has  examined  the  monuments,  taking  Pausanias  as 
her  guide.     It  is  hoped  that  the  book  will  be  found  useful  by  students  at  home  as 
well  as  by  those  who  have  opportunities  of  seeing  the  monuments  for  themselves. 

OBSERVER.—11  Miss  Harrison  is  an  indefatigable  worker,  and  her  volume  is  a 
monument  of  laborious  and  wary  investigation.  It  really  Consists  of  two  distiiH  t 
portions,  though  united  by  a  common  object,  namely,  the  elucidation  of  the  mythology 
of  Athens.  One  of  these  divisions  may  be  briefly  characterised  as  a  comparison  of 
the  'city  of  the  violet  crown'  as  it  now  is,  with  the  city  as  described  in  Pausani.i-. 
...  To  the  scholar  who  visits,  or  has  visited,  or  is  about  to  visit  Athens,  tins  part  of 
her  work  is  specially  delightful ;  but  her  essay  on  Athenian  local  cults  is  perhaps  her 
most  valuable  contribution  to  the  general  science  of  Mythology." 

SPECTA  TOR. — "A  delightful  commentary  dealing  with  all  the  main  points  of 
Athenian  topography,  and  explaining  with  admirable  clearness  the  results  of  the 
Excavations  made,  and  still  being  made,  by  the  Greek  Government.  .  .  .  There  is 
no  English  book  which  can  be  compared  with  the  present  as  summarising  the  latest 
facts  and  views  of  Athenian  Archaeology." 

CAMBRIDGE  REVIEW.— "  In  every  way  satisfactory^  the  style  clear  and 
concise  ;  the  wording  accurately  representing  the  original  text." 

ACADEMY. — "Throughout  the  volume  there  is  abundant  evidence  of  care  ami 
skill,  as  well  as  of  wide  reading.  Miss  Harrison  has  secured  much  valuable  help, 
and  has  duly  acknowledged  it.  She  knows  her  subject  thoroughly,  and  also  knows 
how  to  place  it  before  the  reader  clearly  and  well." 

SPEAKER. "  Every  page  is  well  worth  careful  study.     It  puts  for  the  first  time 

within  reach  of  English  readers  the  Classical  Guide,  improved,  corrected,  and  illus 
trated  by  a  flood  of  modern  excavation  and  research." 

CLASSICAL  REVIEW,  November.—"  For  some  time  to  come  this  book  will 
be  indispensable  to  every  English-speaking  student  of  Athenian  Antiquities." 

SHORTLY.      8VO.      CLOTH. 

DR  SCHLIEMANN'S  EXCAVATIONS  AT  TROY,  TIRYNS, 
MYCENsE,  ORCHOMENOS,  ITHACA,  presented  in  the  light 
of  recent  knowledge.  By  Dr.  CARL  SHUCHHARDT.  Authorised 
Translation  by  Miss  EUGENIE  SELLERS.  With  Appendix  on  latest 
Researches  by  Drs.  SCHLIEMANN  and  DoRPFELU,  and  Introduction 
by  WALTER  LEAF,  Litt.  D.  Illustrated  with  Two  Portraits,  Maps, 
Plans,  and  290  Woodcuts. 

*  *  A  popular  but  strictly  accurate  account  of  the  whole  series  of  Dr.  Schlier 
Excavations  and  their  results.     It  is  believed  that  the  book  will  be  of  great  value  and 
interest  even  to  those  who  already  possess  Dr.  Schliemann  s  larger  works,  not 
bringing  them  up  to  date,  but  also  in  showing  the  connecting  links  1 
discoveries. 

8VO.      CLOTH. 

THE  JOURNAL  OF  HELLENIC  STUDIES.—  Vol.  I.  With 
Plates  of  Illustrations.  3os.-Vol.  II.  3os.  With  Plates  of 
Illustrations.  Or  in  2  Parts,  i5s.  each.-Vol.  Ill  2  I 
With  Plates  of  Illustrations.  155.  each.— Vol.  IV.  2  Parts  With 
Plates.  Part  I.  155.  Part  II.  2is.  Or  complete,  3os.—\  ol.  \  . 
With  Plates.  3os.-Vol.  VI.  With  Plates.  Parti.  155.  Pjjt 
II.  i5s.  Or  complete,  3os.— Vol.  VII.  Parti.  i5s. 

MACMILLAN  AND  CO.,   LONDON. 


153.      Or  complete,  305. — Vol.  VIII.      Parti.      155.      Part  II.      155. 

— Vol.  IX.     2  Parts.      155.  each. — Vol.  X.      305.— Vol.  XI.      Part 

I.      153.        Net. 

***  The  Journal  will  be  sold  at  a  reduced  price  to  Libraries  wishing  to  subscribe, 
but  official  application  must  in  each  case  be  made  to  the  Council.  Information  on 
this  point,  and  upon  the  conditions  of  Membership,  may  be  obtained  on  application 
to  the  Hon.  Sec.,  Mr.  George  Macmillan,  29  Bedford  Street,  Covent  Garden. 


MACMILLAN'S  CLASSICAL  TRANSLATIONS. 

ARISTOTLE— THE  POLITICS.      By  Rev.  J.  .E.  C.  WELLDON, 

M.A.     IQS.  6d. 

THE  RHETORIC.     By  the  same.     7s.  6d. 
THE  ETHICS. .    By  the  same.  [In  preparation. 

CICERO— SELECT    LETTERS.        Translated      from     Watson's 

Edition.     By  Rev.  G.  E.  Jeans,  M.A.     IDS.  6d. 
ACADEMICS.     Translated  by  J.  S.  REID,  M.L.     ss.  6d. 

HOMER— ODYSSEY.     By  Professor  S.  H.  BUTCHER,  M.A.,  and 

A.  LANG,  M.A.     6s. 
ILIAD.     By  A.  LANG,  M.A.,  WALTER  LEAF,  Litt.D.,  and  ERNEST  MYERS. 

THE  ODYSSEY.     BOOKS  I. -XII.     By' the  EARI.  OF  CARNARVON.     8s.  6d. 

HORACE.     By  J.  LONSDALE,  M.A.,  and  S.  LEE,  M.A.     35.  6d. 
HERODOTUS— THE  HISTORY.     By  G.  C.  MACAULAY.     2  vols. 

1 8s. 

JUVENAL  — THIRTEEN   SATIRES.      By   A.    LEEPER,    M.A. 

LIVY.    BOOKS  XXI.-XXV.      By  Rev.  A.  J.  CHURCH,  M.A.,  and 

W.  J.  BRODRIBB,  M.A.     7s.  6d. 
LONGINUS—  ON    THE    SUBLIME.      Translated     by    H.     L. 

HAVELL,  B.A.     With  Introduction  by  ANDREW  LANG.     Crown  8vo.     43.  6d. 
MELEAGER:    FIFTY  POEMS   OF.       Translated    by  WALTER 

HEADLAM.     Fcap.  410.     75.  6d. 
PINDAR— ODES.    By  ERNEST  MYERS,  M.A.     Second  Edition.    5s. 

PLATO— REPUBLIC.        By  J.    LL.   DAVIES,    M.A.,    and    D.   J. 

VAUGHAN,  M.A.     45.  6d. 
EUTHYPHRO,  APOLOGY,  CRITO,  AND  PH^EDO.     By  F.  J.  CHURCH. 

PHJBDRUS,  LYSIS,  AND  PROTAGORAS.      By  Rev.  J.   WRIGHT,  M.A. 
45.  6d. 

POLYBIUS—THE    HISTORIES.       By    E.    S.    SHUCKBURGH. 

2  VOls.       24S. 

SALL  UST—  CA  TILINE  AND  JUG  UR  THA.    By  A.  W.  POLLARD, 

B.A.     6s.     The  Catiline,  35. 

TACITUS.     By  A.  J.  CHURCH,  M.A.,  and  W.  J.  BRODRIBB,  M.A. 

History,  6s.     Annals,  75.  6d.     Agricola  and  Germania.     45.  6d. 

THEOCRITUS,  BION,  AND  MOSCHUS.     By  A.  LANG,  M.A. 

4s.  6d. 

VIRGIL.      By  J.    LONSDALE,  M.A.,   and   S.   LEE,   M.A.      35.  6d. 

THE  ^ENEID.     By  J.  W.  MACKAIL,  M.A.     7s.  6d. 

XENOPHON—  COMPLETE  WORKS.     By  H.  G.  DAKYNS,  M.A. 

With  Introduction  and  Essays.    4  vols.     Vol.  I.,  containing     "The  Anabasis" 
and  "The  Hellenica."     xos.  6d. 

[Vol.  II.  in  the  Press. 

MACMILLAN    AND    CO.,    LONDON. 


BINDING  SECT 

>ECT-     JUl  ?  01980 


PLEASE  DO  NOT  REMOVE 
CARDS  OR  SUPS  FROM  THIS  POCKET 

UNIVERSITY  OF  TORONTO  LIBRARY 


Dyer,  Louis 

Studies  of  the  Gods 
in  Greece.