Skip to main content

Full text of "Art in Egypt"

See other formats


ART  IN  EGYPT 


ARS  UNA:   SPECIES  MILLE 

GENERAL  HISTORY  OF  ART 


Uniform  with  this  Volume 

Already  Published :  — 

THE    HISTORY    OF    ART    IN    GREAT 
BRITAIN  AND   IRELAND. 

By  Sir  Walter  Armstrong. 

THE  HISTORY  OF  ART  IN  FRANCE. 
By  Louis  Hourticq. 

THE  HISTORY  OF  ART  IN  FLANDERS. 

By  Max  Rooses. 

(Director  of  Plantin  Moretus  Museum, 
Antwerp.) 

THE  HISTORY  OF  ART  IN  EGYPT. 

By  G.  Maspero. 
(Director  of  Ghizeh  Museum.) 

THE  HISTORY  OF  ART  IN  NORTHERN 
ITALY. 

By  Corrado  Ricci. 

(Director  General  of  Fine  Arts  and 

Antiquities  of  Italy.) 

THE  HISTORY  OF  ART  IN  SPAIN  AND 
PORTUGAL. 

By  Marcel  Dieulafoy. 

In  Preparation :  — 
BYZANTINE   ART. 
THE  ART  OF  INDIA. 
GERMAN  ART. 
THE  ART  OF  GREECE. 
ART  IN  HOLLAND. 
THE  ART  OF  CHINA  AND  JAPAN. 
ART  IN  NORTH   AMERICA. 
ROMAN  ART. 
THE  ART  OF  SOUTHERN  ITALY. 


UNA:      PECIES   MILLE 
GENERAL,  HISTORY  OF  ART 


BY 


GfTviASPERO 


MEMBER   OF  THE  INSTITUTE 
DIRECTOR   GENERAL   OF  THE   SERVICE   OF  ANTIQUITIES   OF  EGYPT 


fr¥^:# 


LONDON 
WILLIAM    HEINEMANN 

MCMXXI 


This  -volume  is  published  simultaneously  in 
America  by  CHARLES  SCRIBNER'S  SONS,  New 
York;  in  England  by  WILLIAM  HEINEMANN, 
London;  also  in  French  by  HACHETTE  ET  ClE., 
Paris;  in  German  by  JULIUS  HOFFMANN, 
Stutigart;  in  Italian  by  the  ISTITUTO  iTALIANO 
D'  ARTI  GRAFICHE,  Bergamo  ;  in  Spanish  by  the 
LIBRERIA  GUTENBERG  DE  JOSE  Ruiz,  Madrid. 


Printed  by  the  Scribner  Press 
New  York,  U.  S.  A. 


ti 


PREFACE 

THE  art  of  Egypt,  like  all  the  rest  of  its  civilisation,  is  the  pro- 
duct of  the  African  soil.  If  it  occasionally  took  ideas  or  methods 
of  expression  from  the  peoples  with  which  it  was  brought  into 
contact  by  the  Pharaohs,  its  levies  were  not  important  enough 
in  the  first  instance  to  exercise  a  durable  influence  upon  its 
constitution.  In  a  few  years,  it  had  so  completely  assimilated 
the  substance  of  these  as  to  leave  us  hardly  sensible  that  any 
alien  influences  had  interfered,  even  momentarily,  with  its  ho- 
mogeneity. It  was  only  towards  the  end,  when  the  race  whose 
mind  it  had  so  admirably  materialised  and  translated  began  to 
bow  beneath  the  weight  of  over  fifty  centuries  of  existence,  that 
it  too  declined,  lacking  strength  to  defend  its  superannuated 
traditions  successfully  against  the  new  conceptions  of  beauty  set 
before  it  by  younger  races. 

I  have  dealt  briefly  with  the  period  of  its  infancy,  thinking 
the  reader  would  forgive  me,  if  I  devoted  as  little  as  possible 
of  the  limited  space  at  my  disposal  to  dubious  origins.  Besides, 
unlike  some  of  my  confreres,  I  cannot  accept  as  art  the  rude 
images  by  which  every  new-born  people  seeks  to  reproduce  the 
objects  or  the  beings  it  sees  around  it,  and  the  ideas  they  evoke. 
I  have  therefore  dealt  with  it  as  almost  adult,  when  it  had  left 


PREFACE 

its  awkward  age  behind  it.  I  have  indicated  the  religious  and 
social  principles  by  which  it  was  governed  at  the  time  of  the 
Thinite  dynasties,  and  I  have  then  tried  to  determine  the  suc- 
cessive stages  of  its  development  during  the  following  periods. 
It  has  been  very  erroneously  supposed  that  it  presented  a  per- 
fect uniformity  from  beginning  to  end,  and  that  its  character  was 
identical  in  all  the  different  regions  of  the  land,  save  for  certain 
differences  of  handling  which  were  the  results  of  degrees  of  skill 
in  its  artists.  I  have  shown  how,  while  drawing  everywhere 
upon  a  common  fund  of  general  ideas,  it  had  so  far  varied  their 
manifestations  in  different  districts,  as  to  give  birth  to  indepen- 
dent schools,  the  activity  of  which  augmented  or  relaxed  accord- 
ing to  the  varying  fortunes  of  the  cities  which  were  their  homes: 
the  Thinite  School,  the  Memphite  School,  the  Hermopolitan 
School,  the  Tanite  School,  the  secondary  schools  of  the  Said  or 
the  Delta.  The  list  is  as  yet  imperfectly  established,  and  I  have 
not  been  able  to  make  it  complete  in  architecture,  for  lack  of 
a  sufficient  number  of  provincial  documents,  but  it  is  fairly 
satisfactory  as  regards  painting  and  sculpture.  Not  only  is  it 
possible  to  determine  their  principal  characteristics,  but  there 
are  some  among  them,  the  Theban  School,  for  instance,  whose 
fortunes  may  be  easily  followed  from  the  rise  of  the  Eleventh 
Dynasty  to  the  rule  of  the  Caesars.  Its  relics  are  to  be  counted 
by  thousands,  and  each  day  is  marked  by  discoveries  which 
enable  us  to  fill  in  the  lacunae  of  our  science  in  this  connection, 
and  to  give  additional  precision  to  what  we  already  know. 

I  have  put  a  good  deal  of  my  own  into  this  little  volume, 
and  much  that  I  owe  to  others.  We  may  set  aside  those  who 
wrote  when  Egypt  was  a  sealed  book  to  the  modern  world. 
Although  Champollion  had  from  the  outset  very  sound  ideas 
on  the  nature  of  Egyptian  art,  and  thoroughly  appreciated  its 
fine  qualities ,  the  admirable  Emmanuel  de  Rouge  was  the 
first  to  define  its  characteristics  scientifically  and  to  sketch  its 
history,  in  his  Notice  of  1854  on  the  monuments  of  the  Louvre 
(Oeuvres  diverges,  vol.  Ill,  p.  36 — 40)  and  in  his  report  of  1854 

vi 


PREFACE 

(Ibid.  vol.  II,  p.  213—246).  Mariette,  when  in  1864  he  compiled 
his  Catalogue  of  the  Boulak  Museum,  pointed  out  many  traits 
which  his  master  had  not  noted,  but  the  rest  of  the  school,  ab- 
sorbed in  deciphering  texts,  did  not  follow  on  the  path  thus  opened, 
and  for  many  years  only  amateurs  and  classical  archaeologists 
made  their  way  along  it:  Charles  Blanc  (Voyage  dans  la  Haute 
Egypte,  8vo,  Paris,  1870,  and  Grammaire  des  Arts  du  Dessin,  6th 
ed.,  Paris,  1886),  Comte  du  Barry  de  Mervel  (Etudes  sur  V Architec- 
ture Egyptienne,  8vo,  Paris,  1873),  Soldi  (La  Sculpture  Egyptienne, 
8vo,  Paris,  1876),  Marchandon  de  la  Faye  (Histoire  de  I'ArtEgyptien 
d'apres  les  Monuments,  4to,  Paris,  1878),  Perrot-Chipiez  (Histoire 
de  I' Art  dans  VAntiquite,  I' Egypte,  8vo,  Paris,  1880),  Goodyear 
(Grammar  of  the  Lotus,  4 to,  New  York,  1891),  Lubke-Semrau 
(Die  Kunst  des  Altertums,  14th  ed.,  8vo,  Esslingen,  1908),  Springer- 
Michaelis  (Handbuch  der  Kunstgeschichte,  9th.  ed.,  8vo,  vol.  I, 
Leipzig,  1911),  Sybel  (Weltgeschichte  der  Kunst  im  Altertum, 
2nd  ed.,  8vo,  Marburg,  1903),  Woermann  (Geschichte  der  Kunst 
aller  Zeiten  und  Volker,  8vo,  vol.  I,  Leipzig,  1904).  Finally,  how- 
ever, the  Egyptologists  took  the  field,  Maspero  in  Rayet's 
Monuments  de  I' Art  antique,  fol.  vol.  I,  1879 — 1883,  and  Arche- 
ologie  Egyptienne,  8vo,  Paris,  1887,  Flinders  Petrie  in  Egyptian 
Decorative  Art,  8vo,  London,  1895,  Georges  Foucart  in  Histoire 
de  I'ordre  lotiforme,  8vo,  Paris,  1897.  Naville  insisted  on  the 
part  played  by  utilitarian  considerations  in  the  formation  of 
sculptural  and  architectural  types  (L'ArtEgyptien,  8vo,  Paris,  1907), 
while  Bissing  condensed  the  results  of  his  prolonged  studies  in 
a  manual  (Einfiihrung  in  die  Geschichte  der  Agyptischen  Kunst 
von  den  altesten  Zeiten  bis  auf  die  Romer,  8vo,  Berlin,  1908). 
In  the  interval  Steindorff  had  prepared  for  the  curious  who  visit 
the  Nile  each  year  a  substantial  resume  of  our  knowledge  on 
the  subject  (Baedeker's  Handbook,  Egypt  and  the  Sudan,  Leip- 
zig, 1908)  and  Spiegelberg,  reviving  a  theory  propounded  by 
Rouge  and  Mariette ,  sought  to  demonstrate  the  existence  of  a 
popular  art  less  rigid  than  the  official  art  (Geschichte  der  Agyp- 
tischen Kunst,  8vo,  Leipzig,  1903).  As  a  fact,  there  was  never 

vii 


PREFACE 

any  distinction  between  the  two,  but  the  same  artists  were  allow- 
ed more  or  less  liberty  according  to  the  nature  of  the  subjects 
they  treated   and   the  social  condition   of  those  they  portrayed. 
1  have  tried  as  far  as  possible  to  reproduce  and  to  appreciate 
only  things  I  have  myself   seen  and  handled,  and  the  good  for- 
tune which  made  me  twice  the  director  of  the  Service  of  Anti- 
quities  has  greatly  facilitated   my  task.     I  should,  however,  be 
guilty  of  ingratitude  no  less  than  of  injustice  if  I  did  not  acknow- 
ledge my  debt  to  those  collections  of  engravings  and  photographs 
in  black  and  white  and  in  colours,  in  which  so  many  of  us,  and 
I  myself   among   them,   have  been  able  to  study  monuments  to 
which  personal  access  was  denied  us.    The  architectural  drawings 
of  the  French  Commission  have  lost  little  of  their  value,  and  it 
would  be  difficult  to  over-estimate  the  works  of  Cailliaud,  Gau, 
Champollion,   and   Rosellini.     Lepsius'   Denkmaler  has   been   of 
greater  service   to   archaeologists   and   philologists    than    to   art- 
historians.   Weidenbach,  who  executed  the  plates  for  this  with  the 
help  of  his  pupils,  conceived  the  unhappy  idea  of  rendering  bas- 
reliefs  and  paintings  by  a  series  of  stereotyped  designs  which  he 
copied   and   recopied   with   slight   modifications   throughout   the 
work.     We   can   hardly  be  surprised  therefore  if  critics,  finding 
the  same  persons  treated  in  the  same  manner  from  the  Memphite 
period   to  that   of  the  Roman  domination,  accused  Egyptian  art 
of  monotony.     Prisse  d'Avennes,   though  his  drawing  has  more 
warmth   and   flexibility,   was  also  guilty  of  conventionalising  his 
models  excessively   in  his  Histoire   de  I'Art  Egyptien.     We  may 
turn  from  these  approximations  to  the  more  trustworthy  facsim- 
iles  of  Champollion   and  Rosellini   when   we   wish   to   form   an 
opinion  of  monuments  now  mutilated  or  destroyed ;  for  the  others, 
we    may    consult    Maxime  Ducamp's  Voyages   en   Egypte   (fol., 
Paris,    1852),    the   photographs    Banville    tock   in   concert   with 
Emmanuel  de  Rouge  (fol.,  Paris,  1868),  Bechard's  L' Egypte  et  la 
Nubie    (fol.,  Paris,  1887).    the   albums   of   Mariette   (Album   du 
Musee  de  Boulaq,  fol.,  Cairo,  1872),   Borchardt  (Kunstwc.rke  aus 
dew.  Agyptischen   Museum  zu  Cairo,  fol.,  Cairo,  1908),   Capart 


PREFACE 

(L'Art  Egyptien,  4  to,  Brussels,  1909)  and  above  all  the  incom- 
parable atlas  which  Bissing  is  completing  for  the  firm  of  Bruck- 
mann  (Denkmdler  agyptischer  Skulptur,  fol.,  Munich,  1906 — 1911). 
The  Cairo  Museum,  in  the  volumes  of  its  Catalogue  General 
(4to,  Cairo,  1900 — 1911)  has  reproduced  all  the  objects  of  its 
collections  which  are  of  interest  to  artist  or  historian,  and  that 
of  Berlin  has  given  us  a  valuable  selection  of  its  treasures 
(Agyptische  und  Vorderasiatische  Altertiimer  aus  den  Koniglichen 
Museen  zu  Berlin,  fol.,  Berlin,  1897),  but  the  riches  of  the  British 
Museum,  the  Louvre,  Turin,  and  Leyden  have  hardly  been  ex- 
ploited as  yet,  and  how  many  marvels  Egypt  herself  displays  to 
tourists  and  even  to  scholars  who  pass  them  by  indifferently! 

Nevertheless,  Egyptian  Art  is  no  longer  the  exclusive  domain 
of  a  privileged  few.  Artists  —  painters,  sculptors,  architects  — 
blind  at  first  to  its  merits,  have  come  of  late  years  to  perceive 
and  feel  them  keenly;  the  admiration  it  inspires  increases  with 
closer  study.  Men  of  letters  and  the  general  public  are  still 
disconcerted  by  the  strangeness  of  some  of  its  conventions,  and 
a  certain  time  will  no  doubt  have  to  pass  before  they  appreciate 
it  at  its  true  value.  May  this  little  book,  in  which  I  have  follow- 
ed its  fortunes  as  clearly  and  completely  as  lay  in  my  power, 
help  those  who  misjudge  it,  to  understand,  if  not  to  love  itl 


CONTENTS 

PART  I 

PAGE 

THE  BEGINNINGS  OF  ART  IN  EGYPT 
CHAPTER  I 

THINITE  ART    ...............       1 

CHAPTER  II 

MEMPHITE  ART       ..............     26 


CHAPTER  I 

THE  FIRST  THEBAN  AGE  FROM  THE  ELEVENTH  TO  THE  SEVENTEENTH 

DYNASTY          .........        .....      95 

CHAPTER  II 

THE  SECOND  THEBAN  AGE  FROM  THE  EIGHTEENTH  TO  THE  TWENTY- 

FIRST  DYNASTY       .............    124 

PART  III 

THE  SAITE  AGE  AND  THE  END 
OF  EGYPTIAN  ART 

INDEX  .  305 


XI 


FIG.   I. — SHUNET-EZ-ZEBIB,   ABYDOS.     TYPE   OF  THINITE  FORTRESS. 

PARTI 
THE  BEGINNINGS  OF  ART  IN  EGYPT 


CHAPTER  I 
THINITE  ART 

Primitive  Art  in  Egypt  before  Menes  —  Thinite  Art  and  its  Remains:  Architecture,  Fortresses, 
Palaces,  Temples,  Tombs  —  In  it  we  may  trace  the  Principles  and  Forms  which,  de- 
veloping in  the  course  of  centuries,  gave  to  Egyptian  Art  its  characteristic  aspect  — 
Memphite  Art  was  developed  by  contact  with  it. 

THE  most  ancient  tombs,  those  of  the  prehistoric  period,  have 
so  far  yielded  nothing  which  indicates  any  extraordinary 
development  of  the  artistic  sense  among  the  early  Egyptians.  The 
objects  found  in  them  bear  witness  to  a  taste  for  personal  adorn- 
ment, and  for  decorated  arms  and  utensils,  equal,  but  by  no  means 
superior  to  that  of  most  semi-civilised  nations.  They  consist  of 
coloured  pottery,  either  glazed  or  unglazed,  plain,  or  covered 
with  incised  or  painted  ornament,  furniture  of  wood  or  stone, 
jewelry  of  variegated  pebbles,  of  shells,  rough  or  carved,  of  bone, 
ivory,  glassy  pastes,  and  precious  metals;  finally,  figures  of  men 
and  animals,  some  designed  for  personal  use,  such  as  receptacles 
for  cosmetics,  others  reserved  for  funerary  rites.  The  persons 
and  things  represented  on  vases  are  not  grouped  methodically 
in  superposed  rows,  but  are  scattered  irregularly  over  the  sur- 
face at  the  will  of  the  designer,  here  a  house,  there  an  animal, 

1  B 


ART  IN  EGYPT 


FIG.   2. — THE   ABYDOS   BRACELETS. 
(Museum,  Cairo.)     (Phot.  E.  Brugsch.) 


a  palm-tree,  a  boat,  a  few  fish.     They  reveal  facility  in  seizing 
living   forms,    and   a   natural   skill   in   translating    attitudes   and 

movement  by  drawing  and 
modelling.  But  there  is 
nothing  to  compare  with 
the  sculptures  and  paint- 
ings that  contemporaries 
of  the  Reindeer  Period 
were  executing  in  the 
regions  now  known  as 
France  and  Spain. 

Nevertheless,  when  we 
pass  from  these  produc- 
tions to  which  no  exact 
date  can  be  assigned,  to 
those  of  the  historic  dynas- 
ties, we  are  confronted 
by  thousands  of  objects  and  buildings ,  the  execution  of  which 
secures  a  high  place  for  the  Egyptians  among  the  nations  of 
the  East  in  the  realm  of  art.  Where  we  had  found  only  the 
rude  essays  of  laborious  apprentices,  and  the  rudiments  of  a 
craftsmanship  as  yet  uncertain  of  itself,  we  come  suddenly, 
and  almost  without  transition,  to  the  works  of  masters,  and 
to  a  highly  accomplished  technique.  Must  we  conclude  that 

between  the  two  stages, 
alien  races  from  without 
had  dominated  the  na- 
tives, bringing  them  a  con- 
ception of  beauty  and  a 
power  of  realising  it  which 
they  had  lacked  hereto- 
fore? It  seems  improbable 
that  a  su  dden  efflorescence 
of  art  should  have  followed 
on  a  foreign  invasion;  but 
if  there  are  no  extant 
monuments  by  which  we 
may  gauge  the  natural 
evolution  of  the  Egyptian 
genius,  we  are  compelled 
to  recognise  among  the 
artisans  of  the  Thinite  age,  the  inspiration  and  even  the  processes 
of  preceding  generations.  Their  jewelry  had  preserved  the  earlier 

2 


FIG.  3. — IVORY  FEET    OF  A  BED  AND  A   STOOL,. 
(Museum,  Cairo.)    (Phot.  E.  Brugsch.) 


TUINITE  ART 


tradition,  and  their  happiest  effects  were  inspired  by  it.  I  may 
cite  the  four  bracelets  discovered  by  Petrie  in  the  necropolis  of 
Abydos  (Fig.  2),  with  their  alternating  plates  of  graven  gold  and 


FIG.  4.— FIGURINES   OF  ANIMALS,   MONKEY,  LION,   DOGS. 
(Museum,  Cairo.)    (Phot.  E.  Brugsch.) 

of  turquoise  or  light  blue  glass,  their  beads  or  pendants  of  carved 
amethyst,  and  the  chased  floral  ornament,  the  delicacy  of  which 
might  be  envied  by  our  modern  goldsmiths.  The  same  might 
be  said  of  furniture  and  domestic  utensils,  feet  of  bedsteads  or 
stools  in  wood  and  in  ivory  (Fig.  3),  figurines  of  lioiiS,  irAukeys, 
and  dogs  (Fig.  4),  stone 
or  crystal  fish,  statuettes 
of  prisoners  or  slaves,  bone 
tablets  on  which  the  prin- 
cipal episodes  of  royal 
sepulture  were  traced  with 
the  style  (Fig.  5),  cylinders 
bearing  hieroglyphic  le- 
gends or  divine  emblems, 
club-heads,  etc.  In  all 
these  we  recognise  the 
early  ideas  and  conventions, 
with  this  difference,  that 
what  was  the  result  of  pure 
instinct  in  the  beginning 
has  become  that  of  deliber- 
ate intention.  Craftsmen 
or  artists,  the  experience 
of  an  unknown  number  of 
generations  had  taught 

them  gradually  to  bring  out  the  principal  lines  of  their  models, 
to  fix  their  contours,  to  simplify  their  reliefs,  to  co-ordinate  their 

3  B  2 


FIG.  5.— THE   TABLET    OF   AHA. 
(Museum,  Cairo.)    (Phot.  E   Brugsch) 


ART  IN  EG\PT 


FIG.  6.— GROOVED  FACADE  OF  THINITE 

TOMBS  AND  FORTRESSES. 

(After  Garstang.) 


movements  and  their  postures.  They  took  pleasure  in  slow  and 
tranquil  gestures;  if  the  nature  of  their  subjects,  religious 

processions,  hunting  and 
battle  -  scenes  ,  assaults 
on  cities,  the  pursuit  of 
enemies ,  forced  them  to 
express  violent  or  rapid 
action,they  did  their  utmost 
to  minimise  its  hard  ab- 
ruptness. It  will  be  readiiy 
supposed  without  much 
insistence  on  the  point, 
that  an  art  so  well  regulated 
implies  a  long  period  of 
preliminary  effort  and  ex- 
periment. In  spite  of  ex- 
ternal divergences ,  the 
elements  are  those  chosen 

and  employed  by  the  ancestors  of  its  practitioners  from  the 
beginning;  but  the  workman  had  handled  them  so  often  and  for 
so  lu;::;  that  by  dint  of  practice  he  had  at  last  reduced  them  to 
a  system,  and  had  replaced  the  direct  observation  of  nature  by  the 

constant  use  of  decorative  schemes 
or  of  formulae  accepted  in  the 
workshop. 

The  impression  made  by  the 
industrial  arts  is  confirmed  by  the 
rarer  survivals  of  the  higher  arts, 
architecture,  sculpture,  and  paint- 
ing. Very  little  has  come  down 
to  us  of  the  military  and  civil  or 
indeed  of  the  religious  archi- 
tecture; we  have  the  ruins  of  a 
fortress  at  Hieraconpolis,  in  Aby- 
dos  (cf.  Fig.  1)  and  in  one  or 
two  small  townships  of  the  Said, 
while  in  some  of  the  ritual  tablets 
and  the  hieroglyphic  writings  we 
find  incidental  renderings  of  se- 
veral very  ancient  temples  (cf. 
Fig.  8).  The  fortresses,  or  rather 

castles  in  which  the  kings  and  nobles  lived,  are  vast  parallelograms 
of  sun-dried  brick,  the  walls  of  which  are  sometimes  perfectly 

4 


FIG.   7.— PLAN   OF  THE   FORTRESS   CF 
KOM-EL-AHMAR       (After    Quibell  ) 


THINITE  ART 

smooth  and  unadorned  from   one  angle  to  the  other,   sometimes 
divided  into  panels,  the  beds  of  which  are  alternately  horizontal 


FIG.  8.— VARIOUS   TYPES   OF  ARCHAIC    CHAPELS   AND   TEMPLES. 


and  concave,  and  sometimes  finally  present  a  series  of  vertical 
prismatic  grooves  (Fig.  6).  The  principal  doorway  is  generally 
relegated  to  the  end  of  one  of  the  lateral  walls,  and  is  set  in  a 
block  of  masonry  solid  enought  to  defy  sap  and  ram  (Fig.  7). 
Private  persons  inhabited  buildings  of  beaten  earth  or  dried  brick 
similar  to  those  of  the  modern  fellahin,  and  like  these,  generally 
of  a  single  storey.  The  temple  was  an  isolated  cell,  of  variable 
dimensions,  but  always  of 
small  extent,  raised  upon 
an  artificial  mound  at  the 
end  of  a  rectangular  enclosure 
bounded  by  a  low  wall  or 
rows  of  piles;  two  posts 
were  set  up  in  front  of  the 
entrance,  and  the  emblem  of 
the  god  crowned  the  roof,  or 
was  raised  on  a  pole  in  the 
middle  of  the  enclosure 
(Fig.  8).  The  cella  consisted 
at  first  of  four  wooden 
uprights,  connected  by  wicker- 
work  plastered  with  mud; 
the  doorway  was  closed  by  a 
wooden  panel  or  a  hanging 
mat.  In  some  cases ,  the 
roof  was  flat,  with  or  without 
a  cornice;  but  in  general,  it  described  a  peculiar  curve  from 
front  to  back,  the  form  of  which  persisted  after  the  little  build- 


F3G.    Q.— WOODEN  NAOS. 
(Museum    Turin  )     (Phot.  Lanzone.) 


ART  IN  EGYPT 


ing  of  slight  materials  had  become  a  naos  of  wood  (Fig.  9)  or 
of  stone  covered  with  inscriptions  and  hieroglyphic  scenes  (Fig.  10). 
To  this  cell  other  cabins  were  soon  added  for  the  accommodation 
of  auxiliary  gods,  priests,  and  offerings,  and  the  whole,  symmetrically 
arranged  in  an  enclosure,  constituted  a  divine  palace,  analogous 
to  the  royal  dwelling.  At  a  later  stage  the  gods,  dissatisfied 
with  so  poor  a  dwelling,  demanded  thicker  Walls  made  of  bricks, 

and  stone  for  thresholds, 
lintels ,  architraves ,  and 
the  bases  of  columns ;  then 
limestone  or  sandstone 
was  substituted  for  brick, 
with  granite  to  surround 
the  bays,  and  the  perish- 
able huts  of  an  earlier 
age  became  houses  of 
eternity,  without,  however, 
changing  the  main  lines 
of  the  primitive  plan. 
The  fragments  which  we 
possess  of  the  temple 
dedicated  to  the  gods  of 
Hieraconpolis  prove  that 
the  transformation  was 
already  far  advanced  under 
the  Third  Dynasty.  They 
formed  part  of  a  doorway 
of  pink  granite,  the  exterior 


FIG.    10.— SIDE   OF  NAOS   OF  SAFT-EL-HENNEH. 
(Museum,  Cairo.)     (Phot.  L.  Brugsch.) 


faces  of  which  were  decor- 
ated with  royal  legends 
(Fig.  11),  and  with  bas- 
reliefs  which  were  effaced  during  one  of  the  reconstructions  of 
the  building;  the  patterns  made  by  the  hammer-strokes  enable  us 
to  divine  that  these  reliefs  represented  the  sovereign  adoring  the 
divinity  with  the  ritual  familiar  to  us  on  the  monuments  of  the 
Theban  era. 

Like  the  temples,  the  tombs  of  the  Thinite  age  retained  the 
principal  features  of  those  of  the  earlier  period  at  Abydos,  at 
Nakadah,  at  Hieraconpolis,  and  in  all  places  where  they  have 
been  discovered  hitherto.  The  most  famous,  that  of  Nakadah 
(Fig.  12),  did  not  belong,  as  might  have  been  supposed,  to  the 
Menes  who  founded  the  Egyptian  Empire,  nor  to  some  other 
Menes  almost  contemporary  with  him;  it  guarded  the  mummy  of 

6 


THINITE  ART 


some  nameless  lord,  who  ruled  a  portion  of  the  Theban  plain. 
Imagine  a  rectangle  some  176  feet  long  by  88  wide,  running 
diagonally  from  north  to  south.  It  is 
composed  entirely  of  unfired  bricks 
cemented  and  plastered  with  clay, 
without  either  limewash  or  painting. 
The  exterior  surfaces  were  originally 
decorated  with  the  usual  vertical  grooves, 
and  the  plan  of  the  interior  included 
a  large  hall,  separated  from  the  enclosing 
wall  by  a  narrow  passage,  and  divided 
into  five  compartments  ranged  in  a 
line  on  the  main  axis.  The  corpse  was 
laid  in  the  central  compartment,  and 
his  household  goods  were  arranged  partly 
on  the  ground  around  him,  partly  in  the 
four  other  chambers.  When  these  were 
full,  they  were  walled  up,  and  the 
adjoining  passage  was  parcelled  into 
cells  for  the  reception  of  surplus  provi- 
sions, after  which  the  entrance  was 
blocked,  and  the  external  decorations 
were  masked  by  a  facing  of  bricks, 
whitewashed  over.  The  tombs  of  the 
Thinite  Pharaohs  excavated  by  Ame- 
lineau  to  the  west  of  Abydos,  and  by 

Garstang  at  Rekaknah  and  at  Bet-Khallaf  (Figs.  13,  14)  were 
not  all  exactly  similar  in  arrangement;  one  was  shaped  somewhat 
like  a  shuttle,  wider  in  the 
middle  than  at  the  two 
extremities;  another  was 
floored  and  panelled  with 
wood,  and  their  outer 
surfaces  showed  neither 
projections  nor  recesses. 
None  the  less  are  they  of 
the  same  type  as  that  of 
Nakadah,  and  if  we  con- 
sider them  as  a  whole, 
we  are  struck  by  their 
general  resemblance  to  the 

fortresses  of  Abydos  and  Hieraconpolis.     Like  these,   they  have 
in  some  cases  walls  with  prismatic  niches;  they  have  their  store- 


FIG.      II.— DOOR-JAMB   OF  THE 

TEMPLE   OF  KOM-EL-AHMAR. 
(Museum,  Cairo.)  (Phot. L, Brugsch.. 


FIG.  12.— PLAN  OF  THE  TOMB  AT  NAKADAH. 
(After  De  Morgan.) 


ART  IN  EGYPT 


rooms  and  their  lodging  reserved  for  the  chief;  their  doors  are 
hidden  away  at  the  least  accessible  point,  and  they  were  blocked 
up  after  the  deposition  of  the  corpse,  just  as  those  of  the 


FIG.   13.— ROYAL  TOMB  AT  BET-KHALLAF.      (Phot    Garstang.) 

fortresses  were  barricaded  in  the  hour  of  danger  (Fig.  14). 
Thus  the  same  intention  governed  the  construction  of  each. 
Just  as  the  fortress  was  the  residence  of  the  living  lord  or 
sovereign,  the  castle  in  which  he  held  his  court  in  peace,  and 
in  which  he  awaited  behind  barricaded  doors  the  attacks  of  his 
enemies  in  war,  so  the  tomb  was  looked  upon  as  the  castle 
of  the  dead  lord  or  sovereign,  in  which  he  intrenched  himself 
for  all  eternity,  safe  from  the  outrages  of  men  and  of  years.  If 
we  remember  that  the  temple  was  also  a  palace,  the  palace  of 
the  god,  we  shall  be  driven  to  admit  that  identity  of  terms  here 

denotes  identity  of  con- 
ception, and  that  the 
manner  of  life  of  the  lord 
or  the  Pharaoh  before  and 
after  interment,  was  ident- 
ical with  that  of  the  gods. 
Originally,  the  monumen- 
tal tomb  had  been  the 
privilege  of  those  powerful 
enough  to  procure  it,  chiefs 
of  clans,  princes  of  Nomes, 
great  officers  of  the  crown, 
and  kings;  later,  with  the 

FIG.  I4.-PLAN  OF  THE  TOMB     (After  Garstang.)  growth     of    Wealth     in     the 

nation,  the  privilege  was 

extended,  and  was  conferred,  under  the  conditions  we  shall 
presently  note ,  on  those  of  the  people  whose  fortunes  or  the 

8 


THINITE  ART 


favour  of  the   master   encouraged   to  aspire  to   the  luxury  of  an 
independent  future  life. 

The  internal  walls  are  generally  speaking  bare,  but  the  priests 
or  the  relatives  of  the  defunct  stored  up  in  the  vault  or  in  the 
chambers  adjoining  it  the  funereal  trappings,  furniture,  provisions, 
and  simulacra  more  or  less  rude  of  the  dead  man  and  of  the 
servants  who  were  despatched  with  him  to  the  next  world;  they 
were  there  in  numbers,  each  in  the  attitude  proper  to  his  rank 
or  function;  the  master  seated,  or  standing  to  receive  offerings, 
the  servants  engaged  in 
preparing  or  serving  these, 
grinding  corn  (Fig.  15), 
kneading  dough,  brewing 
beer,  plastering  a  jar  (Fig. 
16)  before  putting  wine  in 
it,  busied  with  all  the 
duties  of  the  household, 
apparently  inanimate,  but 
impregnated  with  the 
latent  life  breathed  into 
them  by  virtue  of  the 
rites.  Gradually,  however, 
the  instinct  which  had 
moved  the  primeval  Egyp- 
tians to  decorate  their 
pottery,  led  those  of  the 
archaic  age  to  cover  their 
walls  with  scenes  intro- 
ducing the  persons  whose 
figures  had  heretofore  lain  scattered  on  the  ground.  These  were  mute 
at  first;  later,  they  were  accompanied  by  short  inscriptions  setting 
forth  their  deeds,  their  speeches,  their  names,  the  titles  which 
constituted  their  civil  status,  and  authenticated  their  rights  to 
posthumous  life.  The  earliest  example  is  that  discovered  by 
Quibell  at  Kom-el-Ahmar.  The  motives  were  not  as  yet  arranged 
in  rows  with  methodical  precision,  but  like  those  on  the  vases, 
they  are  scattered  almost  at  random  over  the  surface.  The  first 
essays  are  rude  figures  careless  of  form  and  proportion,  awk- 
wardly drawn  with  red  ochre,  like  those  with  which  our  children 
adorn  the  margins  of  their  books.  With  these  we  find  the  camp 
or  the  village  on  the  edge  of  the  desert,  gazelles  and  oryx 
browsing,  running  on  the  plain,  or  keeping  watch  on  some  rocky 
peak,  men  armed  with  darts  or  clubs,  following  a  trail  or  fighting 

9 


FIG.    15  —WOMAN   GRINDING   CORN. 
(Museum,  Cairo.)   (Phot.  L.  Brugsch.) 


ART  IN  EGYPT 


.with  hyaenas,  all  the  details,  in  short,  which  make  up  the 
hunter's  ideal  of  a  happy  life  and  a  desirable  Elysium.  The  noble 
who  ordered  the  paintings  and  the  artisan  who  daubed  them 
had  no  conception  of  a  life  of  the  manes  differing  in  any  respect 
from  that  of  the  living,  and  they  believed  that  the  surest  way 
of  facilitating  access  to  it  was  to  introduce  paintings  of  it 
in  the  tomb  near  the  corpse.  This  belief  was  the  logical  con- 
clusion of  a  dogma  familiar  to  all  semi- 
barbarians  :  he  who  invents  or  reproduces 
a  figure,  no  matter  what,  immediately 
creates  a  being,  and  if  he  afterwards 
gives  it  the  name  of  a  man,  an  ani- 
mal, or  an  object,  he  endows  it  with 
a  portion  of  soul  stolen  from  that  of  the 
original.  The  life  of  the  simulacrum 
ceases  with  that  of  its  prototype,  but 
the  latter  can  prolong  it  and  himself 
by  means  of  incantations;  and  by  virtue 
of  these  it  may  even  be  transmitted 
to  all  portraits  of  an  individual  executed 
after  his  death.  Stone  or  wooden  images 
took  the  place  of  the  fleshly  bodies  of 
master  and  servants  if  these  were  missing, 
and  thus  assuming  the  function  of  a 
double,  they  guaranteed  its  perpetuity 
as  long  as  they  were  preserved.  The 
decoration  of  the  tomb  produced  an 
identical  effect;  it  prevented  the  annihi- 
lation of  the  servants  and  the  objects 
represented,  and  obliged  them  to  minister 
to  the  comfort  of  the  master,  as  long 
as  substantial  traces  of  what  it  had  reproduced  remained  upon 
the  walls. 

Briefly,  all  forms  of  art,  architecture,  sculpture,  and  painting, 
tended  not  to  the  disinterested  search  after  beauty,  but  to  the 
realisation  of  the  useful.  The  three  categories  into  which  reason- 
able beings  were  divided,  the  living,  the  dead,  and  the  gods, 
shared  an  intense  desire  for  duration;  they  had  bent  all  the  most 
powerful  springs  of  their  minds  in  this  direction,  and  the  effort 
they  had  made  to  achieve  it  had  given  a  special  character  to  the 
arts.  In  the  first  place,  it  had  determined  the  choice  of  materials. 
Men,  whose  years,  however  numerous,  are  but  as  a  moment 
compared  with  the  innumerable  centuries  of  the  dead  and  the 

10 


FIG.    16. 

SERVANT  TARRING   A   JAR. 
(Museum,  Cairo.)  (Phot.  L  Brugsch.) 


THINITE  ART 


FIG,  17.— PAINTINGS  AT  KOM-EL-AHMAR.  (After  Quibell.) 


gods,  were  content  with  slight  materials:  earth,  wood,  and  freestone. 
If  they  made  use  of  metals,  it  was  less  for  their  indestructibility 
than  for  the  charm 
of  their  colour,  the 
fineness     of     their 
texture,  their  ducti- 
lity and  their  value. 
The  richest  of  the 
Pharaohs    did    not 
disdain  to  dwell  in 
houses  of  sun-dried 
bricks,  under  roofs 
of  beaten  mud  sup- 
ported by  wooden 
columns ,    between 
ceilings  and  pave- 
ments      bedaubed 
with  fragile  paint- 
ings;   if  their  palaces  lasted  as  long  as  they  lived,   they   cared 
little  that  they  should  fall  into  ruin  as  soon  as  they  themselves  had 
disappeared.    The  case  was  very  different  when  the  dwelling  was 
destined  for  the  dead  or  for  the  gods;  then  its  longevity  had  to 
rival  theirs,  and  this  result  was  only  to  be  achieved  by  the  help  of 
the  most  solid  materials.     Tombs  and  temples  were  accordingly 
built  of  limestone  or  sand- 
stone of  the  best  quality, 
and   even   of  granite   or 
the  breccia  of  the  Arabian 
mountains,  which,  however, 
was    generally     reserved 
for  the  doors,   the  thres- 
holds,  and  such  portions 
of   the   structure    as   the 
frequent    passage   of   the 
faithful    and    the   perfor- 
mance  of   religious   rites 
tended     to     wear    away 
rapidly;  panegyrics  upon 
the  Pharaohs  extolled  them         FIG.  18.— TEMPLE  OF  KHONSU  AT  KARNAK 
for  having  erected  houses  (Phot.  Beato.) 

of  eternal  stones  for  their 

divine  fathers.     But  as  the  density   and  durability  of  the   blocks 
themselves  was   not   sufficient   to   ensure   the    longevity    of    the 

11 


ART  IN  EGYPT 


FIG.  IQ.— THE   SERVANTS   OF  TI   BRINGING     OFFERINGS. 


building,  care  was  taken  to  choose  those  architectural  forms 
which  promised  the  greatest  stability,  such  as  the  pyramid,  the 

mastaba,  the  rect- 
angular temple,  set 
firmly  upon  the  soil 
under  such  cond- 
itions of  equilib- 
rium that  only  the 
hand  of  man  or 
the  attacks  of  the 
river  could  over- 
throw them;  time 
has  proved  power- 
less against  them. 
The  reliefs  with 
which  they  were 

adorned  were  kept  very  low,  or  "left"  on  sunk  surfaces,  which 
diminished  the  risk  of  damage  from  concussion,  and  protected 
them  from  accidental  mutilations.  It  was  permissible  for  persons 
of  modest  means  to  employ  wood  for  statues,  and  for  divine  or 
funerary  groups;  but  if  economy  was  not  essential,  limestone  or 
sandstone,  alabaster,  schist,  granite,  serpentine,  and  diorite 

were  preferred,  and  he 
who  had  none  of  these 
to  hand  sent  into  the 
desert  to  fetch  them.  Even 
then,  statues  were  given 
one  of  the  three  or  four 
attitudes  which  seemed 
least  fragile;  figures  were 
seated  upon  a  complete 
cube  or  on  a  seat  with 
a  straight  back,  or  they 
'stood  with  legs  pressed 
together,  arms  adhering 
to  the  body,  back  and 
head  engaged  in  a  vertical 
slab.  The  care  for  solid- 
FIG.  20.— HARVEST  SCENES  IN  THE  TOMB  ity  prevailed  over  every 

other  consideration   both 
with  sculptor  and  architect. 

This  gave  their  works  a  unity,  and  also,  we  must  frankly  admit, 
a   constant    uniformity    of    invention    and    execution;    the    very 

12 


THINITE  ART 


FIG.     21.— BRINGING     CORN    TO    THE    GRINDER, 
TREADING  AND  WINNOWING,  IN  THE  TOMB  OF  TI. 


derogations  from  the 
determining  principle 
which  we  seem  to  note 
in  them  prove  to  be  a 
result  and  a  confirmation 
of  it,  when  we  study  them 
attentively.  If  indeed  the 
tomb  and  the  temple, 
supplying  the  same  de- 
mands, consist  on  the 
whole  of  the  same  ele- 
ments as  the  palace,  it 
was  nevertheless  essen- 
tial that  they  should  be 
arranged  to  suit  the  indi- 
vidual conditions  of  posth- 
umous existence  and  of 
divine  life.  Each  contain- 
ed a  personal  dwelling, 

rooms  for  guests  or  slaves,  store-rooms,  and  audience-chambers; 
but  whereas  the  god  consented  to  receive  his  priests  and  to  show 
himself  sometimes  to  the  people,  the  dead  person  was  wholly 
inaccessible,  and  never  again  appeared  in  public,  when  the 
hour  of  his  entry  into  his 
private  apartments  was  past. 
The  sanctuary  was  accordingly 
so  placed  that  it  communicated 
with  the  outside  world,  and 
that  the  living  could  enter  it 
easily  on  prescribed  occasions; 
the  vault,  on  the  other  hand, 
was  never  opened  again  after 
the  corpse  had  been  placed 
in  it,  and  soon,  to  render  the 
dead  more  completely  inac- 
cessible, the  body  was  placed 
underground,  at  the  end  of 
a  passage  or  at  the  bottom 
of  a  well  which  was  subse- 
quently filled  up  to  the  level  of 
the  soil.  Thenceforth,  whereas 
the  temple,  inhabited  by  a  FIG.  22._SLAUGHTER  OF  CATTLE,  IN  THE 
being  whose  nature  did  not  TOMB  OF  TI. 

13 


ART  IN  EGYPT 


FIG.  23. — THE  NOMES  BRINGING  OFFERINGS, 
(Phot.  Beato.) 


AT  EDFU. 


debar   him   from  official   intercourse  with  the  world,   tended   to 
increase  in  size  and  to  expand  in  the  sunlight,  the  tomb,  shunning 

the  light  of  day, 
gradually  diminish- 
ed its  points  of 
external  contact, 
and  ended  by  dis- 
playing only  a 
narrow  fagade,  a 
roughly  hewn 
panel,  the  bay  of 
a  door,  or  a  rect- 
angular orifice  in 
the  rock.  During 
the  process  of  this 
change,  the  common  plan  was  modified.  As  the  princely  dwelling 
was  converted,  on  the  one  hand  into  a  sanctuary,  on  the  other 

into  a  vault,  the  guest- 
chambers,  store-houses,  and 
reception-rooms  were  modi- 
fied and  adjusted  to 
circumstances.  Some  of 
the  store  -  rooms  were 
relegated  to  the  sides,  and 
assimilating  to  some  extent 
with  the  guest  -  chambers, 
they  became,  jointly  with 
these,  depots  for  daily 
offerings,  or  chapels  for 
the  paredri.  The  reception 
rooms  were  then  trans- 
formed into  hypostyle  halls 
or  courts  echeloned  in  a 
line  on  the  longitudinal 
axis  of  the  building,  and 
increasing  in  size  in  propor- 
tion to  their  distance  from 
the  Holy  of  Holies.  Thus 
the  worshippers  of  the  god 
no  more  enjoyed  equal 
facilities  of  approach  than 

the  courtiers  of  Pharaoh;  a  few  were  admitted  to  the  presence, 
many  entered  halls  more  or  less  close  to  him,  as  determined  by 

14 


' 


FIG.   24. — A  WALL  IN   THE  TEMPLE   OF  ISIS,  AT 
PHIL^E.      (Phot.  Beato.) 


THINITE  ART 


the  hierarchy,  and  the  rest  advanced  no  farther  than  the  outer 
courts  or  the  platform  before  the  temple.  A  dead  person  of 
non-royal  race,  who  had  not  a  nation  to  adore  him,  but  whose 
family  and  dependents  were  necessarily  restricted,  would  have 
had  no  use  for  such  vast  spaces;  a  simple  chapel  sufficed  for 
him,  and  three  or  four  little  rooms,  rarely  more,  and  often  less, 
served  him  for  storehouse,  reception-rooms,  and  guest-chambers, 
without  endangering  his  chances  of  immortality. 
When  this  point  had  been  reached,  the  diversity 
between  tomb  and  temple  was  so  great,  that 
all  traces  of  a  common  origin  seem  to  have 
been  effaced,  what  likeness  is  there  at  a  first 
glance  between  a  monument  such  as  that  of 
Khonsu  (Fig.  18)  or  of  Edfu,  and  the  Theban 
hypogea  of  the  Sa'ite  period?  To  perceive 
the  analogy  clearly,  a  long  series  of  deductions 
and  analyses  would  be  requisite. 

The  difficulty  is  hardly  less  when  we  attempt 
to  refer  the  decorations  of  these  buildings 
to  a  single  type;  but  here  again  the  most 
extravagant  divergencies  may  be  explained  by 
the  nature  of  the  personages  concerned. 
Whether  gods  or  mummies,  they  were  incapable 
of  subsisting  by  their  individual  energy,  and 
they  would  have  died  of  starvation  but  for 
the  daily  intervention  of  the  living.  The  latter 
accordingly  endowed  them  with  appanages,  or, 
if  we  prefer  the  term  used  in  Mussulman 
jurisprudence,  dukaf,  which  kept  them  in  good 
case,  as  long  as  the  revenues  were  applied 
to  them.  It  was  not  very  often,  however, 
that  they  profited  long  by  their  endowments. 
Their  descendants  took  them  away  after  a  certain  number  of 
years,  or  the  family  became  extinct,  and  its  heritage  fell  to 
strangers  who  repudiated  the  charges  on  the  property;  the  founder, 
despoiled  of  his  dues,  suffered  the  slow  tortures  of  starvation, 
before  finally  succumbing  to  that  second  death  which  definitively 
annihilated  that  remnant  of  vitality  his  first  death  had  respected. 
The  statuettes  scattered  in  the  rooms,  and  the  scenes  represented 
upon  the  walls  of  the  tomb  averted  the  danger,  and  the  greater 
their  number,  the  more  certain  he  was  to  lack  nothing,  even  if  his 
actual  possessions  had  been  taken  from  him.  Interminable 
processions  of  men  and  women  simulated  his  houses,  his  ponds, 


FIG.   25. — FULL  FACE 
STATUE,     THE    FEET 

JOINED. 

(Museum,   Cairo.) 
(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


FIG.   26.— FULL  FACE 

STATUE,    WALKING. 

(Museum ,    Cairo.) 

(Phot.  E   Brugsch.) 


his  woods,  his  meadows,  and  brought  the 
rents  of  their  farms  to  him  (Fig.  19),  while 
the  serfs  and  servants  of  his  household, 
each  occupied  with  his  special  work, 
manufactured  its  products  for  him  eternally. 
Did  he  want  bread  ?  The  field  was  ploughed 
before  his  eyes  in  simulacrum;  the  ripe 
grain  fell  under  the  sickle  (Fig.  20);  the 
corn  was  trodden  out,  winnowed,  measured, 
and  poured  into  the  granary  (Fig.  21) ; 
then  the  bakers  ground  it,  baked  the 
loaf,  and  presented  it  to  him.  Did  he 
desire  to  eat  meat?  In  other  pictures,  the 
bull  was  coupled  with  the  cow,  the  calf  was 
born,  grew  up,  became  an  ox,  fell  into 
the  hands  or  the  butchers  who  cut  its 
throat,  caught  the  blood,  skinned  and  divided 
the  carcase,  choosing  the  best  pieces  for 
him  (Fig.  22).  The  gods  were  less  in  danger 
of  that  complete  ruin  which  resulted  in 
annihilation;  when  once  their  appanage  had 

been  apportioned,  if  some  impious  person  appropriated  it  in  the 

course  of  time,  a  benefactor  duly  appeared 

to  repair  the  injury.    Nevertheless,  it  was 

considered   prudent   to   assign    to  them 

as  to  the  dead,    a  fictitious   domain  to 

supplement  the  real  one  if  needful;  the 

Nomes    were    accordingly     represented 

bringing    their    tribute    (Fig.   23) ,    and 

sometimes  even  barbarous  peoples  sub- 
ject to  Egypt.    And  as  the  gods  could 

only  be  served  by  the  Pharaoh   or  one 

of  his  family,   on   every  occasion  when 

the  help  of  an  inferior  was  not  indispensable, 

we  see  around  them  on  the  walls  only 

kings  and  queens,  princes  and  princesses; 

these  noble  personages,   however,  were 

not  obliged  to  perform  the  vulgar  tasks 

imposed  on  the  kinsfolk  of  dead  persons. 

It  appears  that  no  effort  was  necessary 

to   obtain  all  that  was  required  in  the 

sphere    in  which    they    moved;    objects 

presented  themselves  all  ready  for  use, 

16 


FIG. 27.— SEATED  STATUE  OF 

THE  PHARAOH   MYCERINUS. 

(Museum,   Cairo.) 

(Phot.  E.  Brugsch.) 


THINITE  ART 


the  ox  decked  for  the  holocaust,  water, 
wine,  milk,  and  oil  poured  for  libation, 
the  incense  lighted  for  sacramental  fumi- 
gation. Nothing  could  be  less  like  the 
varied  episodes  of  human  life  which 
abound  in  the  hypogea  than  these  monoto- 
nous functions  of  divine  life;  and  yet 
the  god  who  might  have  been  suddenly  on 
the  verge  of  starvation  by  reason  of 
some  accident  would  only  have  had  to 
cast  his  eyes  around,  and  the  imagery  of 
the  walls  would  have  nourished  him  with 
delectable  realities;  in  the  temple  as  in 
the  tomb  the  decoration  had  an  important 
utilitarian  function,  and  if  the  elements 
which  composed  the  two  differed  materi- 
ally, the  intention  which  co-ordinated  them 
and  the  conception  from  which  they  were 
deduced  were  one  and  the  same. 

This  insistence  on  a  utilitarian  purpose 
weighed  heavily  upon  the  independence 
both  of  art  and  artist.  The  artist,  obliged 
to  think  above  all  of  the  welfare  of  those 


FIG.  28. — STANDING  STATUE 

BEARING  ENSIGN. 

(Museum ,    Cairo.) 

(Phot.  E.  Brugsch.) 


FIG.    29. — GROUP    OF    PERSONS, 

STANDING  AND    SITTING. 
(Museum,  Cairo.)     (Phot,  E.  Brugsch.) 


for  whom  he  was  working,  was 
not  free  to  abandon  himself  to 
his  inspiration,  nor  to  diverge 
from  the  rules  in  which  religion 
imprisoned  him.  Seeing  that  the 
statue  was  the  body  of  a  double, 
it  was  necessary,  in  order  to 
attach  this  double  to  it,  to  repro- 
duce not  only  the  physiognomy 
and  features  of  the  model,  but 
the  bearing  and  the  costume  of 
his  profession,  to  the  end  that 
in  his  state  of  death  he  should 
remain  what  fortune  had  made 
him  here  below;  it  had  to  be  at 
once  the  portrait  of  an  individual, 
and  a  type  of  the  class  to  which 
he  belonged.  Scribes,  artisans, 
nobles,  priests,  Pharaohs,  and 
gods  had  each  their  prescribed 
17  c 


ART  IN  EGYPT 


formula,  to  which  sculptors  and  painters  were 
required  to  adhere  faithfully,  without,  however, 
modifying  the  characteristic  lines  of  the  indi- 
vidual; some  thirty  positions  were  permitted: 
standing  still  (Fig.  25)  or  walking  (Fig.  26), 
sitting  (Fig.  27),  kneeling,  crouching,  crawling, 
lying,  with  a  hundred  variations  of  costume, 
head-dress  and  insignia.  Persons  were  re- 
presented facing  the  spectator  squarely,  so 
that  a  line  drawn  perpendicularly  through 
the  body  always  divided  it  into  two  equal 
portions,  and  this  law  of  frontality  was 
rigorously  observed  down  to  the  last  days 
of  pagan  Egypt.  Mural  decoration,  by  the 
nature  of  the  episodes  proper  to  it,  admitted 
greater  diversity;  but  here  it  is  necessary  to 
distinguish  between  that  of  tombs  and  that 
of  temples.  It  both  cases  the  moment  chosen 
is  that  of  the  performance  of  the  act  which 
was  to  produce  the  result  enjoined  by  dogma; 
but  whereas  in  the  temple  the  scene,  enacted 
by  gods  and  Pharaohs,  rarely  required  more 
than  three  or  four  persons,  always  of  the  same 

kind,   in  the  tomb  all  Egypt  played  its  part,  men,    beasts   and 

things,    and   the  number  of  the 

participants  had  no  other  limits 

than  those  of  the  surface  to  be 

covered.  The  result  in  the  bas- 
reliefs  of  the  temple  was  a  uni- 
formity of  composition  and  attitude 

almost  equal  to  that  of  the  statues. 

Nothing  was  modified  in  the  course 

of  ages  but  the    distribution  of 

the  pictures  and  the  degree  of 

skill    of   the   execution.      In   the 

tombs,    neither   the  moments  of 

rustic  or  industrial  toil,  nor  those 

of    funereal     offerings     are     so 

inflexibly  treated.  Up  to  a  cer- 
tain point,  the  artist  was  free  to 

introduce   additional  motives,  to 

intermingle    and    disconnect    his 

episodes,  to  alternate  and  break 

18 


FIG.   30. — STATUE    WITH 
BODY  IN  A   SHEATH. 

(Museum,  Cairo.) 
(Phot.  E.  Brugsch.) 


FIG.   31. — THE  DOG  NIBtJ. 
(Museum,  Cairo.)     (Phot.  E.  Brugsch.) 


THINITE  ART 


FIG.   ?2. — THE  LADY  NUTIR. 
(Museum.Cairo.)  (Phot.E.Brugsch.) 


them  up;  but  this  freedom  only  stood 

him  in  good  stead  in  the  somewhat  rare 

cases    when    the    extent    of    the   walls 

allowed  him  to  give  the  decoration  the 

necessary  breadth.      Nine  times  out  of 

ten,   lack  of  space   compelled   him   to 

confine  himself  to  those  themes  which 

contributed  most  to  the  happiness  of 

the  dead  person,    and  to  condense  his 

material    as    much   as  possible.     By   a 

progressive   series  of  eliminations,   the 

work    became    at    last   an   abbreviated 

panorama  of  Egyptian  life,  the  motives 

of  which,  blurred  or  decomposed  slowly 

by  the  lapse  of  centuries,    impress   us 

in  each  epoch  with  a  sense  of  monotony 

hardly  less  than  that  which  we  experience 

in   the   temples.     It  has  not  yet  been 

very  definitely  ascertained   where   this 

system  arose  and  was  developed.    It  is  probable  that  its  materials 

were   evolved   spontaneously  at  first  in  every  quarter,   but  that 

they  were  revised,  subjected  to  a  process  of  selection,  and  co- 
ordinated at  Heliopolis,  in  the  regions  where  the  popular  religions 
were  welded  into  a  body  of  theology  which 
the  whole  nation  accepted.  The  Heliopolitan 
doctrine  unquestionably  governed  the  in- 
stallation of  the  temples,  with  its  Enneas, 
its  insistence  on  the  predominance  of  the 
solar  divinities,  its  ritual  of  prayers  and 
offerings.  And  if  we  pass  in  review  the 
illustration  of  the  tombs,  we  shall  see  that 
the  great  number  of  fishing  and  hunting 
scenes  in  the  marshes  included  in  it  are 
more  in  keeping  with  the  ancient  conditions 
of  the  upper  parts  of  the  Delta  and  of 
the  marshy  districts  adjoining  the  Fayum 
than  with  those  of  the  Said. 

It  would  seem  then  that  it  was  in  the 
Heliopolitan  plain  and  in  the  region  about 
Memphis  that  the  convention  of  the  tombs 
was  instituted,  and  its  northern  origin  will 
explain  why  it  has  not  so  far  been  met 
with  in  the  Thinite  cemeteries  of  Abydos; 
19  C  2 


FIG.      33. — THE      DWARF 

HAPU.    (Museum,  Cairo.) 

(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


FIG.   34.— THE    SOLDIER    ABUNI. 
(Museum,  Cairo.)  (Phot.E.Brugsch.) 


it  had  not  had  time  to  impose  itself 
upon  the  reigning  dynasties  and  their 
subjects,  even  if  it  had  penetrated 
into  the  region.  The  most  richly  or- 
namented sepulchres  here  admitted 
no  elements  of  plastic  art  other  than 
stelae,  or  perhaps  statues  of  the  dead 
and  their  servants.  None  of  the  latter 
have  survived,  but  there  is  no  lack 
of  stelae,  both  royal  and  private. 
The  private  ones,  which  are  by  far 
the  most  numerous,  generally  consist 
of  a  simple  limestone  slab,  roughly 
shaped,  sometimes  rounded  at  the  top, 
sometimes  rectangular ,  sometimes 
irregular  in  shape,  on  which  the  figure 
of  a  man  or  animal  with  an  ins- 
cription has  been  hastily  carved.  It 
may  be  the  dog  Nibu  with  his  straight 
ears  and  pointed  muzzle  (Fig.  31);  the  dwarf  Hapu  standing 
and  presenting  his  misshapen  profile  to  the  spectator  (Fig.  33); 
the  lady  Nutir  crouching  on  the  ground  (Fig.  32);  the  soldier 
Abuni,  also  crouching,  but  grasping  a  bow  (Fig.  34).  Their  lumi- 
nous doubles  were  as  surely  attracted  by 
these  rudimentary  bas-reliefs  as  by  statues, 
and  as  they  were  much  less  costly,  they 
were  a  great  saving  to  the  family;  thanks 
to  them,  the  doubles  followed  the  sovereigns 
on  whose  mastabas  they  were  placed,  and 
by  serving  them,  secured  a  share  in  their 
happy  destinies.  These  stelae  were  executed 
by  workmen,  with  small  pretensions  to  art, 
yet  the  technique  is  already  so  refined 
that  it  almost  compels  us  to  assume  the 
existence  of  a  more  skilful  class  of  crafts- 
men for  aristocratic  patrons.  An  exami- 
nation of  the  royal  stelae  justifies  this 
hypothesis.  They  are  regular  in  form, 
rounded  at  the  top  either  in  a  semi-circle, 
or  a  slightly  curved  outline,  in  imitation, 
I  believe,  of  the  vaulted  chambers  or  corri- 
dors found  in  the  hypogea  of  the  Said. 
As  they  were  set  up  on  the  summit  of 
20 


FIG.  35. — STELE  OF  QA-AU. 
(Museum,    Cairo.) 
(Phot.  E.  Brugsdi.) 


THINITE  ART 


the  tombs,  they  both  simulated  the  entrance  door  of  the  sepulchral 
vault,  and  served  as  indications  to  posterity  of  the  identity  of 
the  Pharaohs  buried  behind  them.  Thus  on  all  of  them  we  find 
the  same  symbol,  the  falcon  of  Horus,  perched  upon  the 
conventional  plan  of  the  eternal  abode,  a  rectangle,  the  lower 
division  of  which  simulated  the  facade  or  doorway  of  a  house, 
while  the  upper  space  bore  the  name  of  the  inhabitant.  The 
Qa-au  stele  (Fig.  35)  suggests  the 
sculptor's'  struggle  with  a  dark  gray 
schist  over  which  his  triumph  was  not 
altogether  complete;  the  outline  of 
the  falcon  is  carved  with  amazing 
precision;  and  its  specific  character- 
istics, the  roundness  of  the  head, 
and  its  attachment  to  the  body,  the 
curve  of  the  back,  the  vigour  of  the 
wing,  the  grip  of  the  claws,  are 
rendered  with  all  the  accuracy  of  a 
naturalist,  but  he  handling  is  still 
somewhat  heavy.  This  is  no  longer 
the  case  in  the  stele  of  King  Serpent 
(Fig.  36).  Here  the  artist  has  mastered 
his  material,  rendering  it  so  supple 
that  we  are  tempted  to  believe  we 
have  before  us  not  the  archaic 
original,  but  a  replica  of  the  time 
of  Seti  I.  If  the  work  is  really  of 
the  First  Dynasty,  as  is  possible,  we 
may  conclude  from  this  solitary  instance 
that  in  the  treatment  of  animals  at 
any  rate,  the  Thinite  masters  had 

attained  a   degree   of   perfection   occasionally  equalled  by  their 
successors,   but  never   surpassed. 

The  same  cannot  be  said  of  the  rare  bas-reliefs  of  Sinai,  nor 
of  those  figured  schist  tablets  the  purpose  of  which  is  still  uncertain, 
though  it  has  been  alternatively  suggested  that  they  were  more 
elegant  variants  of  the  toilet  palettes  of  the  archaic  age,  bases 
for  statuettes  or  divine  emblems,  or  conventionalised  imitations 
of  the  rams'  heads  which  ornamented  the  tombs  of  great  personages. 
The  Sinai  bas-reliefs  prove  that  the  triumphant  attitudes  of  the 
Pharaohs  familiar  to  us  from  later  works  of  art  were  already 
stereotyped  at  this  period,  but  they  have  suffered  so  much  from 
exposure  to  the  weather  that  it  would  be  hazardous  to  pronounce 

21 


FIG.   36.— STELE  OF  KING 

SERPENT. 
(The  Louvre,  Paris.) 


ART  IN  EGYPT 


on  their  style  or  artistic  merit  (Fig.  37).  Some  fifteen  of  these 
tablets,  perfect  or  mutilated,  are  in  the  museums  of  Cairo,  Paris, 
Oxford  and  London.  One  of  the  best  preserved  of  these ,  that 
of  Pharaoh  Neter-baiu  (Betchau),  bears  on  its  summit  on  either 
face  the  royal  name  enclosed  in  its  triangular  cage  and  flanked 
by  two  human  heads  of  Hathor,  with  the  ears  and  horns  of  a 
cow,  the  latter  very  much  twisted.  The  principal  face  is  divided 
into  three  tiers  (rig.  38).  In  the  first,  Pharaoh  (on  the  left), 
crowned  with  the  red  diadem,  clad  in  the  short  skirt  from  which 

hangs  the   fox-tail,  his  feet  bare, 

the  scourge  and  club  in  his  hands, 
advances,  followed  by  his  groom 
bearing  vase  and  sandals,  and 
preceded  by  four  little  figures  to 
whom  the  standards  of  the  four 
quarters  of  the  world  have  been 
confided,  towards  two  rows  of 
the  corpses  of  his  enemies,  on 
the  right;  they  are  laid  flat  on 
the  ground,  in  fives,  their  wrists 
loosely  bound  with  a  cord,  their 
heads  neatly  arranged  between 
their  legs  after  the  Oriental  fashion. 
The  central  compartment  is  occu- 
pied by  two  leopards,  confronting 
each  other;  their  necks,  which  are 
extravagantly  elongated,  curve  and 
interlace  round  the  central  hollow, 
and  are  crowned  by  grimacing 
heads  opposed  one  to  the  other;  their  two  keepers,  in  short 
skirts,  round  wigs,  and  pointed  beards,  strain  on  their  leashes  to 

Erevent  them  from  biting  each  other.  At  the  bottom,  a  sturdy 
ull,  the  symbol  of  the  Pharaohs,  demolishes  with  his  horns  a 
brick  fortress,  and  tramples  on  a  naked  barbarian  who  tries  in  vain 
to  escape.  The  reverse  (Fig.  39)  has  only  two  compartments,  instead 
of  the  three  of  the  principal  face.  In  the  centre  Baiu,  this 
time  mitred  with  the  high  white  cap  and  escorted  only  by 
his  groom,  strikes  down  with  his  club  a  chief  crouching  on  the 
ground  before  him,  and  surmounted  by  a  strange  group:  the 
hieroglyph  of  a  papyrus  marsh  from  which  the  head  of  a  man 
emerges,  and  a  falcon  poised  with  one  foot  on  three  of  the 
stems;  with  the  other,  which  terminates  in  the  arm  and  hand  of 
a  man,  the  bird  holds  a  cord  passed  through  the  nose  of  the 

22 


FIG.   37.— BAS-RELIEF  AT  SINAI. 
(Phot.  Petrie.) 


THINITE  ART 


head;  the  meaning  of  the  whole  is  that  the  god  Horus  delivers 
six  thousand  Northern  prisoners  into  the  hands  of  the  king.  Two 
naked  figures,  running  at  their  utmost  speed,  represent  the  rest 
of  the  defeated  tribes  and  their  flight.  On  the  other  tablets 
episodes  of  war  and  of  the  chase  are  represented,  lists  of  towns 
taken  by  sap,  troops  of  domestic  animals,  oxen,  asses,  sheep, 
goats,  birds,  advancing  in  superposed  rows  towards  a  wood 
(Fig.  40).  Though  there  are  differences  in  these  works  due  to  the 
individuals  who  executed  them,  they  are  all  marked  by  a  real 
sense  of  composition  and  design,  and 
by  thorough  familiarity  with  the  tool 
used,  but  also  by  a  stiffness  and  awk- 
wardness of  which  there  is  no  trace 
in  the  stele  of  King  Serpent;  they 
belong  rather  to  industrial  art  than 
to  Artfpure  and  simple.  Yet  they 
are  interesting,  for  in  them  we  may 
discern  the  chief  characteristics  of 
the  great  sculpture  of  later  ages,  the 
systematic  deformation  of  the  human 
figure,  the  bust  and  eyes  confronting 
the  spectator  while  the  head  and  legs 
are  in  profile,  the  dry  and  angular 
rendering  of  the  shoulder  and  the 
arm,  the  stiff,  almost  benumbed  bearing 
of  many  of  the  persons,  and  at  the 
same  time,  their  gravity  and  their 
purity  of  line,  the  truth  and  spirit  of 
some  of  their  movements,  the  firmness 
of  the  modelling  and  its  learned 
simplicity,  the  systematic  practice  of  keeping  the  relief  low,  and 
of  indicating  the  planes  by  light  touches.  All  this  is  purely 
Egyptian,  without  any  foreign  admixture. 

.  It  would  seem  then  that  the  art  of  Egypt,  having  arisen  and 
developed  in  the  centuries  which  preceded  Menes,  reached  its 
consummation  under  his  descendants;  when  the  Memphite  dynasties 
arose,  it  was  already  in  full  possession  of  its  ruling  ideas,  its 
conventions,  its  formulae,  its  technique,  all  the  features  which 
give  it  originality  and  character.  Perhaps  the  progress  of  discovery 
will  encourage  us  some  day  to  enquire  under  what  influences 
it  flourished,  and  what  were  the  vicissitudes  of  its  childhood 
and  youth;  the  scarcity  of  examples  forbids  any  such  enquiry  at 
present.  The  study  of  later  periods,  however,  justifies  the  belief 

23 


FIG.   38.— TABLET    OF    NETER- 

BAIU  (OBVERSE). 
(Museum,  Cairo.)  (Phot.E.Brugsch.) 


ART  IN  EGYPT 

that  from  the  first,  there  was  no  absolute  uniformity  throughout  the 
land;  each  of  the  sovereign  cities  had  its  schools,  where  architecture, 
sculpture,  and  painting  developed  with  a  vigour  proportioned  to  the 
intensity  of  its  political  or  religious  life,  and  the  characteristics  of 
its  art,  once  determined,  persisted  with  no  serious  modifications, 
to  the  last  years  of  Egyptian  civilisation.  The  history  of  these 
schools  has  been  barely  indicated,  and  their  number  is  uncertain; 
but  their  existence  has  been  notified  at  Memphis,  Abydos,  Thebes, 
Hermopolis,  Tanis,  Sais,  several  minor  towns  of  the  Said  or 
the  Delta,  in  Nubia  and  in  Ethiopia,  and  it  is  probable  that 
future  excavations  will  reveal  others.  The  supremacy  which  their 
rank  as  capitals  finally  secured  for  Memphis  and  Thebes,  gave 
their  schools  a  prestige  and  importance  to  which  the  others 
never  attained;  their  works  account  for  over  three-quarters  of  what 
has  been  saved  of  the  artistic  patrimony  of  Egypt,  and  at  present, 
the  history  of  Egyptian  art  is  mainly  the  history  of  their  art. 


FIG. 39. — TABLET  OF  NETER-BAIU 

(REVERSE). 
(Museum, Cairo.)  (Phot.E.Brugsch.) 


24 


THIN1TE  ART 
BIBLIOGRAPHY  TO  CHAPTER  I  --  PART  I 

Prehistoric  ages  and  archaic  period  in  Egypt:  All  that  is  necessary  to  be  known  of  these 
will  be  found  in  J.  Capart,  Les  debuts  de  iArt  en  Egypte,  8vo.  Brussels  1904,  316  p. 

Thinite  Age.  —  The  existence  of  a  Thinite  art  was  first  clearly  demonstrated  by 
G.  Steindorff,  Eine  neue  Art  Agyptischer  Kunst  in  Aegyptiaca,  Festschrift  fiir  Gcorg  Ebers, 
8vo.  Leipzig  1897,  p.  123-146.  But  the  monuments  of  this  art  only  began  to  be  welt 
known  after  the  excavations  of  Amelineau,  Morgan,  Flinders  Petrie  and  Quibell:  Amelineau, 
Les  Nouvelles  Fouilles  d'Abydos,  8vo.  Paris  18%,  47  p.,  1897,  47  p.  1898,  65  p.,  and  4 to 
I  1899,  XXXII-307  p.  and  XLIII  pi.,  II  1902,  XI-326  p.  and  XXIV  pi.,  Ill  1904,  742  p.  and  LII 
pi.,  and  IV  1905,  to  be  read  in  conjunction  with  Le  Tombeau  d'Osiris,  4to.  Paris  1899, 
155  p.  and  6  pi.  —  J.  de  Morgan,  Recherches  sur  les  origines  de  I'Egypte,  II.  Ethnographic 
prehistorique  et  tombeau  royal  de  Negadah,  8vo.  Paris  1897,  IX-396  p.  —  Flinders  Petrie, 
The  Royal  Tombs  of  the  first  Dynasty  (Egypt  Exploration  Fund,  vol.  XVIII),  4to.  London 
1900,  51  p.  and  LXVII  pi.;  The  Royal  Tombs  of  the  earliest  Dynasties  (Egypt  Exploration 
Fund,  vol.  XX),  4to.  London  1901,  VIII,  60  p.  and  LXIII  pi.;  Abydos  (Egypt  Exploration 
Fund,  vol.  XXII-XXIV),  4  to.  London,  I  1902,  60  p.  and  XXX  pi.,  II  1903,  66  p.  and  LXVI 
pi.,  Ill  1904,  60  p.  and  LX  pi.  —  J.  E.  Quibell,  Hierakonpolis  (Egyptian  Research  Account, 
vol.  IV- V),  4  to.  London,  I  1900,  12  p.  and  43  pi.,  II  1902,  55  p.  and  LXXK  pi.  For  military, 
religious  and  funerary  architecture,  see  in  addition  to  the  works  already  quoted:  J.  Garstang, 


Mahasna  and  Bet  Khallaf  (Egyptian  Research  Account,  vol.  VII),  4  to.  London  1902,  42  p. 
andXLIILpl.;  Tombs  of  the  Third  Egyptian  Dynasty,  4  to.  London  1904,  70  p.  and  XXXIII 
pi.  —  G.  A.  Reisner  and  Mace,  The  Early  dynastic  Cemeteries  of  Naga-ed-Deir,  4 to.  Leipzig, 


I  1908,  160  p.  and  75  pi.,  II  1910,  88  p.  and  60  pi.  —  Flinders  Petrie,  The  Development  of 
the  Tomb  in  Egypt,  in  the  Journal  of  the  Royal  Institute  of  Great  Britain  1898,  session 
of  June  3,  and  on  the  form  of  the  temples,  A.  Jequier,  Les  Temples  primitifs  et  la  per- 
sistance  des  types  archaiques  dans  I'architecture  religieuse ,  in  the  Bulletin  de  I'Institut 
francais  d  Archeologie  orientale,  1908,  vol.  VI,  p.  25-45.  For  stelae  and  palettes,  see  the 
articles  of  J.  E.  Quibell,  Slate  Palette  from  Hierakonpolis,  in  the  Zeitschri/t  fiir  Agyptische 
Sprache  1898,  vol.  XXXVI,  p.  81-84.  —  G.  Benedite,  La  Stele  dite  du  Roi  Serpent,  in  the 
Memoires  de  la  Fondation  Piot,  1906,  vol.  XII,  p.  1-15,  and  Une  nouvelle  Palette  en  schiste, 
in  the  same  Memoires  1904,  vol.  X,  p.  105-122.  Very  good  resumes  of  all  we  know  of  the 
art  of  this  period  in  general  are  to  be  found  in  W.  Spjegelberg,  Geschichte  der  Agyptischen 
Kunst,  p.  7-11,  and  in  R.  Weill,  Les  Origines  de  I'Egypte  pharaonique,  8vo.  Paris  1908, 
p.  443-500. 


FIG.  40. — THINITE  TABLET. 
(The  Louvre,  Paris.)   (Phot.  E.  Brugsch.) 

25 


FIG.  41. — FIELD   OF  THE  PYRAMIDS  OF  GIZEH. 


CHAPTER  II 
MEMPHITE  ART 

It  readies  its  apogee  under  the  Fourth,  Fifth  and  Sixth  Dynasties  —  Architecture:  Houses, 
Palaces,  Mastabas  and  Pyramids,  funerary  Chapels  and  Temples  —  Painting  and 
Sculpture:  the  decoration  of  Tombs  and  Temples  considered  as  a  whole.  Bas-reliefs 
and  Statues  —  The  minor  Arts. 

TOWARDS  the  beginning  of  the  Third  Dynasty  there  were,  in 
the  district  where  the  Pyramids  afterwards  rose,  craftsmen 
capable  of  executing  tombs  like  those  of  Nakadah,  or  carving  a 
seated  or  a  standing  figure  of  a  man  more  or  less  passably,  but 
nothing  that  has  survived  of  their  works  indicates  that  their 
school  would  ever  have  risen  above  mediocrity,  if  the  revolution 
which  brought  about  the  transfer  of  the  royal  residence  had  not 
suddenly  brought  it  into  contact  with  experienced  masters.  The 
architects,  masons,  painters  and  sculptors  who  had  worked  for 
the  Thinite  Court  accompanied  it  in  its  migration  towards  the 
North;  Memphite  art  developed  from  their  teaching  or  examples 
as  a  natural  prolongation  of  Thinite  art.  The  first  buildings  we 
owe  to  it  are  grouped,  some  in  the  mining  region  of  Sinai,  but 
the  greater  part  in  the  neighbourhood  of  Medum ,  Dahshur, 
and  Zawyet-el- Aryan ,  round  the  tombs  which  the  last  king  of 
the  Third  Dynasty,  Neferka-Ra-Huni,  and  the  first  king  of  the 
Fourth,  Seneferu,  had  erected  for  themselves.  At  this  period 
of  history  they  were  few  and  far  between,  but  the  number  in- 
creased from  the  time  of  Cheops  onwards;  towards  the  close  of 
the  Memphite  age,  under  the  Sixth  Dynasty,  they  not  only 
covered  all  Egypt,  but  were  to  be  met  with  beyond  the  cata- 

26 


MEMPHITE  ART 


FIG.  42.— SARCOPHAGUS   OF  KHUFU-ENEKH. 
(Museum,  Cairo.)     (Phot.  E.  Brugsch.) 


racts  of  Assuan,  in  the  northern  districts  of  Nubia  over  which 
the  Pharaohs  had  established  their  domination.  Rude  and  clumsy 
at  first,  they  gradually 
improved  under  the  Third 
Dynasty,  and  reached  their 
highest  perfection  under 
the  Fourth;  they  became 
more  and  more  refined, 
but  lost  something  of  their 
characteristic  simplicity 
and  grandeur  under  the 
Fifth  Dynasty.  Under  the 
Sixth  Dynasty,  the  deca- 
dence had  begun;  the 
little  that  has  come  down 
to  us  from  the  following 
dynasties  betrays  the  hand 
of  the  unskilful  and  unin- 
telligent artisan. 

yThe  architecture  is  known  to  us  mainly  by  the  tombs.  The 
private  houses,  built  of  dried  bricks,  and  perpetually  modified 
or  replaced  for  the  convenience  of  their  inhabitants,  survive 
only  in  shapeless  pieces  of  wall  in  the  deeper  strata  of  the 
existing  towns.  The  palaces,  also  of  brick,  though  they  had  certain 
stone  elements  in  their  doorways  and  internal  colonnades,  have 
proved  hardly  more  dur- 
able. To  judge  by  the 
external  arrangement  of 
the  sarcophagi  of  Khu- 
fu-enekh  (Fig.  42)  and  My- 
cerinus  (Fig.  43),  they  were 
rectangular  masses  with 
vertical  walls  sometimes 
encircled  by  a  beaded 
torus,  and  crowned  by  a 
deep  cavetto.  The  fronts 
were  divided  into  grooved 
panels  like  those  of  the 
fortresses,  and  of  the  tomb 


FIG.  43. — SARCOPHAGUS  OF  MYCERINUS. 
(After  Chipiez,  Hist,  de  1'Art.  vol.  i.  fig.  289). 


of  Nakadah  (cf.    Fig.   6), 
but    more     elaborate     in 
profile,   and  they  were  decorated  towards  the  top  by  an  orna- 
ment of  two  lotus-leaves  with  crossed  stems;  doors  were  pierced 

27 


ART  IN  EGYPT 


FIG.  44. — FACADE  AND  DOOR  OF  A  MEM- 

PHITE  HOUSE,  FROM  THE  STELE  OF  SETI. 

(Museum,  Cairo.)     (Phot.  E.  Brugsch.)  . 


between  the  panels,  and  above 
them  openwork  bays,  or  rows 
of  little  slits  through  which  the 
light  and  air  entered  (Fig.  44). 
The  whole  was  whitewashed, 
and  the  architectural  details 
were  enlivened  with  crude 
colours:  sphinxes  with  lions' 
bodies  and  human  heads  often 
watched  on  either  side  of  the 
door,  or  obelisks,  stones  rising 
from  a  square  base  into  a 
pyramidal  point,  took  their 
place,  proclaiming  the  names 
and  titles  of  the  master.  The 
facades  of  private  houses  were 
probably  similar,  or  at  least 
those  which  belonged  to  per- 
sons of  distinction,  and  had 
any  pretensions  to  elegance. 
The  appearance  of  the  streets  in  certain  African  towns  (Fig.  45),  where 
the  decoration  is  in  mud  or  clay  like  the  houses,  may  give  an  idea 
of  the  rich  quarters  of  Memphis  at  the  time  of  the  Pyramids.  As 
to  the  temples  of  the  city,  which  were  enlarged  or  remodelled  from 

reign  to  reign,  then  pulled  down 
because  of  their  age,  and  set 
up  again  on  new  plans,  all  that 
is  left  of  them  consists  of  frag- 
ments, carved  or  inscribed, 
which  have  been  utilised  in 
buildings  of  recent  date.  We 
should  still  be  ignorant  of  their 
origin,  had  it  not  been  for  the 
fortunate  discovery  of  certain 
funerary  temples  which  were 
attached  to  the  royal  pyramids 
of  the  Fifth  Dynasty;  once  again 
the  fictitious  life  beyond  the 
tomb  has  provided  the  document 
necessary  for  the  reconstrtutidn 
of  real  life. 

The   burial   grounds    of '  the 
Memphite    mountains    contain 


FIG.  45.— AFRICAN  MUD  ARCHITECTURE. 
A  Street  in  Dicnne. 


28 


MEMPHITE  ART 


FIG.  46. — VERTICAL  SHAFT  IN  THE  MAS- 
TABAS   OF  GIZEH  (AFTER  LEPSIUS). 


several  hypogea,    both   vaults  and   chapels,    which    are   entirely 
hollowed   out   in  the  rock;   other  sepulchres    approximate  to  the 
Thinite    type,    but   the    propor- 
tion    between      the      elements 
demanded    by  the   earlier   con- 
ception was  no  longer  observed, 
and    the    internal    arrangements 
were  accordingly  modified.    In- 
deed,   as    the    doctrine   gained 
ground,  according  to  which  the 
images   traced    upon    the    wall 
were    of    equal,     or    even    of 
greater  importance  to  the  dead 
than    real    objects,    the    tomb- 
chamber  was  circumscribed,  and 
the  rooms  composing  the  accessi- 
ble chapel  were  increased.     At 
Memphis  accordingly  the  tomb- 
chamber  is  merely  a  narrow  cell,  more  or  less  deep  beneath  the 
ground,    accessible   until   the  day  of  interment   by  means   of   a 
vertical   well  (Fig.  46),    or  an  oblique  passage  (Fig.  47),  without 
any  decoration  in  the  way  of  figures  or  inscriptions  save  such 
as   were  bestowed    on   the   sarco- 
phagus.    On  the  other  hand,   the 
chapel,    which   had   become    both 
a  reception-room  and  a  storehouse, 
forms  a  building  of  some  importance, 
a  mastaba,  the  visible  bulk  of  which 
was    in    direct  ratio  to   the  means 
possessed  by  the  master  for  ensuring 
a  happy  after-life  to  his  double.  Thus 
it  was  not  open  to  everyone  to  rest 
under  a  mastaba.   It  was  a  privilege 
reserved    for    those    whom    birth, 
talents,    services    rendered   to   the 
state,    or    even    some    momentary 
caprice   had  raised  to   the  summit 
of  the  hierarchy.    As  they  had  been 
permitted   to  approach  the  master 
here   below,    they   desired   not  to 
be  separated  from  him  in  the  other 
world,  that  so  they  might  continue  to  enjoy  his  favour.    However, 
as   they   were  numerous,    and   space  was  limited,    Pharaoh   was 

29 


f  r—  -.- 


FIG.  47.— THE   SLOPING  PASSAGE 

IN  THE   TOMB   OF  TI,   DRAWN  BY 

A.   BAUDRY. 


ART  IN  EGYPT 


S. — A  •  CORNER    OF    THE    NECROPOLIS    AT 
GIZEH,    RESTORED    BY    CHIPIEZ. 

(Hist,  de  1'Art,  vol.  i.  i\g.  108.) 


obliged  to  allot  it  with  discretion,  if  he  wished  to  satisfy  his 
courtiers.  Concessions  of  ground  were  made  methodically,  on  a 

predetermined  plan ,  and 
the  mastabas  were  ranged 
in  regular  lines  (Fig.  48), 
the  larger  ones  divided 
one  from  another  by  lanes, 
the  smaller  combined  into 
islets  of  two,  three,  or  more ; 
when  his  hour  had  come, 
Pharaoh  had  distributed 
several  hundreds  of  these, 
which  formed  a  city  of  the 
dead  around  him.  Its 
appearance  was  monoto- 
nous. These  houses,  or, 
if  we  prefer  to  call  them 
so,  these  palaces  of  the 

necropolis  would  have  been  much  like  those  of  living  cities,  if 
their  facades,  instead  of  being  straight,  had  not  inclined  symmetrically 
backwards,  which  gave  them  a  certain  vague  likeness  to  an 
unfinished  pyramid  (Fig.  49).  Some  few  of  these  were  of  sun-dried 
brick;  the  majority  were  of  freestone,  or  small  dressed  stones, 
with  bare  plain  facings,  the  door  on  the  east  or  the  north,  and 
in  some  cases,  a  row  of  apertures  just  below  the  line  of  the 
summit.  Some  were  from  thirty  to  forty  feet  high,  one  hundred 
and  fifty  feet  in  width,  and  about  seventy-five  in  depth; 
but  this  was  not  usual,  and  examples  occur  no  more  than  about 
nine  feet  high  by  fifteen  wide.  Some  are  crowned  by  a  cavetto 
and  an  entablature;  but  the  majority  terminate,  without  any 

transition  in  the  last 
course,  in  an  earthen 
platform ,  the  soil 
mixed  with  fragments 
of  limestone ,  and 
dotted  with  terracotta 
jars,  buried  up  to  the 
neck.  To  tell  the  truth, 
I  see  in  them  the 
regularisation  and  con- 
solidation of  those 
heaps  of  sand  and  pebbles  which  the  primitive  Egyptians  piled  over 
their  graves;  the  architect  had  little  to  do  with  their  actual  form; 

30 


FIG.  4Q. — A  MASTABA  AT  GIZEH  (After  Lepsius). 


MEMPHITE  ART 


tradition  had  imposed  it  upon  him,  and  he  was  obliged  to  repeat 
it  servilely  on  the  outside. 

But  there  was  compensation  in  the  amount  of  liberty  permit- 
ted him  in  the  interior.  In  the  beginning,  the  Memphite  mastaba 
had  been  solid  like  the 
tumulus,  whether  it  was 
built  up  over  the  vault,  or 
constructed  beforehand,  and 
pierced  with  a  tunnel  which 
was  filled  up  on  the  eve- 
ning or  the  morrow  of  inter- 
ment; in  either  case  the 
architect  was  careful  to 
indicate  on  the  eastern  front, 
by  means  of  a  panel  simu- 
lating a  door,  the  place 
where  the  double  was  sup- 
posed to  go  out  and  return 
(Fig.  50).  This  feigned  en- 
trance was  often  of  natural 
size ,  and  it  would  have 
resembled  a  practicable  door 
in  every  way,  but  for  the 
fact  that  the  back  was 
always  closed.  It  was  some- 
times doubled,  at  first 
only  for  the  king,  but  later, 
when  private  persons  ven- 
tured to  imitate  the  king, 
for  those  noble  or  wealthy 
individuals  who  were  con- 
cerned that  their  souls 
should  not  lose  any  of  the 
offering;  one  of  the  two 
was  dedicated  to  the  north 
and  its  tribute,  the  other 

to  the  south,  and  its  productions.  The  decoration  was  sober 
at  first:  the  name  of  the  master  on  the  tympanum  over  the 
opening,  his  titles  and  image  on  the  jambs  sufficed  as  long 
as  clients  were  content  with  a  plain  door.  But  very  soon,  pursu- 
ing the  imitation  of  what  had  at  first  been  the  privilege  of  the 
sovereign,  they  required  that  the  slab  should  represent  not  merely 
the  door  of  a  dwelling,  but  the  entire  building,  and  the  model 

31 


FIG.  50. — STELE  IN  THE  FORM  OF  A  FALSE 
DOOR.     (Museum,  Cairo.)     (Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


adopted  was  what  is  known  as  the  royal  banner,  in  other  words, 
the  rectangular  structure  in  which  the  Pharaohs  enclosed  their 
name,  Hortis.  It  consisted  of  two  parts;  in  the  lower  of  these 
was  the  facade  of  a  house  with  a  closed  door,  in  the  upper 
one  an  empty  space,  a  chamber  in  which  the  signs  which  constituted 
the  name  were  written  (Fig.  51).  In  imitation  of  this,  the 
slab  was  divided  into  two  registers,  one  above  the  other,  enclosed 
in  a  flat  band  which  formed  a  frame  common 
to  both  (Fig.  52).  The  lower  compartment 
answered  to  the  false  door  of  the  earlier  period, 
often  so  modified  as  to  be  almost  unrecog- 
nisable. The  panels  of  the  rebates  were  brought 
forward,  the  jambs  were  flattened,  and  the 
reliefs  as  a  whole  were  only  a  few  millimetres 
above  the  surface.  In  the  upper  compartment, 
which  corresponds  to  the  tomb  -  chamber ,  the 
dead  man  was  seated  at  a  round  table,  laden 
with  the  foods  and  ornaments  he  might  require 
in  the  other  world.  These  were  conveyed  to  him 
invisibly  by  means  of  a  special  apparatus,  a 
stone  table  originally  fixed  between  the  uprights 
of  the  door,  and  afterwards  placed  on  the  ground, 
against  the  stele.  The  celebrant  heaped  on  this 
all  the  objects  of  offering,  and  the  doubles  of  these, 
detached  by  virtue  of  his  prayers,  were  projected 
upon  the  round  table  destined  to  receive  them. 
This  ritual  of  the  dead  was  carried  out  in  the 
open  air  (Fig.  53),  in  the  sight  of  all,  and  though 
in  theory  this  unrestricted  publicity  did  not 
affect  its  efficacy,  in  practice  the  result  was, 
that  when  the  congregation  had  dispersed,  the 
offerings  were  at  the  mercy  of  marauders,  human  and  animal;  the 
person  for  whom  they  were  intended  ran  the  risk  of  losing  the 
best  part  of  them  before  he  had  secured  his  ration.  Two  de- 
vices were  accordingly  adopted  for  their  protection:  an  enclosure 
of  bricks  was  built,  projecting  from  the  east  wall,  square  in  the 
mastaba  of  Kaapiru  (Fig.  54),  irregular  in  that  of  Neferhetep 
(Fig.  55),  at  Sakkarah;  but  the  more  important  measure  was 
the  imbedding  of  the  false  door  in  the  masonry  in  such  a  manner 
as  to  bring  it  to  the  back,  sometimes  of  a  niche,  sometimes  of 
an  actual  room.  This  was  very  often  unique,  and  so  small  that 
it  looks  to  us  drowned  in  the  general  mass.  It  is  a  minute  cell, 
the  longer  axis  of  which  is  parallel  to  the  facade;  if  the  false  door 

32 


FIG.  51. 

THE  NAME  HORUS 
OF  CHEPHREN. 


MEMPHITE  ART 


be  placed  at  one  of  the  extremities,  the  ground-plan  forms  a  figure 
like  a  double-headed  hammer  (Fig.  56) ;  if  it  be  hollowed  out  opposite 
the  entrance,  it  suggests  a  cross  the  head  of  which  is  cut  out 
more  or  less.  Such  simple  arrangements  are  found  principally 
in  the  more  archaic  quarters,  such  as 
Dahshur,  Medum,  Gizeh  and  Sakkarah, 
side  by  side  with  more  complex  types. 
In  the  latter,  the  single  chapel  was 
first  enlarged,  then  doubled,  and  re- 
doubled (Fig.  57),  until  the  mastaba 
became  a  series  or  a  labyrinth  of  rooms 
large  and  small:  that  of  Mereruka, 
under  the  Sixth  Dynasty,  contained 
over  thirty  compartments  (Fig.  58);  some 
of  these  were  passages  concealed  in 
the  thickness  of  the  structure,  sometimes 
blind,  sometimes  communicating  with 
the  world  by  conduits  so  narrow  that 
it  is  difficult  to  thrust  the  hand  into 
them;  these  were  the  serdabs,  in  which 
the  statues  of  the  deceased  and  of  his 
servants  were  imprisoned,  to  preserve 
them  from  possible  destruction.  Several 
were  used  as  warehouses  or  store-rooms, 
and  for  ceremonies,  there  were  rooms 
upheld  by  square  piers,  or  by  columns 
with  lotus-bud  capitals;  the  entrance  is 
sometimes  preceded  by  a  porti  co  (Fig.  59). 
In  the  course  of  time,  the  false  door 
lost  its  original  character;  its  hollows 
were  attenuated,  its  projections  flattened, 
and  it  was  finally  resolved  into  an 
upright  slab,  on  which  the  design  of  a 
door  was  indicated  on  the  surface  by 
almost  impalpable  reliefs;  in  a  word,  it 
became  a  stele,  towards  which  all  the  several  parts  of  the  tomb 
converged,  just  as  if  it  had  remained  the  actual  door  which 
had  formerly  led  to  the  vault.  Occasionally,  however,  its  ancient 
character  was  revived,  at  least  in  appearance.  Thus  in  the  tomb 
of  Mereruka  (Fig.  60),  the  life-size  statue  of  the  master  was 
introduced  into  the  bay,  a  flight  of  three  steps  was  at  his  feet, 
by  which  he  was  supposed  to  come  down  into  the  chamber, 
to  take  the  offerings  left  by  the  celebrants.  In  the  mastaba  of 

33  D 


FIG.     52. — STELE-DOOR     OF 

USIRU.     (Museum,  Cairo.) 

(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


Atoti  (Fig.  61),  the  statue  is  set  against  the  stone  which  fills  up 
the    doorway,    rather   than   enframed    in    the   bay.      In   that    of 

Neferseshemptah  (Fig.  62),  the 
conception  is  more  complex; 
the  bust  of  the  dead  man  rises 
above  the  closed  door  and  its 
lintel,  to  see  what  is  happening 
in  the  chamber,  while  right  and 
left  two  statues  of  him  stand 
erect  against  the  facade,  as  if 

FIG.  53.-FA<:ADK  OF  THE  MASTABA  OF     keeping  guard  over  him.   Later 
MENEFER.    (After  Mariette.)  again,    the  head   alone  appears 

over   the   panel    (Fig.  63).    The 

stele  thus  loses  its  independent  character  to  become  a  mere 
element  of  the  decoration  (Fig.  64).  The  table  of  offering  rests 
on  the  ground  before  it,  and  sometimes,  as  among  the  living, 
it  was  flanked  by  two  miniature  obelisks,  on  which  the  name 
and  titles  of  the  master  were  proclaimed  in  large  letters  (Fig.  65). 
There  is  no  evidence  that  this  evolution  was  based  on  a  pre- 
conceived idea,  nor  that  it  culminated  in  the  creation  of  a  typi- 
cal mastaba,  all  the  internal  parts  of  which  were  deduced  one 
from  another  in  logical  order.  In  the  mastaba  of  Ti,  at  Sak- 
karah  (Fig.  66)  there  is,  indeed,  a  veritable  progression  in  the 
successive  apartments,  from  the  entrance  portico  to  the  point 
where  the  stele  rises  towards  the  south-western  angle:  first 
there  is  a  hall  with  pillars,  where  the  passage  leading  up  from  the 

vault  reached  the  level, 
then  a  corridor  divided 
by  a  door  into  two  une- 
qual lengths,  a  little  room 
on  the  right  for  the  dead 
man's  wife,  at  the  end  the 
chapel  with  its  two  stelae, 
and,  parallel  with  its  south 
wall ,  a  serdab  in  which 
the  statues  were  conceal- 


ed.    But  this  was  an  ex- 
ception.   In  nearly  every 
other  instance,  the  arch- 
itect did  not   trouble  to 
arrange  the  rooms  metho- 
dically, provided  he  placed  his  chapel  as  far  west  as  possible;  he 
was  only  concerned  to  increase  their  number,  and  consequently, 

34 


FIG.   54.— CELL  AND   FORE-COURT  OF  THE 
MASTABA  OF  KAAPIRU      (After  Mariette.) 


MEMPHITE  ART 


to  develope  to  the  utmost  surfaces  capable  of  receiving  decoration. 

In  this  he  was  influenced  by  those  utilitarian  principles  which  had 

regulated  the  arrangement  of  the 

tomb  from  the  first;    for  as  the 

decorator  was  thus  free  to  repeat 

the  principal    scenes    in    several 

rooms,   if  one  was  defaced,  and 

lost  its  efficacy,  the  replica  took 

its  place  and  continued  to  supply 

the  deceased   with  his  revenues. 

Towards  the  end  of  the  Memphite 

age,  he  further  realised  that  the 

tomb-chamber,  buried  beneath  the 

mastaba  and  separated  from  it  by 

a  conglomerate  cement,    offered 

a  greater  chance   of  inviolability 

than  the  super-structure,  and  he  conceived  the  idea  of  laying  up 

a  reserve  of  pictures  here,  in  case  those  above  should  fail.     He 

accordingly  devised  a  new   model    (Fig.  67),    good  examples  of 

which  are  to  be  found  among  the    brick   tombs  discovered  from 

1881  onwards  about  the  Pyramid  of  Pepi  II.    Built  of  dried  bricks, 

upon   the   sand   itself,    this   mastaba,    like    those   of  the  earliest 

period,   at  first  presents   to   the  spectator  either  a  stele  on  the 

west  face,    or  a  niche   in   front   of   which  the  family  assembled 

to    offer  sacrifice.     In  the  interior,  the  well  was  replaced  by  a 

sort  of  court,  in  the  western  part  of  which    a  place  was  reserved 

for  the  tomb-chamber.    This  consisted  of  a  long,  low  cell,  formed 

of  five  limestone  slabs;   a  brick  vault,  with  a  radius  of  from  20 


FIG.  55.— FORE-COURT  OF  THE  MASTABA 
OF  NEFER-HETEP.     (After  Marietta.) 


FIG.gO. — THE  MASTABA  OF  ZAZAMENEKH. 
(After  Mariette.) 


FIG.  57. — THE  MASTABA 
OF  KHABEUPTAH.      (After  Mariette.) 


to  24  inches,  relieved  the  upper  slab  of  the  weight  of  the 
successive  strata  rising  above  it  to  the  level  of  the  terminal  plat- 
form. The  lateral  walls,  carved  and  painted,  were  receptacles 

35  D2 


ART  IN  EGYPT 


FIG.   58.— THE  MASTABA    OF 
MERERUKA.     (After  Morgan.) 


for  articles  of  food  and  clothing;  the  coffin  was  pushed  along  be- 
tween them,  then  the  opening  in  front  was  walled   up,  and  the 

little  court  was  filled  in.  When  the 
dead  man  had  a  stone  sarcophagus, 
instead  of  a  wooden  coffin,  the  whole 
building  was  placed  upon  this,  as  of- 
fering a  more  solid  foundation;  the  lid 
formed  the  floor  of  the  chamber  above, 
and  on  this  statues  and  offerings  were 
heaped  pell-mell  before  the  mummy  was 
consigned  to  his  eternal  rest. 

We  find  then  that  the  Memphite 
mastaba,  in  its  final  development,  allow- 
ed of  two  decorated  elements,  the 
chapel  and  the  vault,  which  were  con- 
nected by  a  well,  or  an  unornamented 
passage.  It  was  almost  universally 
adopted  among  the  wealthier  classes, 
from  the  Fayum  to  the  centre  of  the 
Delta,  but  it  never  became  general  in 
districts  where  the  court  did  not 
habitually  reside.  The  prevalent  form 
in  Upper  Egypt,  in  the  places  where  vast  sandy  spaces  at  the 
foot  of  the  mountain-boundaries  invited  the  great  nobles  to 
construct  mastabas,  was  derived  both  from  the  Thinite  and 
Memphite  type,  but  was  more  akin  to  the  first  than  to  the 

second.  It  is  partly  buried 
in  the  ground,  as  at  El- 
Kab  or  Denderah  (Fig.  68), 
and  access  to  it  is  ob- 
tained by  an  inclined  plane, 
or  by  a  staircase;  the 
corpse  is  buried  not  very 
deep  below  the  soil  in  a 
cavity  more  like  a  pit  than 
a  chamber.  Preference 


was  given  to  the  hypogeum 
hewn  in  the  mountain 
side,  in  a  vein  of  lime 
or  sand  stone,  solid  and 
fine,  running  horizontally 

at  a  good  height  above  the  plain.    When  such  a  spot  had  been 
selected,    the  members  of  a  family  and   their  servants  were  laid 

36 


FIG.  59. — PORTICO  OF  THE  MASTABA  OF  TI. 


MEMPHITE  ART 


FIG.  60.— FALSE  DOOR  OF 
MERERUKA.     (Phot.  E.  Brugsch.) 


there  from  generation  to  generation, 
the  mighty  side  by  side  in  a  row 
on  the  same  level,  the  plebeians 
haphazard  on  the  slope,  in  front 
of,  and  so  to  speak,  below  them, 
as  if  to  maintain  the  hierarchical 
distance  even  in  death.  The  plan 
necessitated  the  three  divisions  of 
the  mastaba,  the  vault,  the  well 
or  sloping  passage,  and  the  chapel; 
but  the  material  in  which  the  archi- 
tect had  to  cut  them  obliged  him 
to  modify  the  detail ,  at  least  as 
far  as  the  chapel  was  concerned. 
As  the  expense  would  have  been 
very  great  if  he  had  carried  one 
of  those  series  of  chambers  such 
as  we  find  in  the  mastaba  of 
Mereruka  at  Memphis  right  through 
the  living  rock,  he  was  nearly  al- 
ways content  with  a  single  room  for  receptions  and  worship; 
when  more  were  demanded,  they  were  rarely  more  than  two  or 
three  in  number.  In  the  mastaba  of  Zauti,  at  Kasr-es-sayad,  we 
find  surbased  vaults;  elsewhere,  and  indeed  generally,  the  ceiling 
was  flat,  like  that  of  the  stone  mastabas. 
Where  the  orientation  allowed  it,  the 
stele  was  placed  opposite  the  en- 
trance, in  a  niche  cut  towards  the 
centre  of  the  back  wall;  if  this  was 
impracticable  it  was  carved  on  the 
west  wall,  or  set  against  it,  and  the 
niche  served  as  serdab  for  the  statues 
of  the  double,  these  being  either 
separate,  or  cut  in  the  mass  of  the 
rock.  The  hypogeum  thus  carried 
out  became  a  systematised  fragment 
of  a  quarry ,  as  at  Kom  -  el  - Ahmar, 
Koseir-el-Amarna,  Meir,  near  Akhmim, 
Der-el-Melak,  and  at  Kasr-es-sayad. 
Sometimes,  however,  an  artistic  in- 
tention is  revealed.  That  of  Afai, 
at  Kau-el-Kebir,  has  ceilings  the  reliefs 
of  which  imitate  the  palm  -  trunks 

37 


FIG.  6l.— FALSE  DOOR  OF  ATOTI. 
(Museum  Cairo).  (Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


which  supported  the  roofs  of  houses.  The  chapels  of  the  princes 
of  Elephantine  are  veritable  temple  halls ;  that  of  Mekhu  (Fig.  69), 
with  its  triple  row  of  six  columns,  its  niche,  and  its  stele  to  which 
a  little  staircase  gives  access,  its  stone  three-legged  table  of  offerings, 
standing  between  two  of  the  columns  of  the  central  aisle;  that  of 
Sabni  with  its  fourteen  square  pillars  in  two  rows,  and  its  narrow 
door,  interrupted  at  about  one  third  of  its  height  by  a  triangular 
lintel  which  crowns  a  smaller  orifice  in  the  monumental  door. 
All  this  is  rude,  ill-proportioned  and  barbarous  compared  with 

what  we  see  at  Memphis; 
but  the  excellence  of  the 
site  shows  in  those  who 
selected  it  a  delicate  feel- 
ing for  nature,  and  makes 
amends  to  some  extent 
for  their  artistic  short- 
comings. From  the  terrace 
which  runs  along  the  storey 
reserved  for  the  patricians, 
the  eye  travels  freely  to 
the  horizon ;  the  double, 
escaping  from  the  dark- 
ness of  the  vault,  was  able 
to  take  in  at  a  glance  the 
whole  expanse  of  the  do- 

FIG.62.— FALSE  DOOR   OF  NEFERSESHEM-PTAH.  •      U      U     J        1     J    J.U 

(Phot.  Abbe  Thedenat.)  mam  he  had  rulea.  the great 

river     with     its    hurrying 

waters,  its  changing  islands,  the  open  country  invaded  by  the 
sands  of  the  desert,  the  villages  among  the  palm-trees,  and  in 
the  distance,  the  mountains  to  which  he  had  so  often  been  lured 
by  the  pleasures  of  the  chase. 

The  tombs  of  the  Memphite  Pharaohs  were  also  so  placed  as 
to  command  the  valley  from  afar.  They  stretch  out  in  a  line 
on  the  edge  of  the  Libyan  desert,  and  succeeding  one  another 
from  Abu-Roash  to  Gizeh,  from  Gizeh  to  Sakkarah,  from 
Sakkara  to  Dahshur  and  thence  to  Medum ,  they  pursue  the 
traveller  who  is  going  up  the  Nile  for  days  together.  Zoser  of 
the  Third  Dynasty,  in  his  character  of  King  of  Upper  Egypt, 
possessed  a  huge  brick  mastaba  of  the  Thinite  type  at  Bet- 
Khallaf  (cf.  Fig.  12);  as  King  of  northern  Egypt,  he  built  him- 
self a  second  tomb  of  a  novel  type,  at  least  to  us  (Fig.  70). 
The  apartments  of  the  double  are  here  cut  in  the  rock  on  a 
system  similar  to  those  of  Bet-Khallaf,  but  much  more  com- 

38 


MEMPHITE  ART 


plicated  (Fig.  71);  it  is  entered  by  no  less  than  four  doors,  the 
principal  one  being  on  the  north.  After  traversing  a  labyrinth 
of  passages,  low  chambers,  and  hypostyle  galleries,  we  find 
ourselves  on  the  brink  of  a  central  well,  at  the  bottom  of  which 
is  a  pit,  sealed  with  a  kind  of  stone  stopper;  this  was  doubtless 
the  receptacle  for  the  funerary  treasure.  The  superstructures 
are  not  of  brick,  but  of  the 
coarse  lime-stone  of  the 
surrounding  mountains, 
and  it  was  perhaps  this 
fact  which  gave  rise  to  the 
legend  that  Zoser  was  the 
first  Egyptian  who  built 
with  stone.  The  base  on 
which  these  upper  build- 
ings rest  is  a  parallelo- 
gram of  about  390  by  350 
feet,  and  their  appearance 
would  lead  us  to  suppose 
that  they  were  composed 
of  five  blocks  of  masonry 
with  sloping  surfaces,  each 
receding  some  6'/2  feet 
from  the  lower  stage,  and 
so  diminishing  gradually 
as  they  rise  to  terminate 
in  a  platform  about  190 
feet  above  the  level  of  the 
ground.  But  this  appear- 
ance is  deceptive.  They 
are  not  mastabas  of  decre- 
asing dimensions  piled  one 

above  the  other;  the  core  was  raised  uninterruptedly,  and  then 
dressed  on  its  four  sides  with  parallel  courses  of  masonry,  which, 
ceasing  four  times  at  different  levels,  formed  the  four  successive 
storeys.  The  monument  is  known  at  present  as  the  Step  Pyramid, 
but  it  is  not  a  true  Pyramid.  When  he  ordered  his  architects 
to  undertake  it,  Zoser  wanted  a  tomb  superior  to  those  of  his 
predecessors,  and  even  to  the  one  he  was  preparing  at  Bet- 
Khallaf,  on  the  ancient  plan.  Now  the  slopes  of  the  mastaba 
approached  too  closely  to  the  perpendicular  to  allow  of  bringing 
them  up  to  the  required  height  without  risking  the  downfall  of 
the  whole,  when  the  small  blocks  of  dressed  stone  which  were  in 

39 


FIG.   63. — STELE-DOOR    OF  NIBERA 
(Museum,    Cairo.)     (Phot.    E.    Brugsch.) 


ART  IN  EGYPT 


general  use  at  this  period  were  used.  It  was  to  obviate  this 
danger  that  they  were  compressed  between  the  four  graduated 
facings;  these,  buttressing  them  and  each  other,  ensured  the 
stability  of  the  whole.  This  elevated  and  reinforced  variant  of 
the  mastaba  continued  in  favour  for  several  generations.  Nearly  a 

century  and  a  half  after 
Zoser,  Seneferu,  the  first 
sovereign  of  the  Fourth 
Dynasty ,  still  retained 
it.  The  false  Pyramid  at 
Medum  (Fig.  72) ,  where 
he  had  taken  up  his  resi- 
dence as  King  of  Upper 
Egypt ,  is ,  in  fact ,  com- 
posed like  the  so-called 
Step  Pyramid,  of  vast 
cubes  of  masonry  with 
sloping  faces,  each  slightly 
smaller  than  the  one  be- 
low; they  are,  however, 
square  instead  of  rectan- 
gular, and  are  only  four 
in  number.  Their  progres- 
sive diminution  is  a  pure 
caprice  of  the  architect's, 
no  longer  justified  by  a 
technical  necessity.  The 
core  of  the  fabric  is  not 
artificially  built  up;  it  is  a 
natural  hill,  the  solidity  of 
which  was  beyond  question, 
and  the  masonry  which 
masks  it  consists  of  magnificent  limestone.  The  four  diminishing 
cubes  are  independent  one  of  the  other,  and  it  is  probable  that 
the  last  was  never  finished. 

After  Medum,  the  Pharaohs  built  nothing  but  Pyramids  through- 
hout  the  Memphite  age.  In  my  opinion,  the  pyramid  was  not 
derived  from  the  oblong  mound  with  an  almost  perpendicular 
incline  and  a  flat  top  which  was  the  origin  of  the  mastaba,  but 
from  a  stone  tumulus,  pointed  at  the  top  and  sloping  gently 
upwards,  which  was  peculiar  to  the  northern  districts.  There 
was  a  tradition  that  the  fourth  king  of  the  First  Dynasty,  Uen- 
nephes,  was  the  author  of  the  one  which  existed  at  Kokome  in 

40 


FIG.  64.— STELE   OF  NIKHAFITKA. 
(Museum,    Cairo.)     (Phot.    E.   Brugsch.) 


MEMPHITE  ART 


FIG.   65.— OBELISKS   OF  PTAH-HETEP 

AND   HAITI. 
(Museum,  Cairo.)     (Phot.  E.  Brugsch.) 


We  know 


the  Greek  period,  and  this  is 
possible;  but  the  earliest  known 
to  us  is  at  Dahshur,  where  Sene- 
feru  built  it  for  himself  as  king  of 
Lower  Egypt.  Even  then,  it  did 
not  constitute  the  entire  sepulchre; 
it  was  accompanied  by  a  chapel 
with  subterranean  store-houses, 
and  a  paved  temenos  surrounded 
it,  protected  by  a  square  or 
rectangular  enclosure  (Fig.  73). 
A  causeway  connected  the  whole 
with  a  temple  situated  in  the  royal 
town,  towards  the  fringe  of  culti- 
vated land.  The  apartments  of 
the  double  were  concealed  in 
or  under  the  pyramid;  Seneferu 
received  his  revenues  in  the  ad- 
joining chapel,  and,  as  a  living 
god,  he  was  associated  with  the 
other  gods  in  the  temple  of  the  city  (cf.  Fig.  41). 
what  the  Pyramids  are.  The  heap  of  stones  composing  them 
rests  on  a  square  base;  the  faces  confront  the  four  cardinal 
points  as  in  the  mastabac ,  but  no  more  precisely  than  in  the 
majority  of  these.  Their  height 
varies  from  482  feet,  as  in  that 
of  Cheops  (Fig.  74)  to  62  feet 
in  that  of  Unas;  but  whatever 
the  individual  dimensions,  the 
general  plan  was  marked  out 
once  for  all  before  the  work 
was  begun,  and  the  architect 
proposed  to  carry  it  on  without 
modification  to  the  end.  It  did 
sometimes  happen ,  however, 
that  it  was  altered  in  the  course 
of  building,  and  that  the  propor- 
tions of  the  whole  were  increased, 
which  neccessitated  changes  in 
the  arrangement  of  the  interior; 
this  was  probably  the  case 
with  the  Pyramids  of  Chephren 
and  Mycerinus,  whereas  that  of 


41 


FIG.   66. — PLAN   OF  THE  TOMB  OF  TI. 
(After  Mariette.) 


ART  IN  EGYPT 


Cheops  seems  to  have  been  built  uninterruptedly  on  the  site  of 
an  earlier  tomb,  the  materials  of  which  it  absorbed  in  its  rough 
masonry.  This  was  afterwards  covered 
with  a  facing  of  massive  blocks  (Fig.  75). 
These  modifications  in  the  course  of  exe- 
cution were  most  frequent  under  the 
Fourth  Dynasty,  when  the  constructive 
formula  had  not  yet  been  fixed  by  prolonged 
experience,  and  when  architects  perhaps 
allowed  themselves  to  be  carried  away 
by  daring  experiments  which  imperilled 
the  solidity  of  their  work.  Those  of 
Cheops,  fearing  that,  the  tomb-chamber 
might  succumb  under  the  weight  of  over 
300  feet  of  stone,  built  five  chambers, 
superposed  along  the  central  axis,  to  relieve 
it;  this  device,  which  carried  the  greater 
part  of  the  central  pressure  out  to  the 
lateral  surfaces,  did,  in  fact,  save  the 
building;  but  it  was  not  repeated.  From 
the  middle  of  the  Fifth  Dynasty  onward, 
the  majority  of  the  Pyramids  are  almost 
identical  in  their  plan,  in  which,  though 
the  dimensions  were  reduced,  the  inviola- 
bility of  the  mummy  was  none  the  less 
A  sloping  passage,  rising  to  the  level  of  the 


FIG.67-— MASTABA  WITH 
HOLLOW  CHAMBER  AND 

DISCHARGING   ARCH. 

(Drawing     by     Bourgoin, 

communicated   by  Mess's. 

A.  Picard.) 


assured  (Fig.  76). 
soil  under  the  centre  of 
the  north  side,  just  at 
the  height  of  the  first 
course,  led  to  a  low  ante- 
room ;  this  gave  access  to 
a  horizontal  passage,  bar- 
red almost  in  the  centre 
by  granite  portcullises, 
and  a  vestibule  (Fig.  77), 
communicating  on  the 
right  with  the  tomb- 
chamber,  on  the  left  with 
a  serdab  or  a  store-room. 
The  vestibule  and  the 
tomb-chamber  were  crown- 
ed by  a  pointed  roof, 
consisting  of  three  courses 


42 


3.— TOMB   OF  ADU,  AT  DENDERAH. 
(After  Petrie.) 


MEMPHITE  ART 


FIG.   69.— TOMB   OF  MEKHU  AND   SABNI 
AT  ASSUAN.     (After  Morgan.) 


of  limestone  beams,  leaning  one  against  the  other  at  the  top. 
Sometimes  the  superstructures  were  composed  entirely  of  large 
blocks  of  fine  limestone,  as  in  the  Pyramid  of  Unas;  but  more 
commonly  they  have  a 
core  of  coarse  limestone 
from  the  neighbouring 
mountains,  with  a  facing 
of  fine  limestone. 

Little  as  geometrical 
figures  are  calculated,  in 
general,  to  evoke  a  sen- 
timent or  give  artistic 
enjoyment ,  the  pyramid 
as  realised  by  the  Egyp- 
tians on  their  native  soil 
never  fails  to  move  those  who  see  it  for  the  first  time  pro- 
foundly. And  when,  instead  of  the  finished  work,  we  have  as 
it  were  the  sketch  only  before  our  eyes,  we  are  hardly  less 
deeply  impressed.  The  Pharaoh  Nefer-ka-Ra,  of  the  Third  Dynasty, 
has  left  such  a  sketch  at  Zawyet-el-Aryan ,  in  the  trenches 
destined  for  the  super-structures  of  his  tomb,  and  the  inclined 
plane  with  its  slides  over  which  the  blocks  passed  while  awaiting 
the  construction  of  the  passage  leading  to  the  mortuary  chambers 
(Fig.  78).  The  works  ceased  at  the  moment  when  the  lower 
courses  of  granite  had  been  set,  and  there  is  nothing  above 
the  surface  of  the  ground  but  an  admirable  oval  basin  destined 
for  libations  (Fig.  79).  The  whole  is  merely  a  T-shaped  ditch, 

some  100  feet  deep;  and 
yet  the  impression  it 
makes  when  one  goes 
down  into  it  is  unforget- 
table. The  richness  and 
the  cuttingof  the  materials, 
the  perfection  of  the 
joints  and  sections,  the 
incomparable  finish  of  the 
basin,  the  boldness  of 
the  lines  and  the  height 
of  the  walls  all  combine 
to  make  up  a  unique 

creation.  The  chapels  of  the  completed  pyramids  are  not  marked 
by  this  almost  brutal  strength.  They  were  buildings  of  medium 
height,  which,  projecting  from  the  eastern  facade,  extended  to 

43 


FIG.  70.— THE  STEP  PYRAMID     (Phot  Diimichen.) 


ART  IN  EGYPT 


FIG.   71. —SECTION   OF   THE    STEP-PYRAMID. 
(After  Howard-Vyse.) 


the  enclosing  wall.  The  only  one  which  has  come  down  to  us  complete, 
that  of  Seneferu,  consists  of  two  dark  little  rooms,  without  ornaments 

or  inscriptions,  a  courtyard 
behind  them,  and  in  the 
courtyard,  two  bare  stelae 
rising  boldly  against  the 
pyramid.  Only  the  pave- 
ment of  the  chapel  of  Cheops 
remains,  and  the  chapel  of 
Chephren  is  in  ruins  (Fig. 
80);  but  recently,  several 
of  those  which  belonged  to 
the  sovereigns  of  the  Sixth 
Dynasty  have  been  ex- 
cavated, notably  those  of 
Sahu-Ra  and  Ra-en-user.  They  were  approached  by  propylaea  built  at 
the  foot  of  the  plateau  (Fig.  81 ),  beyond  which  a  long  incline  rose  to  the 
body  of  the  building,  the  arrangements  of  which  varied.  In  the  chapel 
of  Sahu-Ra  (Fig.  82),  for  instance,  there  was  a  dark  passage,  then  a 
colonnaded  court,  then  a  complicated  series  of  cells  and  storehouses,  and 
in  the  obscurity  of  the  background,  the  stele,  in  the  form  of  a  closed 
door,  where  the  office  of  the  dead  king  was  celebrated.  It  was  a  temple, 
lacking  the  Holy  of  Holies,  or  rather  the  tomb-chamber  was  to  it 
what  the  Holy  of  Holies  was  to  the  real  temple,  the  dwelling  where 
the  master  of  the  house  was  lodged,  safe  from  attacks  from  with- 
out; the  stele  represented  the  mysterious  door  which  could  no  longer 
be  opened,  but  on  the  threshold  of  which  offerings  were  heaped. 
Among  the  inscriptions  of  Sakkarah,  the  student  occasionally 

comes  upon  a  certain 
curious  hieroglyph,  a 
truncated  pyramid 
surmounted  by  an 
obelisk ,  and  accom- 
panied by  a  solar  disc, 
which  seems  sometimes 
to  be  poised  upon 
the  point  of  the  obe- 
lisk. It  indicated  a 
temple  which  Pharaoh 
had  dedicated  to  Ra, 
the  Sun  of  Heliopolis, 

in  his  royal  city,  near  his  tomb;  but  it  seemed  uncertain  whether 
it  was  an  exact  figure  of  this,  or  only  a  graphic  combination  of 

44 


FIG.   72.— THE    FALSE   PYRAMID   AT  MEDUM. 


MEMPHITE  ART 


elements  really  separate.  The  German  excavations  near  Abusir 
have  brought  to  light  fragments  which  prove  that  the  obelisk 
rose  upon  the  pyramid 
itself,  and  not  beside  it. 
An  inclined  plane  be- 
tween two  parapets  led 
from  the  palace  of  Ra~-en- 
user  to  the  temple,  which 
consisted  of  a  rectangular 
court,  about  325  feet  long 
by  280  wide,  the  main 
axis  of  which  ran  from  east 
to  west;  it  was  surrounded 
by  a  brick  wall,  which  had 

J  e  i  .  .1  FIG.   73-— THE   PYRAMID   OF  RA-EN-USER. 

a     sort    Ot    pylon    in     the  As  RESTORED  BY  L.  BOHCHARDT. 

middle   of   the   east  face. 

The  pyramid  covered  nearly  all  the  western  half  of  the  enclosed 
area;  it  was  not  a  classic  pyramid  like  those  of  Gizeh,  but  a 
square  mastaba,  analogous  to  those  of  which  the  monument  of 
Medum  is  composed  (Fig.  83).  It  measured  probably  from  60  to 
100  feet  in  height,  with  a  base  of  about  130  feet,  and  three  of 
its  faces  were  bare;  by  a  door  pierced  in  a  chapel  attached  to 
the  fourth,  that  of  the  south,  access  was  obtained  to  a  staircase 
which  led  to  the  platform.  Here  the  obelisk  rose,  or  rather, 
the  facsimile  in  brick  of  an  immense  stone  in  the  shape  of  a 
squat  obelisk,  the  point  of  which  rose  to  about  120  feet.  The 
platform  in  front  of  this 
strange  monument  was 
bordered  on  the  east,  the 
south,  and  the  north  by 
long  vaulted  corridors ;  that 
on  the  south  led  to  the 
chapel,  where  the  staircase 
debouched,  that  on  the 
north  to  cells  where  pro- 
visions were  stored,  and 
where  the  officiating  priests 
were  lodged,  together  with 
the  materials  for  worship. 
In  the  court  itself,  a  deeply 
grooved  pavement  forms 
a  parallelogram  upon  the  ground,  terminating  on  the  east  with 
a  row  of  nine  alabaster  basins,  while  on  the  west,  almost  at 

45 


FIG    74. — THE   GREAT  PYRAMID  AND  THE 
SPHINX.     (Phot.  Beato.) 


ART  IN  EGYPT 


- 


FIG.  75.— THE   FACING    SLABS    STILL   INTACT 
OF  THE  GREAT  PYRAMID.     (Phot.   Covington.) 


the  foot  of  the  pyramid,  an  immense  table  for  offerings,  also 
of  alabaster,  stood  in  a  little  court  surrounded  by  low  walls.  Many 

details  in  the  arrange- 
ment of  the  various 
parts  are  inexplicable  or 
obscure;  but  one  point 
is  now  firmly  established, 
namely,  that  the  obelisk 
stood  both  for  the  sanc- 
tuary and  for  the  god 
who  was  worshipped 
there.  We  are  so  much 
accustomed  to  consider 
the  Egyptian  divinities  as 
beings  of  flesh  and  blood, 
men  or  animals,  that  we 
are  surprised  when  one 
of  them  is  revealed  as 
an  inanimate  object.  Here, 

however,  the  obelisk  was  the  god  himself,  and  what  is  more, 
the  Sun-god,  and  as  if  to  leave  no  room  for  doubt  on  this  score, 
the  Egyptians  fashioned  near  the  south  front  a  brick  model  of 
one  of  the  solar  boats,  with  its  special  design,  and  its  cargo  of 
sacred  insignia  (Fig.  84).  At  the  first  blush,  there  seems  some- 
thing paradoxical  in  the  idea  of  imitating  in  a  heavy  substance 
a  thing  as  light  and  rapid  as  a  boat,  and  setting  it  motion- 
less upon  the  sand  of  the  desert.  And  yet  this  was  but  the 
inevitable  outcome  of  that  longing  for  a  future  life  which  is 

manifested  in  all  their 
works.  However  carefully 
preserved ,  the  wooden 
vessel  of  the  god  was 
destined  some  day  to 
crumble  into  dust,  and 
perhaps  the  circumstances 
of  the  moment  would  be 
such  that  it  would  be 
impossible  to  make  a  new 
one;  then  the  brick  vessel 
would  take  its  place,  with 
more  chance  of  being  useful, 
in  that  it  would  be  more  difficult  to  damage  or  destroy;  and  would 
continue  its  function  for  Ra  as  long  as  a  fragment  of  it  subsisted. 

46 


FIG.  76. — PLAN  OF  THE  PYRAMID  OF  UNAS. 

(From    a    drawing    by    Maspero,     communicated    by 

Mess''s.  A.  Picard.) 


MEMPHITE  ART 


FIG.  77.— DOOR  BETWEEN  ANTE-ROOM 

AND  TOMB-CHAMBER  IN  THE  PYRAMID 

OF  UNAS. 


The  monument  of  the  Sphinx 
(Fig.  85)  was  no  more  a  real 
temple  than  that  of  Abusir. 
It  was  a  kind  of  waiting 
hall,  built  on  the  edge  of  the 
plain  in  front  of  the  second 
pyramid,  and  connected  with  it 
by  a  causeway  the  line  of 
which  is  still  perceptible;  it  was 
the  starting-point  of  the  proces- 
sions which,  on  fixed  days,  went 
to  the  chapel  of  Chephren,  to 
perform  the  worship  of  the  so- 
vereign. The  plan  is  very 
simple,  yet  it  is  not  easy  to 
determine  the  uses  of  the  various 
parts.  The  centre  is  occupied 
by  a  hall  79  feet  long  by  23 
wide,  with  a  row  of  six  mono- 
lith pillars  in  the  middle.  A 

second  hall,  571/z  feet  by  29,  adjoins  this,  forming  a  T  on 
plan;  it  was  ornamented  with  a  double  row  of  five  pillars,  and 
lighted  by  oblique  vent-holes  at  the  top  of  the  walls.  The  hall 
with  the  six  columns  is,  as  it  were,  the  pivot  round  which  the 
rest  of  the  building  was  set;  it 
communicates  on  the  east  with  a 
gallery,  which,  running  parallel 
with  the  facade  (Fig.  86),  termi- 
nates at  the  two  extremities  in 
a  rectangular  cabinet ;  at  the  south- 
west angle,  the  hall  has  a  cell 
containing  six  niches  in  super- 
posed pairs,  at  the  north-west 
angle  it  communicates  with  the 
sloping  passage  which  leads  by 
a  gentle  incline  to  the  plateau. 
The  core  of  the  masonry  is  of 
Turah  limestone;  the  facings,  the 
pillars,  the  ceilings,  the  fagade 
(Fig.  87)  are  of  gigantic  granite 
or  alabaster  blocks,  polished  and 
adjusted  to  perfection,  but  not 

j  i     J       LJ  •       il  f  FIG.   78.— INCLINED  WAY  AT 

decorated.  Here,  as  in  the  unfm-        ZAWYET-EL-ARYAN.  (Phot.  Oropesa.) 

47 


ART  IN  EGYPT 


FIG.    79. — FLOOR   AND    LIBATION-TROUGH    OF   THE 
UNFINISHED    PYRAMID    AT   ZAWYET-EL-ARvAN. 

(Phot.   Oropesa.~). 


ished  tomb  of  Nefer-ka-Ra,    the  architecture  produces  its  effect 
without   any   adventitious   aid,   by    purity   of   line   and    Tightness 

of  proportion.  The  build- 
ing is  almost  complete, 
and  it  might  be  supposed 
that  by  the  help  of  the 
data  gleaned  at  Abusir 
and  in  the  chapels  of  the 
Pyramids,  we  should  be 
able  to  re-  establish  the 
general  plan  of  the  ordinary 
temples  from  this  example ; 
such,  however,  is  not  the 
case,  and  the  problem  is 
still  obscured  by  too  many 
unknown  issues  to  be 
solved.  I  think  I  may 
venture  so  far  as  to  say 
that  they  lacked  certain 
features  proper  to  those  of  the  later  periods,  such  as  pylons, 
with  their  high  bay  flanked  by  two  massive  towers.  The  doors 
opened  directly  in  the  enclosing  wall  as  did  later  those  of 
Thebes,  in  the  Saiite  period.  They  were  accompanied  by  a 
portico,  and  followed  by  a  court,  round  and  in  the  midst  of 
which  the  offering-tables  and  the  materials  of  worship  were 
disposed.  The  main  body  of  the  building  rose  at  the  end,  but 
we  are  unable  to  say  how  the  different  apartments  were  arranged. 

The  sanctuary  was  assu- 
redly quite  at  the  back, 
but  it  is  a  question 
whether  the  rooms  which 
flanked  it  right  and  left 
were  already  assigned  to 
the  mother-goddess  and 
the  child.  The  point  most 
clearly  established  by  the 
ruins  is  that  a  good  many 
elements  very  frequent  at 
a  later  period  were  alrea- 
dy in  use,  among  them 
the  cornice  with  its  curved 
demi-lions  on  the  stone 
the  square  pillars, 


FIG.  80.— PLAN   OF  THE  CHAPEL  OF  CHEPHREN. 
(After  Steindorff.) 


gorge,    the    gargoyles    of    projecting 
facings,  the  images  of  guardian  lions  (Fig. 

48 


MEMPHITE  ART 


FIG.    8l. — PLAN    OF   THE    PROPYLj<EA. 

(After  Borchardt.) 


the  palm  or  lotus  column.  The  former  was  in  favour  under  the 
Fifth  Dynasty,  rather  heavy  in  the  temple  of  Sahu-Ra,  light  and 
slender  in  that  of  Unas  (Fig.  89). 
Its  capital  is  formed  of  a  bunch 
of  palm-leaves,  attached  to  the 
shaft  by  four  ribbons,  and  bend- 
ing gracefully  under  the  weight 
of  the  abacus.  The  lotus-like 
columns  of  Sahu-Ra  and  Shepses- 
ptah  (Fig.  90)  are  circular,  where- 
as those  of  Ra- en -user  are 
rectangular  at  the  base,  but 
are  gradually  rounded  as  they 
rise  till  they  become  almost 
circular  at  the  summit  (Fig.  91). 
Save  for  some  slight  variations, 
they  consist  of  four  or  six  lotus- 
stems  in  fasces,  bulbous  at  the 
foot,  and  adorned  with  triangular 
leaves;  the  buds,  bound  to  the 
neck  of  the  shaft  by  four  or 
five  bands,  are  grouped  into  a 
bouquet  to  form  the  capital, 
and  sometimes  young  buds,  inserted  between  the  half-open 
ones,  fill  the  spaces  above  the  ligatures.  Examination  of  the  ruins 
leads  to  the  conclusion  that  Memphite  architecture,  though  it 
inclined  to  the  gigantic  for  the  tombs  of  its  kings,  did  not 
desire  it  for  the  temples 
of  the  gods;  it  aimed  here 
at  strength  and  elegance 
rather  than  at  immensity. 
The  surfaces  it  offered 
nevertheless  afforded 


an 

almost  boundless  field  for 
the  activity  of  sculptor 
and  painter.  Generally 
speaking ,  the  Egyptians 
would  not  allow  even  the 
most  beautiful  stone  to 
remain  bare,  while  on 
the  other  hand,  painting 
alone  without  sculpture 
beneath  it,  did  not  approve 


FIG.  82. 

SEPULCHRAL,  CHAPEL    OF  RA-EN-USER. 
(After  Borchardt.) 

49  E 


ART  IN  EGYPT 


FIG.  83.— SOLAR  TEMPLE   OF  RA-EN-USER, 
AS  RESTORED  BY  BORCHARDT. 


itself  to  them  as  possessing  the  enduring  qualities  required  for 
the  adornment   of  temples  and  tombs;    with  few  exceptions,   it 

was  tolerated  only  in- 
houses  and  palaces.  The 
Pharaohs,  indeed,  had 
an  instinctive  repugnance 
to  taking  up  their  abode 
in  dwellings  where  others 
had  lived  before  them; 
they  generally  abandoned 
these  to  their  progeny, 
and  improvised  new  dwell- 
ings for  themselves,  which 
always  seemed  to  them 
good  enough  if  they  were 
of  a  nature  to  last  as 
long  as  themselves.  For 
such  ephemeral  buildings 

they  were  content  with  a  perishable  ornamentation  of  simple 
painting  on  the  ceilings,  pavements,  and  walls,  and  the  same 
latitude  obtained  perforce  in  hypogea  carved  in  a  rock  unfit  for 
sculpture,  as  well  as  in  the  chapels  of  sun-dried  bricks  which 
villagers  too  poor  to  use  stone  raised  for  their  gods;  everywhere 
else,  colour  is  only,  so  to  speak,  the  complement  of  relief,  but  a  comple- 
ment so  indispensable  that 
it  is  difficult  to  imagine 
a  building  without  it. 
We  understand  now  why 
painting  in  Egypt  never 
acquired  the  personal 
development  and  com- 
plexity which  characterises 
it  in  our  own  countries. 
It  laid  flat  tints  on  the 
work  of  the  sculptor,  and 
indicated  the  details  of 
costume  and  the  acces- 
sories which  he  had  not 
noted.  The  artist's  work 
was  therefore  rather  that 
of  an  illuminator  than  a 

painter,    and   the  necessity   of   reliefs   to   cover   was  so    impera- 
tive in  his  eyes  that  he  did  his  utmost  to  suggest  them,   even 

50 


FIG.  84.— BRICK  BOAT  OF  RA-EN-USER. 
(After  Borchardt.) 


MEMPHITE  ART 


"FIG,  85.— TEMPLE    OF   THE  SPHINX. 
(After  Marietta.) 


where  they  did  not  exist;  he  surrounded  his  figures  with  a 
red  or  black  outline  which  defines  the  contours  as  sharply  as  if 
he  had  cut  them  with  a  style. 
The  deliberate  neglect  of  half- 
tones and  of  their  infinite  variety, 
led  him  to  choose  for  each  ob- 
ject or  person  a  tone  which, 
without  deviating  too  widely 
from  nature,  sometimes  made 
no  attempt  to  approach  it  very 
closely.  Thus  men  are  represented 
with  skin  of  a  more  or  less  dark 
brown,  while  women  are  light 
yellow;  a  blue,  either  pure  or 
streaked  with  black,  was  reserved 
for  the  sea,  a  bright  green  for 
grass  and  foliage,  and  a  dirty 
yellow  dotted  with  red  stood 
either  for  corn  piled  in  heaps 
or  for  the  sand  of  the  desert.  With  conventions  so  harassing, 
and  means  so  restricted,  artists  nevertheless  managed  to  produce 
works  of  striking  truth  and  senti- 
ment. Such  was  the  tomb  of  the 
time  of  Seneferu  at  Medum, 
where  Vassalli  saved  the  famous 
geese  which  are  now  in  the  Museum 
of  Cairo  (Fig.  92).  The  movement 
is  excellent,  and  the  characteristics 
which  distinguish  the  male  and 
female  in  each  couple  are  noted 
with  an  accuracy  which  surprises 
naturalists;  a  Chinese  or  Japanese 
artist  could  have  done  no  better. 
Unfortunately,  this  is  an  excep- 
tional example;  the  painting  of 
the  Memphite  age  rarely  rose  to 
the  dignity  of  an  autonomous  art; 
it  was  a  servile  dependent  of 
sculpture. 

Such  being  the  case,  decoration, 
whether  of  tomb  or  temple,    was 
considered  an  immense  composition,  every  part  of  which  converged 
to  the  same  point:    in  the  temple,  to   the  wall  at  the   back   of 

51  E2 


FIG.  86.— INTERIOR  OF  THE  TEMPLE 
OF  THE  SPHINX  (Phot.  E.  Brugsch). 


ART  IN  EGYPT 


FIG.   87. — FACADE   OF  THE   TEMPLE   OF  THE  SPHINX 
(Phot.  Steindorff). 


the  sanctuary,  In  the  tomb  to  the  stele  which  had  replaced  the 
door  of  the  vault.    It  is  true  that  every  room,  and  in  every  room 

each  wall ,  and  on 
every  wall  each  picture 
constitutes  a  whole 
where  the  various 
persons  mingle  and 
confront  each  other 
in  such  a  manner, 
that  if  some  are  ad- 
vancing to  this  kib- 
lah L,  others  seem 
either  to  be  going 
away  from  it,  or  at 
least,  not  to  be 
making  their  way 
towards  it;  but  this 
contrariety  of  move- 
ment ,  which  might 
seem  to  stultify  the 

principle  just  laid  down,  is  explained  when  we  examine  the 
conditions  under  which  it  is  produced.  In  the  temple,  it  is 
always  the  god,  the  supreme 
deity  of  the  place,  and  the  divin- 
ities of  his  family  or  of  his  suite 
who  move  in  the  opposite  direct- 
ion to  the  rest:  the  oblationist, 
priest  or  king,  always  advances 
in  the  normal  direction.  Occasion- 
ally, but  infrequently,  a  single 
scene  occupies  the  entire  wall; 
more  often,  it  is  divided  into 
panels.  Thus  the  ritual  of  divine 
worship  was  resolved  into  a 
definite  number  of  ceremonies, 
which  were  at  will  isolated  from 
their  neighbours,  or  grouped  in 
processions  more  or  less  long. 
When  at  the  beginning  of  the 
sacrifice  Pharaoh  washed  the  altar, 
lighted  the  fire,  burnt  the  incense,  poured  the  libations  of  water, 
wine,  milk  and  essences,  he  provided  the  material  for  so  many 

1  The  Arab  prayer-niche,  facing  towards  Mecca. 

52 


FIG.  88.— HEAD  OF  A  LION  FROM 

SAKKARAH  (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


MEMPHITE  ART 


distinct  scenes.  And  as,  to 
ensure  the  complete  efficacy  of 
these  operations,  he  had  to  per- 
form them  once  as  the  king  of 
the  South,  and  again  as  the  king 
of  the  North,  the  artist  was  also 
obliged  to  depict  them  twice,  but 
at  the  same  time  to  distribute 
them  symmetrically  from  room  to 
room,  so  that  at  last  the  temple 
came  to  be,  as  it  were,  cut  into 
two  parallel  sections  with  corre- 
sponding decorations;  in  the 
right  hand  section  the  sovereign 
officiated  in  the  name  of  Upper 
Egypt,  in  the  left  in  that  of 
Lower  Egypt.  He  thus  pro- 
ceeded from  without  to  within 
until  he  reached  the  Holy  of 
Holies,  and 
at  each  stage, 
the  god  rose 


FIG.  89. — TWO  COLUMNS  WITH  PALM- 
LEAF  CAPITALS  FROM  THE  FUNER- 
ARY CHAPEL  OF  UNAS 
(Museum,  Cairo).  (Phot.  E.  Brugsch.) 


FIG.  go. 

LOTUS    COLUMN    OF 

SHEPSES-PTAH 

(Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


before  him,  like  some  great 
lord  coming  out  to  meet  his  subjects,  con- 
fronting them  and  receiving  their  homage 
from  station  to  station.  The  concatenation 
is  far  less  strict  in  the  tombs,  for  here  the 
deceased  plays  a  double  part,  and  whereas 
in  some  places  he  passively  awaits  the  results 
of  the  labours  his  posterity  performs  for  his 
benefit,  in  others  he  behaves  as  if  he  were 
still  reckoned  among  the  living;  he  passes 
through  his  fields  or  workshops  to  see  what 
is  being  done,  and  superintend.  The  con- 
tradiction in  the  two  parts  is  accordingly 
translated  by  a  similar  contradiction  in  the 
orientation  of  his  images;  some  of  these  move 
or  stop  facing  the  kiblah,  like  the  faithful  who 
have  come  from  the  outer  world  to  do  honour 
to  their  ancestor,  but  the  majority  turn  their 
backs  on  it  and  seem  to  be  advancing  from 
it,  as  becomes  the  master  of  the  house.  If, 
disregarding  these  exceptions,  which  are  the 
result  of  the  ideas  held  by  they  Egptians  as 
53 


ART  IN  EGYPT 


to  the  material  conditions  of  the  after-life,  we  take  the  pictures 
of  the  hypogeum  as  a  whole,  we  must  admit  that  they  tend  uni- 
formly towards  the  stele,  and  that  they  illustrate  by  their  suc- 
cession the  mystic  drama,  the  episodes  of  which  are  evolved 
from  the  threshold  of  the  chapel  to  the  sup- 
posed door  of  the  tomb-chamber. 

Artists  registered  them,  and  as  they  were 
used  in  the  same  manner  in  the  temples,  they 
finally  became  a  series  of  designs  containing 
all  the  elements  necessary  to  decorate  the 
house  of  the  dead  or  that  of  the  god.  It  is 
probable  that  in  the  beginning  each  town  had 
its  cartoons ,  in  which  the  characteristic  features 
of  its  religion  and  its  burial  rites  were  repro- 
duced, but  at  the  moment  when  history  begins 
for  us,  local  diversities  persisted  only  in  a 
slight  degree,  and  two  general  types  prevailed, 
one  for  the  tomb,  the  other  for  the  temple; 
the  numerous  examples  of  each  which  we 
possess  were  distinguished  only  by  the  details 
of  the  names  and  figures.  As  I  have  already 
said,  several  evidences  lead  one  to  conclude 
that  they  were  definitively  fixed  in  the  schools 
of  the  Delta,  and  this  I  consider  a  proved 
fact  in  the  case  of  the  tombs;  it  is,  indeed, 
in  the  Delta,  and  in  the  Delta  alone,  that 
the  papyrus  reed  grows  to  an  immense  height, 
and  forms  those  vast  thickets  into  which  the 
holy  dead  penetrate  at  will,  to  harpoon  fish, 
or  hunt  water-fowl  and  hippopotamus  (Fig.  93). 
After  the  priests  of  Heliopolis  had  codified 
the  principles  of  the  worship  of  the  gods  and 
of  the  dead,  they  came  almost  inevitably  to 
lay  down  rules  for  the  composition  and  exe- 
cution of  the  pictures  in  which  they  represent- 
ed it.  They  permitted  no  variation  in  aught 
relating  to  the  gods,  for  when  man  was  invoking  these,  his 
most  insignificant  acts  and  attitudes  had  their  importance.  To 
carry  out  his  work  to  perfection,  the  artist  should  have  de- 
composed the  slightest  gestures  of  the  celebrant  into  as  many 
distinct  images,  but  the  theologian  did  not  insist  on  this.  He 
merely  required  the  artist  when  rendering  each  of  the  episodes, 
the  sequence  of  which  constituted  the  ritual,  to  express  the 

54 


FIG.QI.— LOTUS-COL- 
UMN OF  RA-EN-USER 
(Museum ,  Cairo). 
(Phot.  E.  Brugsch.) 


MEMPHITE  ART 


critical  moment  when  the  act  was  accomplished  which  produced 
the  maximum  effect.  In  earlier  ages,  the  chief  or  king  himself 
lassoed  the  almost  wild  bull  in  the  fields  for  sacrifice  (Fig.  94). 


FIG.  Q2.— THE  GEESE  OF  MEDUM  (Museum,  Cairo).     (Phot.  E.  Brugsch.) 

He  felled  it,  tied  its  hoofs  together,  and  then  killed  it  with  a 
wooden  pole-axe,  partially  sharpened,  with  which  he  dealt  it 
a  blow  on  the  skull  between  the  eyes.  At  a  later  period,  they 
cut  the  beast's  throat  instead  of  dealing  it  a  death-blow,  but 
the  antique  weapon  and  the  gesture  it  demanded  were  preserved 
and  inscribed  on  the  walls  as  the  characteristic  emblem  of 
the  rite.  Where  we  see  the  king  standing  before  the  god  and 
presenting  the  pole-axe  to  him,  we  are  contemplating  the  sacri- 
fice, although  the  victim  is  absent;  when  once  the  scene  had 
been  thus  symbolised,  it  was 
transmitted  from  generation  to 
generation  in  the  same  form, 
varying  only  in  the  accessories, 
and  it  was  to  be  found  at 
Kom-Ombo  under  the  Antonines 
just  as  it  was  under  the  earliest 
Pharaohs.  If  we  now  return  to 
the  mastabas  and  examine  the 
same  motive  there,  we  shall 
suddenly  perceive  that  it  is  not 
treated  in  accordance  with 
an  immutable  formula.  The 
draughtsman  expands  or  con- 
denses it  regardless  of  the 
theologian;  he  multiplies  or 
suppresses  supernumeraries,  re- 


FIG.  QJ.— HIPPOPOTAMUS  HUNT  IN  THE 
TOMB  OF  TI. 


laxes  or  stiffens  their  gestures, 

combines    their    efforts;     if    so 

disposed,     he     devotes     entire 

panels  to  the  ultimate  fate  of  the  bull,   the  cutting-up,   and  the 

presentation  of  the  pieces  to  the  master.     And  what  is  true  of 


ART  IN  EGYPT 


FIG.    04. — THE   SACRIFICIAL  BULL,  LASSOED    BY 
THE  KING,  AT  ABYDOS   (Drawing-  by  Boudier). 


the   sacrifice   applies  equally   to  all   the   rest;    composition   and 
rendering  are   no   less  varied   in   the   book   of   the   tombs   than 

they  are  uniform  in  that 
of  the  temples.  Dogma, 
which  prescribed  to  the 
artist  the  choice  and  treat- 
ment of  the  scenes  in 
which  the  gods  were  visibly 
present,  allowed  him  much 
more  liberty  in  dealing 
with  the  dead. 

Incoherent  as  they  are, 
the  fragments  of  the 
chapels  of  Unas,  Ra-en- 
user  and  Sahu-Ra  which 
have  come  down  to  us, 
suffice  to  prove  that  the  book  of  the  temples  comprised  even 
at  this  period  the  same  kind  of  pictures,  connected  almost  in 
the  same  manner,  as  those  we  find  under  the  Second  Theban 
Empire.  The  decorative  scheme  changed  its  nature  as  it  pro- 
gressed from  without  to  within.  In  the  places  accessible  to 
the  public,  in  the  col- 
umned hall  which  served 
as  vestibule,  and  under 
the  porticoes  bordering 
the  entrance  court,  the 
warlike  deeds  of  the  so- 
vereign were  set  forth, 
or  at  least  those  for  which 
he  gave  glory  to  the  god, 
and  the  spoils  of  which 
had  helped  to  build  or 
restore  the  temple.  Thus 
Sahu-Ra  was  shown  on 
the  south  side  of  his 
hypostyle  hall  striking 
down  a  king  of  Libya 
who  is  prone  at  his  feet 
(Fig.  95) ;  further  on,  three 
daughters  of  a  Libyan 
chief  implored  his  mercy, 
captive  herds  of  oxen,  asses,  goats  and  sheep  advanced  in  four 
rows,  while  at  the  base  of  the  wall,  beneath  the  animals,  the 

56 


FIG.    95. — A    LIBYAN    CHIEF    STRUCK    DOWN    BY 
SAHU-RA  (Museum,   Berlin).     (Phot.  L.  Borchardt.) 


MEMPHITE  ART 


family  of  the  vanquished 
wept  over  the  fate  of  its 
chief  in  the  presence  of 
Amentit,  Regent  of  the  West, 
and  Ashu,  Lord  of  the  Desert. 
Elsewhere,  Pharaoh  is  en- 
gaged in  a  naval  expedition 
against  Asiatics;  his  fleet  ad- 
vances towards  him  in  two 
lines,  amidst  the  clamour  of 
the  crews.  Or  he  is  hunting 
in  the  desert,  where  he  pur- 
sues birds  through  the  pa- 
pyrus. All  these  recollect- 
ions of  his  princely  life  cease 
when  he  crosses  the  thresh- 
old of  the  inner  chambers. 
An  escort  of  offering-bearers 
accompanies  him  thither  for 
a  few  moments,  but  these 


soon     leave 
turn,  and   he 


him 


remains 


FIG.g6. — KING  SAHU-RA  ADOPTED  BY  THE 

GODDESS  IN  THE  PRESENCE  OF  KHNEMU 

(Museum,  Cairo).     (Phot.  L,  Brugsch.) 


FIG.   Q7. — HENKHAU-HERU 
(The  Louvre,  Paris). 


their 
alone 
with 

his  divine  fathers;  the  goddesses  adopt 
him  as  their  son,  suckling  him  from  their 
breasts  (Fig.  96) ,  and  the  gods  receive 
wine  and  water  from  him,  and  perfumed 
oil,  tribute  by  which  he  hopes  to  gain 
their  goodwill.  Several  of  these  motives 
we  have  already  seen  in  the  mastabas; 
but  until  we  have  studied  them  oh  the 
royal  monuments,  we  can  form  no  idea 
of  the  perfection  with  which  the  Memphite 
artists  have  treated  them.  The  Menkhau- 
Heru  of  the  Louvre  (Fig.  97)  had  already 
shown  us  with  what  charm  they  were 
able  to  invest  the  images  of  their  Pharaohs, 
but  this  was  but  an  isolated  fragment; 
on  the  great  bas-reliefs  of  Abusir,  each 
figure,  from  head  to  foot,  and  when 
several  figures  in  conjunction  are  in 
question,  each  group  of  figures,  is  drawn 
with  a  continuous  line,  traced  upon  the 
57 


ART  IN  EGYPT 


?.— TRIUMPHAL  BAS-RELIEF  OF  SENEFERU 
AT  SINAI.     (Phot.  Petrie.) 


stone   with   an  assurance  and  freedom   that  never   falter  for  an 
instant.     The    background    is   hollowed  imperceptibly    along  this 

line,  to  accentuate  the 
relief,  but  so  subtly  is 
it  done  that  we  can  only 
perceive  it  by  an  effort; 
the  subject  is  by  this 
means  placed  in  an  at- 
mosphere which  softens 
its  contours  more  than 
might  have  been  thought 
possible  with  a  relief 
kept  so  low.  The  inner 
details  show  a  mingling  of 
definite  lines  and  almost 
imperceptible  modellings; 
the  individual  elements 
of  the  face,  the  eyes,  the 
nose,  the  mouth,  and  the 
chin  are  indicated  with  a 

vigorous  point,  and  with  sharp  edges  which  accentuate  the  form; 

but  the  elasticity  of  muscles 
and  flesh  is  expressed  by  mellow 
strokes  and  touches  which  coun- 
teract the  hardness  of  the  rest 
(Fig.  99).  Beings  of  supernatural 
proportions,  kings  or  gods,  had 
eyes  of  enamel,  and  this  device 
gave  them  an  appearance  of  life 
which  was  enhanced  by  the  paint- 
ing. The  colour  has  fallen  off 
nearly  everywhere,  but  where 
it  has  been  preserved ,  it  is  ad- 
mirably fresh  and  harmonious.  It 
completes  the  work  of  the  sculptor, 
and  adds  to  this  a  precision  which 
the  chisel  could  hardly  have 
achieved  without  heaviness;  thus 
it  clothes  Uzueri,  the  god  of  the 
sea,  with  a  tunic  of  undulating 
blue  stripes,  symbolising  the  ocean 
or  covers  the  god  of  cereals,  Napriti,  with  a  sprinkling  of 
brownish  yellow  oblong  grains,  typifying  corn  (Fig.  100). 

58 


FIG.    99. — PORTION     OF     A     FEMALE 
FIGURE   (Tomb  of  Geranikai). 


MEMPHITE  ART 

All  these  were  produced  in  the  royal  workshops,  like  the 
triumphal  bas-reliefs  of  Sinai  (Fig.  98),  and  also,  probably  as  a 
result  of  royal  favour,  certain  funerary  bas-reliefs  of  tombs  in 
which  friends  of  the  sovereign  were  buried.  In  my  opinion,  we 
should  include  in  this  category  the  admirable  wood -carvings  of 
Hesi  (Fig.  102),  one  panel 
at  least  of  which  (Fig.  101), 
ranks  among  the  most 
astonishing  manifestations 
of  Memphite  art.  It  is 
not  surprising  that  these 
workshops ,  installed  as 
they  were  in  the  royal 
residence,  in  the  richest 
and  most  highly  civilised 
centre  of  the  age,  staffed 
by  families  attached  for 
generations  to  the  service 
of  the  sovereign,  and  con- 
stantly recruited  from  all 
the  best  elements  of  the 
popularworkshops,  should 
have  produced  these  fine 
things;  but  the  level  of 
artistic  excellence  sinks 
as  soon  as  we  turn  away 
from  them,  and  in  certain 
provinces  it  falls  so  low 
that  it  is  hardly  superior 
to  that  of  the  most  bar- 
barous people.  The  local 

schools,  though  they  had  adopted  the  decorative  system  of 
Heliopolis,  had  not  cast  aside  their  individual  characteristics, 
and  these  are  clearly  manifested  in  private  tombs.  Those  of  the 
Said  have  left  us  but  a  few  specimens  of  their  respective  art, 
and  it  would  perhaps  be  imprudent  to  judge  them  from  the 
examples  we  have  at  present.  Two  or  three  full  length  por- 
traits of  the  barons  of  Elephantine,  incised  on  the  facades  of 
their  hypogea,  are  fairly  correct  in  treatment  (Fig.  103),  as  are 
also  their  bas-reliefs  (Fig.  104),  but  the  rest  are  merely  rude  dis- 
jointed figures  with  ill-matched  arms  and  legs,  rugged,  twisted, 
and  loaded  with  crude  colour.  A  stonemason  turned  sculptor 
would  give  a  better  account  of  himself  after  a  fortnight's  study, 

59 


FIG.  I  CO.— BAS-RELIEF  OF  THE  CHAPEL,  OF  SAHC- 
RA  (Museum,  Cairo).  (Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


and  we  should  readily  attribute  them  to  a  very  primitive  period, 
if  we  did  not  know  from  their  inscriptions  that  they  were  exe- 
cuted under  the  Sixth  Dynasty. 
The  persons  who  worked  in  the 
mastabas  of  Denderah  hardly  show 
a  more  highly  developed  artistic 
sense  than  those  of  Elephantine, 
although  they  prove  themselves 
more  skilful  craftsmen.  They  encir- 
cled the  human  face  with  two  stiff 
lines,  uniting  at  an  almost  insen- 
sible angle  towards  the  tip  of  the 
nose;  they  furnished  the  mouth 
with  lips  of  equal  thickness  from 
end  to  end ;  they  set  the  almond- 
shaped  eye  between  two  pads  which 
are  comic  as  indications  of  human 
eyelids.  The  slope  of  the  shoulder 
is  over-round  in  their  figures,  the 
elbow  too  pointed,  the  knee  too 
knotty;  the  leg  is  swollen  with 
muscles  which  defy  the  laws  of 
anatomy.  We  divine  a  strong 
ambition  to  excel,  but  feeling  and  technique  are  not  on  a  level 
with  aspiration.  Some  few  miles  west  of  Denderah,  we  enter 
suddenly  into  a  world  with  higher  aptitudes  for  the  plastic  arts. 

Here  the  unity  of  style 
reveals  unity  of  tradition; 
and  in  fact,  one  single 
school,  theThinite,reigned 
supreme  from  Kasr  -  es- 
Sayad  to  the  burial- 
grounds  of  Heracleopolis 
in  Abydos,  to  Akhmim, 
to  Kau-el-Kebir,  to  Suit, 
to  Beni  -  Mohammed  -  el- 
Kufur,  to  Kom-el-Ahmar, 
everywhere  save  at  Her- 
mopolis.  Hermopolis,  the 
city  of  Thoth,  had  been 
from  the  most  remote 

antiquity,    a    centre    of   religious   speculation ,   where  theories  as 
to  the  creation   of  beings  and   the   essence   of  things  were  ela- 

60 


FIG.   IOI. — ONE   OF  THE  FIGURES  OF 

HESI-RA   (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


FIG.  102. — WOODEN  PANELS  FROM  THE  TOMB  OF 
HESI-RA  (Museum,  Cairo).     (Phot.  E.  Brugsch.} 


MEMPHITE  ART 


FIG.    103.— A  PRINCE    OF    ELEPHAN- 
TINE.    (Phot.  J.  de  Morgan.} 


borated;  nevertheless,  having  arrived  at  reflection  and  a  system 

after  Heliopolis,  it  had,  in  the  main  accepted  the  doctrines  and 

funerary  decorations  of  the  latter, 

and  its  originality  is   revealed   to 

us    less    by    the    concept    it    may 

have  formed  of  the  tomb,-  than  by 

the  details  of  its  scenes  and  their 

material  execution.  Its  draughtsmen 

were    remarkable   for    their    sense 

of  life,  the  intensity  and  diversity 

of    their   movement,    and    a   good 

humour,  the   expression   of  which 

sometimes    verges    on    caricature. 

Thus ,    in    one    of   the    tombs    of 

Meir,  there  are  persons  evidently 

suffering  from  famine;  reduced  to 

positive  physiological  distress,  their 

bones  are  coming  through  the  skin; 

this  is  the  procession  of  the  lean 

(Fig.  105).    Another  artist  near  by 

has  reserved   his   wall   for  the   fat 

and  well -liking,  both  of  man   and 

beast;    it   shows   a   kind   of    carnival   of  the    obese    (Fig.  106) 

Anatomical  accuracy  is  scrupulously  observed  in  both  cases,  but 

the   lean   are  perhaps   superior  to   the   fat;   they   come   and   go 

with  an  angular  vivacity  which  would  befit  the  skeletons  of  our 

dances  of  death. 

The  Thinite  School  is  only  to  be  distinguished  by  its  air  of 
provincial  stiffness, 
or  rather,  the  Mem- 
phite  School  is  in 
sculpture  as  in  archi- 
tecture the  con- 
tinuation of  the 
Thinite.  The  royal 
workshops  of  This, 
transferred  to  the 
North  at  the  be- 
ginning of  the  Third 
Dynasty,  taught  their 
methods  to  the 
natives,  and  soon  these,  gaining  in  refinement  by  practice, 
became  capable  of  executing  the  commissions  of  princes  and 

61 


FIG.    104. — BAS-RELIEF  IN  THE  TOMB   OF  MEKHU  AT 
ELEPHANTINE.      (Phot.  Couyat.) 


ART  IN  EGYPT 


private  persons.  They  were  at  their  full  maturity  as  early  as 
the  time  of  Cheops,  and  their  prosperity  endured  until  the  end 
of  the  Memphite  empire.  True,  there 
is  not  a  general  level  of  excellence 
throughout  their  burial-places;  but  if 
there  is  a  good  deal  of  poor  work  to 
be  found,  there  is  still  more  that  is 
good,  and  examples  of  a  very  high 
quality  are  not  uncommon,  even  setting 
aside  those  mastabas,  alloted  to  their 
masters  by  the  king's  favour,  which,  are 
the  actual  work  of  the  royal  artists. 
The  groups  of  sculptured  tombs  follow 
each  other  regularly  enough  in  chrono- 
logical order;  the  earliest,  at  Medum 
and  Dahshur,  rose  under  the  protection 
of  Seneferu,  the  next  towards  Gizeh, 
in  the  shadow  of  the  great  Pyramids, 
the  rest  on  the  sandy  plateaux  of  Abusir 
and  Sakkarah,  together  with  the  Pharaohs 
of  the  Fifth  and  Sixth  Dynasties;  and, 
as  we  descend  from  one  group  to  another, 
the  scheme  of  decoration  expands  and 
becomes  more  complex.  In  the  first,  at  Medum  and  Dahshur, 

the  mastabas,  colossal  as  they  some- 
times are  in  the  mass,  contain  but 
a  restricted  surface  of  ornamented 
wall  (Fig.  107).  The  draughtsman 
has  been  content  to  make  a  choice 
among  the  operations  most  favourable 
to  the  future  life ;  generally  speaking, 
these  elements  are,  in  addition  to 
the  stele  which  has  the  dimensions 
of  a  palace  door,  the  procession  of 
domains  bringing  tribute,  the  voyage 
in  a  ship  on  the  waters  of  the  West, 
the  sacrifice  of  the  bull,  the  dead 
man  seated  before  the  table  awaiting 
offerings,  the  principal  scenes  of 
the  obsequies,  and  nothing  more. 
They  are  spaced  out  widely,  with 
but  few  figures  in  each,  and  the 
air  circulates  freely  in  them.  The 
62 


FIG.  lOg.  — ONE  OF  THE  LEAN 

MEN   OF   MEIR 
(Drawing  by  Cledat). 


FIG.  106. — ONE  OF  THE  FAT  MEN 
OF  MK1R  (Drawing  by  Cledat). 


MEMPHITE  ART 


FIG.  107. ^A  WALL  IN  THE  TOMB  OF  RA- 
HETEP  AT  MEDUM  (After  Petrie). 


relief  is  fairly  high,  the  modelling  precise  and  supple,  the  writing 
careful;  each  of  the  hieroglyphs  is  worked  with  as  much  delicacy 
as  if  it  were  an  intaglio  on  a 
precious  stone,  and  to  make 
the  colour  more  durable,  they 
are  sometimes  enlivened  with 
incrustations  of  stones  or  of 
paste  made  of  tinted  glass.  At 
Gizeh,  a  few  years  later,  the 
tendency  to  enrich  the  com- 
position is  already  perceptible; 
it  becomes  more  and  more 
marked  under  the  Fifth  Dynasty, 
and  under  the  Sixth,  at  Abusir 
and  Sakkarah,  the  entire  book 
of  the  tombs  is  in  use.  Here 
the  artist  no  longer  contents 
himself  with  an  abridged  re- 
presentation of  the  actual  rite 
of  sacrifice  (Fig.  108)  and  of 
homage  (Fig.  109);  he  traces 
at  great  length  and  with  infinite 
prolixity  the  cycle  of  operations  leading  up  to  the  consummation ; 
thus,  dealing  with  stuffs  and  ornaments,  he  shows  on  the  one 
hand  the  reaping  of  the  flax,  the 
stripping  of  the  stalks,  the  spin- 
ning and  glazing  of  the  thread, 
and  the  weaving  of  linen;  on 
the  other,  the  weighing  of  precious 
metals,  their  fusion  in  the  cru- 
cible, the  making  of  necklaces 
and  bracelets,  and  finally,  the 
delivery  at  the  shop  of  chests 
containing  pieces  of  stuff  and 
jewels,  introducing  here  and  there 
comical  episodes  which  relieve 
the  austerity  of  the  place,  such 
as  that  of  the  tame  monkey  who 
has  fallen  out  with  a  bearer  of 
offerings,  and  seizes  him  by  the 
leg  (Fig.  110);  there  is,  in  fact, 
no  longer  any  limit  to  the  number 
of  the  pictures,  save  that  of  the 

63 


FIG.    108. — SACRIFICE    IN    THE    TOMB 
OF    PTAH-HETEP,    AT    SAKKARAH. 


ART  IN  EGYPT 


time  or  money  allowed  to  the  artist;  and  in  order  to  multiply 
them  without  unduly  increasing  the  surfaces,  the  number  of  the 
registers  was  augmented,  and  the  inscriptions  and  figures  crowded 

together  and  piled  one 
above  the  other.  The  tombs 
look  as  if  they  were  hung 
in  the  interiors  with  im- 
mense tapestries ,  not  an 
inch  of  which  has  been 
left  bare,  and  if  there 
are  unornamented  panels 
and  chambers,  it  is  because 
death  snatched  away  the 
master  before  he  had 
finished  his  "eternal  dwell- 
ing". The  effect  upon  the 
modern  spectator  who 
enters  these  sepulchres 
for  the  first  time  is  that 
of  stupefaction  rather  than 
admiration.  His  eye,  daz- 
zled by  the  flash  of  colours 
and  the  exuberance  of 
episode,  fails  to  grasp 
the  whole;  the  general 
theme  escapes  him,  and 
he  perceives  only  the 
amusing  detail. 

The  whole  is,  however, 
less  homogeneous  than  he 
might  suppose,  if  he  trust- 
ed to  his  own  impres- 
sion. The  small  and  medi- 
um sized  tombs  were,  no 
doubt,  decorated  at  a 
breath,  so  to  speak,  and 
we  recognise  in  them  the 
hand  of  a  single  crafts- 
man, or  at  least,  the  impress  of  a  single  enterprise;  but  this 
is  by  no  means  true  of  the  larger  ones;  in  every  period,  but 
more  especially  under  the  Fifth  and  Sixth  Dynasties,  we  find 
in  these  from  room  to  room,  or  even  in  the  same  room  from 
wall  to  wall  and  from  register  to  register,  enough  characteristic 

64 


FIG.    IOQ.— A  WALL  IN  THE  TOMB    OF    SABU 
(Museum,  Cairo). 


:  MEMPHITE  ART 

peculiarities  to  show  that  one  or  more  companies  of  craftsmen 
co-operated.  In  the  Tomb  of  Ti  there  is  identity  of  work- 
manship in  the  two  chapels,  and  diversity  in  the  corridor,  the 


FIG.     110.— HUMOROUS  EPISODE  (Museum,  Cairo).     (Phot.  E.  Brugsch.) 

hypostyle  hall,  and  the  exterior  portico;  but  the  divergence  is 
of  the  kind  we  notice  in  persons  formed  in  the  same  school, 
and  does  not  force  us  to  the  conclusion  that  there  was  a  colla- 
boration of  two  independent  schools;  we  may  rather  suppose 
that  while  the  principal  chambers  occupied  the  most  dexterous 
chisels  of  the  company,  the  less  important  rooms  were  left  to 
less  skilful  workmen.  Such  inequalities  of  treatment  are  more 
strongly  marked  in  the  Tomb  of  Mereruka,  and  this  is  hardly 
surprising,  when  we 
remember  that  this 
contained  over  thirty 
chambers;  three  com- 
panies at  least  shared 
the  work  of  deco- 
ration, and  if  these 
comprised  some  good 
craftsmen ,  they  had 
also  a  proportion  of 
very  indifferent  ones. 
The  examination  of 
some  thirty  mastabas 
scattered  amonsr  the 


FIG.    III.— THE   SCULPTOR  PTAH-ENEKH. 
(Phot.  E.  Brugsch.) 


sands     of     Sakkarah 
enables  me  to  affirm 
the    existence    of  five,    and    perhaps  even   six  workshops,  which 
flourished   under   Unas   and   the   two  Pepis,  each  possessing  its 

65  F   " 


ART  IN  EGYPT 


own  version  of  the  Book  of  the  Dead,  its  own  fashion  of  posing 
figures  and  distributing  accessories,  its  own  manner  of  preparing 

the  drawing  and  then  of 
attacking  the  stone,  even 
its  special  colour.  Ob- 
viously, these  were  not 
the  only  ones,  and  others 
existed  which  will  be  re- 
vealed to  us,  when  the 
hypogea  which  are  not 
yet  destroyed  have  all 
become  accessible ,  and 
we  are  able  to  study  their 
technique  in  the  originals, 
and  not  only  in  pencil 
sketches  and  photographs 

FIG.  112.— COW  TURNING  ROUND  TO  HER  CALF.          which    fail     tO      CXDrCSS     its 

subtleties.    Meanwhile,  we 

claim  the  right  to  assert  that  the  differences  exist  solely  in 
slight  degrees  and  that  all  were  inspired  by  the  same  traditions; 
they  formed  a  powerful  school ,  the  seat  of  which  was  in  the 
Memphite  plain,  near  the  royal  residences.  A  few  of  the  masters 
it  produced  are  known  to  us,  such  as  that  Ptah-enekh,  who  re- 
presented himself  as  the  guest  of  Ptah-hetep,  served  by  the 
servants  of  his  patron  (Fig.  Ill),  and  that  other  who,  taking 

advantage  of  an  unoccu- 
pied panel  in  the  tomb 
where  he  was  working, 
used  it  for  his  own  por- 
trait; seated  before  his 
easel,  his  brush  and  his 
pot  of  colour  in  his  hands, 
he  paints  industriously, 
but  he  has  omitted  to 
tell  us  his  name.  These, 
however,  are  exceptions, 
and  the  finest  works  of 
the  Memphite  age  have 
no  responsible  authors, 


FIG.    IIJ.— PERSPECTIVE   OF  REGISTERS  IN  THE 
TOMB  OF  PTAH-HETKP   (After  Dumichen). 


far 


as   we    are    con- 


cerned. 

The  examples  known  to  us   are,    however,   so   numerous   now 
that  there  is  no  longer  any  difficulty  in  defining  the  characteristics' 

66 


MEMPHITE  ART 


of  the  school.  In  the  first  place,  their  technique  is  extraordina- 
rily perfect,  even  in  hastily  executed  hypogea,  and  we  are  in- 
clined to  wonder  in  our  surprise,  what 
kind  of  discipline  the  heads  of  work- 
shops can  have  accepted  for  them- 
selves and  imposed  upon  their  pupils, 
to  produce  such  confidence  and  pre- 
cision in  the  handling  of  brush  and 
chisel.  The  line  with  which  they  en- 
velope bodies  and  objects  is  not  stiff 
and  inflexible  as  we  might  think  at 
a  first  glance;  it  swells,  diminishes, 
and  contracts  according  to  the  nature 
of  the  forms  it  indicates  and  the 
movements  which  animate  them.  Not 
only  do  the  flat  surfaces  contain  the 
summary  indication  of  the  bony  struc- 
ture and  the  large  planes  of  the  flesh, 
but  the  muscles  are  suggested ,  each 
in  its  place ,  by  projections  so  slight 
and  depressions  so  delicate  that  we 
fail  to  understand  how  the  craftsman 
can  have  produced  them  with  the 

poor  tools  at  his  command;  the  fine  white  limestone  of  Turah 
could  alone  have  enabled  him  to  preserve  them  in  a  relief  which 
in  parts  is  no  more  than  two  millimetres  high.  The  science  of 
the  composition  is,  unfortunately,  greatly  inferior  to  that  of  the 
material  execution. 
In  most  cases  the 
participants  in  a 
common  action, 
who  would  be  in- 
termingled by  an 
artist  of  our  own 
times,  are  ar- 
ranged separately 
one  after  another, 
as  in  a  procession. 
Men  or  beasts,  they 
present  themselves 
in  profile  against 
the  background,  their  faces  turned  to  the  point  of  common 
interest  or  attraction,  save  in  cases  where  an  accidental  neces- 

67  F2 


FIG.   114. — THE  MEMPHITE  FOR- 
MULA.     PTAH-HETEP    AND    HIS 
WIFE  (After  Prisse  d'Avesnes). 


FIG.   IIS.— HERDSMEN  DRIVING  BITLLS.     TOMB  OF 
PTAH-HETEP.     (Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


sity  forces  them  to  inflect  some  portion  of  the  body  in  the 
opposite  direction,  as  when  the  reaper  talks  with  his  neighbour 
in  the  interval  between  two  strokes  of  his 
sickle,  or  the  cow  turns  her  head  to  look 
at  the  calf  she  is  suckling  (Fig.  112),  or 
the  herdsman  who  is  milking  her;  here  the 
head  and  neck  are  thrown  back  upon  the 
shoulder  with  such  force  that  they  would 
be  dislocated  permanently  if  the  animal 
were  thus  posed  in  reality.  When  it  was 
impossible  to  bring  all  the  figures  to  the 
front  without  destroying  the  unity,  and  con- 
sequently the  ritual  efficacy  of  the  scene, 
the  artist  made  no  attempt  to  fix  their 
relative  positions  by  any  artifice  of  drawing 
or  perspective,  but  planted  them  one  against 
the  other,  as  if  they  had  all  been  standing 
upon  the  same  vertical  plane.  The  deceased 
recognised  the  propriety  of  this  device  in 
dealing  with  all  the  episodes  of  the  posthu- 
mous life  and  the  details  of  sacrifice;  but 
he  would  not  tolerate  it  in  the  vast  pano- 
FIG.  iie.-NEFER-  ramas  which  professed  to  display  to  him 

SESHE'MPTAH  WALKING      the  sum  of  pleasures  or  occupations  neces- 
(After  Capart).  sarv  to    nis   eternal   happiness.     The   artist 

decomposed    these    into   groups    which    he 

staged  one  above  the  other;  those  which  with  us  would  occupy 
the  foreground  were  placed  at  the  bottom  of  the  wall,  and  the 


FIG.   117.— BRAWL  ON  THE  WATER   (Museum,   Cairo).      (Phot.  E.  Brugsch.) 

more   distant  episodes  at  the  top :    boatmen   quarrel  on  a  pond 
or  a  canal,  fowlers  snare  birds  in  the  thickets  of  the  shore,  and 

68 


MEMPHITE  ART 


carpenters  build  boats  above  the  fowlers,  while  hunters  press  the 
animals  of  the   desert   up  against  the  ceiling  (Fig.  113).     These 


FIG.    Il8. — BRAWL  BETWKN  BOATMEN.  TOMB   OF  PTAH-HETEP.    (Phot.  E.  Brugsch.) 

are  awkwardnesses  which  we  should  wonder  to  find  persisting 
among  the  Memphites,  if  we  did  not  know  that  at  the  other 
extremity  of  the  Oriental  world  such  consummate  draughtsmen 
as  the  Chinese  and  Japanese  were  long  the  slaves  of  conventions 
no  less  puerile.  It  would  seem  as  if  when  once  certain  habits 
of  seeing  and  transcribing  the  object  have  been  contracted,  the 
eye  of  the  races  most  susceptible  of  progress  is  for  ever  sealed 
to  other  impressions,  and  that  it  becomes  incapable  of  con- 
ceiving representations  more  consonant  with  reality  than  those 
which  sufficed  it  in  the  beginning.  The  Memphite  School,  perhaps 
the  most  gifted  of  those 
which  flourished  on  Egyp- 
tian soil ,  accepted  the 
abnormal  structure  of  the 
human  person  imposed 
upon  it  by  its  Thinite  or 
pre-historic  precursors,  in 
default  of  knowing  how 
to  present  the  truth  cor- 
rectly on  a  flat  surface; 
it  continued  obediently 
to  plant  a  head  in  profile 
with  an  eye  full  to  the 
front,  upon  a  bust  facing 
the  spectator,  and  sur- 
mounting an  abdomen  threequarters  to  the  front  supported  by 
legs  in  profile,  and  this  formula,  legitimised,  as  it  were,  by 


FIG.  Iig. — DANCERS  IN  THE  TOMB  OF  ANKHMABA 
(After  Capart). 


ART  IN  EGYPT 


the  talent   of  those  who  employed  it,  was  perpetuated  without 
any  modifications  to  the  end  (Fig.  114).    Nevertheless,  a  certain 


FIG.  120.— CRAMMING  GEESE,  IN  THE  TOMB  OF  TI. 

liberty  of  action  is  allowed  in  the  case  of  secondary  personages, 
workmen,  peasants,  scribes,  fishermen  and  hunters,  servants  and 
slaves,  whose  mode  of  life  necessitated  attitudes  that  varied  from 
moment  to  moment,  attitudes  which  the  craftsman  was  not,  in- 
deed, always  capable  of  expressing  correctly,  as  in  the  case  of 

a  man  walking  (Fig.  115),  which  he 
has  only  succeeded  in  rendering  by 
dislocating  the  legs,  or  by  violently 
twisting  the  shoulder  nearest  to  the 
spectator  and  pressing  it  flat  upon 
the  torso  (Fig.  116). 

These  are  faults  very  well  calcu- 
lated to  repel  the  modern.  But  if  we 
make  an  effort,  and  force  ourselves  to 
overcome  this  initial  repugnance,  it  is 
impossible  not  to  be  fascinated  by 
the  merits  we  discover  when  we  analyse 
these  awkward  compositions.  As  the 
decoration  of  the  tombs  did  not,  like 
that  of  the  temples,  depict  grave  and 
sedate  personages,  who  could  not  un- 
bend without  disrespect  to  the  majesty 
of  the  gods,  the  artists  who  worked 
on  the  former  have  allowed  their  figures 
full  liberty  of  action,  and  have  drawn 
them  with  a  fidelity  which  astounds 

the  student,  who,  knowing  how  closely  the  Egypt  of  the  past 
resembles  that  of  the  present,  is  able  to  appreciate  the  truth 

70 


FIG.    121.— THE  AGED 

CHEPHREN   (Museum,   Cairo). 

(Phot.  E.  Brugsch.) 


MEMPHITE  ART 


FIG .  122 . — RA-NEFER 

(Museum,   Cairo). 
(Phot.  E.  Brugsdi.) 


of  their  observation.  These  ancient  people 
of  the  hypogea,  intent  on  their  tasks  for 
centuries,  scribes  or  servants,  shoemakers, 
goldsmiths,  joiners,  potters,  are  with  us  still 
in  their  offices  or  their  sheds;  we  recognise 
their  manner  of  walking  or  crouching,  of 
preparing  their  work  and  handling  their  tools. 
And  if,  passing  from  the  towns  where  gesture 
is  apt  to  become  constrained  and  the  body 
to  become  heavy,  we  note  the  outdoor  pur- 
suits which  necessitate  incessant  vigour  and 
flexibility,  could  there  be  a  more  rhythmic 
march  or  a  more  lively  impulse  than  among 
those  reapers  who  advance  in  a  line,  cutting 
down  the  corn  (cf.  Fig.  20),  or  those  moun- 
tain hunters  with  arrow  strung  to  pierce  the 
prey,  or  lasso  coiled  to  entangle  it  (Fig.  11 3). 
Take  any  one  of  the  brawls  between  boatmen, 
that  in  the  Museum  of  Cairo  (Fig.  117),  or 
that  which  we  admire  in  the  Tomb  of  Ptah- 
hetep  (Fig.  118).  Three  boats  are  engaged,  that  in  the  middle 
against  the  two  others,  and  while  several 
of  the  crew  exchange  blows,  others 
continue  to  work  the  craft.  One  is 
planted  firmly  upon  his  left  leg,  his 
chest  expanded,  his  neck  stiffened, 
his  hand  thrown  back  vigorously  behind 
his  head,  and  we  await  the  blow  with 
which  he  intends  to  strike  down  his 
antagonist;  the  latter,  however,  is 
ready  for  him,  and  with  his  knee 
against  the  prow,  he  thrusts  his  weapon 
straight  at  his  assailant's  side.  In  the 
Tomb  of  Ankhmara  there  are  dancing 
girls  who,  balanced  steadily  on  the 
right  leg,  bend  back  their  bodies  and 
kick  with  the  left  foot  above  their 
heads  (Fig.  119).  All  the  bodies  are 
strained,  all  the  muscles  work;  the 
figures  straddle,  lean  back,  thrust  them- 
selves forward,  shove  with  the  boat- 
hook,  stretching  wide  their  arms  or  throwing  back  their  legs, 
and  among  all  these  violent  attitudes,  there  is  not  one  which 

71 


FIG.    123. — THE  BREWER 
(Museum,  Cairo).  (Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


<r 


II 


FIG.    124. — WOMAN   GRINDING  CORN 
(Museum,  Cairo).     (Phot.  E.  Brugsdi.) 


does  not  correspond  accurately  with  the  effort  made.    Our  modern 
sculptors    might    treat    the    subject    differently;    they   could    not 

treat  it  better,  and  how 
many  among  them  could 
render  the  aspect  of 
animals  with  so  much 
sincerity?  Here,  in  the 
Tomb  of  Ti,  are  ducks 
and  geese  which  their 
keepers  are  fattening  by 
cramming  them  with  large 
pellets  of  some  appar- 
ently unsavoury  com- 
pound; the  ordeal  ever, 
they  are  walking  about 
to  get  over  their  agitation 
(Fig.  120).  The  artist  has 
noted  the  sex  charac- 
teristics so  well  that  we 
are  able  to  distinguish 
his  males  from  his  fe- 
males by  the  carriage  of  the  head  or  the  outline  of  the  body, 
and  in  addition,  he  has  marked  the  wagging  of  tails,  the  arching 

of  necks,  the  preen- 
ing     of      feathers, 

the    stretching    out 

of  beaks   in   which 

they    betray    their 

feelings,   and  their 

delight     at    having 

got    over    the    evil 

moment.  The  geese 

of  Medum  are   fa- 
mous (Fig.  92),  and 

they  show  us  what 

painting  might  have 

done  if  its  fragility 

had  not  discredited 

it  in  the  eyes  of  a 

nation    where    no- 
thing  impermanent 

was  esteemed.    The 

sculptorhasrecount- 
72 


FIG.  125. — HUSBAND    AND 

WIFE   STANDING. 

(Museum,   Cairo). 

(Phot.  E.  Brugsdi.) 


FIG.  126. — HUSBAND  AND 

WIFE  SEATED 
(Museum,  Cairo). 
(Phot.  E.  Brugsdi.) 


ed  the  life  of  the  desert  beasts  with  the  utmost  rres\ness  oT 
design,  showing  the  hare  crouching  behind  a  tuft  ol  jrass,  th,-: 
hedgehog  emerging  from  his  hole  to  catch  a  grasshopper,  the 
gazelle  suckling  her  fawn,  the  oryx  in  full  flight  and  the  grey- 
hound pulling  him  down;  as  to  the  domestic  animals,  he  who 
has  seen  the  Egyptian  flocks  of  to-day  returning  from  pasture, 
the  sheep  and  goats  in  dusty  disarray,  the  donkeys  trottin  and 
shaking  their  ears,  the  slow,  ruminating  oxen, 
outlined  in  a  dry  silhouette  against  the  slope, 
has  also  seen  at  a  glance  the  finest  bas-reliefs 
of  Ti  or  Mereruka. 

Statuary  developed  in  a  domain  less  vast 
and  consequently  with  less  freedom  of  inspir- 
ation than  bas-relief.  The  attitudes  between 
which  the  utilitarian  tendencies  of  religion 
permitted  a  choice  were  of  two  kinds,  and 
these  were  determined  by  the  condition  of 
the  model:  either  he  was  noble,  and  his  statue 
represents  him  seated  or  standing,  in  the  cos- 
tume of  his  class,  or  he  was  of  plebeian 
origin,  and  in  this  case  it  showed  him  in 
the  most  significant  of  his  professional  atti- 
tudes. There  were,  however,  exceptions  to 
this  rule:  it  happened,  perhaps,  that  some  noble 
attached  to  the  King's  household  agreed  to 
be  represented  in  a  posture  characteristic  of 
his  office,  and  not  in  that  proper  to  his  rank, 
while  a  low-born  scribe  or  even  an  artisan 
might  claim  the  semblance  of  a  person  of 
rank  for  his  stone  double.  But  in  no  case, 
not  even  when  workers  were  represented,  was 
it  legitimate  to  give  to  statues  those  con- 
torted and  ill-balanced  attitudes  which  abound  in  the  bas-reliefs. 
They  continue  almost  invariably  to  observe  the  law  of  frontality, 
a  convention  due,  not  to  the  incompetence  of  the  craftsman, 
but  to  ritual  obligation.  They  confront  the  spectator,  and  the 
top  of  the  skull,  the  junction  of  the  neck,  the  navel  and  the 
fork  of  the  legs  are  in  a  line  on  the  same  vertical  plane,  without 
the  slightest  deviation  to  right  or  left.  The  Egyptians,  in  fact, 
were  a  leisurely  race,  upon  whom  the  fevers  of  our  age  would 
have  had  little  hold,  and  to  them  gravity  carried  to  the  verge 
of  hieratic  immobility  was  the  supreme  mark  of  birth  and  autho- 
rity. The  effigy  of  the  prince  was  expected  to  be  what  the 

73 


FIG.    127.— HUSBAND 
AND     WIFE    OF    UN- 
EQUAL HEIGHTS 
(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


FIG.  128. — THE  MOST 

FREQUENT  TYPE  OF 

THE  SEATED  STATUE 

(Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


prince  himself  had  been, 
at  least  on  days  of  cere- 
monious reception,  se- 
rious,   impassible,    the 
chin     held  .  high ,     the 
bust  upright,  the  thighs 
parallel ,    and    the  feet 
firmly   planted    on   the 
same     line,     if    seated 
(Fig.  121),    or  the   left 
leg   advanced    and    all 
the  weight  concentrated 
on    the    right    leg,    if 
standing  (Fig.  122).  The 
plebeian  and  the  slave 
imitated  the  bearing  of 
courtiers    and    nobles, 
and   their   images   per- 
form  their    tasks   with 
a    calm    and    sobriety 
scarcely  inferior  to  the 
composure    of   their   masters,   whether   they 
toil    at    the    kneading   trough   (Fig.  123)   or 
kneel  over  the  stone  to  grind  corn  (Fig.  124). 
Women  were  treated  according  to  the  class 
to    which    they    belonged,    and    the    king's 
daughter    or   the    great    lady    invested    with 
rights   equal  to   those  of  her  husband   pos- 
sessed  like  him,  her  independent  image,  or, 
if  they  were  associated  in  a  group,  she  stood 
(Fig.  125)   or  sat   on  the  bench  beside  him, 
laying  her  arm  across  his  shoulders  in  token 
of  affection  (Fig.  126).  Nevertheless,  as  he  was 
the  head  of  the  family,  round  whom  all  the 
other    members   gathered   for   worship,    she 
allowed  herself  to  be  represented  either  of 
the    same    dimensions    as   he,    but   standing, 
while  he  was  seated  on  the   chair   of  state, 
or  on  a  much  smaller  scale,  her  back  against 
the  front  of  the  seat,  with  her  children,   or 
nestling  affectionately  against  his  leg  (Fig,  1 27). 
She   is   always    clothed,    but    the    boys    and 
even   the   men,    both    free    and    slaves,    are 

74 


FIG.    129. 
CHEOPS  IN  IVORY 

(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


TIG.   1JO. 

IVORY    BAS-RELIEF. 
(Phot.  Bounant.) 


MEMPHITE  ART 


sometimes  naked;  this  may  have 
been  in  obedience  to  some  religious 
prescription,  or  perhaps  upon  certain 
occasions  these  nude  figures  were 
dressed  in  real  garments,  like  the 
Madonnas  of  the  present  day  in 
Italy.  Broadly  speaking,  it  may  be 
said  that  there  are  only  some  fifteen 
attitudes,  some  of  which  are  very 
rare,  among  this  nation  of  statues 
derived  from  Memphite  tombs,  and 
it  is  hardly  surprising  that  the 
visitors  to  our  museums  should 
end  by  feeling  a  certain  weariness 
as  he  confronts  them  (Fig.  128).  This 
is  not  altogether  the  fault  of  the 
Egyptians;  we  ourselves  are  to  blame 
for  having  crowded  together  in  two 
or  three  gloomy  rooms  works  ori- 
ginally dispers- 


FIG.    TJT.— BUST   OF  A  WOODEN 

STATUE  (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


FIG.    132. — THE  TWO 
BRONZE    STATUES 
(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


ed  in  a  hundred  different  places.  Those  who 
visit  the  galleries  in  the  Louvre  devoted  to 
Greek  and  Roman  sculpture  are  sometimes 
oppressed  by  a  kindred  sense  of  monotony 
and  disgust,  in  spite  of  the  greater  variety 
of  types  and  movements. 

Stone  was  the  favourite  material,  pink 
or  black  granite,  diorite,  green  breccia,  schist, 
red  sandstone,  alabaster,  the  white  limestone 
of  Turah,  and  the  Memphites  cut  the  hardest 
of  these  with  a  dexterity  which  amazes  us, 
when  we  remember  that  they  had  no  know- 
ledge of  steel,  and  that  their  tools  were  of 
flint,  bronze,  and  untempered  iron.  It  was 
therefore  no  lack  of  manual  dexterity  which 
caused  them  not  to  disengage  certain  statues 
and  groups  entirely,  but  to  keep  them  nearly 
always  with  their  backs  against  a  rectangular 
slab,  which  protrudes  sometimes  on  either 
side  like  a  wall  against  which  they  are  leaning, 
and  sometimes  is  reduced  to  the  semblance 
of  a  pillar  terminating  squarely  at  the  level 
of  the  shoulders  or  the  neck,  or  in  a  point 
75 


ART  IN  EGYPT 


which  is  lost  in  the  hair.     They  had  no  difficulty  in  suppressing 
this  buttress  when  they  pleased,    and  if  they  generally  retained 


FIG.  133. — BUST  OF  THE  STATUE  OF  PEPI I. 
(Museum,  Cairo).  (Phot.  E.  Brugsch.) 


FIG.  134.— HEAD  OF  THE  STATUETTE 
(Museum,  Cairo).  (Phot.  E.  Brugsch.) 


it,  it  was  out  of  respect  for  a  tradition  established  at  a  time 
when  the  artist  would  have  feared  to  weaken  his  work  and 
diminish  its  chances  of  duration  by 
omitting  it.  They  accordingly  continued 
to  the  end  not  to  separate  the  arms 
from  the  trunk,  and  to  retain  a  solid 
partition  between  the  leg  on  which  the 
body  rested  and  that  which  was  in 
advance.  I  am  inclined  to  believe 
that  the  types  in  which  these  imper- 
fections occur  are  the  most  ancient  of 
those  which  were  invented  for  the 
double,  but  that,  on  the  other  hand, 
those  in  which  we  do  not  find  them 
were  created  later,  when  the  school, 
after  long  practice,  had  so  far  gained 
confidence  in  its  strength  as  to  discard 
them.  The  ritual,  though  it  regulated 
artistic  themes  very  strictly  in  the  be- 
ginning, did  not  define  those  of  more 
ecent  invention  with  the  same  rigour  ; 

76 


FIG.  135.— THE  KHASAKHMUI 
AT  CAIRO.   (Phot.  E.  Brugsch.) 


MEMPHITE  ART 


FIG.  136. — THE  KNEELJNG  MAN 
AT  CAIRO  (Phot.  E.  Brugsch.) 


FIG.IJ7.-STATUE  NO. I  AT 
CAIRO.  (Phot.  E.  Brugsch.) 


great   personages    accordingly    continued    to   be    repre? 
statues  partially  engaged  in  the  stone,  while  the  other 

the     servants     of 
the  double,  millers, 
bakers ,     brewers, 
mourners,  and  do- 
mestic dwarfs  had 
free  statues.  Very 
soon,  too,  the  sup- 
porting slabs  were 
used  for  the  ben- 
efit of  individuals; 
their  names,  titles, 
parentage,  the  for- 
mulae   of    incant- 
ation     were      in- 
scribed upon  them, 
and    the     advan- 
tages they  derived 
from  this  practice 
in  their  life  beyond 
the  tomb  was  no 
doubt  a  factor  in  the  retention  of  these  surfaces.    Wood,  ivory, 
and   metal   had   never   been   regarded  with    the 
same  distrust  as  stone,  and  their  firm  yet  flexible 
texture  enabled   the  artists  who   used   them   to 
disengage   their  works   entirely;    yet  they,    too, 
submitted    to    technical    exigencies   which    must 
be  noted.     Ivory  was  only   used    for  small  bas- 
reliefs   and    statuettes,    such   as   the  Cheops  at 
Cairo,  discovered  by  Petrie  at  Abydos  (Fig.  129), 
and  the  bas-reliefs    of   the  Fifth  Dynasty,  frag- 
ments of  which  were  found  at  Sakkarah  (Fig.  130). 
Both  are  very  carefully  worked,    but  they  have 
no   great    artistic  merit.     Egypt   produces   little 
wood  fit  for  carving,  and  that  which  was  bought 
in  Syria  or  Caramania,  pine,  cedar,  and  cypress, 
arrived  in  beams  and  blocks  too  small  to  serve 
for  the   carving  of  a  life-size  figure.     A  trunk, 
a  head,  and  sometimes  legs  were  obtainable,  but 
the  arms,    unless   these  were   incorporated  with 
the  body,  and  generally  speaking  the  legs,  were 
joined  to    the   rest  (Fig.  131);    the   pieces  were 
77 


FIG.  138.— THE 
LADY  NASI   (The 

Louvre,  Paris). 
(Phot.  Bouriant). 


ART  IN  EGYPT 


FIG.  I3Q. 

ARCHAIC   STATUETTE   OF  A 

WOMAN   (After  Pleyte) 

(Museum,  Turin). 


fitted  together 
by  means  of  rect- 
angular tenons, 
and  as  the  whole 
was  lightly  over- 
laid with  stucco 
and  then  painted, 
the  joints  dis- 
appeared. Metal, 
gold  or  silver, 
bronze  or  copper 
would  have  easily 
furnished  large 
pieces  all  in  one 
if  the  art  of 
the  founders  had 
been  more  ad- 
vanced ;  but  it 
seems  evident  to 
me  that  they 
only  dared  to 


FIG.     I4O. — ARCHAIC 

STATUETTE      OF      A 

WOMAN  (AfterCapart) 

(Museum,  Brussels). 


operate  on  modest  quantities,  and  that  they  did  not  known  how 
to  prepare  large  moulds.  Figurines  and  amulets  were  accordingly 
cast  whole,  but  statues  were  partially  hammered  out.  The  face, 

hands  and  feet,  all  the  parts  which 

demanded  delicacy,  were  made 
in  moulds.  The  bust,  the  arms 
and  the  legs  were  merely  repousse 
plates,  mounted  upon  a  common 
core,  and  put  together  with  rivets. 
It  was  thus  that  the  statue  of 
Pepi  I. ,  and  the  statuette  found 
with  it  at  Hierakonpolis  (Fig.  132) 
were  ompossed.  The  framework 
of  these  was  of  wood,  the  petti- 
coat of  gold,  and  the  headdress 
of  lapis-lazuli.  As  was  to  be  ex- 
pected, the  apron  and  the  wig 
have  disappeared;  their  material 
value  tempted  thieves  in  ancient 
times.  In  spite  of  the  rudeness 
of  the  technique  and  the  muti- 
lations they  have  suffered,  they 
78 


FIG.    141. — THE    SPHINX    OF    GIZEH. 
(Phot.  E.  Brugsch.) 


MEMPHITE  ART 


FIG.  142.— THE  GREAT 

CHEPHREN  AT  CAIRO. 

(Phot.  E.  Brugsdi.) 


are  two  very  remarkable  examples  (Figs.  133 
to  134),  which  hold  their  own  even  beside 
works  like  the  diorite  Chephren. 

The     earliest     statues     belong     to     two 
schools;    those  of  Pharaoh  -  Khasakhmui   and 
the  crouching  man  of  Kom-el-Ahmar  to  the 
Thinite,    the    No.    1    Cairo    statue    to    the 
Memphite.       The     Khasakhmui      at      Cairo 
(Fig.  135),  the  finer  of  the  two,  is  of  schist,  half 
the  size  of  life,   and  though  its  author  had 
not  thrown  off  a  certain  stiffness   and  awk- 
wardness, it  bears  witness  to  a  dexterity  of 
no  mean  order  in  the  use  of  the  chisel.    The 
king  is  dressed  as  Osiris  for  his  deification 
in    the  festival  of  habi  sadu ,  the  high  white 
cap   on    his   head,    his    short    cloak    drawn 
closely  about  him,    and  while   his   left   arm 
and    hand   are   defined   under   the   drapery, 
the  right   hand   and   arm  are  laid  along  the 
knee.    The  head  has  lost  the  right  half,  but 
if   we   reconstruct    it  with    the    limestone   fragment   at   Oxford, 
we     divine     the     true     portrait,     modelled    with    a    somewhat 
rude    touch,   but   with    a    perfect    comprehension    of    anatomy, 
and    of   the    processes    required    for 
its   faithful   expression.     It  is  a  good 
example,  which  I  ascribe  to  the  royal 
workshop,  and  its  merits  are  the  more 
striking  when  we  compare  it  with  the 
crouching  man  (Fig.  136).    This  is  the 
product   of  a  private  workshop,   and 
the  style  is  so  rough   and  heavy  that 
we  might   naturally    suppose  it  to  be 
earlier  than   the  Pharaoh.     But   close 
examination    shows     that    its    short- 
comings are  due  less  to  archaism  than 
to   provincial   clumsiness,    and  I  hold 
the  same  opinion  of  the  granite  statue 
No.  1  at   Cairo   (Fig.  137);    here  the 
head  is  too  large,  the  neck  too  short, 
the  torso  too  thickset,   the  leg  badly 
formed,  the  foot  perfunctory.     These 
faults  are  repeated  in  varying  degrees 
in  the  similar  statues  or  groups  from 

79 


FIG.  143.— THE  ALABASTER 
CHEPHREN  AT  CAIRO 

(Phot.  E.  Brugsdi.) 


ART  IN  EGYPT 


Sakkarah  and  Gizeh  scattered  in  European  museums.  The  most 
famous  are  the  Sapui  and  the  Nasi  in  the  Louvre  (Fig.  138), 
but  there  are  others  at  Turin  (Fig.  139),  at 
Naples,  at  Munich,  at  Brussels  (Fig.  140) 
and  at  Leyden.  They  have  certain  stylistic 
features  in  common,  the  short,  thick  neck, 
the  head  pressed  down  between  the  shoul- 
ders, the  round,  massive  body,  the  ill-drawn 
leg  and  foot.  We  shall  realise  the  contrast 
between  this  provincial  art  and  the  art  of 
the  Court,  if  we  compare  these  works  with 
the  Ra-hetep  and  the  Neferet  of  Medum, 
their  contemporaries  within  a  few  years  (See 
Frontispiece).  These  date  from  the  time  of 
Seneferu,  the  century  in  which  political  vicis- 
situdes transported  the  royal  workshop  from 
This  to  the  Memphite  plain.  The  bearing  of 
the  man,  with  his  intelligent  face,  his  broad 
shoulders,  his  slender  torso,  and  slim  legs 
is  full  of  spirit  and  vivacity,  but  the  woman 
is  a  masterpiece,  perhaps  the  masterpiece  of 
this  archaic  sculpture.  Not  only  do  the  head 
and  face  stand  out  in  the  most  vigorous 
manner  from  the  enframing  wig,  but  the 
bust  and  hip  are  revealed  with  discreet  ele- 
gance beneath  the  white  mantle.  The  colour 
and  the  enamelled  eyes  contribute  to  the  effect  of  reality,  and 
almost  produce  the  illusion  of  life.  Should  the  Sphinx  of  Gizeh 

be  assigned  to  the  same 
period  and  the  same  school? 
It  has  been  the  fashion  for 
the  last  twenty  years  to 
rejuvenate  the  monuments 
to  which  the  Egyptologists 
of  the  first  two  generations 
assigned  great  antiquity.  The 
Sphinx  (Fig.  141)  has  not 
been  spared,  and  several 
scholars  have  brought  it 
down  to  the  Eighteenth  Dy- 
nasty. It  is  true  that  it  has 
undergone  countless  restor- 
ations in  the  course  of  its 


FIG.  144.—  REISNER'S 
MYCERINUS  (Museum, 
Cairo).(Phot.E.Brugsch). 


FIG.    145. — MYCERINUS  AND  HIS  WIFE 
(Museum,  New  York).  (Phot.  Reisner.) 

80 


MEMPHITE  ART 


existence,    but   patched    though   it  be,   it  retains  enou 
primitive  appearance  to  entitle  it  to  be  classed  as  contc 
with   the    Pyramids,    if   not   anterior   to 
them.     In  spite  of  the  mutilations  which 
have    disfigured    it,    I  believe  I  can   re- 
cognise  in  it   the   characteristics   of  the 
two  statues   of   Medum,    works   of   the 
Thinite  School  at  its  apogee. 

There  is  always,  even  in  the  most 
accomplished  works  of  the  Thinites,  a 
something  stiff  and  angular;  the  Memphite 
artists  whom  the  Pharaohs  summoned  to 
the  royal  workshops  soon  lost  their 
awkwardness,  but  preserving  that  tenden- 
cy to  roundness  shown  in  their  early 
productions,  they  evolved  a  fat  and 
supple  touch  which  distinguishes  them 
from  their  masters.  They  had  that  re- 
spect for  material  truth  which  was,  in- 
deed, enjoined  by  their  religion,  but 
at  the  same  time  they  permitted  them- 
selves to  idealise  the  features  of  their 
models  as  far  as  this  was  compatible 
with  the  exigencies  of  likeness.  They 
delicately  attenuated  certain  curves  of 
nose  and  chin  which  seemed  to  them 
ungraceful,  they  filled  out  hollow  cheeks,  refrained  from  sinking 
the  eye  too  deeply  in  the  orbit,  sloped  the  shoulders  slightly, 
and  modified  the  prominence  of  the  muscles  on  the  arms,  legs, 
and  bust.  The  best  among  them 
thus  succeeded  in  creating  sta- 
tues or  groups  of  much  harmony 
and  nobility,  in  which  energy 
was  not  wanting  upon  occasion. 
Their  qualities  proclaim  them- 
selves as  early  as  the  middle 
of  the  Fourth  Dynasty,  in  the 
admirable  series  of  royal  effigies 
preserved  in  the  Cairo  Museum. 
The  great  Chephren  (Fig.  142) 
discovered  by  Mariette  in  1859 
in  the  temple  of  the  Sphinx  is 


FIG.  146. — ONE   OF  THE  TRI- 
ADS OF  MYCERINUS  (Museum, 
Cairo).     (Phot.  E.  Brugsch.) 


in   diorite,   the   most  obdurate 


81 


FIG.    147. — HEAD   OF  DIDUFRIYA 

(Museum,  Cairo).    (Phot,  E.  Brugsch.) 


ART  IN  EGYPT 


FIG.    148. — RA-NEFER 

(Museum,   Cairo). 
(Phot.  E.  Brugsdi.) 


FIG.  I4Q.— THE 
DWARF  OF  GIZEH 

(Museum,  Cairo). 
(Phot.  E.  Brugsdi.) 


material  imaginable;  it  is  attacked  here 
with  so  much  boldness  that  it  seems  to 
have  lost  its  hardness.  Like  the  majority 
of  statues  in  dark  stone,  such  as  black 
or  red  granite,  or  green  breccia,  it  was 
only  painted  in  parts;  parts  of  the  face, 
the  eyes,  the  nostrils,  the  lips,  and  certain 
details  of  the  costume  were  heightened 
with  red  and  white.  The  polish,  and 
the  multiplicity  of  glazes  it  entailed, 
masks  the  modelling  a  little:  it  is  ne- 
cessary to  study  it  for  a  long  time  and 
in  a  variety  of  lights  to  perceive  its 
perfection  and  its  masterly  simplicity. 
What  again  can  be  said  of  the  manner 
in  which  the  king  is  set  on  his  low- 
backed  seat,  while  the  hawk  behind  him 
spreads  its  wings  to  shield  his  head  and 
neck?  Rarely  has  royal  majesty  been 
rendered  with  so  much  breadth.  The 
sculptor,  while  faithfully  reproducing  the 
features  of  the  reigning  Pharaoh,  has 

further    succeeded    in 

rendering  the  idea  of 

sovereignty    itself ;    it 

is  not  only  Chephren 

whom  he  calls  up  be- 
fore   our    eyes,     but 

Pharaoh     in    general. 

The    same   expression 

of     serene     grandeur 

reappears ,   though   in 

a    lesser    degree,     in 

the  alabaster  statuette 

(Fig.  143),  the  statue 

in  green  breccia  which 

shows  Chephren  when 

a     little     older     (cf. 

Fig.  121)  and  the  ala- 
baster    and     granite 

statuettes    of    Mycer- 

inus,  Ra-en-user,  and 

Menkhau  -  Heru.     The 
82 


FIG.     IgO. 

THE    DWARF    KHNEMU- 

HETEP  (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


FIG. 151. — THE  SHEIKH- 
EL, -BELED    (Museum, 
Cairo).  (Phot.E.Brugsch.) 


alabaster  sta- 
tue of  the  seat- 
ed Mycerinus, 
which  Reisner 
collected  piece 
by  piece  in  1908 
near  the  third 
Pyramid  is  re- 
markable above 
all  for  the  beau- 
ty of  the  stone 
(Fig.  144);  the 
figure  is  not 
well  balanced 
upon  the  seat, 
and  the  head 
is  too  small  for 
the  body.  It  may 
be ,  however, 


FIG.    152.— THE    SHEIKH-EL- 

BELED  IN    PROFILE. 

(Phot.  E.  Brugsch.) 


that     here     the 

sculptor  faithfully  reproduced  a  peculiarity  of  the  sitter,  for  the 
other  statues  of  Mycerinus  show  the  same  disproportion.  Apart 
from  this,  it  must  be  admitted  that  the  group  in  schist  which 
represents  him  side 
by  side  with  his  wife 
(Fig.  145)  and  the 
four  geographical  tri- 
ads in  which  he 
stands  between  the 
Goddess  Hathor  and 
one  of  the  Nomes 
of  the  SaYd  deserve 
nothing  but  praise 
(Fig.  146).  The  sta- 
tues of  Didufriya, 
the  fruits  of  -exca- 
vations made  by 
Chassinat  at  the  Pyra- 
mid of  Abu-roash, 
were  almost  equal 
to  the  Chephren,  and  FIG.  154.— SUPPOSED 
may  have  been  by  WIFE  OF  THE  SHEIKH- 

.,     •'  ,  J  EL-BELED  (Museum, 

the    same    sculptor;      Cairo). (Phot.E.Brugsch.) 
83  G2 


FIG.    153. 

MYERS'  STATUETTE; 
(After  Capart), 


FIG.     I55-— WOODEN 

STATUE    AT    CAIRO. 

(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 

the  heads,  which  are  all  that  has  survived, 
have  been  so  outrageously  mutilated  that  it 
would  be  imprudent  to  make  any  assertions 
in  this  sense,  So  far,  we  know  of  no  stone 
statues  of  the  last  kings  of  the  Fifth  Dynasty, 
or  of  those  of  the  Sixth;  they  were  not  in- 
ferior to  those  of  their  predecessors,  if  we 
may  judge  by  the  contemporary  statues  of 
private  persons  which  have  come  down  to  us. 
It  is  probable  that  several  of  these  were 
executed  in  the  royal  workshops,  notably  the 
Cairo  Ra-nefer,  whose  lofty  majesty  is  almost 
comparable  to  that  of  the  Chephren  (Fig.  148); 
but  the  majority  must  be  attributed  to  the 
private  ateliers  of  the  Memphite  plain,  and 
as  the  sitters  belonged  to  all  classes  of 
society,  they  present  a  greater  variety  of  types 
than  the  royal  iconography.  Firstly,  we  have 
the  courtier  and  the  baron,  standing  to  re- 
ceive offerings,  with  arms  hanging  down,  and 
the  left  foot  advanced;  the  Ti  of  the  Cairo 
Museum  is  a  good  example,  almost  equal  to 
the  Ra-nefer,  but  others  are  interesting  chiefly  as  curiosities;  such 
are  the  circumcised  priest  Anisakha,  who  is  completely  naked, 

and  the  two  dwarfs  whose  defor- 
mities are  rendered  with  medical 
exactness,  without  any  touch  of 
caricature  (Figs.  149 — 150).  These 
are  in  white  Turah  limestone, 
heightened  with  vivid  colours. 
Kaapiru ,  the  famous  Sheikh  -  el- 
beled  (Fig.  151)  is  in  wood,  which 
enabled  the  sculptor  to  project 
the  left  arm  with  the  ceremonial 
wand,  and  to  give  lightness  to  the 
gait  by  detaching  the  legs  one 
from  another.  The  Sheikh-el-beled 
marks  the  apogee  of  Memphite  art, 
and  if  some  exhibition  of  the 
world's  masterpieces  were  to  be 
inaugurated,  I  should  choose  this 
work  to  uphold  the  honour  of 
Egyptian  art.  It  is  not  only  the; 
84 


FIG.  156.— BUST  OF  FIG.  155 
(Communicated  by  Messrs.   A.  Picard). 


ART  IN  EGYPT 


FIG.    Igg.— THE    CROUCHING    SCRIBE 
AT  CAIRO.     (Phot.  E.  Brugsch.) 


of  small  size,  the  finish  of  the  touch  corrects  the  trivial  and 
impersonal  quality  of  the  conception.  Our  Neferu  at  Cairo 

(Fig.  158),  Neferu  the  cooper,  who 
would  not  be  admired  at  all  if 
he  were  life-size,  appears  charming, 
thanks  to  his  small  dimensions.  Many 
visitors  would  like  to  take  him 
away  and  set  him  up  as  an  orna- 
ment on  their  shelves. 

The  crouching  scribe  and  the 
reading  scribe  are  sometimes  not 
easily  distinguishable  on  from 
another;  they  are  differentiated  by 
the  head,  which  is  more  inclined, 
and  the  crossing  of  the  legs,  which 
is  flatter  in  the  reader,  but  very 
often  the  sculptor  has  not  insisted 
on  these  differences,  and  the  types 
are  interchangeable,  or  nearly  so. 
They  serve,  however,  to  establish 
the  link  between  the  aristocracy 

and  the  commonalty,  citizens,  merchants  and  workpeople.  It  even 
happened  that  a  person  of  high  rank,  who  held  the  post  of 
secretary  to  the  Sovereign,  chose  the  attitude  of  a  professional 

scribe  for  his  double.  It  was 
in  itself  ungraceful  enough, 
reducing  the  individual  to 
about  half  his  height,  and 
replacing  the  slender  curves 
of  the  leg  by  a  sort  of  flat 
angular  sole  over  wich  the 
bust  was  planted.  The  Egyp- 
tians nevertheless  succeeded 
in  evolving  a  very  presentable 
type  from  these  mediocre  pre- 
mises. They  chose  the  mo- 
ment when  the  man,  having 
taken  up  his  position  on  the 
ground,  his  legs  bent  under 
him,  his  skirt  drawn  tightly 
over  his  thighs  and  his  arms 
stretched  across  his  lap  to 
counterbalance  the  bust,  pre- 


FIG.    l60.— THE  CROUCHING   SCRIBE 
(The  Louvre,  Paris). 


86 


MEMPHITE  ART 


FIG.    157. 

HEAD   IN   THE   LOUVRE 
(Drawing  by  Faucher-Gudin). 


head    which    is    so    perfect    here   (Fig.   152);    the    modelling   of 
the   body   has   been   elaborated   con   amore,   and   the   execution 
has  been  carried   as  far  as  that  of  the 
most  realistic  of  our  contemporary  sculp- 
tors.     The    man    was  a  rustic,   smooth 
shaven,    thickset,    short   in   the   leg,  of 
a   vigorous    but    plebeian    aspect;    he 
lived  in    offices  more  than   in  the  open 
air,  and  having  passed  his  fiftieth  year, 
he  suffered  from  the  superabundant  flesh 
usual  among   persons   of   his   class   and 
temperament.  Illustrations  give  but  little 
idea   of  him:   he   must   be   seen  in  his 
place   in   our   Museum   to    be   properly 
appreciated.     Both  back  and  front,   the 
artist  has  noted  the  tokens  of  approaching 
age  with  a  curious   insistence,    but   he 
has   stopped   short   at  the  point  where 
truth  threatened  to  trench  on  brutality. 
The   bust    formerly    in    the    Myers   col- 
lection (Fig.  153),    the   two  Cairo   torsoes,    one   of   a   man   (cf. 
Fig.  131),  the  other  of  a  woman  erroneously  called  the  wife  of 
the  Sheikh-el-beled  (Fig.  154),    and   the   statue   of  an  unknown 
young  man  (Figs.  155,  156)  are  certainly  less 
distinguished;  the  wood  is  carved  more  drily, 
and  the  whole    makes   an  impression   of  hard- 
ness which  was  not  perhaps  apparent  in  anti- 
quity, when  the  form  was  veiled   by  painting. 
For  the  rest,  it  may  be  said  that  the  majority 
of  the  stone  statues  or  groups  in  our  museums 
do    not    rise    above    mediocrity:     portions    of 
these  are  often  excellent,   the  heads  in   parti- 
cular (Fig.  157),    but   very    often    the    bodies 
are  imperfect,    with  the   feet   and   legs  barely 
indicated,    the  arrangement    of  the   persons  is 
ungraceful ,    and    the    gestures    by    which    the 
women    and    children    manifest    their   affection 
for  the  head  of  the  family  are  too  stiff  to  be 
elegant.     This   is   because   we    have    in    these 
objects  of  current  commerce,  manufactured  in 
the  shops  of  funeral  undertakers  by  sound  and 
well-trained    workmen     quite     devoid     of    in- 
spiration.   Sometimes,  indeed,  when   they  are 

85 


FIG.    158.— NEFERU 

(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


MEMPHITE  ART 


FIG.    l6l.— SADUNIMAT 
(Museum,   Cairo).     (Phot,  E.  Brugsch.J 


pares  to  read  or  write.    Sometimes  he  holds  an  unfolded  papyrus 

or  a  tablet  before  him,  and,  his  right  hand  resting  on  the  margin 

waits  for  the   dictation  to  begin; 

sometimes  again  he  has  laid  aside 

the  scroll,  and  is  meditating.    The 

crossing    of    the    legs    is    usually 

execrable.  The  sculptor  has  treated 

it  as  a  kind  of  reinforcement  of 

the   base,   and   has   neglected   it. 

On  the  other  hand,   the   torso  is 

generally  most   carefully  treated; 

it  is   either  slightly  hunched,   as 

in  the  Cairo  scribe,  or  drawn  up 

firmly   above   the   haunches,   like 

that   in    the   Louvre.    The   Cairo 

Scribe    (Fig.    159)    is    admirable, 

with  his  pitiful  mien,  his  peevish 

mouth,  his  large  eyes  which  seem 

to  meet  those  of  the  visitor  with 

a   kind   of   malevolence,    but  the 

Louvre  example  (Fig.  160)  surpasses  it  in  every  way,  and  if  we 

were  called  upon   to  classify  masterpieces  it  might  fairly  claim 

a  place  not  much  below  the  Sheikh-el-beled.    He  is  the  typical 

scribe,  vigorous.heal- 

thy,  and  sufficient- 
ly    provided     with 

the  stock  of  intelli- 
gence necessary  for 

his  craft;  he  smiles 

slightly ,     and     his 

features,   as  far  as 

they    express     any- 
thing,   suggest   but 

little  interest  in  his 

task,    and   a   good 

deal    of    boredom. 

The   seated  scribes 

and  readers  in  pink 

or  black  granite  at 

Berlin  or  Cairo,  Sad- 

unimat    (Fig.   161), 

or    Ra-hetep     have 

the    same   peculiar- 
87 


FIG.    162. — KNEELING 
SCRIBE  AT  CAIRO    (Draw- 
ing- by  Faucher-Gudin). 


FIG.  163. 

THE   COOK  AT  CAIRO. 
(Phot.  E.  Brugsdi.) 


ART  IN  EGYPT 


FIG.    164. — SERVANT    CARRYING 

HIS  MASTER'SBAGGAGE(Museum, 

Cairo).     (Phot.  E.  Brugsch.) 


ities  more  or  less  accentuated;  their  faces  are  stolid,  their  bodies 
inert,  and  in  spite  of  the  excellence  of  the  technique,  they  suggest 

the  hand  of  the  craftsman  rather  than 
that  of  the  artist.  The  kneeling  scribe 
(Fig.  162)  is  on  a  different  plane  alto- 
gether; the  sculptor  to  whom  we  owe 
this  must  certainly  have  studied  his 
model  very  closely,  for  he  has  brought 
out  all  the  professional  traits  of  the 
physiognomy.  Here  is  the  true  scrivener 
of  the  mudirieh,  with  his  resigned 
air  and  timid  mien,  his  hands  folded 
on  his  lap  in  sign  of  submission,  his 
back  bent  as  if  anticipating  blows.  On 
the  last  rung  of  the  social  ladder,  the 
slaves  of  the  dead  man,  or  sometimes 
the  dead  man  himself  assuming  the 
function  of  a  slave  to  serve  a  god, 
carry  on  their  various  occupations, 
and  they  would  have  given  opportu- 
nities for  endless  variations,  had  they 
been  confided  to  the  head  of  a  work- 
shop; but  whether  in  wood  or  stone, 

they  were  generally  entrusted  to  the  inferior  craftsmen.  This  explains 
why  the  majority  of  them,  grinders  of  corn,  male  or  female  (cf. 
Fig.  15  and  124)  brewers  (cf.  Fig.  123),  glazers  of  pottery  (cf. 
Fig.  16),  crouching  mourners,  cooks  trussing  or  roasting  a  goose 
(Fig.  163)  are  merely  plebeians  by  the  dozen,  correct  in  structure,  but 

devoid  of  any  in- 
dividual accent; the 
only  one  with  any 
originality  of  ap- 
pearance, as  far  as  I 
know,  is  that  wood- 
en servant  in  our 
Museum,  who  walk- 
ed along  following 
his  master,  a  bag 
over  his  shoulder, 
and  his  sandals  in 
his  hand.  (Fig.  164). 

Towards  the  close  of  the  Memphite  age,  these  figures  of  vassals 
and  slaves  multiplied,  and  formed  episodical  groups;  here  again 

88 


FIG.   165.— BAKERS   (Museum,  Cairo).     (Phot.  E.  Brngsch.) 


FUNERARY  LANDSCAPE 

Painted  on  the  Stele  of  Zadamonefonukhu 
(Museum,  Cairo) 


MEMPHITE  ART 


the  arrangement  was  a  utilitarian  one.  It  was  expensive  to  pre- 
pare a  tomb  with  carved  or  painted  walls,  so  the  after-life  remain- 
ed the  privilege 
of  rich  men  and 
nobles ;  to  extend 
it  to  the  greatest 
possible  number, 
the  pictures  which 
covered  the  walls 
were  reproduced  in 
the  round  by  means 
of  small  wooden 
dolls.  As  the  ob- 


ject was  to  procure        FIG.  166.— THE  KITCHEN  (Museum,  Cairo).  (Phot.E.Brugsch.) 

a  cheap  immortality 

for  the  poor,  they  are  nearly  always  rudely  executed,  and  must 
rank  rather  as  funerary  industry  than  as  art.  Thus  v/e  have  gardens 
and  arbours  where  the  double  sits  to  take  the  air  when  he  pleases, 
houses  and  granaries  where  coopers  and  scribes  gauge  the  corn, 
breweries,  bakeries  (Fig.  165),  and  kitchens  (Fig.  166).  A  narrow 
wall  separates  the  building  from  the  street;  its  rustic  door  is 
placed  near  the  corner;  butchers  kill  cattle,  and  cooks  roast 
geese  before  a  shed  arranged  as  a  storehouse,  in  which  we  see 
isolated  vessels  in  the  background,  and  in  the  front,  groups  of 
jars  for  corn,  barley,  wine,  and  oil.  A  little  further,  we  are  present 
at  a  concert  (Fig.  168).  The  dead  man  is  enthroned  in  a  kind 
of  stall,  and  at  his  right,  a  little  to  the  front,  a  young  woman, 
dressed  in  the  apron 
with  braces,  is  seat- 
ed on  a  chair;  two 
harpists ,  posted  on 
either  side ,  sing, 
clapping  their  hands. 
These  festive  episodes 
are  rare,  but  there 
are  innumerable  in- 
dustrial scenes,  where 
the  little  figures  are 
working  busily  for 
the  benefit  of  the 
deceased.  Joiners  saw 

beams   for   his   furniture.     Potters   turn   the   wheel    and  put  his 
crockery  into  the  furnace.    A  procession  of  yellow  women,  each 

89 


FIG.    167.— BAND    OF    OFFERING-BEARERS 
(Museum,  Cairo).     (Phot.  P.  Brugsch.) 


ART  IN  EGYPT 

flanked  by  a  small  brown  boy,  defiles  with  the  produce  of  his 
eternal  domains  (Fig.  167),  and  boats  await  him,  should  he  feel 
inclined  to  go  upon  the  river.  On  some  of  these  the  sails  are  set,  to 
go  up  the  Nile  favoured  by  the  "soft  wind  of  the  North."  Others 
have  taken  down  the  mast,  for  the  downward  passage;  the  sailors 
paddle,  and  the  pilots  are  at  their  posts.  All  these  were  sold  whole- 
sale, and  kept  in  the  workshops  in  sections;  the  customer  ordered 
at  will,  according  to  the  sum  at  his  disposal,  a  full  granary,  cooks, 
one  or  two  butchers,  brewers,  a  company  of  archers  or  of  heavily 
armed  soldiers,  vessels  with  a  more  or  less  numerous  crew,  and 
the  salesman  arranged  the  scenes  according  to  the  instructions 


FIG.   168. — A  CONCERT  (Museum,  Cairo).     (Phot.  E.  Brugsch.) 

received.  It  happened  sometimes  that  the  sailors  were  too  large 
for  the  boat  that  had  been  chosen,  or  that  the  coopers  were  not 
in  proportion  with  the  house,  but  no  one  was  disturbed  by  these 
inequalities;  when  once  they  had  been  blessed  and  shut  up  in 
the  tomb,  badly  composed  scenes  were  just  as  efficacious  as  the 
others.  They  are  amusing  to  us  in  spite  of  their  shortcomings, 
and  they  are  the  great  delight  of  visitors  to  museums;  the  room 
in  our  Museum  at  Cairo  where  the  archers  and  pikemen  of  Meir 
are  exhibited  (Fig.  169)  is  always  crowded.  They  have,  indeed, 
traces  of  the  qualities  we  find  in  the  bas-reliefs  of  which  they 
are  copies.  They  live,  they  act,  they  move,  they  adapt  themsel- 
ves to  one  another,  and  even  when  their  modelling  is  summary, 
we  feel  that  the  workmen  who  carved  them  had  been  trained  in 
a  good  school ;  by  nature  and  education,  they  tended  to  produce 
works  of  art,  even  when  they  were  working  at  modest  prices  for 
the  poor  and  humble. 

90 


MEMPHITE  ART 

Examples  of  the  minor  arts  are  not  numerous,  or  at  least  of 
such  as  have  some  claim  to  beauty  as  well  as  to  utility.  Domestic 
pottery  is  for  the  most  part  coarse;  certain  forms  in  use  through- 
out centuries  in  an  earlier  age  persisted,  notably  the  red  variety 
with  a  black  border,  but  others  had  disappeared,  and  had  not 
been  replaced  by  more  refined  types.  We  can  scarcely  venture 
to  include  among  works  of  art  the  aediculae  of  red  terra-cotta 
which  are  found  in  the  tombs,  and  are  supposed  to  furnish  the 
soul  with  a  dwelling  duly  provided  with  all  the  necessaries  of 
life.  They  are,  in  fact,  simulacra  of  houses  with  a  court,  a 
portico,  lofty  chambers,  store-rooms,  and  on  the  ground  of  the 


FIG.    169. — INFANTRY,   FROM  HEIR   (Museum,  Cairo).      (Phot.  E.  Bmgsch.) 

court,  opposite  to  the  entrance,  a  complete  meal  of  bread,  vege- 
tables, meats,  cakes,  and  various  liquors,  the  whole  of  the  rudest 
and  most  naive  description,  and  quite  lacking  in  artistic  value. 
The,  potters,  however,  had  already  learnt  to  cover  the  clay  with 
a  vitreous,  semi-transparent  glaze,  tinted  with  various  colours. 
The  polychrome  tiles  facing  the  walls  of  King  Zoser's  mortuary 
chamber  in  the  Step  Pyramid  at  Sakkarah  (Fig.  170),  and  the 
fragments  of  green  plaques  found  by  Petrie  among  the  ruins  of 
Abydos,  show  that  enamelled  earthenware  was  used  for  the  deco- 
ration of  buildings  under  the  Thinite  dynasties,  while  enamelled 
beads  for  necklaces,  fragments  of  vases,  yellow,  green,  and  blue 
bricks  have  come  down  to  us  from  the  Memphite  dynasties;  but 
it  is  nevertheless  evident  that  the  more  luxurious  table  utensils 
were  of  stone  or  metal.  The  Egyptians  had  brought  the  art  of 
piercing  stone,  and  of  cutting  and  polishing  it  to  the  highest 
degree  of  perfection;  not  only  the  softer  kinds  such  as  limestone 

91 


ART  IN  EGYPT 


and  alabaster,  but  granite,  breccia,  diorite,  cornelian,  onyx  and 
lapis  lazuli  became  flexible  under  their  fingers,  and  assumed  the 
most  varied  and  graceful  forms.  We  have  bronze  bowls  and 
ewers  which  have  been  discovered  in  mastabas,  but  none  of  those 
golden  and  silver  vessels  mentioned  in  contemporary  texts,  or  in 
those  of  the  period  immediately  following:  these  all  passed  into 
the  melting-pot,  and  we  can  only  wonder  by  what  happy  chance 
the  admirable  golden  hawk's  head  which  Quibell  found  at  Kom- 
el-Ahmar  escaped  the  common  lot  (Fig.  171).  Its  design  is 

no  less  remarkable 
than  its  technique; 
the  physiognomy 
of  the  bird  is  as- 
toundingly  vigor- 
ous and  exact, 
and  the  use  of  red 
jasper  for  the  eyes 
gives  it  an  extra- 
ordinary vitality. 
The  body  was  of 
bronze ,  but  the 
pieces  were  too 
much  oxydised  to 
allow  of  its  recon- 
struction; only  the 
statuette  of  Phara- 
oh which  was  rest- 
ing against  its  breast  has  been  preserved.  The  jewels  no  doubt 
equalled  those  we  have  discovered  in  the  tombs  of  the  first  two 
Dynasties,  but  the  specimens  we  possess  are  of  the  most  trivial 
kind,  strings  of  enamelled  or  stone  beads,  imitations  of  sea-shells 
in  gold,  gold  or  silver-gilt  amulets,  plain  or  ribbed  gold  beads; 
there  are,  however,  at  Cairo  some  little  figures  of  gazelles,  goats 
and  oxen,  repousse  in  gold  leaves,  and  then  retouched  with  the 
point  to  serve  as  clasps  or  pendants  which  are  above  the  general 
level  of  mediocrity.  We  know  even  less  of  the  furniture  than 
of  the  jewelry,  for  we  are  reduced  to  seeking  information  from 
bas-reliefs  as  to  the  appearance  of  linen-chests,  jewel-boxes,  seats, 
beds  and  tables;  the  representations  of  these  objects  give  us  a 
good  idea  of  the  inventive  taste  and  skill  of  the  Egyptian 
joiners.  To  sum  up,  the  more  we  study  the  relics  of  this 
age  of  the  Pyramids,  the  more  convinced  we  become  that  its 
industrial  art  was  not  unworthy  of  its  higher  art;  the  joiners, 

92 


FIG.   I7O. — ENAMELLED   CHAMBER  OF  KING  ZOSER 
(From  the  Drawing-  by  Segato). 


MEMPHITE  ART 

founders,  goldsmiths  and  potters  who  catered  for  the  masses  had 
the  same  instinctive  sense  of  grace  and  harmony  to  which  I  have 
called  attention  in  the  creations  of  the  painters  and  sculptors  who 
worked  for  Pharaoh. 


BIBLIOGRAPHY  TO  CHAPTER  II  —  PART  I 

Architecture:  A.  The  Memphite  Mastabas.  The  classical  work  for  the  architectonic 
study  of  the  Mastabas  is  si  ill  Mariette's,  Les  Mastabas  de  I'ancien  Empire,  vol.1,  in  quarto. 
Paris,  1882-1886,  592  p.;  this  must  be  supplemented  by  Mariette,  Voyage  de  la  Haute- 
Egypte,  vol.  I,  p.  31-44  and  pi.  3-14,  and  Flinders  Petrie,  Dendereh  (Egypt  Exploration 
Fund,  vol.  XVII)  in  quarto.  London  1900,  78  p.  and  78  pi.  for  the  Mastabas  of  the  Said, 
and  J.  de  Morgan,  De  la  frontiere  de  I'Egypt  a  Kom-Ombo,  in  quarto.  Cairo,  1894,  XI, 
212  p.  and  N.  de  G.  Davies,  The  Rock  Tombs  of  Sheikh  Said  (Archaeological  Survey  of 
Egypt,  vol.  X)  in  quarto.  London,  1901,  XII,  46  p.  and  34  pi.  for  hypogea  in  the  rock.  — 
B.  The  Pyramids:  Lepsius'  theory  of  the  construction  of  the  Pyramids,  combated  by 
Maspero,  Archeologie  Egyptienne,  1 st  ed.,  p.  127-128,  has  been  revived  by  L.  Borchardt, 
Lepsius's  Theorie  des  Pyramidenbaues,  in  the  Zeitschrift  fur  Agyptische  Sprache,  1891, 
vol.  XXIX,  p.  102-106,  and  his  conclusions  have  been  adopted  by  W.  Spiegelberg,  Ge- 
schichte  der  Agyptischen  Kunst,  p.  17-19.  The  questions  of  the  construction  and  alterations 
of  the  pyramids  of  Gizeh  and  Sakkarah  towards  the  Sa'ite  period  are  discussed  by 
L.  Borchardt,  Zur  Baugeschichte  der  Stufenpyramide  bei  Sakkarah,  also  Zur  Baugeschichte 
der  dritten  Pyramide  bei  Gizeh,  and  Zur  Baugeschichte  der  zweiten  Nebenpyramide  neben 
der  dritten  Pyramide,  in  the  Zeitschrift  fur  Agyptische  Sprache,  1891,  vol.  XXIX,  p.  87,  94, 
98,  100;  for  the  general  mass  of  building  constituting  a  royal  tomb  at  Sakkarah,  see 
Barsanti  -  Maspero,  Fouilles  autour  de  la  Pyramide  d'Ounas,  in  octavo,  Cairo,  1902-1906 
(Extract  from  the  Annales  da  Service  des  Antiquites),  175  p.;  for  the  group  of  Zawyet- 
el-Aryan,  Barsanti,  Fouilles  de  Zaouiyet-el-Aryan,  in  the  Annales  du  Service  des  Antiquites, 
1906,  vol.  VII,  p.  257-286  with  three  plates;  1907,  vol.  VIII,  p.  201-210;  for  the  group  of 
Abusir,  Fr.  W.  v.  Bissing,  Re-Heiligtum  des  Konigs  Ne-woser-Re  Rathoures:  I.  L.  Borchardt, 
Der  Bau,  in  quarto,  Berlin,  1905,  87  p.  and  7  pi.;  —  L.  Borchardt,  Das  Grabdenkmal  des 
Konigs  Ne-user-re  (Abusir  I),  in  quarto,  Leipzig,  1907,  184  p.  and  20  pi.,  Das  Grabdenkmal 
des  Konigs  Nefer-ir-ke-re  (Abusir  V),  in  quarto,  Leipzig,  1909,  VI,  91  p.  and  10  pi.,  Das 
Grabdenkmal  des  Konigs  S'ahu-re  (Abusir  VI),  in  quarto,  Leipzig,  1910,  162  p.  and  16  pi., 
and  more  especially  for  the  solar  or  funerary  temples:  G.  Foucart,  Un  temple  solaire  de 
I'Ancien  Empire,  in  the  Journal  des  Savants,  1906,  p.  360-370;  —  Holscher-Steindorff,  Die 
Ausgrabungen  des  Totentempels  der  Chephren-Pyramide  durch  die  Sieglin-Expedition  1908, 
in  the  Zeitschrift  fur  Agyptische  Sprache  vol.  XLVI,  p.  1-12;  —  L.  Borchardt,  Der  Toten- 
tempel  der  Pyramiden,  in  the  Zeitschrift  fur  Geschichte  der  Architektur,  vol.  Ill,  p.  65-88. 

Painting  and  Sculpture.  —  Several  of  the  Mastabas  of  the  Ancient  Empire  have  been 
reproduced  entirely  in  works  by:  Flinders  Petrie,  Medum,  in  quarto,  London,  1892,  52  p. 
and  XXXVI  pi.:  —  Paget-Pirie-Quibell,  The  Tomb  of  Ptah-hetep  (Egyptian  Research 
Account,  vol.  II),  in  quarto,  London,  1898,  p.  25-34  and  pi.  XXXI-XLI:  —  N.  de  G.  Davies, 
The  Mastaba  of  Ptahhetep  and  Akhhetep  at  Saqqareh  (Archaeological  Survey  of  Egypt, 
vol.  VIII-X),  in  quarto,  London,  I,  1900,  42  p.  and  XXXI  pi.,  II,  1901,  19  p.  and  XXV  pi.; 
—  M.  A.  Murray  and  Hilda  Petrie,  Saqqara  Mastabas  (Egyptian  Research  Account,  vol. 
XX-XXI),  in  quarto,  London,  1905,  I,  50  p.  and  XLV  pi.;  —  Fr.  W.  v.  Bissing,  Die  Mastaba 
des  Gemnikai,  in  quarto,  Berlin,  I,  1905,  VIII,  42  p.  and  XXXIII  pi.,  II,  1911,  30  p.  and 
XXXV  pi.;  —  J.  Capart,  Une  rue  de  tombeaux  a  Sakkarah,  in  quarto,  Brussels  1907,  I, 
79  p.,  vol.  II,  2  p.  and  CVII  pi.,  Chambre  funeraire  de  la  VI"  Dynastic  aux  Musees  royaux 
du  Cinquantenaire,  in  quarto,  Brussels  1907,  26  p.  and  5  pi.  For  the  few  names  of 
sculptors  given  on  the  monuments,  see  A.  Erman,  Ein  Kiinstler  des  Allen  Reiches, 
in  the  Zeitschrift  fur  Agyptische  Sprache,  1894,  vol.  XXXII,  p.  97-99,  and  for  the  bas-reliefs 
of  the  Hermopolitan  School:  J.  Cledat,  Notes  sur  quelques  figures  egyptiennes,  in  the 
Bulletin  de  I'lnstitut  francais  d' Archeologie  orientale,  1901,  vol.  I,  p.  21 — 24.  —  The  archaic 
statuary  of  the  Meiiphite  age  has  been  studied  by  W.  Pleyte,  L'Art  antique  egyptien  dans 
le  Musee  de  Leide,  in  the  Verhandlungen  des  VII.  Orientalisten-Kongresses ,  agyptisch- 
afrikanische  Sektion  (1888),  p.  47-54,  —  G.  Steindorff,  Archaische  dgyptische  Statuen,  in 

93 


ART  IN  EGYPT 

the  Archdologischer  Anzeiger,  1898,  p.  64-66;  —  Grebaut  -  Maspero,  Le  Musee  Egyptien, 
1890-1900,  p.  12-13  and  pi.  XIII;  —  Wiedemann,  Zwei  dgaptische  Statnen  des  Museums  zu 
Leiden,  in  the  Orientalische  Literaturzeitung,  1898,  vol.  I,  p.  269-273  and  pi.  I,  II,  and  Die 
dgyptische  Statue  A 39  des  Louvre,  in  the  same  journal,  1901,  vol.  IV,  p.  41-43,  —  Bissing-, 
Denkmaler  agyptischer  Skulptur,  1906-1911,  in  folio,  Munich,  pi.  1-6;  —  J.  Capart,  Recueil 
de  Monuments  egyptiens,  in  quarto,  Brussels,  1902,  pi.  H-III,  VI,  LI,  and  the  corresponding- 
portions  of  text;  —  R.  Weill,  Les  Origines  de  I  Egypte  pharaonique,  1908,  p.  143-146, 
181-188,  255-260  and  pi.  I-II,  V-VL  For  the  statues  of  the  fine  period,  consult  in  addition 
to  the  works  already  quoted  Fr.  W.  v.  Hissing,  Denkmaler  agyptischer  Skulptur,  pi.  7-18 
and  text;  —  E.  de  Rouge,  Album  photographique,  No.  89-108;  —  Mariette,  Voyage  de  la 
Haute-Egypte,  vol.  I,  p.  47  and  pi.  16  and  Album  du  Musee  de  Boulak,  pi.  18-21,  25-27;  — 
Borchardt,  Kunstwerke  «us  dem  Agyptischen  Museum  zu  Kairo,  pi.  1-5,  20-22,  32  and  3-5, 
10-11,  14{  J.  Capart,  Recueil  de  Monuments  egyptiens,  1908,  in  quarto,  Brussels,  pi.  IV-XIII, 
LJI-LV,  and  the_corresponding-  text,  and  L'Art  Egyptien,  pi.  11-20,  26,  the  separate  articles 
L.  Borchardt,  Uber  das  Alter  des  Sphinxes  bei  Gizeh,  in  the  Sitzungsberichte  der  K.  Pr. 
Akademie  der  Wisse"schaften,  1897,  p.  752-760.  Die  Dienerstatuen  aus  den  Grdbern  des 
Alien  Reiches  and  Uber  das  Alter  der  Khephrenstatue'  in  the  Zeitschrift  fur  Agyptische 
Sprache,  1897,  vol.  XXXV,  p.  119-134  and  1898,  vol.  XXXVI,  p.  1-18;  —  Chassinat,  Les 
Fouilles  d'Abu-Rodsh,  in  the  Comptes  rendus  de  PAcademie  des  Inscriptions  et  Belles- 
Lettres,  1901,  p.  616-617;  —  Daressy,  Sur  I'age  du  Sphinx,  and  L'Age  du  Sphinx,  in  the 
Bulletin  de  I'Institut  Egyptien,  1906,  vol.  VII,  p.  93-97,  and  1909,  vol.  Ill,  p.  35-38;  — 
Maspero,  in  O.  Rayet,  L'Art  Antique,  vol.  I,  5  pi.  and  corresponding  text,  Le  Nouveau 
Scribe  du  Musee  de  Gizeh,  in  the  Gazette  des  Beaux-Arts,  3Qd.  series,  1893,  vol.  IX,  p.  265-270. 
Le  Scribe  accroupi  de  Gizeh,  in  the  Monuments  et  Memoires  Plot,  1894,  vol.  I,  p.  1-16  and 
pi.  H,  Le  Musee  Egyptien,  in  quarto,  1890-1900,  vol.  I,  pi.  Vffl-XII,  XIV,  XXVI  and  p.  9-12, 
13-14-15,  and  vol.  U,  1901-1907,  pi.  XI  and  XVII,  and  p.  30-33,  47-48.  For  the  Schools  of 
Egyptian  Statuary,  see  Maspero,  La  Statuaire  egyptienne,  in  the  Journal  des  Savants,  1908, 
p.  5-17,  and  for  groups  of  wooden  figurines  representing-  funereal  or  domestic  scenes, 
Maspero,  Sur  les  figures  et  sur  les  scenes  en  ronde-bosse  qu'on  trouve  dans  les  tombeaux 
egyptiens,  in  the  Bulletin  de  I'Institut  Egyptien,  1904,  vol.  IV,  p.  367-384,  Le  Musee  Egyptien, 
vol.  I,  pi.  XXXHI-XLIII,  and  p.  30-40,  and  Causeries  d'Egypte,  in  octavo,  Paris,  1907,  p.  351-357. 
For  polychromy  in  statues,  cf.  Fr.  W.  v.  Bissing,  Zur  Polychromie  der  altdgyptischen  Skulptur. 
in  the  Recueil  de  Travaux,  1898,  vol.  XX,  p.  120-124. 


FIG.    171. — GOLDEN 

HAWK' S  HE  AD,  AT  CAIRO. 

(Phot.  E.  Brugsch.) 

94 


FIG.    172. — KARNAK.      THE    GREAT  TEMPLE    OF  AMON    SEEN  FROM 
THE   SOUTH  IN  1804.     (Phot,  de  Banville) 


PART  II 
THEBAN  ART 

CHAPTER  I 


The  Art  of  the  first  Theban  Age  —  Civil,,  religious  and  funerary  Architecture  —  Painting 
begins  to  detach  itself  from  Sculpture,  at  least  in  the  Tombs  —  The  provincial  Schools 
of  Sculpture:  Theban  School,  Hermopolitan  School,  Tanite  School  —  The  minor  Arts: 
Goldsmith's  Work. 

THE  weakness  of  the  Pharaohs  who  followed  Pepi  II.,  was 
such  that  many  of  the  great  lords  between  whom  Upper 
Egypt  was  divided  made  themselves  almost  independent;  one 
Kheti  dethroned  the  Memphites,  and  reigned  over  the  whole 
valley;  after  four  or  five  generations,  the  Theban  barons  revolted 
against  his  descendants,  and  fought  for  the  crown.  They  were 
at  last  victorious,  and  their  hegemony  lasted  from  fifteen  to 
twenty  centuries,  almost  without  interruption.  The  first  period 
was  a  term  of  feudality,  during  which  the  local  tyrants  exercised 
in  their  own  domains  and  the  fiefs  attached  to  them  an  author- 
ity almost  as  complete  as  that  of  the  suzerain  dynasty.  Memphis, 
fallen  from  her  rank  as  capital ,  witnessed  the  gradual  decline 
and,  at  intervals,  the  almost  complete  extinction  of  her  artistic 
activities;  but,  on  the  other  hand,  the  remaining  cities  of  Middle 

95 


ART  IN  EGYPT 


and  Upper  Egypt,  Heracleopolis,  Minyeh,  Hermopolis,  Cusae,  Siut, 
Abydos,   Coptos,   Thebes,   and  Elephantine,   mingled  more  and 

more  happily  with  the  artistic  life 
of  the  nation.  They  became  for 
the  most  part  the  seats  of  special 
schools,  some  of  which  derived 
from  the  Thinite  or  Memphite,  while 
others  were  the  result  and  the  cul- 
mination of  provincial  schools  hither- 
to embryonic  for  lack  of  resources. 
Relics  of  these  are  by  no  means 
so  numerous,  as  yet,  as  those  of 
the  earlier  periods.  There  are,  how- 
ever, enough  to  enable  us  to 
determine  the  general  tendencies 
of  each.  That  of  Thebes  predo- 
minated in  the  circle  of  the  Pharaohs, 
as  was  natural,  but  its  influence 
Over  its  rivals  was  restricted,  and 
their  originality  did  not  suffer  from 
its  preponderance. 

There  is  little  to  say  of  the 
temples.  The  Pharaohs  of  the  Eigh- 
teenth and  Nineteenth  Dynasties  demolished  them  for  the  most 
part,  or  preserved  only  insignificant  portions  of  them.  It 
may  be ,  however ,  that  the  chapel  at  Kom-es-Sagha ,  on  the 
ancient  northern  slope  of  Birket-el-Kurun  (Figs.  173 — 174)  is  a 

specimen,  so  far  unique. 
This  seems  probable  when 
we  consider  the  elegant 
shaping  of  the  limestone 
blocks,  and  the  care  with 
which  they  are  laid  together, 
but  as  they  bear  neither 
sculptures  nor  inscriptions, 
we  are  not  justified  in  pro- 
nouncing finally  on  this 
point.  The  remains  brought 
to  light  by  Petrie  at  Aby- 
dos and  Sinai,  and  by 
our  own  Service  at  Hermo- 
polis, seem  to  prove  that  the  plan  generally  adopted  was  similar 
to  that  used  by  the  architects  of  the  following  age.  The  walls 

96 


FIG.    173.— INTERIOR   OF  THE 

TEMPLE    OF    KOM-ES-SAGHA. 

(Phot.  Schweinfurt.) 


FIG.    174.— TEMPLE   OF  KOM-ES-SAGHA. 
(Phot.  Schweinfurt.) 


THE  FIRST  THEBAN  AGE 


FIG.    175. — HATHOR   CAPITAL 
(Museum,  Cairo).  (Phot.  E.  Brugsch.) 


and   their   facings   were   of  limestone   or  sandstone,   the   doors, 
sphinxes,  and  obelisks  of  black  or  pink  granite.     The  palm-leaf 
and     the     lotus     capital    continued 
to     be    generally     used ,     as    well 
as      the      Hathor       capital ,       but 
this    consisted    sometimes    of    two 
(Fig.  175),  sometimes  of  four  heads 
of   the    goddess   soldered    together 
at  the  back,  and  surmounted  by  a 
somewhat  low  abacus.     The  excav- 
ations   at     Karnak    have     restored 
to    us    some    simple    pillars   admir- 
able   in    style ,     which    Sesostris    I. 
erected    in    the    temple    of   Amon 
(Fig.  176),  and  those  of  his  pyramid 
at     Lisht ,     some     Osirian     pillars , 
while    throughout   the  valley,    from 
Assuan  to  the  marshes  of  the  Delta, 
fragments    of    various    shapes    bear 
witness  to  the  constructive  zeal  of  the   first  Thebans;   but  even 
when  brought  together   and    combined,   they  fail  to  furnish  any 
data  as  to  the  appearance  of  the  temple  as 
a  whole.    Did  the  pylon  already  exist  in  its 
classic  form,  a  doorway  between  two  towers? 
It  is  doubtful  at  least,   and   so  far  no  trace 
of   it   has  been  found.     We  know,  however, 
that    one   element,    formerly    optional,    had 
become    a   regular    feature   of   the    external 
decoration,    a   large    obelisk   modelled  upon 
the  minute   obelisk  of  the  Memphite  tombs. 
Sesostris  I.,  when  he  restored  the  Temple  of 
the  Sun  at  Heliopolis,  erected  two,  and  one 
of    these    is    still    standing   in    the    midst   of 
the  plains  of  Matariyeh  (Fig.  177).     It  is  of 
red  granite  from  Syene ;  it  measures  66  feet 
in  height,  and  the  point  was  crowned  by  a 
pyramidium    in    copper    which    was    still    in 
existence    in   the  fourteenth    century   of   our 
era.  The  type  of  the  obelisk  as  guardian  of  the 
temple,  and  emblem  of  the  founder,  was  immut- 
ably fixed  from    the    beginning   of  this  first 
Theban  age,  just  as  we  shall  find  it  down  to  the  Roman  period. 
Monumental    tombs   abound ,    though    they   are   less  numerous 

97  H 


FIG.   176.— A  PILLAR 
AT  KARNAK. 

(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


than  in  the  Memphite 
age,  and  many  of  them 
are  amazingly  well  pre- 
served. The  rulers  had 
not  renounced  the  pyra- 
mid for  this  purpose,  but 
they  modified  its  struc- 
ture in  various  ways.  In 
the  beginning ,  whereas 
the  Heracleopolitans  of 
the  Ninth  and  Tenth  Dy- 
nasties clung  to  the  tra- 
the  Eleventh,  anxious  to 


FIG.    178. — PYRAMIB-MASTABA    OF  DRAH-ABU'L- 
NEKKAH  (After  Prisse  d'Avennes). 


FIG.    177.— THE   OBELISK   OF  HELIOPOLIS. 
(Phot.  Beato.) 

ditions  of  the  Sixth,  the  Thebans  of 
appropriate  a  form  o:  i- 
ginally  reserved  for  royal- 
ty, and  not  daring  to 
usurp  it  just  as  it  stood, 
conceived  the  idea  of 
placing  it  upon  the  mas- 
taba. It  was,  we  re- 
member, by  a  combination 
of  this  kind  that  the  Helio- 
politans  of  the  Memphite 
age  had  created  the  form 
of  the  solar  temple  in  use 
under  the  Fifth  Dynasty, 
grafting  the  obelisk  of  the  Sun  upon  the  mastaba.  The  semi- 
independent  nobles ,  or  the  courtiers  who  were  buried  near 

Abydos  or  at  Drah-abu'l- 
Nekkah  (Fig.  178)  seem  to 
have  furnished  the  first 
examples.  These  are  build- 
ings of  coarse,  unbaked 
bricks,  consisting  of  a 
mastaba,  square  or  rect- 
angular on  plan,  the 
longest  side  or  which  was 
rarely  more  than  fifty  feet 
in  extent;  the  pyramid 
was  implanted  in  this  as 
upon  a  plinth  about  30 
or  40  feet  high  at  the 
most.  Sometimes  a  single 


FIG.    179. — SECTION   OF  ONE   OF  THE 

PYRAMID-MASTABAS    OF   ABYDOS 

(After  Mariette). 

98 


THE  FIRST  THEBAN  AGE 


chamber  vaulted  on  corbels  occupies  the  interior  alike  of  mastaba 
and  pyramid,  and  the  sarcophagus  or  the  coffin  was  deposited 
here ;  very  often  again,  the 
vault  was  constructed 
in  the  mastaba,  and  the 
pyramid  contained  only  a 
vaulted  space  designed  to 
bear  the  weight  (Fig.  179). 
When  the  Theban  barons 
usurped  the  royal  power, 
they  substituted  stone  for 
brick,  and  increased  the 
proportions  of  their  monu- 
ments. That  Menthu-hetep 


FIG.    ISO. — PLAN   OF  THE  TOMB  OF  MENTHU- 
HETEP  (After  Naville  and  Hall). 


who  united  all  Egypt 
under  his  sway  installed 
his  sepulchre  in  the  south- 
ern hollow  of  the  circus  of  Der-el-Bahari  (Fig.  180).  It  was 
approached  on  the  level,  through  a  court  bounded  on  the  west 
by  two  porticoes  of  square  pillars;  between  the  two  was  an 
inclined  causeway  leading  to  a  terrace  partly  made  of  stones 
fitted  together,  partly  hewn  out  of  the  solid  rock.  The  mastaba 
rose  in  the  centre,  a  rectangle  some  130  feet  long,  faced  with 
slabs  of  carved  limestone  and  furnished  at  the  sides  with  porticoes 
corresponding  to  those  of 
the  lower  floor.  The  pyra- 
mid crowned  the  mastaba, 
so  to  speak,  but  it  was 
solid ;  the  royal  vault  was 
concealed  underground, 
and  was  approached  by 
a  secret  gallery,  the  door 
of  which  opened  some 
way  off  on  the  plain,  in 
front  of  the  building.  Be- 
hind the  pyramid,  in  the 
temple  itself,  the  sepul- 
chral chapels  of  the  women 
of  the  harem  were  ranged 
in  rows,  and  behind  them 
a  second  court  with  porti- 
coes extending  westward,  was  supported  against  the  cliff  (Fig.  181). 
Was  it  a  chamber  attached  to  the  sepulchre?  or  a  mysterious 

99  H2 


FIG.  l8l.— TOMB  OF  MENTHU-HETEP,  RESTOR- 
ATION BY  SOMERS  CLARKE. 


ART  IN  EGYPT 


FIG.  182.— PYRAMIDIUM  OF  DAHSHUR 

(Museum  Cairo). 
(Phot.  E.  Brugsch.) 


sanctuary,  this  cell  of  granite   and   alabaster,    marvellously   pro- 
portioned, which  we  reach  after  traversing  a  corridor  over  550  feet 

long,  whose  door  stands  open 
at  the  end  of  the  court.  In  any 
case,  it  dates  from  the  same 
period  as  the  rest.  The  Pharaohs 
of  the  Thirteenth  Dynasty  and 
of  the  succeeding  Dynasties  who 
rest  at  Thebes,  were  buried  in 
mastabas  with  pyramids  down 
to  the  inauguration  of  the  New 
Empire;  those  of  the  Twelfth 
and  Thirteenth  Dynasties,  who 
lived  in  Middle  Egypt,  preferred 
simple  pyramids  in  the  Memphite 
style,  with  the  paved  temenos 
and  chapels  turned  to  the  east. 
The  external  constituents  have  suffered  a  good  deal,  but  the 
bulk  of  the  tombs  proper  still  subsist  at  Dahshur,  Lisht,  and 
Ellahun,  near  Hawara.  Those  of  Amenemhat  I.  and  of  Sesostris  I. 
at  Lisht  are  of  limestone  or  granite.  Those  of  their  successors 
at  Dahshur  and  in  the  Fayum  are  of  unbaked  brick  with  a  peak 
of  black  granite  (Fig.  182),  but  they  were  perhaps  originally 
faced  with  limestone.  They  differ  from  their  Memphite  models 
by  details  of  internal  arrangement,  designed  to  render  access 
to  the  sarcophagus  even  more  difficult  than  in  the  past,  and 

they  are  for  the  most  part 
so  decayed  that  they  make 
no  artistic  impression  upon 
the  spectator.  Only  one 
among  them,  the  northern 
stone  pyramid  at  Dahshur, 
called  theBluntedPyramid, 
manifests  some  attempt  at 
originality.  Half  way  up, 
the  facades  are  interrup- 
ted, and  the  angle  passes 
suddenly  from  54"  41'  on 
the  horizon  to  42°  59';  we 
have  here  a  mastaba  with 
a  gigantic  mansard  -  roof . 
Private  persons  remained  no  less  faithful  than  the  Pharaohs 
to  the  local  fashions,  and  retained  the  mastaba  of  the  old  type 

100 


FIG.    l8j. — REMAINS   OF  THE  PORTICO   OF 
SA-RENPUT  I.     (Phot.  Morgan.) 


THE  FIRST  THEBAN  AGE 


FIG.  184.  —  PLAN  OF 
THE  TOMB  OFSA-REN- 
PUT  II.  (AfterMorgan). 


in  the  Memphite  necropolis,  the  pyramid-mastaba  at  Thebes  and 
Abydos ,  and  hypogea  in  the  mountains  elsewhere.  These  vary 
in  arrangement  according  to  the  district.  At 
Assuan,  the  sepulchre  of  Sa-Renput  I.  (Fig.  183) 
was  preceded  by  a  portico;  six  pillars,  cut 
out  in  the  rock,  upheld  the  architraves  and 
the  ceiling  of  dressed  stone.  The  door  gives 
access  to  a  first  chamber,  whence  a  vaulted 
passage  leads  to  the  chamber  of  statues  sub- 
stituted for  the  serdab  of  the  Memphites. 
The  portico  is  absent  in  the  hypogeum  of  Sa- 
Renput  II.  (Fig.  184),  and  we  pass  directly 
into  a  hall  with  pillars  continued  by  a  passage 
with  three  niches  on  either  side,  containing 
the  mummy-statues  of  the  master.  Another 
pillared  room  follows,  with  a  niche  for  the 
funerary  stele.  The  vault  is  without  any 
decoration,  and  the  wells  which  go  down  to 
it  are  flush  with  the  ground,  sometimes  in 
one  of  the  rooms,  sometimes  in  the  open  air 
on  the  esplanade  outside.  In  Middle  Egypt,  at  Siut,  Bersheh, 
and  Beni- Hasan,  the  plan  differs  only  in  detail  from  one  place 
to  another.  Hapsefai's  entrance  is  sheltered  by  a  veritable  porch 
with  a  rounded  arch,  about  22  foet  high;  the  first  and 
second  hall  are  connected  by  a  vaulted  passage,  but  they  them- 
selves have  flat  ceilings.  At  Beni -Hasan  the  two  hypogea  of 
Khnemuhetep  and  Ameni 
confront  the  valley, 
their  porticoes  upheld  by 
two  polygonal  columns 
(Fig.  185).  The  chapel 
consists  of  a  hypostyle  hall 
divided  into  three  vault- 
ed aisles  by  two  double 
rows  of  columns;  the 
central  aisle  terminates 
in  the  niche  where  statues 
are  seated  awaiting  offer- 
ings (Fig.  186).  The  hypo- 
geum No.  7 ,  which  was 
originally  a  vaulted  hall 
very  much  surbased,  supported  by  six  columns  in  three  rows,  was 
enlarged  to  the  right  subsequently,  and  the  new  excavation, 

101 


FIG.    185.— THE    HYPOGEUM    OF  KHNEMU-HETEP 
(After  Lepsius). 


ART  IN  EGYPT 


returning  in   a   square   towards   the   west,   there    forms    a    wing 
with  a  flat  ceiling  resting  on  four  columns.    All  these  monuments 

show  a  tendency 
to  replace  the  ceil- 
ings of  the  hypo- 
gea  and  the  Mem- 
phite  mastabas  by 
a  curved  roof,  and 
the  same  tendency 
makes  itself  felt 
in  the  stelae.  This 
is  a  natural  conse- 
quence of  progress 
in  religious  ideas ; 
as  the  stele  no 
longer  represented 
only  the  door  of 


FIG.    l86.— PLAN   OF  THE   HYPOGEUM   OF  AMENI 
AT  BENI-HASAN  (After  Newberry). 


the     dead     man's 
it  was   logical   to 
While  in  Memphis 


apartment,    but    his  whole   house,    the  tomb 

suggest   this  by  giving  it  the   same   aspect. 

sculptors  remained  faithful  to  the  square  form  deduced  from  the 

mastaba,  in  Upper 
Egypt ,     and 
Abydos , 


even 
they 

preferred  the  round- 
ed summit  which 
recalled  the  vaults  of 
the  hypogea.  Other 
points  on  which 
architects  were 
agreed  were  the 
number  and  the 
variety  of  the  sup- 
ports. In  the  Mem- 
phite  burial  grounds 
there  are  scarcely 
more  than  two  or 
three  examples  of 
columns ;  at  Beni- 
Hasan  and  at  Ber- 
sheh,  there  is  no 

tomb  of  any  importance  without  several.    By  cutting  off  the  angles 
of  a  square   pillar   architects    transformed   it   into   an   octagonal 


FIG.  187.— PAINTED  INTERIOR  OF  THE  TOMB  OF  HERU- 
HETEP  (Museum,  Cairo).  (Phot.  E.  Brugsch.) 


THE  FIRST  THEBAN  AGE 


FIG.    188. — PAINTING  WITHOUT  A 

SCULPTURED   GROUND 
(Museum,  Cairo).     (Phot.  E.  Bragsch.) 


prism ,  and  by  repeating  the  process  on  this ,  they  obtained  a 
prism  of  sixteen  facets;  these  polygonal  shafts,  imbedded  in  a 
low  base  rounded  to  a  disc, 
and  completed  by  a  square 
abacus  uniting  them  to  the 
architrave ,  constitute  what 
Champollion  calls  by  a  rough 
analogy  primitive  Doric.  They 
appear  first  at  Beni- Hasan, 
side  by  side  with  lotiform 
capitals  of  a  particular  type; 
in  two  or  three  tombs,  the 
base  has  been  left  unhewn, 
or  roughly  shaped  in  such  a 
manner  as  to  suggest  the 
junction  of  a  palm-trunk  and 
its  roots.  Was  this  an  acci- 
dent, or  did  the  sculptor  wish 
to  make  his  shaft  look  like  a 
tree  ?  This  form  has  only 
been  found  at  Beni-Hasan. 

We  are  better  informed  as 
to  the  painting  and  sculpture, 

and  what  strikes  us  at  once  is  that  they  seem  to  have  loosened 
the  bonds  which  held  them  together  in  primitive  times;  not  that 
the  statue  or  the  bas-relief  is  no  longer  coloured;  but  after  the 
Sixth  Dynasty,  painters  were  emboldened  to  suppress  the  sculp- 
tured foundations  which 
had  seemed  indispensable 
to  their  masters  of  the 
Memphite  age.  We  must 
not ,  however ,  suppose 
that  the  emancipation  of 
painting  was  complete ; 
the  temples  did  not  permit 
it,  and  the  evidences  of 
it  are  only  to  be  seen 
on  the  walls  of  tombs 
(Figs.  187,  188),  or  in 
the  coarse  decorations  of 
certain  private  houses 

(Fig.  189).    It  is  to  be  accounted  for  by  purely  material  causes; 
the  rock  in  which  the  hypogea  were  cut  did  not  offer  the  sculptor 

103 


FIG.    l8g.— PAINTED  DECORATION   OF  A 
PRIVATE  HOUSE   (After  Petrie). 


ART  IN  EGYPT 


those    homogeneous    surfaces    which    mastabas    built    of   blocks 
of    dressed    limestone    afforded    him.      But    the    brush     worked 


FIG.  190.— WAR  DANCE  (After  Champollion). 

with  ease  where  the  chisel  would  have  got  mediocre  results 
or  have  failed  entirely;  a  picture  without  a  foundation  in 
relief  was  also  more  quickly  executed  and  cheaper.  We 
can  understand,  therefore,  why  the  Thebans  and  Heracleo- 
politans,  people  of  modest  means,  were  often  content  with 
painted  tombs.  Let  me  hasten  to  say,  that  if  the  execution 
varied  sometimes,  the  underlying  principle  remained  unchanged, 
and  as  in  the  past,  the  advantage  of  the  master,  god  or  dead 
man,  was  the  first  consideration ;  nevertheless,  its  application  was 
modified,  at  least  in  the  tombs,  under  the  influence  of  political 
circumstances  or  contemporary  ideas.  In  the  neighbourhood  of 
the  cataract,  where  the  nobles  had  no  great  battalions  at  their 
disposal,  they  repeated  the  domestic  or  agricultural  themes  which 
had  satisfied  their  ancestors.  At  Siut,  at  Bersheh,  at  Beni-Hasan, 
where,  associated  by  their  geographical  position  with  the  struggles 


FIG.   igi. — SIEGE  OF  A  FORTRESS  (After  Champollion). 

between  Heracleopolis  and  Thebes,    they   were  obliged  to   keep 
their  troops  on  a  war-footing,  preoccupation  with  military  matters 

104 


THE  FIRST  THEBAN  AGE 


FIG.   IQ2.— SOLDIERS  AT   SIUT.     (Phot.  Insinger.) 


appears  more  or  less  insistently  in  the  majority  of  their  hypogea. 
Tefyeb,  Kheti,  and  Beket,  who  had  been  generals  of  renown 
in  their  lifetime, 
wished  to  parade 
in  the  other  world 
escorted  by  the 
soldiers  who  had 
made  their  glory 
in  this,  and  they 
demanded  to  be 
represented  among 
the  bands  of  vassals 
whose  doubles  they 
took  away  with 

them.  The  drill  of  recruits,  racing,  jumping,  war-dances  (Fig.  190), 
wrestling,  battles,  the  siege  of  fortresses  (Fig.  191)  were  introduced 
into  artists'  sketch-books,  and  as  in  many  a  tomb  there  was  not 
space  enough  for  all  these  novelties,  they  either  ousted  an  equi- 
valent number  of  pacific  scenes,  or  reduced  them  to  their  simplest 
expression ;  agriculture  was  so  indispensable  that  no  one  dared  to 
curtail  its  episodes  over-boldly,  but  there  were  no  such  scruples 
to  interfere  with  the  abridgment,  upon  occasion,  of  the  bringing 
of  offerings,  processions  of  territories,  sailors'  brawls,  and  the 
various  handicrafts.  For  here  the  evil  was  very  slight,  since  it 
was  permissible  to  substitute  the  wooden  groups  of  the  tomb- 
chamber  for  the  bas-reliefs  and  paintings  of  the  chapel. 

The  necropolis  of  Siut  has  been  so  devastated  by  quarriers 
that  it  is  hardly  possible 
to  appreciate  the  merits 
of  the  artists  who  decor- 
ated it ;  no  doubt  also 
some  of  their  defects  are 
due  to  the  bad  quality 
of  the  stone.  As  far  as 
I  have  been  able  to  judge, 
they  appear  to  have  been 
inspired  by  the  Memphite 
School,  or  the  School  of 


FIG.    193.— SHOULDER-MOVEMENT,   TOMB   OF 
KHNEMU-HETEP  (After  Champollion). 


Abydos,  which,  after  the 
end  of  the  Thinite  dynas- 
ties,   was    practically    an 
annexe   of   the   Memphite   School.     The   technical  processes   are 
the  same  in  each,  as  also  the  proportions  of  the  figures  and  the 

105 


ART  IN  EGYPT 


FIG.    194. — SQUATTING  PEASANT. 
(After  Champollion). 


distribution  of  the  episodes;    the   military   novelties   are    treated 
in  accordance    with    current    conventions   round   Memphis   under 

the  Sixth  Dynasty ,  and 
Kheti's  heavy  infantry 
marches  along ,  dragging 
shields,  with  no  more  vi- 
vacity than  the  ancient 
processions  of  offering- 
bearers  (Fig.  192).  The 
contrast  is  very  striking 
when  we  pass  from  Siut  to 
Beni-Hasan  and  Bersheh. 
These  two  places  and  all 
the  surrounding  district 
were  indeed  under  the 
influence  of  that  School 
of  Hermopolis ,  which, 
several  centuries  earlier, 
showed  such  marked  ori- 
ginality in  the  drawing  of 
the  fat  and  the  lean.  I  think  that  these  masters  and  their  pupils 
cannot  have  had  a  facility  with  the  chisel  comparable  to  that 
of  the  Memphites,  for  they  often  avoided  carving  their  scenes, 
and  were  content  to  draw  and  paint  them;  the  use  of  brush 
and  colour  permitted  a  freedom  of  action,  of  which,  however, 

they  did  not  avail  them- 
selves equally  in  all  the 
subjects  they  had  to  treat. 
They  followed  the  old 
methods  for  the  fundamen- 
tal themes,  those  which 
filled  the  sketch-books 
their  ancestors  had  be- 
queathed to  them ,  save 
that  they  occasionally  in- 
troduced modifications, 
particularly  in  perspec- 
tive. Thus  in  the  tomb 
of  Khnemu-hetep,  a  good 
many  of  the  secondary 
figures  have  silhouettes 
more  consistent  with  reality;  placed  in  profile,  the  bust  is 
sometimes  foreshortened  accurately,  or  at  least  sometimes  one 

106 


FIG.    195. — CAT  WATCHING  FOR  PREY 
(After  H.  Carter). 


THE  FIRST  THEBAN  AGE 


and  sometimes  the  other  shoulder  is  brought  forward,  according 
to  the  gesture  to  be  expressed  (Fig.  193).  We  have  seen  that 
the  Memphites  rarely  did 
as  much,  save  in  the  later 
period  of  the  Sixth  Dy- 
nasty; the  Hermopolitans 
almost  transformed  what 
had  been  the  exception 
hitherto  into  a  current 
rule.  Their  attempts  of 
this  nature  were  of  the 
happiest,  and  it  is  pleasant 
to  note,  in  the  midst  of 

the  conventional  poses,  FIG  ig6_SCENE  IN  A  SIEGE)  AT  DESHASHEH 
attitudes  which  a  painter  (After  Petrie). 

of   our  own   times  would 

not  treat  otherwise.  Such  is  the  action  of  that  peasant,  who, 
in  the  Tomb  of  Khnemu-hetep ,  is  about  to  seat  himself  on 
the  neck  of  a  gazelle  to  force  it  to  crouch  beside  him ;  the  action 
of  the  arms,  the  curve  of  the  loins,  the  sweep  of  the  back,  the 
effacement  of  the  shoulders,  and  the  protuberance  of  the  breast 
are  all  rendered  with  almost  faultless  precision  (Fig.  194).  Even 
in  passages  where  tradition  is  rigorously  observed,  the  drawing 
differs  in  many  respects  from  that  of  the  Memphites.  It  is  less 
refined,  less  sure,  less  uniformly  equal  in  quality,  but  also  more 
varied ,  more  expressive ,  more  eager  to  suggest  truth ;  if  the 
draughtsman  respects  the  general  formula  and  transcribes  it  in 
accordance  with  the  consecrated  models,  he  at  least  strives  to 


FIG.   197. — WRESTLING,  AT  BENI-HASAN  (After  Champollion). 

improve  the  details  and  to  copy  nature  more  faithfully.  He  is 
more  successful  with  animals  than  with  human  beings;  who  has 
ever  rendered  the  cat  lying  in  wait  among  the  reeds  at  once 

107 


ART  IN  EGYPT 


FIG.    IQS. — PORTRAIT  OF  SIESIS 
(Museum,    Cairo).     (Phot.  E.  Brugsch.) 


with  greater  realism  and  greater  brilliance  ?    Every  characteristic 
has  been  seized,  the  extension  of  the  neck,  the  quivering  of  the 

spine,  the  contractions  of  the  tail, 
the  slight  recoil  of  the  body  before 
springing  upon  the  prey,  and  the 
fixed  intensity  of  gaze  which  arrests 
and  fascinates  the  victim.  (Fig.  195). 
The  merits  of  the  school,  how- 
ever, are  nowhere  more  strikingly 
shown  than  in  its  dealings  with 
martial  subjects,  for  here,  indeed, 
it  was  not  hampered  by  a  long 
routine.  The  Memphites  had  al- 
ready profited  by  the  liberty  due 
to  the  absence  of  religious  obli- 
gation, to  interpret  the  quarrels 
and  encounters  of  boatmen  upon 
the  canals;  in  these  they  showed 
a  knowledge  of  the  human  form, 
and  a  sense  of  composition  with 
which  we  could  not  have  credited 
them  on  the  evidence  of  their 
severely  conventional  scenes  of  agriculture  and  industry.  The 
Hermopolitans  commissioned  to  paint  the  lives  of  soldiers,  had 

models  for  the  actual  mo- 
ment of  battle  either  in 
the  temples,  or  in  certain 
earlier  tombs,  such  as  that 
at  Deshasheh  (Fig.  196), 
where  a  prince  of  the  Fifth 
Dynasty  had  recorded  his 
exploits;  thus  the  shock 
of  armies',  the  attack  on 
fortresses,  the  transport  of 
dead  and  wounded  as 
treated  by  them  have  no 
characteristic  or  arresting 
features.  The  soldiers  who 
exchange  blows  with  the 
axe  and  flights  of  arrows, 
show  hardly  less  compo- 
sure in  their  approach 

FIG.   109. — STATUES   OF  SESOSTRIS  I.  ,,  ,          w,-,       ..,  i  !• 

(Museum,  Cairo).    (Phot.  E.  Brugsch.)  than    do    K.heti  s    soldiers 

108 


THE  FIRST  THEBAN  AGE 


on  the  inarch  at  Siut.    On  the  other  hand,  the  athletic  exercises 
by  which  the  recruits  trained    their  bodies   during   their  term  of 


FIG.  20O.— ONE  OF  THE  STATUES 

OF  SESOSTRIS  (Museum,  Cairo). 

(Phot.  E  Brugsch.) 


FIG. 201.— COLOSSAL  STATUE 

OF     SESOSTRIS    I.     (Museum, 

Cairo).     (Phot.  E.  Brugsch.) 


instruction  show  them  to  us  in  a  variety  of  attitudes  which  had 
never  been  noted  before.  In  the  tomb  of  Kheti  at  Beni-Hasan, 
and  in  the  tomb  of  Beket  there 
are  over  120  of  these  groups, 
which  reproduce  as  if  in  a  cine- 
matograph the  successive  move- 
ments of  these  duels  (Fig.  197). 
In  obedience  to  a  somewhat 
childish  convention,  the  two 
combatants  are  not  the  same 
colour;  one  is  painted  black, 
the  other  red,  to  avoid  any 
confusion  in  the  interlacement 
of  their  limbs.  We  see  them 
approaching,  touching,  and  seizing 
each  other,  relaxing  or  stiffening 
their  muscles  alternately,  in  order 
to  escape  the  grip  of  an  ad- 
versary or  to  bring  him  down;  one  of  them,  seized  by  the 
middle  of  the  body,  is  hurled  to  the  ground,  but  as  his  shoulders 

109 


FIG.  202.— WOODEN  HEAD  FROM  LISHT 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


FIG.   203.— PHARAOH 

HORUS  (Museum ,   Cairo). 

(Phot.  E.  Brugsdi.) 


have  not  touched  it,  the  struggle  continues. 
A  professional  athlete  would  easily  name 
the  various  falls,  and  he  would  perhaps  also 
recognise  certain  tricks  of  hands  or  legs 
which  are  no  longer  tolerated  in  modern 
arenas ;  what  we  can  never  sufficiently  admire 
is  the  skill  and  facility  with  which  the  play 
of  the  limbs  and  their  interlacement  have 
been  analysed,  and  fixed  on  the  wall.  Before 
achieving  such  mastery,  the  draughtsman 
must  have  made  prolonged  studies  from 
life,  and  have  spent  much  time  in  the 
palestra;  he  could  not  otherwise  have  follow- 
ed the  lines  and  emphasised  the  decisive 
moments  of  each  bout.  No  one  familiar 
with  Egyptian  monuments  would  have  been 
surprised  to  find  some  ten  or  even  twenty 
groups  well  drawn,  but  what  is  really 
amazing  is  that  out  of  over  a  hundred  and 
twenty,  there  are  scarcely  half  a  dozen 

which  are  incorrect  or  badly  balanced.  Not  only  in  bas-relief 
on  the  flat  stone  did  contemporary  artists  venture  to  combine 
these  violent  attitudes;  they  dared  to  realise  them  in  the  round. 
The  wrestlers  of  the  Munich  Museum  show  the  same  correctness 
and  animation  which  characterise  those  of  the  hypogea,  in  spite 

of  a  certain  rudeness  of  technique. 
At  Syene,  Beni-Hasan,  and  Siut, 
workshops  depending  upon  the  great 
nobles  flourished  during  the  last 
years  of  the  Eleventh  and  the  first 
years  of  the  Twelfth  Dynasties.  The 
royal  ateliers  made  their  appearance, 
when,  the  Twelfth  Dynasty  having 
firmly  established  its  suzerainty  over 
the  great  feudal  families,  the  whole 
country  was  peacefully  united  under 
a  single  chief.  Naturally  enough,  the 
Memphite  School  remained  predo- 
minant in  Middle  Egypt,  in  Abydos, 
at  Heracleopolis  and  in  the  Fayum. 
Its  traditions  had  not  been  allowed 
to  fall  into  decay  during  the  inter- 
mediate period;  it  imposed  them  on 
110  N 


FIG.  204.— HEAD  OF  THE  STATUE 

OF  PHARAOH  HORUS 
(Museum,  Cairo).  (Phot.  E.  Brugsdi.) 


THE  FIRST  THEBAN  AGE 


the  Theban  kings  when,  quitting  their  southern  abode,  these 
installed  themselves  in  the  palaces  of  their  northern  predecessors. 
We  have  but  a  small  number  of  bas-reliefs  to  attribute  to  it; 
but  the  fragments  of  the  chapel  of 
Amenemhat  I.  at  Lisht  and  the  mastabas 
of  Dahshur  are  certainly  equal  to  the 
best  works  of  the  Fifth  Dynasty.  It 
continued  to  keep  the  relief  very  low 
on  the  surface  of  the  stone  and  aimed 
more  than  ever  at  elegance  and  deli- 
cacy of  contour,  though  this  did  not 
prevent  it  from  rendering  its  sitters 
with  an  almost  brutal  realism  upon 
occasion ,  as  in  the  portraits  of  Siesis 
at  Dahshur  (Fig.  198).  Its  statues  are 
penetrated  by  the  same  spirit  as  its 
bas-reliefs.  Those  of  Sesostris  I.,  which 
come  from  Lisht,  are  marked  by  great 
dignity,  and  they  keep  the  attention  of  the  visitor  long  fixed 
upon  them ,  in  spite  of  the  sense  of  monotony  induced  by  the 
repetition  of  the  same  attitude  eleven  times  (Fig.  199).  He  will 
soon  perceive,  however,  that  their  sculptors  were  swayed  by 
school  tradition ,  when  they  gave  these  figures  their  short,  oval 


FIG.  205.  PAINTED  BAS-RELIEF 

FROM  THE  TOMB  OF  MENTHU- 

HETEP    (After  Mme.  Naville). 


FIG.   206.— ONE    OF   THE    SIDES    OF   PRINCESS   KAUIT'S    SARCOPHAGUS 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 

faces,  their  smiling,  good-humoured  expression,  their  placid  eyes 
(Fig.   200),     and    broad,    plump    bodies.     They   were    obsessed 

111 


ART  IN  EGYPT 


FIG.  207. — PART  OF  A  STELE  OF  SESOSTRIS  HI. 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


by  the  Memphite  type  of 
the  Pharaoh  fixed  by  the 
masters  of  the  Fifth  Dy- 
nasty. The  same  conven- 
tion prevailed  in  the  work- 
shops of  Abydos;  the 
colossal  Osirian  statues 
of  Amenemhat  and  of 
Sesostris  I.  are  so  com- 
monplace in  conception, 
that  even  the  excellence 
of  their  workmanship  fails 
to  redeem  them  (Fig.  201). 
This  languid  and  imper- 
sonal manner  persisted 
under  the  Thirteenth  and 
Fourteenth  Dynasties,  and 
is  noticeable  even  in  pieces 
not  devoid  of  merit,  such 
as  the  little  wooden  head  at  Cairo  with  its  plaster  wig  (Fig.  202) 
and  the  wooden  statue  of  the  little  King  Horus  at  Dahshur 
(Fig.  203).  He  is  naked,  and  walks  with  a  stride,  his  left  arm 
advanced.  He  is  an  agreeable  figure  enough,  but  how  insipid 

he     seems     when 

compared  with  the 

Sheikh  -  el  -  beled. 

The  torso  is  light, 

the  hip  slender,  the 

leg  long  and  slight, 

the  features  regu- 
lar (Fig.  204),  but 

the  grace  is  purely 

superficial,  and  the 

first     impulse     of 

admiration       does 

not  stand  the  test 

of  careful  exami- 
nation. After  this, 

we   shall    not    be 

surprised    to    find 


FIG.    2O8.— MENTHU-HETEP    IN 
THE  COSTUME  OF  APOTHEOSIS 

(Museum,  Cairo). 
(Phot.  E.  Brugsch.} 


in  museums  many 
statues  and  statu- 
ettes, manufactur- 
112 


FIG.   2OQ. 

ANONYMOUS   STATUE 
(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


THE  FIRST  THEBAN  AGE 


FIG.   2IO.— HEAD    OF   SESOSTRIS  I. 
AT  KARNAK.     (Phot.  Legrainj 


ed  according  to  the  best  rectipfe,  but  never  rising  above  the  level 
of  commercial  imagery;  when  the  inscriptions  reveal  their  origin, 
we  recognise  regretfully  that  they  belong  for  the  most  part  to 
the  inferior  Memphite  School  of 
the  first  Theban  period ,  or  to 
its  Abydonian  branches. 

The  Theban  School,  at  first 
hardly  distinguishable  from  the 
local  schools  which  vegetated 
obscurely  at  Denderah,  Coptos, 
Nakadah  and  Erment  during 
the  age  of  the  Great  Pyra- 
mids, did  not  repudiate  its 
technique  and  principles,  when 
the  rise  of  the  Antef  power 
roused  it  from  its  torpor  to 
interpret  the  artistic  aspirations 
of  a  new  royalty.  Its  first  known 
works,  the  stele  of  Prince  Antef  and  the  bas-reliefs  of  the  pyramid- 
mastaba  of  Menthu-hetep  at  Der-el-Bahari  (Fig.  205)  mark  an  im- 
mense advance  from  the  one  to  the  other.  The  stele  is  still  bar- 
barous in  style:  figures  distributed  unsymmetrically  in  the  registers, 
bodies  badly  proportioned,  attitudes  laboured,  gestures  angular. 
On  the  other  hand ,  those  of  the  bas-reliefs  which  are  not  so 
mutilated  as  to  make  an  appreciation  impossible,  show  drawing 
as  correct  and  a  touch  as  firm  as  the  good  Memphite  sculptures 
of  the  Fifth  Dynasty;  the  scenes  on  the  sarcophagus  of  Princess 
Kauit  are  good  examples  (Fig.  206).  At  the  same  time,  the 
relief  is  stronger,  the  contour  bolder  and  more  animated,  the 
man  is  sturdier,  and  is 
planted  more  solidly  on 
the  line  of  ground,  the 
woman  is  shorter,  and 
fuller  in  the  hips  and 
bosom.  When  they  had 
reached  this  point,  the 
Thebans  progressed  rapid- 
ly. Among  the  broken 
blocks  which  Thothmes  III. 
used  at  Karnak  to  raise 
the  level  of  one  of  the  courts  was  a  square  pillar  which  came 
from  the  limestone  temple  of  Sesostris  I.  (cf.  Fig.  176).  On  it 
Pharaoh  and  the  god  Ptah,  standing,  nose  to  nose,  inhale  each 

113  I 


FIG.  211. — HEAD   OF  A   COLOSSAL   STATUE   OF 
SESOSTRIS  I.  (Museum,  Cairo).    (Phot.  Legrain.) 


ART  IN  EGYPT 


FIG.  212. — HEAD  OF  A  COLOS- 
SAL STATUE  OF  SESOSTRIS  III. 
(Phot.  Legrain.) 


other's    breath   in   accordance  with   the 
etiquette  observed    by  persons   of  rank 
in  salutation.  The  profiles  are  vigorously 
marked,  the    relief   is   stronger   than   at 
Memphis,    and  consequently,    it  throws 
a  stronger  shadow,  and  stands  out  from 
the    background    more    decisively    than 
in    the   pictures   of  Gizeh  or  Sakkarah. 
The   scenes    of   the    three    other    faces 
reveal  an  art  no  less  exquisite,    and   it 
is  much  to  be  regretted  that  the  example 
is,    so  far,    unique;    if   all  the  building 
was   decorated  as  happily  as  this   frag- 
ment,   the  Twelfth  Dynasty   must  have 
raised  at  Thebes  a  monument  compar- 
able   to    the    noblest    achievements    of 
the  Eighteenth  and  Nineteenth  Dynasties. 
The   only  work  I   can   set   beside   it    is 
the  stele    of  the    same   king,  which  he 
dedicated   to   the   memory  of  his   pre- 
decessor Menthu-hetep  (Fig.  207)  in  the  pyramid-mastaba  of  Der- 
el-Bahari.      The    graving    both    of    picture    and    inscriptions    is 
marvellous ;    the  figures  of   the   two  Pharaohs   and  of  Amon-Ra 
are    cut   with    as    much    delicacy   as    an 
intaglio    on    a    precious    stone,    but    the 
minuteness  of  the  detail  does  not  detract 
from  the   breadth  of  the    execution.     All 
this  was  produced  in  the  royal  workshop 
at  Thebes  and  is  the  best  work  of  the 
school;     the    fairly    numerous     examples 
produced      in      private      workshops      — 
stelae,  sarcophagi  and  bas-reliefs  —  are 
by  no  means  on  the  same  level.    The  most 
careful    of    them     are    disfigured    by    a 
stiffness   of  attitude   and   a   heaviness   of 
chiselling  which  proclaim  them  the  direct 
products  of  the  old  provincial  academies, 
innocent   of   Memphite    influences.     It   is 
probable  that  the  first  Antefs,   conscious 
of    the     defects    of    their     national    art, 
summoned     draughtsmen     and     sculptors 
from  Abydos   or   some  other  city  to  in- 
struct the  natives;    these  assimilated  the 
114 


FIG.  213. — HEAD    OF 

SESOSTRIS   IV. 
(Museum,   Cairo). 
(Phot.  Legrain). 


THE  FIRST  THEBAN  AGE 


more  refined  processes  of  their  masters,  and  the  fusion  of  this 
acquired  dexterity  with  their  instinctive  rudeness  produced  the 
Theban  style. 

Its  characteristics  are  yet  more  strongly  marked  in  their  statues. 
When  about  to  undergo  the  ceremonies  of  deification  which 
usage  prescribed  for  the  Pharaohs  after  the  first  years  of  their 
reign ,  Menthu-hetep  ordered  the  statues  which  were  to  replace 
him,  or  rather,  duplicate  him,  during  the  mysteries  of  his  identi- 
fication with  Osiris  (Fig.  208).  They  carved  one  in  limestone, 
a  lofty  visionary 
creation  with  its 
massive  feet  and 
knees,  its  heavy 
hands,  its  per- 
functory bust, 
broadly  indicated 
features,  harsh  il- 
lumination, black 
flesh,  crude  white 
costume,  and  the 
dark  red  cap 
prescribed  by  rit- 
ual ;  the  whole 
is  savage ,  but 
with  a  deliberate 
savagery,  design- 
ed for  religious 
effect.  If  a  Mem- 
phite  had  treat- 
ed the  subject, 
he  would  have  done  his  utmost  to  soften  the  lines  and  harmonise 
the  colours;  he  would  unconsciously  have  approximated  his 
model  to  the  ideal  type  of  humanity  which  pleased  his  school, 
at  the  risk  of  robbing  him  of  his  native  energy.  The  Theban, 
on  the  other  hand,  strove  only  to  transcribe  reality  as  it  present- 
ed itself  to  him ,  and  this  preoccupation  dominated  those  who 
succeeded  him  to  the  end.  They  sought  resemblance,  with  a 
determination  to  exaggerate  rather  than  attenuate  the  character- 
istic peculiarities  of  the  sitter,  and  in  their  pursuit  of  it,  they 
were  not  repelled  either  by  harshness  of  handling  or  violence 
of  colour.  This  is  well  shown  in  the  statue  of  an  unknown  person, 
seated,  and  wrapped  in  his  mantle,  who,  with  his  bold  glance, 
seems  still  to  exhal§  a  breath  of  intense  life,  in  spite  of  the 

115  I  2 


FIG.  214.— STATUETTE 
OF  AMENEMHAT  III. 

(Museum,    Cairo). 
(Phot.  E.  Brugsch.) 


FIG.  215. —  STATUE   OF 

AMENEMHAT  III. 

(Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


FIG.  216.    SPHINX  OFAMENEMHAT  HI. 
(Museum,  Cairo).    (Phot.  E.  Brugsch.) 


mutilations  of  his  face  (Fig.  209). 
Other  examples  even  better  are 
the  colossal  statues  of  Sesostris  I. 
(Figs.  210,  211)  and  of  Sesostris  III. 
found  by  Legrain  at  Karnak.  The 
body  differs  from  that  of  the  Mem- 
phite  statues  only  by  its  more 
slender  proportions,  but  what  of 
the  head?  The  artist  chosen  by 
Sesostris  III.  reproduced  line  by 
line  the  long,  thin  face  of  the 
prince ,  his  narrow  forehead ,  his 
high  cheek-bones,  his  bony,  almost 
bestial  jaw.  He  hollowed  the 
cheeks,  enframed  the  nose  and 
mouth  between  two  furrows,  com- 
pressed and  thrust  out  the  lip  in 
a  disdainful  pout  (Fig.  212);  he 
thus  fixed  the  true  image  of  the 
individual  Sesostris  where  the  Memphite  and  the  Abydonian, 
imbued  with  the  opposite  principles,  would  have  evolved  from 
the  stone  an  effigy  of  Pharaoh ,  idealised  as  much  as  it  was 
possible  to  idealise  without  entirely  destroying  the  likeness.  Two 
or  three  centuries  later,  at  an  advanc- 
ed stage  of  decadence,  the  same 
characteristics  persisted  in  the  colossal 
statue  of  Sesostris  IV.  (Fig.  213).  This 
is  the  only  e>tant  portrait  of  the  king, 
but  no  one,  looking  at  it,  can  doubt 
that  he  has  the  man  himself  before  his 
eyes.  The  living  monarch  was  certainly 
what  the  stone  tells  us  he  was,  a  jovial 
and  sensual  rustic.  It  is  not  improbable 
that  the  will  of  the  sovereign  some- 
times influenced  the  manner  of  the 
artist  I  am  inclined  to  believe  that 
this  was  the  case  with  Amenemhat  III. 
The  majority  of  his  statues  bear  the 
impress  of  the  school,  and  are  the  work 
of  Thebans,  whether  they  originated  at 
Thebes  (Fig.  214)  or  in  the  Memphite 
districts;  I  cannot  except  even  that  of 
Hawara  (Fig.  215).  The  king  wears  on 
116 


FIG.   217.— QUEEN  NEFERT 

(Museum.  Cairo). 
(Phot.  E.  Brugsch.) 


THE  FIRST  THEBAN  AGE 


FIG.  2l8.— HEAD  OF  A  COLOSSAL, 
ROYAL    STATUE    OF    THE    THIR- 
TEENTH DYNASTY 
(Egyptian  Museum). 
(Phot.  E.  Brugsch.) 


his  head  the  kufieh  and  the  uraeus, 

and    his    face    has    nothing   of  the 

usual   conventional  cast.     The   eye 

is  small  and  prominent,  slightly 
I  veiled  at  the  extremity,  the  eyelid 

heavy,  the  nose  straight  and  short, 
1  the  cheek  hollow,  the  cheek-bone 
I  very  strongly  marked,  the  mouth, 

with  its  thin,  compressed  lips,  firm 

and   scornful,    the    chin   hard    and 

obstinate,  the  neck  thin,  the  chest 
j  flat,  the  leg  sinewy,  the  foot  ner- 
I  vous.  The  whole  reveals  a  very 

remarkable  technique,  though  cer- 
tain details  of  the  lower  extre- 
mities are ,  as  usual ,  summary. 

The  sovereign  must  have  insist- 
[  ed  upon  a  realistic  rendering, 

and   nowhere    are    the   naturalistic 

tendencies    of    the     school     more 

strongly   marked   than    in    this   admirable   piece   of   work. 

I  To  him  again  we  undoubtedly  owe  those  sphinxes  of  Tanis 
erroneously  ascribed  by  Mariette  to  the  Hyksos  Pharaohs  (Fig.  2 16). 
We  can  understand,  when  we  see  them, 
that  Mariette  should  have  been  misled, 
and  that  he  should  have  hesitated  to 
believe  in  their  Egyptian  origin.  There 
is  a  superabundant  energy  in  these 
nervous  leonine  bodies,  which  are  sturdier 
and  more  compact  than  those  of  the 
ordinary  sphinxes.  The  face  is  bony, 
the  nose  aquiline,  the  nostril  slightly 
flattened;  the  lower  lip  is  thrust  out, 
a  bull's  ear  emerges  from  the  lion's 
mane  which  enframes  the  face ,  and 
drapes  the  neck  and  shoulders.  The 
technique  is  that  of  the  Thebans,  and 
Thebes  I  believe  to  be  the  source  of 
this  Tanite  School ,  but  we  are  con- 
scious of  an  inspiration  as  yet  undis- 
ciplined and  almost  barbarous.  The 
semi-civilised  inhabitants  of  the  eastern 
marshes  of  the  Delta  imposed  a  certain 

117 


FIG.2IQ. — COLOSSALSTATUE 

OF      MIRMASHAU       (Museum, 

Cairo).     (Phot.  E.  Brugsdi.) 


ART  IN  EGYPT 


FIG.   22O. 

SEBEK  -  HETEP 

(The  Louvre,  Paris). 

(Phot.  Gaudier  -  Gudin.) 


brutality  peculiar  to 
them  on  the  works 
of  their  masters.  The 
sphinxes  of  Amenem- 
hat  III.  show  us  plainly 
enough  the  ideal  they 
had  in  view;  they  produc- 
ed nothing  to  surpass 
these,  but  we  have  sever- 
al remarkable  works  by 
them  executed  under 
the  Twelfth  Dynasty, 
and  also  between  the 
Twelfth  Dynasty  and 
the  rise  of  the  second 
Theban  Empire.  The 
black  granite  statues  of 
Nefert,  wife  of  Sesos- 
tris  II.  (Fig.  217)  have 
a  special  charm,  in  spite 
of  the  ungraceful  Ha- 


rm. 221. 

SEBEK-EMSAF 
(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


thorian  headdress  which  makes  the  face  heavy.  The  colossal  head 
from  Bubastis  (Fig.  218)  and  the  colossal  statues  of  Mirmashau 
(Fig.  219),  now  in  the  vestibule  of  the  Cairo  Museum,  were  hewn 
in  broad  planes  from  a  recalcitrant  granite,  and  the  modelling 
was  not  carried  .very  far;  it  is,  however,  so  correct  as  to  be 
comparable  to  the  best  Theban  pieces.  The  face  is  mutilated; 

but  on  what  remains  of  it 
we  divine  a  vigour  equal  to 
that  in  the  faces  of  the 
sphinxes.  It  would  be  inter- 
esting to  see  more  exam- 
ples which  would  throw  light 
on  the  destinies  of  this 
school ;  but  unfortunately 
the  relics  of  the  Thirteenth 
and  Fourteenth  Dynasties 
are  so  scanty,  that  it  is  im- 
possible to  deduce  even 
the  elements  of  a  history 
of  art  from  them.  As  far 
as  we  may  safely  conjec- 

FIG.  222.— THE  TWIN   STATUES   AT    CAIRO.  <,  J       ..       J  . 

(Phot  E.  Brugsch.)  ture ,   a   uniform   mediocrity 

118 


THE  FIRST  THEBAN  AGE 


gradually  invaded  the  whole  of  Egypt. 
Neither  the  Sebek-hetep  of  the  Louvre 
(Fig.  220)  the  Sebek-emsaf  (Fig.  221), 
the  twin  kings  of  Cairo  (Fig.  222)  nor 
the  colossal  statues  usurped  by  Rameses  II. 
are  bad,  and  yet  no  one  would  venture 
to  pronounce  them  good.  The  royal 
workshops  whence  they  came  had  lost 
little  of  their  manual  facility ,  but  they 
no  longer  formed  artists  capable  of 
competing  with  those  who  fashioned  the 
colossal  statues  of  Sesostris.  The  private 
workshops  were  very  unequal  in  their 
productions.  The  statues  we  have  from 
them  in  the  Cairo  Museum  (Figs.  223, 
225),  are  coarse  and  heavy,  but  the 
majority  of  their  customers  ordered  only 
statuettes,  many  of  which  are  no  larger 
than  figurines ;  these 


FIG.  223. — STATUE   OF  A 
PRIVATE  PERSON 
(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


FIG.  224.  — SKBEK-EMSAF 
(Museum,  Vienna). 
(Phot.  Bergmaitn.) 


they    treated    with 
brilliant     dexterity. 

The  Sebek-emsaf  at  Vienna  (Fig.  224)  owes 
a  rather  ridiculous  rotundity  to  his  horrible 
petticoat;  his  little  person  is  nevertheless, 
interesting  for  the  knowledge  of  the  human 
structure  it  reveals.  The  dainty  walking 
scribe  at  Cairo  would  take  his  place  among 
the  most  delicate  works  of  the  Twelfth 
Dynasty,  if  the  inscriptions  we  read  upon 
him  did  not  compel  us  to  refer  him  to 
the  Thirteenth.  We  know  scarcely  anything 
of  the  period,  and  each  time  an  attempt 
is  made  to  re -construct  it  from  existing 
data,  new  documents  come  to  light,  which 
overthrow  systems  to  all  appearance  most 
solidly  built  up.  I  have  given  the  results 
of  my  examination  of  all  that  is  known; 
I  refrain  from  positive  conclusions  which 
might  be  demolished  to-morrow. 

Furniture,    domestic    pottery,    and    table 

utensils  of  stone  or  metal,  textiles,  embro- 

deries,     in    a    word,    the    minor    arts,    all 

flourished     under     the     Theban     Pharaohs, 

119 


ART  IN  EGYPT 


FIG.    225.— STATUE 
OF  A  PRIVATE  PER- 
SON (Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


although  our  museums  contain  but  few  spe- 
cimens. The  discovery  of  the  treasures  of 
Dahshur  has  given  us  so  many  precious  objects 
that  we  are  able  to  form  a  well-grounded 
opinion  of  the  art  of  jeweller  and  goldsmith. 
Three  harems  of  the  Twelfth  and  Thirteenth 
Dynasties  have  combined  to  bequeathe  us 
almost  complete  sets  of  jewels  belonging  to 
queens  and  princesses.  Their  necklaces,  their 
mirrors,  their  rings,  their  bracelets  and  their 
crowns  are  heaped  pell-mell  beside  pectorals 
bearing  the  names  of  their  fathers  and  hus- 
bands; he  who  would  wish  to  give  an  idea 
of  the  elegance  of  their  forms,  and  the 
harmonious  vivacity  of  their  colours  would 
have  to  describe  everything,  or  rather  repro- 
duce everything  in  coloured  facsimiles.  The 
principal  pectoral  of  Sesostris  III.  (Fig.  226), 
simulates  a  naos  in  gold  with  lotus -columns, 
the  field  of  which  is  occupied  in  the  centre 
by  a  vulture  hovering  over  a  cartouche;  two 

griffins ,    emblems    of    Mentu ,     the    god    of    war ,    strike    down 

Asiatics   right  and    left   of  the  cartouche.     The   breast -plate    of 

Amenemhat   III.    (Fig.   227)    is   also    a   naos,    but   the   Pharaoh, 

twice    represented    upon 

it,  brandishes  a  club  over 

a  kneeling   prisoner   who 

begs   in   vain   for  mercy. 

Gold  chains,  filagree  stars, 

medallions  of  glass  mo- 
saic, necklaces  with  golden 

pendants   in   the    form  of 

shells  (Fig.  228)  we  pass 

from  one  piece  to  another, 

unwearying  in  admiration. 

One      of      the       crowns 

(Fig.  229)    is    formed   of 

rosettes   and  lyre -shaped 

Ornaments  surmounted  by  FIG.  ^.-PECTORAL  OF  SESOSTRIS  m. 

eight     Upright      florets      in  (Museum,  Cairo).     (Phot.  E.  Brugsch.) 

gold ,     lapis    lazuli ,    red 

jasper  and  green  felspar;   a  vulture  of  gold  and  precious  stones 
with  outspread  wings  accompanied  this,  and  an  aigrette  of  gold, 

120 


THE  FIRST  THEBAN  AGE 


™E?  --  • 


iSF™80 

~3^***      -A 


m  mBMMngrg:)  :••>     » 


FIG.   227.— PECTORAL  OF  AMENEMHAT   III. 
(Museum,  Cairo).  (Phot.lll.  E.  Brugsch.) 


representing  a  spray  with  golden  leaves  and  trusses  of  flowers. 
The  other  crown  (Fig.  230),  is  an  interlacement  of  delicate  threads 
of  gold,  on  which  six 
Maltese  crosses  in  gold, 
with  centres  of  cornelian 
and  blue  limbs,  are  set 
at  regular  intervals;  a 
handful  of  little  blossoms 
with  red  hearts  and  blue 
petals  arranged  in  a  star 
is  scattered  between  the 
florets.  Nowhere  in  Egypt, 
or  throughout  the  antique 
world  do  we  find  a  richer 
design,  a  more  skilful  dis- 
tribution, a  truer  sense 
of  colour.  The  faults 
that  have  been  pointed 
out ,  the  superabundance 
of  heavy  enamels  and  the  slightness  of  the  mounting  are  the 
results  of  causes  which  explain  and  perhaps  excuse  them.  The 
Egyptians  were  richly  adorn- 
ed ,  not  only  during  their 
lifetime,  but  after  their  death ; 
their  mummy-jewelry,  how- 
ever, destined  for  a  motion- 
less body,  did  not  need  to 
be  so  solid  as  that  of  a 
living  person ,  continually 
shaken  by  the  movements 
of  the  wearer.  If  our  crown 
had  adorned  the  head  of 
Khnemit  during  the  court 
ceremonies,  it  would  not 
have  lasted  more  than  a 
few  days  or  perhaps  a 
few  hours;  the  enamelled 
flowers  and  crosses  weigh- 
ing on  the  gold  threads, 
would  have  broken  them 
promptly.  They  were  de- 
signed for  the  coffin,  and  SELECTION  or, FROM  DAHSHUR 
the  eternal  inertia  to  which  (Museum,  Cairo).  (Phot.  E.  Brugsch.) 

121 


ART  IN  EGYPT 

they  were  to  be  consigned  encouraged  the  artist  to  consider 
only  his  own  taste  and  fancy.  The  Greek  goldsmiths  reasoned 
in  the  same  manner,  when  they  worked  under  similar  con- 
ditions, and  the  workmanship  of  their  funerary  jewels  is  as 
frail  as  that  of  the  crowns  of  Dahshur. 


FIG.  229.— ONE   OF   KHXF.MIT'S   CROWNS 
(Museum,   Cairo).     (Phot.  E.  Brugsch.) 


BIBLIOGRAPHY  TO  CHAPTER  I  —  PART  II 

The  first  Theban  age  is  the  one  of  whose  art-history  in  Egypt  we  know  least,  that  on 
which  there  are  fewest  books  and  articles. 

Architecture.  — •  For  civil  architecture,  cf.  Flinders  Petrie,  Hawara  Biahmu  and  Arsinoe, 
in  quarto,  London,  1889,  66  p.  and  XXIX  pi.,  Illahun  Kahun  and  Curob,  in  quarto,  London, 
1891,  55  p.  and  XXXIII  pi.,  Kahun,  Gurob  and  Hawara,  1890,  53  p.  and  XXVIII  pi.,  and 
more  especially  on  the  decoration  of  houses:  L.  Borchardt,  Das  agyptische  stddlische 
Wohnhaus  mil  besonderer  Beriicksichtigung  der  inneren  Dekoration,  in  the  Deutsche  Bau- 
zeitschrift,  vol.  XXVII  ,  p.  200;  for  private  hypogea,  Marittte,  Voyage  de  la  Haute-Egypte, 
vol.  I,  p.  49-53,  and  pi.  17;  —  P.  E.  Newberry,  Beni- Hassan  (Archaeological  Survey  of 
Egypt,  vol.  I,  II,  V,  VII),  in  quarto,  London,  I,  1893,  87  p.  and  XLVII  pi.,  II,  1894,  87  p.  and 
XXXVIII  pi.,  Ill  18%,  42  p.  and  X  pi.,  IV  19CO,  9  p.  and  XXVII  pi.;  Bersheh  (Archaeo- 
logical Survey  of  Egypt,  vol.  III-IV),  in  quarto,  I  ondon,  I,  1893,  40  p.  and  XXXIV  pi.,  II 
1895,  71  p.  and  XXIII  pi.;  —  N.  de  G.  Davies,  The  Rock-tombs  of  Deir  el  Gebrawi  (Archaeo- 
logical Survey  of  Egypt,  vol.  XI),  in  quarto,  London,  1902,  43  p.  and  26  pi.;  —  J.  de  Morgan, 
De  la  frontiere  d'Egypte  a  Kom-Ombo,  in  quarto,  Vienna,  1894,  VIII-212  p. ;  for  royal  pyramids 
of  the  Memphite  type:  Flinders  Petrie,  Kahun  Illahun  and  Hawara,  in  quarto,  London, 
1890,  53  p.  and  XXVIII  pi.,  and  Illahun,  Kahun  and  Gurob,  1891,  56  p.  and  XXXIII  pi. ;  — 
J.  de  Morgan,  Fouilles  de  Dahchour,  in  quarto,  Vienna,  I,  1895,  IV-165  p.  and  40  pi.,  II 
1903,  V1II-119  p.  and  27  pi.  —  J.-E.  Gautier  and  G.  Jequier,  Memoires  sur  les  fouilles  de 
Licht  (Memoires  de  1'Institut  francais  d'archeologie  orientale,  vol.  VI),  in  quarto,  Cairo, 
1902,  107  p.  and  30  pi.;  for  pyramid-mastabas :  Mariette,  Abudos,  in  folio,  Paris,  1870, 
vol.  II,  p.  38-45  and  pi.  XLVI-XLVII;  —  Naville,  The  XIth  Dynasty  Temple  at  Deir  el 
Bahari  (Egypt  Exploration  Fund,  vol.  XI-XII),  in  quarto,  London,  I,  1907,  75  p.  and  XXI  pi., 
II,  1910,  29  p.  and  XXIV  pi.;  for  the  points  of  pyramids:  H.  Scheefer,  Die  Spitze  der  Py- 
ramide  des  Konigs  Amenemhats  III.  in  the  Zeitschrift  fur  Agyptische  Sprache,  1900, 
vol.  XLI,  p.  84-85. 

122 


THE  FIRST  THEBAN  AGE 

Painting  and  Sculpture.  —  The  history  of  the  Theban  School  for  this  and  the  following 
period  has  been  established  by  G.  Maspero,  La  Cachette  de  Karnak  et  iEcole  de  Sculpture 
thebaine,  in  the  Revue  de  I'Art  ancien  et  moderne,  1906,  vol.  XX,  p.  241-252,  337-348. 
For  the  whole  field  of  artistic  activity,  cf.  in  addition  to  the  works  of  Davies,  Gautier- 
Jequier,  Newberry  and  Petrie  quoted  in  reference  to  sculpture,  the  following:  Flinders 
Petrie,  Tanis  (Egypt  Exploration  Fund,  vol.  II  and  V),  in  quarto,  London,  I,  1885,  63  p.  and 
XIX  pi.,  II,  1888,  116  p.  and  LXIII  pi.,  Koptos,  in  quarto,  London,  18%,  38  p.  and  XXVIII  pi.; 
—  Fr.  W.  v.  Bissing,  Denkmaler  dgyptischer  Skulptur,  pi.  19-35,  40 a,  77  a;  —  I.  Capart, 
Recueil  de  Monuments  egyptiens,  1902-1905,  in  quarto,  Brussels,  pi.  XV,  XVII,  XXIV-XXXlII- 
LIX-LXII,  and  the  corresponding  text;  —  L.  Borchardt,  Kunstwerke  aus  dem  dgyptischen 
Museum  zu  Kairo,  pi.  6-7,  23,  p.  5,  11 ;  —  E.  de  Rouge,  Album  photographique  de  la 
Mission,  No.  109-120;  —  G.  Legrain,  Statues  et  Statuettes  de  Rois  et  de  particuliers  (Cata- 
logue general  du  Musee  du  Caire),  in  quarto,  Cairo  I,  1906,  p.  1-29,  and  pi.  I-XXVI;  — 
G.  Maspero,  Le  Musee  egyptien,  in  quarto,  Cairo,  1904,  vol.  II,  p.  25-30,  34-35,  41-45  and 
pi.  IX-X,  XIII,  XV.  Sur  trois  Statues  du  premier  Empire  thebain,  in  the  Annales  du  Service, 
1902,  vol.  Ill,  p.  94-95  and  1  pi.  —  On  the  special  question  of  the  so-called  Hyksos  Sphinxes 
of  Tanis,  cf.  W.  Golenischeff,  Amenemha  III  et  les  Sphinx  de  San,  in  the  Recueil  de 
Travaux,  1893,  vol.  XV,  p.  131-136  and  5  pi. 

For  goldsmith's  work  and  the  minor  arts,  see  in  addition  to  J.  de  Morgan's  work  on 
the  excavations  at  Dahshur,  the  treatises  of  E.  Vernier,  La  Bijouterie  et  la  Joaillerie  egyp- 
tiennes  (Memoires  de  1'Institut  francais  d'archeologie  orientale,  vol.  II),  in  quarto,  Cairo, 
1907,  VII-156  p.  and  XXV  pi.,  and  Bijoux  et  Orfevreries  (Catalogue  general  du  Musee  du 
Caire),  in  quarto,  Cairo,  1907-1909,  200  p.  and  XXXVII  pi.,  also  the  work  of  Schaefer, 
Agyptische  Goldschmiedearbeiten  (with  the  collaboration  of  G.  Moller  and  W.  Schubart, 
forms  vol.  I  of  Mitteilungen  aus  der  Agyptischen  Sammlung),  in  folio,  Betlin,  1910,  p.  16-19 
and  pi.  3 ;  —  L.  Borchardt,  Kunstwerke  aus  dem  dgyptischen  Museum  zu  Kairo,  pi.  41-42 
and  p.  17-18. 


FICi.  230.— DIADEM  OF  K.HNEMIT  (Museum,    Cairo).     (Phot.  E.  Brugsch.) 


123 


FIG.  231. — ATLANTES   OF   THE   FORE-COURT  AT  MEDINET-HABU.      (Phot.  Beato.) 


CHAPTER  II 

THE  SECOND  THEBAN  AGE  FROM  THE  EIGHTEENTH 
TO  THE  TWENTY-FIRST  DYNASTY 

Renaissance  of  Art  at  the  beginning  of  the  Eighteenth  Dynasty  —  The  Temple  and  its 
various  Types  —  The  Hemispeos  and  the  Speos  —  The  royal  Workshops  of  Painting 
and  Sculpture  at  Thebes  and  in  the  Provinces:  the  decoration  of  Tombs  and  Hypogea 
—  Goldsmith's  Work,  Jewelry,  and  the  minor  Arts. 

WHEN  we  compare  the  works  of  the  Thirteenth  and  Seven- 
teenth Dynasties,  the  differences  between  them  appear 
so  slight  that  we  are  almost  tempted  to  believe  them  contem- 
porary. This  is  more  especially  true  of  the  statues  and  statu- 
ettes; that  of  the  Shepherd  King  Khayanu  might  have  been  exe- 
cuted for  one  of  the  Sebek-heteps,  and  the  mutilated  bust  which 
Mariette  discovere  in  the  Fayum  (Fig.  232)  bears  a  most  decep- 
tive resemblance  in  technique  to  the  Tanite  sphinxes  of  Amen- 
emhat  III.  It  must  be  admitted,  however,  that  a  judgment  based 
wholly  upon  these  official  examples  might  be  unfair.  At  periods 
of  political  abasement,  the  court  workshops  were  maintained  with 
great  difficulty  on  the  scanty  resources  at  the  disposal  of  their 
masters,  and  they  were  reduced  to  servile  reproduction  of  the 

124 


THE  SECOND  THEBAN  AGE 


types  and  technique  of  happier 
periods.  It  seems  probable,  how- 
ever, that  they  did  not  remain 
altogether  stationary,  any  more 
than  the  private  workshops.  Un- 
civilised as  the  Hyksos  are  sup- 
posed to  have  been  at  the  be- 
ginning of  their  domination,  they 
had  nevertheless  brought  with 
them  not  only  material  elements 
of  progress,  such  as  the  horse, 
the  chariot,  the  quiver,  the  bronze 
squamate  cuirass  and  weapons  of 
a  new  type,  but  also  habits  and 
modes  of  thought  novel  to  the 
Egypt  of  their  day.  True,  the 
leaven  of  originality  they  intro- 
duced into  the  ancient  mass  was 
not  so  active  as  to  change  its 


FIG.  233.— THE  TWO  OBELISKS  OF  KARNAK. 
(Phot.  Beato.) 

125 


FIG.  232.— BUST  OF  A   STATUE  OF 

THE  HYKSOS  PERIOD 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


entire  nature:  but  it  had 
strength  enough  to  burst  the 
ancient  moulds  in  many  direc- 
tions. Indications  of  .Asiatic 
and  European  influences  in 
furniture ,  goldsmith's  work, 
and  pottery  mark  their  advent; 
and  it  is  not  unlikely  that 
we  may  be  obliged  to  enlarge 
very  greatly  the  share  assigned 
to  the  foreigner  in  the  consti- 
tution of  Theban  art  of  the 
second  period,  when  exca- 
vations shall  have  brought  to 
light  the  monuments  of  Meso- 
potamia, Syria,  Asia  Minor, 
and  the  ./Egean  peoples. 

A.  ARCHITECTURE 
We  are  now  no  longer  ob- 
liged   to    judge    of    architec- 
ture by  mere  fragments;   our 
relics  of   the  earliest  periods 


ART  IN  EGYPT 


FIG.    2J4- — RUINS    OF   THE    SECOND   PYLON    OF   HERU- 
EM-HEB  AT  KARNAK.      (Phot.  Beato.) 


are  chiefly  sculpture,  but  thenceforward  architecture  predominates. 
Many   of   the    temples    built   under   the    second   Theban    Empire 

still  exist ,  more 
or  less  complete, 
and  reveal  to  us  its 
conceptions,  plans, 
and  methods  of 
execution.  It  is 
very  possible  that 
the  architecture 
of  this  period 
counted  several 
schools,  but  we  do 
not  yet  know  how 
to  define  them. 
Nearly  all  of  its 
surviving  works 

are  situated  at  Thebes  itself,  and  in  regions  under  the  artistic 
control  of  Thebes,  i.  e.  Southern  Egypt,  and  Ethiopia.  If  any  examples 
of  those  which  embellished  Memphis  and  the  cities  of  the  Delta 
at  this  time  had  come  down  to  us,  we 
should  no  doubt  see  in  them  peculiarities 
which  would  enable  us  to  settle  the 
question  in  one  way  or  another;  but  so 
far,  the  fragments  which  survive  are  not 
sufficiently  characteristic  to  give  us  the 
right  to  say  that  another  school,  distinct 
from  the  Theban  School ,  flourished  in 
the  north. 

Two  elements  seem  to  have  been  in 
common  use  at  this  period ,  which  were 
either  unknown ,  or  very  rarely  used  in 
the  earlier  ages,  the  pylon  with  its 
customary  pair  of  preliminary  obelisks 
(Fig.  233),  and  the  hypostyle  hall.  The 
pylon  '  (Fig.  234)  is  a  straight  monu- 
mental door,  surmounted  by  a  massive 
cornice,  and  enclosed  between  two  rect- 
angular towers  with  sloping  walls.  It 
is  the  face  which  the  temple  turns  to 
the  outer  world ,  and  it  was  through 

it  that  Pharaoh  and  the  faithful  passed  in  state,  when  they  went 
to   enter  into  official  relation  with   the   god.     Each  temple  was 

126 


i 


FIG.  235. — PLAN  OF  THE 
PRINCIPAL  TEMPLE  AT 
KARNAK.  (Phot.  Beato.) 


THE  SECOND  THEBAN  AGE 


supposed  to  have  but  one; 
at  Thebes,  however,  and 
also,  no  doubt,  at  Memphis 
and  other  important  cities, 
the  kings,  anxious  to 
enlarge  the  divine  house, 
constructed  other  pylons 
in  front  of  the  principal 
one,  and  these  were  made 
gradually  wider  and  wider, 
and  higher  and  higher; 
as  the  number  was  not 
limited  by  any  law,  it 
increased  almost  indefin- 
itely, till  it  was  checked 
only  by  the  poverty  or 
insignificance  of  the  sover- 
eign. At  Karnak  there 
are  six  from  west  to  east 
(Fig.  235)  and  four  from 
north  to  south ;  they  were 
separated  from  each  other 
or  from  the  body  of  the 
building  by  a  court  bor- 
dered with  a  portico  on 
three  sides,  north,  east,  and  west.  A  hypostyle  hall  generally 
intervened  between  the  last  of  them  and  the  actual  dwelling 
of  the  god.  It  consisted  of  a  central  nave  upheld  by  two  rows 
of  columns,  and  two  side- 
aisles,  the  number  of  rows 
in  which  was  variable,  two 
and  two  at  Medinet-Habu, 
three  and  three  in  the 
western  Ramesseum,  seven 
and  seven  at  Karnak.  The 
columns  of  the  central 
aisle  (Fig.  236)  are  often 
higher  than  those  of  the 
laterals  (Fig.  238)  and  the 
architect  utilised  the  re- 
sulting difference  of  levels 

in    the   ceilings    to    light       FIG        _ATLANTES  OF  THE  RAMESSEUM  AT 
the   interior;   he   pierced  THEBES.    (Phot.  Beaio.j 

127 


FIG.  236.— CENTRAL  AISLE  OF  THE  HYPOSTYLE 
HALL  IN  THE  RAMESSEUM.   (Phot.  Beato.) 


ART  IN  EGYPT 


FIG.   238.— TRANSVERSE   SECTION   OF  THE  HYPOSTYLE  HALL  AT  KARNAK 
(After  Chipiez)  (Hist,  de  I'Art.  vol.  I,  pi.  V). 


the  vertical  wall  that  united  them  with  a  stone  clerestory,  through 

which  the  light  entered 
(Fig.  239).  The  side -aisles 
were  illuminated  only  by  nar- 
row slits  in  the  ceiling;  as  one 
receded  from  the  centre,  the 
light  diminished,  and  semi- 
darkness  reigned  against  the 
lateral  walls,  even  at  the 
most  brilliant  hours  of 
the  day.  Very  often,  a 
single  chamber  not  being 
considered  enough,  the  archi- 
tect placed  two  or  even 
three  in  a  line,  and  in  this 
case  he  also  doubled  the 
sanctuary,  which  was  then 
composed  of  a  hypostyle 
hall  with  four  columns  for 
the  sacred  boat,  and  beyond 
this,  of  one  or  several 
rooms  where  the  god  re- 
ceived daily  worship.  For 
the  rest  it  would  seem  that 
the  Second  Empire  added 


FIG.  23Q. — CLERESTORY  OF  THE  HYPOSTYLE 
HALL  AT  KARNAK.      (Phot.  Beato.) 

128 


THE  SECOND  THEBAN  AGE 


FIG.  240. — PHOTO-DORIC  COLUMNS  AT 
KARNAK.  (Phot.  Beato.) 


little  to  what  had  been  in  use  under  the  First.  Its  architects 
employed  pillars,  either  bare,  or  with  Atlantes  against  them 
(Fig.  231,  237),  and  hexa- 
gonal columns  of  the 
kind  called  proto- Doric 
(Fig.  240),  at  first  lavishly, 
and  then  more  sparingly, 
though  they  were  never 
dispensed  with  alto- 
gether. They  persisted, 
indeed,  to  a  later  period, 
and  there  are  examples  at 
Elephantine  dating  from 
Amasis,  and  in  the  temple 
of  the  Theban  Ptah,  dating 
from  the  Ptolemies.  On 
the  other  hand,  the 

column  with  a  bell-shaped  capital  became  more  frequent,  and 
vied  in  popularity  with  the  lotus -bud  and  palm -leaf  capitals. 
The  Hathor  column  was  reserved  for  the 
buildings  sacred  to  the  goddess  Hathor  at 
Bubastis  and  at  Der-el-Bahari ;  the  capital 
was  composed  of  two  masks  of  the  goddess 
set  back  to  back  and  encircled  at  the 
neck  by  a  simple  band  where  they  im- 
pinge upon  the  shaft.  Once  only,  in  the 
ambulatory  of  Thothmes  III.  at  Karnak 
(Fig.  241),  a  variation  more  eccentric  than 
ingenious  was  tried,  in  which  the  bell  was 
reversed,  and  the  thinner  end  of  the 
shaft  was  sunk  in  the  base,  while  the 
thicker  one  was  set  into  the  mouth  of  the 
bell.  It  would  seem  that  this  combination, 
in  which  all  the  elements  were  transposed, 
had  no  success,  for  we  find  no  trace  of 
it  elsewhere.  The  three  usual  columns 
are  not  found  indifferently  everywhere; 
the  bell-shaped  capital  was  used  preferably 
for  the  central  aisle  of  hypostyle  halls, 
while  the  lotus-bud  form  was  relegated  to 
the  exterior  porticoes,  the  interior  rooms, 
or  the  side  aisles  of  the  hypostyle  halls,  and  the  palm  column 
reigned  in  the  porticoes.  These  customs  were,  however,  tenden- 

129  K. 


FIG.  241. — ONE  OF  THE 
COLUMNS  OF  THE  AM- 
BULATORY AT  KARNAK 
(After  Chipiez)  (Hist,  de 
1'Art,  vol.  I,  p.  572). 


ART  IN  EGYPT 


ties  rather  than  a  rule;  broadly  speaking,  it  may  be  said  that 
the  lotiform  order  was  the  one  most  in  favour  during  the  Second 

Theban  Empire. 

Some  of  the  temples 
such  as  those  of  Thoth- 
mes  III.  and  of  Amen- 
ophis  III.  at  Elephantine 
contained  only  the  number 
of  rooms  strictly  necessary 
for  the  wants  of  the  god. 
One  of  these  (Fig.  242)  was 
merely  a  sanctuary  of  sand- 
store,  about  14  feet  high, 
by  40  feet  long,  and  30 

FIG.  242.— THE  EAST  TEMPLE  AT  ELEPHAN-     r   ,    -J     TJ.  L  J    L 

TINE  (After  Chipiez'  restoration)  ««t  Wide.      It    had    a  base- 

(Hist  de  1'Art,  Vol.  I,  p.  402).  ment   of   masonry  with  a 

slight  parapet,   sustaining 

a  portico  composed  on  each  side  of  five  square  pillars,  enclosed 
between  two  large  corner-pillars,  and  on  each  facade,  of  two 
columns  with  lotus  capitals.  It  was  entered  on  the  east,  where  a 
flight  of  ten  or  twelve  steps  mounted  to  the  portico  and  to 
the  cella  between  the  two  columns ;  another  door  opened  at  the 
western  extremity  (Fig.  243).  It  was  a  peripteral  temple,  and  the 
Pharaohs  of  the  Eighteenth  Dynasty  had  a  certain  predilection 
for  the  plan,  for  it  recurs  at  Karnak  and  at  Medinet - Habu. 
That  of  Medinet-Habu,  almost  identical  with  that  of  Elephantine 
in  dimensions,  was  founded  by  Thothmes  II.  and  Queen  Hat- 
shepset,  but  it  would  seem  that  before  its  completion,  a  second 
building  was  added  towards  the  west  —  perhaps  by  Thothmes  III. 

—  consisting  of  six 
rooms  arranged  in 
three  rows  of  two  each, 
the  sanctuary  at  the 
end,  and  the  chapels 
of  the  paredri  at  the 
sides.  Construction 
and  decoration  betray 
negligence,  or  rather 
lack  of  skill,  on  every 
hand,  and  this  is  not 
surprising,  if  we  re- 

member  that  at  the  time  Aat-tcha-Mutet  —  our  Medinet-Habu  — 
was  a  little  provincial  town.  To  give  but  one  instance,  the  slabs 

130 


FIG.  243. — LONGITUDINAL,  SECTION  OF  THE 
EAST  TEMPLE: 


THE  SECOND  THEBAN  AGE 


FIG.  244. — RUINED  FACADE  OF  THE  TEMPLE  OF 
AMADA.  (Phot.  Oropesa.) 


of  the  roof  are  so  badly  adjusted,  that  it  was  found  necessary 
to  prop  up  those  of  the  portico  by  means  of  columns  placed 
at  the  angles,  regard- 
less of  symmetry.  They 
seem  to  be  there  more 
or  less  by  accident, 
and  yet  they  harmon- 
ise so  well  with  the 
whole,  that  their  pre- 
sence does  not  shock 
the  spectator;  they 
appear  as  a  singu- 
larity, or  a  graceful 
audacity,  rather  than 
a  constructive  error. 
The  temple  of  Amen- 
bphis  III  at  El-Kab, 
almost  as  simple  as 
those  of  Elephantine 

and  Medinet-Habu,  is  on  a  different  plan.  It  has  two  compart- 
ments at  present,  but  the  first,  which  is  a  portico,  was  built 
under  the  Ptolemies,  and  formed  no  part  of  the  original  arrange- 
ment; only  the  room  of  the  sacred  boat,  the  sanctuary,  dates 
from  the  Theban  period.  It  is  oblong  in  shape,  and  is  sustained 
by  four  Hathor-pillars ;  a  niche,  which  was  approached  by  four 
steps,  is  hollowed  out 
in  the  end  chamber, 
and  this  was  the  Holy 
of  Holies,  the  retreat 
in  which  the  divine 
statue  was  concealed. 
In  general ,  we  note 
two  distinct  types  for 
the  most  simple  form, 
the  oratory,  under  the 
Eighteenth  Dynasty : 
that  of  the  single 
chamber  with  or  with- 
out columns  in  the 
interior,  and  that  of 
the  peripteral  temple, 
which,  though  it  does  not  lend  itself  to  scientific  combinations, 
may,  if  judiciously  treated,  produce  true  masterpieces.  The 

131  K    2 


FIG.  245. — COURT  OF  THE  TEMPLE  OF 
RAMESES  HI.  AT  KARNAK.  (Phot.  Legrain.) 


ART  IN  EGYPT 


FIG.  246. — TEMPLE  OF  KHONSTT  AT  THEBES'; 
SECTION  CUT  THROUGH  ITS  GRAND  AXIS 
(After  Chipiez)  (Hist,  de  1'Art,  vol.  I,  p.  355). 


chapel  of  Khnum  at  Elephantine  was  certainly  the  most  finished 
example  of  the  latter:    the  relative  proportion   of  the  parts  was 

calculated  so  scientifically 
that  the  artists  of  the 
French  expedition  never 
wearied  in  their  admi- 
ration of  its  perfection. 
It  is  a  surprise  to  those 
who  are  accustomed  to 
consider  Egyptian  archi- 
tecture a  massive  and 
colossal  art,  to  find  it 
producing  works  posi- 
tively Greek  in  their 
precision  and  elegance. 
There  is  reason  to  be- 
lieve that  this  peripteral  form  was  unknown  in  the  first  Theban 
period,  and  that  it  was  invented,  or  at  least  brought  to  per- 
fection, at  the  beginning  of  the  second.  Almost  at  the  same  time 
there  appeared  a  more  developed,  though  as  yet  restricted  model, 

which  I  will  call  the  temple  of 
the  small  town.  That  of  Amada, 
which  dates  from  the  time  of 
Thothmes  III.  and  Amenophis  II., 
consists  of  three  long  parallel 
ducts  (Fig.  244^,  in  the  centre  the 
sanctuary  of  Amon-Ra  and  Ra- 
Harmachis,  and  on  either  side  two 
little  rooms  in  a  line.  Originally 
these  were  disconnected,  and  access 
to  the  further  of  the  two  could 
only  be  obtained  from  the  Holy 
of  Holies,  but  later,  doors  were 
pierced  between  the  partition-walls, 
and  the  rooms  were  made  to  com- 
municate. The  three  aisles  lead 
into  a  transverse  vestibule  occupy- 
ing the  entire  breadth  of  the  build- 
ing, and  preceded  by  a  portico 
with  four  proto- Doric  columns. 
Ending  here,  the  temple  was  com- 
plete, but  Thothmes  IV.,  the  successor  of  Amenophis  II.,  inter- 
posed, between  the  portico  and  the  brick  enclosing  wall,  a 

132 


FIG.  247. — COLOSSI  IN  FRONT 

OF    THE    TEMPLE    OF  LUXOR. 

(Phot.  Beato.) 


THE  SECOND  THEBAN  AGE 


L 


FIG.  248.— NORTH   SIDE   OF    THE    AVENUE    OF    RAMS 

AT  KARNAK   DURING   THE    INUNDATION. 

(Phot.  Legrain.) 


hypostyle  hall  of  twelve  square  pillars  in  four  rows,  the  last 
two  of  which,  right  and  left,  were  connected  by  party  walls; 
later  again,  Seti  I.  re- 
placed the  plain  wall 
on  which  the  hall 
abutted  towards  the 
east,  by  a  composite 
pylon ,  consisting  of 
a  sandstone  gateway 
between  two  brick 
towers.  Even  with  all 
these  additions,  Ama- 
da  is  very  small,  for 
it  measures  barely 
30  feet  in  width  by 
72  feet  in  depth,  and 
the  height  is  about 
15  feet,  but  the  execution  is  very  careful,  and  does  credit  to 
the  provincial  Theban  art  of  the  Eighteenth  Dynasty.  The  blocks 
are  accurately  adjusted,  the  sculpture  is  delicate,  and  the  paint- 
ing brilliant;  the  brush  has  accentuated  the  work  of  the  chisel, 
and  has  expressed  the  details  of  figures  and  hieroglyphics  with 
great  elaboration.  The  temple  of  Ptah  at  Karnak,  built  by  the 
the  Amenemhats 
and  reconstructed 
by  Thothmes  HI., 
was  rebuilt  so  ex- 
tensively under  the 
Ptolemies  that  it 
would  be  impru- 
dent to  insist  upon 
its  original  form; 
I  think ,  however, 
that  the  arrange- 
ment must  have 
resembled  that  of 
Amada.  On  the 
other  hand ,  the 
temple  of  Rame- 
ses  III.  at  Karnak 


of 


a   more 


de- 


FIG.  240.— COURT  OF  AMENOPHIS  IH.  AT  LUXOR  SEEN 
FROM  THE  NORTH  EAST.   (Phot.  Beato.) 


veloped  type.    The  chevet  is  here  divided  into  three  compartments 
with  the  sanctuary  in  the  middle,  and  just  as  at  Amada,  the  two 

133 


ART  IN  EGYPT 


FIG.  250. — SOUTH-EAST  OF  THE  TEMPLE  OF  AMENO- 
PHIS  III.  AT  LUXOR.   (Phot.  Beato.) 


rooms  which  terminate  the  wings  open   only  into  the  sanctuary, 
but  the  remaining  space  contains,  besides  the  chapels  dedicated 

to  the  goddess  and 
the  divine  son, 
serving-rooms,  one 
of  which,  that  on 
the  west ,  does 
duty  as  the  cage 
of  the  staircase 
which  led  to  the 
terraces.  In  ad- 
dition ,  the  trans- 
verse vestibule  of 
Amada  has  be- 
come a  hypostyle 
hall  with  two  rows 
of  columns,  and  the 
pronaos  is  arrang- 
ed, as  is  also  the  pylon,  on  a  new  plan  which  was  applied  on 
a  larger  scale  at  Medinet  -  Habu.  It  is  on  a  higher  level  than 
that  of  the  court,  and  is  reached  from  the  latter  by  an  inclined 

plane;  towards  the 
east  and  the  west 
it  adjoins  porticoes 
which  terminate 
against  the  pylon, 
and  colossal  statues 
of  the  king  as  Osi- 
ris are  set  against 
the  pillars  which 
border  these  porti- 
coes (Fig.  245).  It 
is  permissible  to 
suppose  that  the 
temple  of  Mentu, 
built  by  Ameno- 
phis  III.,  was  simi- 
lar in  arrangement, 

GREAT  COLONNADE   OF  HERU-EM-HEB   AND   SETI  I.  "UI  tnCSC  TUIHS  IiaVC 

AT  THE   TEMPLE   OF  LUXOR.  not    yet    been     Suffi- 

ciently    studied   to 

justify   an   assertion.     What  we  may,   however,   affirm   without 
rashness   is   that   the   temple   of   the   small   town,   as  we  see  it 

134 


THE  SECOND  THEBAN  AGE 


FIG.  252.— SOUTH-WEST  ANGLE  OF  THE 

COURT  OF  RAMESES  II.  AT  THE  TEMPLE 

OF  LUXOR.  (Phot.  Beato.) 


at   Amada,     was    simply    a    reduction    of    the    temple    of    the 
great    city,     and    that    the    arrangements    were    in     the    main 
the   same  in  both.     The 
sanctuary    was     at     the 
end,    against    the    back 
wall,  between  two  rooms, 
or  two   series   of  rooms, 
the      dwellings     of     the 
other  gods  of  the  triad, 
used    for    the    secondary 
services  of  worship.     A 
vestibule  extending  right 
across  the  building  divid- 
ed these   intimate  apart- 
ments    from     those     re- 
gions   accessible    to    the 
public ,     hypostyle    halls, 
courts,  monumental  gate- 
ways flanked  by  towers; 
obelisks   or  a  guard    of   sphinxes  rose   on   the   terrace  in  front. 
Well-preserved  examples  of  this  type  are  so  rare  that  we  cannot 
exactly  follow  its  evolution  be- 
tween the  Eighteenth  and  Nine- 
teenth Dynasties.    It  culminated, 
towards   the   end   of  the  Nine- 
teenth Dynasty,  in  a  conception 
of  which  the  temple  of  Khonsu 
at  Thebes  is  the  most  lucid  and 
complete  realisation.   (Fig.  246). 
The    distinction    between     the 
private    dwelling    of    the    god, 
and    the    space    open    to    the 
public  is  clearly   defined.    The 
one  is  separated  from  the  other 
by  a  wall  in  which   two   doors 
are  pierced;    the   first,   on  the 
longitudinal   axis,   was   a  state 
portal,   for  solemn  ceremonies, 
when  Khonsu   came  out  of  his 
sanctuary,   and   for   the  official 
visits   of  the  Pharaohs;   the  se- 
cond, placed  towards  the  western  extremity,  was  the  household 
postern,     by    which   the   priest   came  and  went  every  day,  and 

135 


FIG.  253.— ONE   OF  THE   CHAPELS   OF 
GEBEL-SILSILEH.     (Phot.   Thedenat.) 


ART  IN  EGYPT 


FIG.  254. 

PLAN    OF  THE   SPEOS  OF  HERU- 
EM-HEB  AT  SILSLLEH. 


was  also  used  by  the  sovereign  when  he  visited  the  god  in- 
formally. Beyond  this  barrier,  we  find  the  tripartite  arrangement 
I  have  described  above;  in  the  middle,  the  shrine  of  Khonsu, 

and  on  either  side,  the  chapels 
of  the  paredri,  then  the  serving- 
rooms,  but  with  new  combina- 
tions. In  the  temple  of  Ra- 
meses  III.  at  Karnak,  the  Holy 
of  Holies  was  a  single  chamber, 
in  which  not  only  the  idol  was 
enclosed,  but  also  the  bari  on 
which  the  idol  was  seated  when 
it  left  its  retreat  to  show  itself 
in  public.  In  the  temple  of 
Khonsu,  it  consisted  of  three 
chambers  in  a  line  on  the  axis:  first,  against  the  end  wall,  a 
dark  cabinet  which  was  the  mysterious  residence  of  the  master, 
then,  in  front  of  this,  an  anteroom  with  four  columns,  and 
finally,  in  front  of  the  anteroom,  a  vast  hall  in  the  centre  of 
which  rose  the  pink  granite  cell  which  contained  the  ark.  This 
cell  had  a  back  door  by  which  the  image  was  brought  out  on 
specified  days  for  embarkation,  and  a  front  door  from  which  it 
emerged  in  state.  The  chambers  of  the  side-aisles  communicated 
with  one  or  the  other  of  these  three  chambers,  according  to 
the  use  for  which  they  were  destined,  those  of  the  paredri 
with  the  ante-room,  the  others  with  the  shrine  of  the  boat. 
The  staircase  which  led  to  the  terraces  was  concealed  on  the 

right,    in    the  angle  form- 


ed by  the  exterior  east 
wall,  and  the  interior  par- 
tition wall.  Beyond  this, 
the  public  parts  of  the  build- 
ing began,  and,  in  the 
first  place,  the  hypostyle 
hall  which  traversed  it  from 
east  to  west.  The  central 
aisle  was  defined  by  four 
columns  with  bell -shaped 
capitals  23  feet  high,  and 
the  wings  contain  two  lotus 
columns  18  feet  high;  the  light  is  furnished,  as  at  Karnak,  by 
a  clerestory  between  the  terrace  of  the  central  aisle,  and  the 
lateral  platforms.  The  pronaos  is  supported  on  twelve  columns 

136 


FIG.  255- 

FACADE    OF    THE   SPEOS    OF  HERU- 
EM-HEB.     (Phot.   Thedenat.) 


THE  SECOND  THEBAN  AGE 


in  two  rows;  an  inclined  way  descends  from  it  into  the  court, 
which  is  bordered  south,  east,  and  wes  by  a  double  row  of  lotus  col- 
umns. Access  was 
freely  accorded  by 
four  lateral  pos- 
terns ,  and  by  a 
pylon  measuring 
104  feet  long, 
60  high  and  33  wide. 
It  is  solid,  save 
for  the  staircase, 
which  runs  straight 
from  the  north- 
eastern corner  of 
the  block  to  the  plat- 
form over  the  door 
and  thence  to  the 
summits  of  the  two 
towers.  The  facade  FIG.  256.— FACADE  OF  THE  HEMI-SPEOS  OF 

was    grooved    with  BET-EL-WALI.    (Phot.  Oropesa.) 

four  cavities  to  hold 

the  masts  for  pennons.  A  pair  of  obelisks,  and  colossal  statues  rose 
in  front,  their  backs  to  the  pylon,  their  faces  to  the  city  (Fig.  247), 
often  precedel  by  long  avenues  of  sphinxes  or  rams  (Fig.  248),  and  all 
these  protected  the 
god  against  evil  in- 
fluences. It  is  pro- 
bable that  the 
majority  of  the 
Ramesside  temples 
were  built  on  this 
plan  with  slight  var- 
iations ;  it  per- 
sisted during  the 
centuries  which 
followed  the  fall 
of  the  Second  The- 
ban  Empire ,  and 
in  its  main  lines, 
to  the  end  of  the 
pagan  period. 

In  addition  to  these  regular  buildings,  the  constituents  of  which 
were    brought   together   more  or  less  on  fixed  principles,    there 

137 


FIG.   257.— AVENUE   OF    SPHINXES  AT  WAD!  SABU'A. 
(Phot.  Oropesa.) 


ART  IN  EGYPT 


were  some  at  Thebes,  and  no  doubt  in  other  great  cities  of 
Egypt,  the  arrangement  of  which  does  not  agree  with  any  recog- 
nised type,  Luxor,  as  conceived  by  Amenophis  III.,  is  inspired 
by  the  same  idea  as  the  temple  of  Khonsu.  It  was  to  have  a 


FIG.  258. 
SECTION  OF  THE  HEMI-SPEOS  OF  GARF-HUSEN   (After  Gau). 

pylon  turned  to  the  north-east,  a  court  with  porticoes,  at  the 
end  of  which  was  the  pronaos  with  its  eight  rows  of  lotus  columns 
(Fig.  249),  (almost  a  hypostyle  hall  left  open  in  front),  then  behind 
this  pronaos,  the  true  hypostyle  hall,  which  has  lost  its  columns, 
and  no  longer  extends  right  across  the  building;  it  is  flanked 
right  and  left  by  dark  rooms  used  for  the  most  part  as  auxiliary 
chapels.  As  usual,  the  hypostyle  hall  and  its  annexes  terminated 
the  public  part  of  the  temple,  and  a  wall,  pierced  with  a  state 
doorway  and  a  service-postern,  separated  them  from  the  actual 
abode  of  Amon.  This  comprised  in  its  axis  two  rooms  with  four 
columns  each,  the  farther  one  of  which  contained  the  sacred  ark, 
then  a  second  hypostyle  hall,  and  against  the  back  wall,  a  final 

room  with  columns,  which 
was  the  sanctuary.  Right 
and  left  of  this  row  of 
apartments  were  succes- 
sive chapels,  in  one  of 
which,  on  the  east,  the 
marriage  of  Queen  Mut- 
emua  with  Amon,  and  the 
birth  of  Amenophis  III 
were  described  and  pic- 
tured; along  the  east  and 
west  walls  little  rooms, 
or  rather  closets,  were  ranged,  the  uses  of  which  are  not  cer- 
tainly known,  but  in  which  it  is  probable  that  clothes,  jewels, 
perfumes,  furniture,  and  gold  and  silver  plate  were  stored 
(Fig.  250).  The  building  was  almost  finished  when  the  king, 

138 


FIG.  259.— FACADE  OF   THE    LITTLE    SPECS    AT 
ABU  SIMBEL.     (Phot.  Oropesa.) 


THE  SECOND  THEBAN  AGE 


FIG.  26O. — FACADE   OF  THE   GREAT   SPEOS   OF 
ABU   SIMBEL.     (Phot.   Oropesa.) 


modifying  the  design,  replaced  the  pylon  by  a  thick  wall  and 
laid  on  the  north  the  foundations  of  another  hypostyle  hall,  which, 
had  it  been  com- 
pleted, would  have 
had  no  parallel.  Only 
the  central  aisle  with 
its  columns  52  feet 
high  was  erected 
(Fig.  251),  and  the  dis- 
turbances in  the  reign 
of  Amenophis  IV. 
compelled  the  archi- 
tect to  stop  the  decor- 
ation. When  it  was 
resumed,  the  course 
of  the  Nile  had  de- 
viated eastward,  and 
Heru  -  em  -  heb  was 
obliged  to  deflect  the 

main  axis  to  find  room  for  the  new  court  (Fig.  252)  and  for 
a  pylon  which  Rameses  II. 
finished  and  faced  with  sculp- 
tures. Karnak  shows  more 
irregularity  and  incoherence 
even  than  Luxor,  and  this 
is  not  surprising,  when  we 
remember  that  all  the  Theban 
Pharaohs  from  the  Seven- 
teenth to  the  Twentieth 
Dynasty  vied  with  each  other 
in  enlarging  it  without  any 
definite  plan.  A  big  book 
would  not  be  too  much  to 
devote  to  its  history,  and 
even  this  could  not  be  com- 
plete, for  lack  of  evidences 
bearing  on  the  earlier  periods. 
The  original  building,  that 
of  the  Twelfth  Dynasty,  has 
disappeared,  and  we  do  not 
know  what  were  its  main 
features.  The  Ahmessids, 
from  Amenophis  I.  to  Thoth- 


FIG.  26l.—  NORTHERN  EXTREMITY  OF 
THE  ESPLANADE  BEFORE  THE  GREAT 
SPEOS  OF  ABU  SIMBEL.  (Phot.  Oropesa), 

139 


ART  IN  EGYPT 


FIG.  262. 

THE  PYRAMID-MASTABA  OF  AN  APIS  AT 
SAKKARAH  (After  Mariette). 

old  buildings,  the  most  imposing  of 
which,  an  audience-chamber,  bears  the 
traditional,  but  inaccurate,  name  of 
ambulatory;  he  then  enclosed  the  whole 
with  a  stone  wall,  dug  out  the  lake 
on  the  south,  and,  anxious  to  provide 
a  triumphal  entrance  for  the  god,  erected 
two  enormous  pylons  on  the  Luxor 
road,  to  which  Heru-em-heb  soon  added 
two  others.  Thothmes  IV.  and  Amen- 
ophis  III.  erected  a  still  more  massive 
pylon  in  front  of  those  on  the  west, 


mes III. .surrounded it  wit! 
buildings  which  in  some 
ways  repeat  the  combi- 
nations at  Luxor,  with  a 
room  for  the  boat  in 
the  centre,  and  auxiliary 
chapels  in  the  side  aisles, 
but  also  with  a  perfect- 
ly novel  element,  three 
pylons  rising  one  behind 
the  other  from  east 
to  west.  Thothmes  III., 
having  reached  this  point, 
returned  to  the  east,  and 
there  re-constructed  some 


FIG.   263.— A  THEBAN  TOMB 
"WITH  A  PYRAMIDAL  SUMMIT. 


FIG.  264.— TOMB    OF    THOTHMES   HI. 

140 


which  Rameses  I.  preceded 
by  another  yet  more  gi- 
gantic: between  the  two 
he  built  the  famous  hypo- 
style  hall  which  Seti  I., 
Rameses  II. ,  and  the  Ra- 
meses of  the  Twentieth 
Dynasty  finished  deco- 
rating. Karnak  is  not,  , 
strictly  speaking,  a  single 
temple;  it  is  a  haphazard 
mass  of  temples  and  store- 
houses (cf.  Fig.  172).  It 
must  be  looked  upon  in 


THE  SECOND  THEBAN  AGE 


FIG.   265.— DECORATION   OF  THE  END  WALL  IN  THE   TOMB-CHAMBER  OF 
RAMESES  V.      (Phot.   Golenischeff.) 

the  history  of  Egyptian  art,  not  as  a  normal  creature,  long  con- 
sidered, and  produced  on  a  preconceived  plan  but  rather  as 
a  marvellous  monster, 
whose  limbs  are  grafted 
on  to  the  original  body 
fortuitously ,  regardless 
of  logic  and  symmetry. 
Taken  in  detail,  the 
parts  are  often  admirable 
in  execution;  when  we 
attempt  to  coordinate 
them,  we  find  it  im- 
possible to  reduce  them 
to  unity. 

With  the   ideas  which 


FIG.  266. — PLAN  OF  THE    HVPOGEUM  OF 
AMENOr-HIS    II. 


prevailed  in  Egypt  on 
the  nature  of  the  temple 
and  the  tomb,  it  was 

inevitable   that  sooner   or   later   it  would   be   proposed    to  instal 
the  house  of  the  god  in  the  rock.    We  have  as  yet  no    authority 

141 


ART  IN  EGYPT 


FIG.   267. — GENERAL  VIEW  OF  DER-EL-BAHAKI. 
(Phot.  Beato.) 


for  saying  whether 
this  came  about  un- 
der the  Memphite 
or  the  first  The- 
ban  Empire ;  the 
most  ancient  sub- 
terranean temples 
known  to  us,  speos 
and  hemi- speos, 
date  from  the 
Eighteenth  Dynas- 
ty. Queen  Hat- 
shepset  had  a  vestb 
bule  with  eight 
pillars,  a  passage 
and  an  inner  cham- 
ber, which  was  the 
sanctuary ,  cut  in 
the  rock  near  Beni- 
Hasan ,  in  honour  of  the  lioness-goddess  Pekhet ;  two  centuries 
later  Seti  I.  hollowed  the  chapel  of  Redesiyeh  on  the  road  to 
the  gold  mines.  When  we  examine  these  carefully,  we  find 
that  the  architect  took  the  isolated  temple  of  the  small  town, 
and  imbedded  it  in  the  mountain.  Occasionally  it  is  only  a  single 

apartment,  with  a 
fagade  set  between 
columns,  as  at  Sil- 
sileh  (Fig.  253), 
but  the  type  of 
El  -  Kab  prevailed 
in  general ,  and  if 
the  door-way  con- 
necting vestibule 
and  sanctuary  was 
elongated,  and 
transformed  into 
a  passage,  it  was 
partly  because  the 
safety  of  the  faith- 
ful required  a  stone 
partition  more  so- 

FIG.  268.— NORTH-WEST    ANGLE   OF  THE  ESPLANADE  ,,          -      * 

AT  DER-EL-BAHARI.    (Phot.  Beato.)  nary  wall   or    ma- 

142 


THE  SECOND  THEBAN  AGE 


sonry.  Indeed,  when  we  compare  the  cavern -temple  in  general 
with  the  disengaged  temple,  we  must  recognise  that  the  arrange- 
ment of  the  two  is  in  the  main  

identical;  the  differences  are  the 
result  of  special  conditions  which 
the  new  surroundings  imposed  on 
the  architect,  and  are  not  more 
marked  than  those  which  distinguish 
the  free  mastaba  from  the  sepulchral 
chapel  hollowed  in  the  rock.  The 
taste  for  the  specs  developed  towards 
the  end  of  the  Eighteenth  Dynasty, 
and  two,  not  the  least  interesting 
among  them,  were  the  work  of 
Heru-em-heb.  At  Silsileh  (Fig.  255) 
the  specs  is  a  long  gallery  sup- 
ported by  four  massive  pillars  left  in 
the  rock,  with  the  sanctuary  adjoining 
it  at  right  angles  (Fig.  254).  Aba- 
huda,  a  little  to  the  north  of  the 
Second  Cataract,  has  no  true  fagade, 
but  a  portion  of  the  cliff  was  planed 
vertically,  a  few  steps  were  cut  in 

front,  and  a  high,  narrow  door,  hardly  more  than  a  slit,  was 
pierced  in  the  rock.  The  hypostyle  hall,  supported  by  four  poly- 
gonal columns,  leads  to  the  three  usual  chapels;  these,  however, 
instead  of  being  arranged 
in  a  line,  parallel  one  with 
another,  are  placed  on 
the  three  sides,  the  sanc- 
tuary at  the  end,  facing 
the  entrance,  the  cham- 
bers of  the  mother  and 
son  right  and  left  of  the 
hypostyle  hall.  Rameses  II 
showed  a  special  pre- 
ference for  this  type  of 
building,  and  Nubia  is 


FIG.   269. 

PLAN    OF    THE    MEMNONIUM   OF 
SETI  I.  AT  ABYDOS. 


FIG.  270. — ONE  OF  THE    HYPOSTYLE    HALLS    IN 
THE  MEMNONIUM  OF  SETI  I.     (Phot.  Beato.J 


full    of   those    which    he 
dedicated    ostensibly    to 
his  father  Amon,  but  in 
reality   to  his  own   divinity.     The   oldest   and   the   most  elegant 
of  all  these,  the  hemi-speos  of  Bet-el -Wali,  has  a  deep  vestibule, 

143 


ART  IN  EGYPT 


FIG.  271.— FA9ADE  OF  A   SEPULCHRAL  TEMPLE  AT  KURNAH.      (Phot.  Beato.) 

suggested  on  the  facade  by  two  square  pillars,  and  covered 
with  a  roof  not  cut  in  the  rock,  but  vaulted  with  bricks 
(Fig.  256).  Three  doors  —  those  on  the  sides  are  later  than 
the  central  one  —  lead  to  the  transverse  vestibule,  where  two 
rather  squat  proto-Doric  pillars  have  been  cut  out  in  the  rock; 
the  Holy  of  Holies  contains  three  statues  which  represent  the 
three  gods  of  the  local  triad.  At  Wadi-es-Sabu'a,  at  Der,  at 
Garf-Husen,  the  excavation  and  its  outworks  of  masonry  attained 
the  dimensions  of  the  isolated  temple  of  a  large  town.  The  pro- 
pylaea  of  Sabu'a,  recently  exhumed,  form  a  magnificent  array 
of  colossal  figures  and  sphinxes  with  human  faces  or  falcons' 
heads  (Fig.  257).  Garf-Husen  possessed  a  sanctuary,  two  hy- 

postyle  halls,  the  larger 
upheld  by  pillars  adorned 
withAtlantes,  a  court  with 
porticoes  of  the  same  type 
as  that  ot  the  Ramesseum, 
a  pylon,  courts,  and 
an  avenue  of  sphinxes 
(Fig.  258).  The  little  speos 
of  Abu  Simbel  is  less 
complex  in  design.  Its 
fagade  towards  the  river 
is  decorated  with  six 
colossal  standing  figures 
in  niches,  four  for  Ra- 

meses  II ,  two  for  his  wife  Nefert-ari  (Fig.  259).  The  hypostyle 
hall  has  six  polygonal  pillars,  on  the  summits  of  which  heads 

144 


FIG.  272. — GENERAL  VIEW  OF   THE  RAMESSEUM 
OF  THEBES.     (Phot.  Beato.) 


THE  SECOND  THEBAN  AGE 


FIG.   273.— STOREHOUSES    WITH    BRICK    VAULTS    IN 
THE  RAMESSEUM.      (Phot.  Baraize.) 


of  Hathor  are  placed  instead  of  capitals.  It  communicates  with 
the  vestibule  by  three  doors,  and  three  chapels  are  connected 
with  the  vestibule, 
the  sanctuary  in  the 
centre,  facing  the  en- 
trance, the  other  two 
at  the  two  extremi- 
ties. The  large  adjoin- 
ing speos  (Fig.  260) 
is  a  complete  temple, 
built  in  the  spirit 
which  governed  the 
plan  of  the  isolated 
temples,  and  contain- 
ing all  the  consti- 
tuent parts  of  these. 
First  of  all  there  is  an 
esplanade  of  beaten 
earth;  a  short  flight 

of  steps  connects  it  with  a  terrace,  bordered  by  a  solid  balus- 
trade, behind  which  rose  in  a  single  line  twenty  figures  of  alter- 
nate Osiris-mummies  and  falcons  (Fig.  261),  eight  to  the  right 
and  eight  to  the  left  of  the  central  landing.  Behind  this  line, 
the  slanting  pylon, 
cut  in  the  rock, 
presents  its  vast 
surface ,  the  four 
prescribed  colos- 
sal statues  watch- 
ing impassibly 
along  it.  Beyond 
the  pylon,  in  the 
place  of  the  co- 
vered court,  was 
a  hall  130  feet 
long ,  bordered, 
like  the  court 
of  Rameses  III.  at 
Karnak,  by  eight 
square  pillars,  each 
with  an  Osiris  set 

against     it.    This    sort    of    covered    yard    was    followed    by    the 
hypostyle  hall,  and,  at  the  end  of  this  was  the  sanctuary  between 

145  L 


FIG.  274.  — SECOND  COURT  OF  THE  TEMPLE  OF 
MEDINET-HABU.  (Phot.  Beato.) 


ART  IN  EGYPT 


FIG.  275.— FACADE  OF  THE  PALACE  OF  RAMESES  III. 

TOWARDS  THE  FIRST  COURT  OF  THE  TEMPLE  AT 

MEDINET-HABU.  (Phot.  Beato.) 


the  cells  of  the  paredri.  Eight  crypts,  on  a  lower  level  than  the 
central  nave ,  were  distributed  unequally  on  either  side ,  simu- 
lating the  accessory 
chambers.  The  dif- 
ferences and  inequa- 
lities of  the  arrange- 
ment are  explained 
by  the  necessity 
imposed  upon  the 
builder  of  choosing 
the  most  solid  strata 
in  the  stone,  and 
of  making  sure  that 
his  work  should  not 
be  crushed  by  the 
mountain. 

If  the  isolated  tem- 
ple thus  buried  itself 
in    imitation    of    the 
old    sepulchral    cha- 
pels,  these,   by   an  inverse  phenomenon,   were   often   detached 
from  the  hypogea  to  which  they  belonged,  and  became  isolated 

temples.  Tombs  of  private  per- 
sons were  of  two  sorts,  as  in 
the  first  Theban  period;  one, 
hollowed  out  entirely  in  the 
cliff,  the  other  in  the  manner 
of  pyramid-mastabas ,  but  with 
important  modifications  in  the 
relative  importance  of  the  pyra- 
mid and  the  mastaba.  The  latter, 
which  at  first  had  been  gradually 
decreased  till  it  became  merely 
an  insignificant  base,  steadily 
grew  until  it  almost  recovered 
its  original  size,  while  the  pyra- 
mid shrank  to  the  dimensions 
of  the  pyramidium  on  an  obelisk. 
There  is  only,  as  far  as  I  know, 
a  single  specimen  of  the  kind, 
the  chapel  of  Apis  discovered 
by  Mariette  in  the  Serapeum  sixty  years  ago  (Fig.  262).  The 
mastaba  is  still  in  existence,  a  chamber  of  masonry  perched  on 

146 


•HB 


FIG.  276. — DECORATION   OF  THE 

CEILING  AND  ONE  OF  THE  ROOMS 

OF  THE  PALACE  OF  AMENOPHIS  III. 

AT  MEDINET-HABU. 


THE  SECOND  THEBAN  AGE 


FIG.  277. 

PAVEMENT  OF  ONE  OF  THE  HALLS  OF  THE  PALACE 
OF  AMENOPHIS  IV.  AT  EL-AMARNA. 


a  solid  basement,  adorned  on  the  outside,  towards  the  corners, 
with  polygonal  engaged  columns ,  and  crowned  with  a  cavetto. 
The  pyramidium 
has  disappeared 
almost  entirely, 
and  the  vault  is 
under  the  building, 
but  independent 
of  it;  it  is  ap- 
proached by  an  in- 
clined plane  which 
descends  into  the 
ground  a  little 
way  in  front  of 
the  door  of  the 
mastaba.  Monu- 
ments of  this  kind 
abounded  at  The- 
bes, but  they  have 

all  been  destroyed,  and  we  should  not  suspect  their  existence  were 
they  not  frequently  represented  in  paintings.  The  pyramidium 
was  more  or  less  pointed,  and  it  was  built  of  brick;  a  gable- 
window  was  occasionally  pierced  in  it  which  gave  light  to  the 
interior,  and  it  terminated  in  a  point  of  black  stone,  either 
granite  or  schist  (Fig.  263).  The  hypogea  properly  so-called, 
with  which  the  Theban  mountains  are  riddled,  so  to  say,  still 
followed  the  tradition  of 
the  Twelfth  Dynasty  in 
so  far  as  to  retain  the 
hypostyle  hall  behind  the 
fagade,  but  the  available 
space  was  restricted,  and 
in  order  to  economise 
this,  a  less  ambitious  plan 
was  adopted  from  the 
beginning  of  the  Eigh- 
teenth Dynasty.  It  shows 
generally  an  open  court, 
roughly  quarried  in  the 
hillside,  where  the  pre- 
liminary rites  of  burial 
were  performed ,  then  a  long ,  narrow  ante-room ,  to  the  end 
of  which  the  stele  was  often  relegated;  its  decoration  included 

147  L  2 


FIG.  278. 

DETAIL  OF   A  PAVEMENT  IN  THE  PALACE  OF 
AMENOPHIS  III.  AT  MEDINET-HABU. 


ART  IN  EGYPT 


FIG.  27Q.— THE  MIGDOL   AT   MEDIXET-HABU. 
(Phot.   Thedenat.) 


representations  of  the 
various  scenes  of  burial, 
the  funeral  banquet,  the 
music,  the  dances,  even 
the  fishing,  hunting,  and 
agricultural  labours  which 
ensured  the  nourishment 
of  the  deceased.  On  rare 
occasions,  the  form  of  a 
Greek  cross  was  admitted 
for  the  arrangement  of 
the  chapel ;  but  nearly 
always  it  was  merely  an 
oblong  cell,  or  even  a 
blind  alley,  at  the  end 
of  which  the  deceased  and 
his  family  sat,  carved 
out  in  the  rock.  The  vault 
was  concealed  somewhere 
below,  sometimes  in  one  place,  sometimes  in  another;  it  was 

approached  either  by  a  perpen- 
dicular shaft,  a  passage,  or  a  steep 
staircase.  Large  fortunes  were 
evidently  rarer  at  Thebes  than  they 
had  formerly  been  at  Memphis, 
but  competences  abounded,  and 
their  possessors  peopled  the  ceme- 
teries of  Asasif ,  Sheikh  -  Abd- 
el-Kurnah,  Der-el-Medinet,  and 
Kurnet-Murrai ;  nearly  all  the  gaily 
painted  and  delicately  carved  hy- 
pogea  which  visitors  admire  in 
these  places  were  the  work  of  artists 
paid  by  these  people. 

The  tombs  of  the  first  Pharaohs 
of  the  Eighteenth  Dynasty,  Kames, 
Aahmes,  Amenophis  I.,  were  pyra- 
mid-mastabas  of  the  same  kind  as 
those  of  Menthu-hetep  at  Der-el- 
Bahari ,  but  situated  near  Drah- 
abu'l-Nekkah ,  on  the  boundaries 
of  the  cultivated  land ;  with  Thoth- 
mes  I.  the  conception  changed,  and 
148 


FIG.  28o. — SKETCH  IN  BLACK  ON 
ONE  OF  THE  PILLARS  OF  THE  HYPO- 
GEUM  OF  SETI  I.  (Phot.  Insinger.) 


THE  SECOND  THEBAN  AGE 


FIG.   28l. — SKETCH"  IN  THE   TOMB 

OF  KHA-EM-HET 
(After  Prisse  d'Avennes). 


a  new  system  was  established,  which  was  faithfully  observed  for 
nearly  five  centuries.     The  parts  of  the  tomb  were  divided  into 

two    groups ;     the    subterranean     . 

chambers  were  exiled  to  the 
desert,  behind  the  heights  which 
bounded  the  plain  on  the  north, 
in  the  valley  now  called  Biban- 
el-Muluk,  the  Gate  of  the  Kings; 
the  visible  elements  remained  on 
the  southern  slope  of  the  moun- 
tains and  in  the  plain,  at  Der- 
el-Bahari,  Sheikh-Abd-el-Kurnah, 
and  Medinet-Habu.  Like  the  main 
body  or  subterranean  portion  of 
the  earlier  Memphite  pyramid, 
the  Theban  hill  contains  only  the 
vault  and  the  passages  leading 
to  it.  At  first  this  kind  of  hy- 
pogeum  was  fairly  small;  that 
of  Thothmes  I.  is  concealed  at 
the  bottom  of  a  hole,  in  the 
base  of  the  cliff  itself.  A  steep 
staircase  brings  one  down  into  a  square  ante-room ,  where  a 
second  staircase  leads  to  the  vault.  This  is  an  elongated  par- 
allelogram, the  angles  rounded  in  such  a  manner  as  to  resemble 
a  cartouche ;  a  cell  cut 
out  towards  the  end,  in 
the  left  wall ,  formed  a 
kind  of  serdab.  The  en- 
trance is  on  the  east,  but 
the  axis  of  the  passages 
and  rooms  deviates  con- 
tinuously to  the  left  in 
such  a  manner  as  to 
bring  back  the  coffin  to 
the  west  by  a  kind  of 
imperfect  arc  of  a  circle, 
and  the  same  orientation 
persisted  until  the  end 
of  the  dynasty,  regulated 
however  in  such  a  manner 
that  the  axis,  instead  of  describing  a  curve,  traces  two  straight 
lines  which  join  at  a  more  or  less  acute  angle.  The  passages 

149 


L J 


FIG.  282.— THE  KING  CHARGING  (Ostrakon  in  the 
Museum,  Cairo).     (Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


and  the  descending  staircases  led  to  a 
square  well,  generally  some  30  feet  deep, 
designed  not  only  to  check  the  advance 
of  the  violators  of  tombs,  but  to  receive 
water  that  might  invade  them  during 
storms.  Beyond  this  obstacle,  just  op- 
posite the  passage  of  approach,  and  on 
a  level  with  it,  a  door  was  masked  in 
the  wall,  giving  access  to  a  room  with  two 
pillars:  it  was  here,  generally  speaking, 
that  the  axis,  turning  back  upon  itself, 
deviated  to  the  west.  Another  staircase 
was  set  in  the  left  angle  of  the  hypo- 
style  room ;  it  was  followed  by  a  slightly 
inclined  passage,  leading  to  the  vault; 
this  sometimes  has  sustaining  pillars,  and 
it  is  flanked  by  rooms  used  for  de- 
positing offerings,  generally  four  in 
number.  The  tombs  of  Hatshepset,  of 
Thothmes  III.  (Fig.  264),  and  IV.,  of 
Amenophis  II.  (Fig.  266)  and  HI. ,  of 
Heru-em-heb ,  and  later ,  that  of  Rameses  II. ,  were  all  on  this 
plan,  with  variations  more  or  less  marked  in  the  number  of  the 


FIG.   283. 

THE    KING    BRINGING    IN   A 
PRISONER    (Ostrakon    in    the 

Museum,  Cairo). 
(Phot.  E.  Brugsdi.)    ; 


FIG.   284.— A   PRIEST   PRAYING 

(Ostrakon  in  the  Museum,   Cairo). 

(Phot.  E.  Brugsdi.) 


150 


FIG.  285.— TWO  WRESTLERS 

(Ostrakon  in  the  Museum,   Cairo). 

(Phot.  E.  Brugsch.) 


THE  SECOND  THEBAN  AGE 

rooms,  and  the  extent  of  the  different  parts;  the  corridor  leading 
to  the  chambers  of  Hatshepset  is  700  feet  long.  Towards  the 
beginning  of  the  Nineteenth  Dynasty, 
architects  simplified  the  lines,  and 
straightening  the  axis,  .they  made  it 
run  directly  from  one  end  to  the 
other,  or  with  a  slight  deviation  to 
the  side  where  the  well  was  to  be 
placed.  The  best  known  example  is 
the  hypogeum  of  Seti  I.,  with  its 
string  of  rooms ,  the  last  of  which, 
left  unfinished,  is  325  feet  from  the 
outer  door.  Each  Pharaoh  made  a 
more  or  less  happy  variation  on  this 
theme  for  himself;  some,  like  Ra- 
meses  III.,  multiplied  the  cells  right 
and  left  of  the  first  passage;  others, 
on  the  contrary,  reduced  their  number, 
like  the  last  Rameses  of  the  Twentieth 
Dynasty.  The  chief  interest  does 
not,  however,  lie  in  the  architectural  arrangements,  which  are, 
as  we  have  seen,  simple  enough,  but  in  the  carved  and  painted 
decorations,  and  in  the  scenes  and  inscriptions  (Fig.  265)  which 
give  us  definite  inform- 
ation 
tical 


FIG.  286.— HEAD  DRAWN  IN  OUT- 
LINE IN  BLACK,  HEIGHTENED 
WITH  RED  (Ostrakon  in  the  Mu- 
seum, Cairo).  (Phot.  E.  Brugsch.) 


as     to     the     mys- 
doctrines     of     the 
Ramesside  age. 

The  plan  of  the  chapels 
was  so  ambitious  that  they 
are  for  the  most  part 
temples  comparable  to 
those  of  the  right  bank. 
The  oldest,  that  of  the 
two  Thothmes  and  Queen 
Hatshepset,  one  of  the 
most  original  and  finished 
works  of  Egyptian  art,  is 
what  is  now  known  as 
Der-el-Bahari,  from  the 
Coptic  monastery  founded 
among  its  ruins  in  the  sixth 
century  of  our  era  (Fig.  267).  The  sanctuaries  it  contains  — 
the  central  sanctuary  of  Amon,  of  Thothmes  and  of  the  Queen, 

151 


FIG.    287.  — A    DANCING    GIRL     (Ostrakon    in 
the  Museum,  Turin).     (Phot.  Lanzone.) 


ART  IN  EGYPT 


FIG.  288.— A  COUPLE   OF  THEBAN  CITI- 
ZENS (After  a  water-colour  by  H.  Carter). 


the  sanctuary  of  Hathor,  the  sanctuary  of  Anubis,   are   cut  out 
in  the  rock,    and  give  it   the  character   of   the   hemi-speos;    the 

rest  is  of  detached  masonry. 
Placed  beside  the  pyramid- 
mastaba  of  Menthu  -  hetep  ,  it 
might  be  supposed  that  it  would 
have  borrowed  certain  elements 
from  this;  but  such  is  not  the 
case,  and  never  were  two  build- 
ings more  dissimilar.  It  filled 
all  the  northern  hollow  of  the 
valley,  and  presented  the  most 
imposing  appearance  to  those 
who  approached  it  by  the  plain. 
Three  terraces  rise  in  recession 
one  above  the  other,  connected 
by  two  gently  inclined  planes 
along  which  ran  a  serpent  with 
scaly  folds  sculptured  in  the 
limestone.  The  lower  terraces 
were  adorned  with  porticoes 
on  three  sides,  those  on  the 

west   supported  by    square   pillars ,    and    that   on  the  north   by 

polygonal  columns  of  daz- 
zling whiteness :  the  spirit 
of  the  design  is  so  noble, 
and  the  contour  so  pure, 
that  it  might  almost  be 
a  Greek  colonnade  trans- 
ported from  the  Parthenon 
to  the  heart  of  the  Thebaid 
(Fig.  268).  The  third  ter- 
race was  enclosed  in  front 
by  a  straight  limestone 
wall,  behind  which  the 
sanctuary  extended  freely. 
Like  the  private  precincts 
of  the  non-funerary  tem- 
ples, it  was  divided  into 
three  compartments  paral- 
lel one  to  another.  In  the 
centre  was  the  abode  of  the  god,  with  its  hypostyle  hall  (now 
destroyed),  and  its  mysterious  chambers  in  the  rock.  On  the 

152 


FIG.    28g.— TWO    MUSICIAN-PRIESTESSES 
a  water-colour   by  H.  Carter). 


(After 


THE  SECOND  THEBAN  AGE 


FIG.    2QO. — THE   TWO   YOUNG   DAUGHTERS 
OF  AMENOPHIS  IV.   AT   EL-AMARNA. 


north  side,  the  apart- 
ments of  the  living  Horus, 
identified  with  Pharaoh, 
are  grouped  round  a  court 
in  which  the  altar  of 
Harmachis  is  still  standing. 
In  the  left  aisle ,  sacred 
to  the  divine  son,  the  dead 
king  lodged  when  he  pleas- 
ed. By  a  fiction  which 
will  not  seem  strange  to 
those  familiar  with  the 
Egyptian  doctrines,  Pha- 
raoh's double,  when  weary 
of  his  solitude  at  Biban- 
el-Muluk,  left  it  at  will 
for  his  sepulchral  temple, 
where  he  enjoyed  the  so- 
ciety of  the  priests  who 
administered  his  worship.  A  kind  of  temporary  lodging  was 
accordingly  provided  for  him,  a  set  of  half-a-dozen  rooms,  one 
at  least  of  which  had  an  arched  ceiling,  while  in  parts  its  decor- 
ations recall  those  of  the  Memphite 
mastabas.  The  vault  was  not,  indeed, 
a  true  vault  with  a  central  keystone, 
but  was  upheld  by  corbels.  It  was  no 
easy  task  to  arrange  this  corner  of 
the  mountain  in  a  manner  which,  while 
satisfying  all  the  demands  of  ritual 
and  doctrine,  did  not  completely  rob 
it  of  its  wild  grandeur;  here  again  the 
Theban  architects  showed  with  what 
technical  skill  and  feeling  for  nature 
they  adapted  the  style  of  their  works 
to  the  landscape  in  which  they  set 
them. 

It  is  with  Der-el-Bahari  that  the 
Memnonium  of  Seti  I.  at  Abydos  has 
most  analogy  (Fig.  269).  It  is  not  a 
hemi-speos,  but  something  intermediate 
between  this  and  the  isolated  temple: 
a  temple  set  against  a  low  hill  and 
partly  imbedded  in  it.  Seti  had  con- 

153 


FIG.  201.— SKETCH,  WITH  AN 
OUTLINE  IMITATING  A  SCULP- 
TURED CONTOUR  IN  ONE  OF 
THE  TOMBS  OF  SHEIKH-ABD- 

EL-K.URNAH.     (Phot.   Insinger.) 


ART  IN  EGYPT 


tented ,  himself  for  his  Theban  worship  with  a  place  in  the 
building  which  his  father  Rameses  I.  had  built  at  Kurnah 
(Fig.  271),  but  it  pleased  him  to  have  at  Abydos,  not  exactly 
a  cenotaph,  but  a  resting-place  where  his  double,  escaping  from 
the  darkness  of  his  Theban  syrinx,  might  shelter  at  leisure  under 
the  protection  of  the  tomb  of  Osiris.  He  therefore  retained  the 
features  essential  to  a  temple:  pylon,  courts,  a  portico,  two 
hypostyle  halls  (Fig.  270);  then,  as  there  was  not  room  to 
continue  the  building  to  the  west  without  rasing  the  hill  entirely, 

he  reduced  the  sanctuary 
and  its  adjoining  cham- 
bers to  two  rows  parallel 
with  the  facade,  the  last 
abutting  on  the  masses 
of  sand.  The  dividing  wall 
is  accordingly  pierced 
with  seven  doors  at  equal 
distances,  which  lead  into 
as  many  oblong  chambers 
vaulted  on  corbels;  six 
of  them  are  closed  at  the 
back,  but  the  third  on  the 
right  is  open  each  end, 
as  befits  a  chamber  for 
the  sacred  boat.  Passing 
through  it,  we  come  to 
the  little  hypostyle  hall 
essential  to  the  plan,  and 
the  mysterious  chambers, 
instead  of  being  grouped  behind  it,  are  ranged  right  and  left  at 
the  sides.  And  as  if  this  were  not  enough,  the  refuge  of 
the  deceased  sovereign  with  all  its  dependencies  was  thrown 
out  on  the  left,  in  a  special  wing,  detached  from  the  main  building, 
and  forming  a  square.  Some  hundred  metres  away,  Rameses  II., 
choosing  a  piece  of  ground  which  was  less  uneven,  erected  a 
second  resting-place  on  the  regular  plan,  the  same  used  for  his 
funerary  chapel  at  the  Ramesseum  in  the  Theban  plain  (Fig.  272). 
Chance,  which  does  not  always  favour  the  excavator,  has  preser- 
ved for  us  with  this  last  the  crowd  of  storehouses  (Fig.  273), 
stables,  houses  of  priests  or  artisans,  festival  or  assembly  halls, 
which  clustered  round  the  temples  and  made  each  of  them  the 
kernel  of  a  veritable  city;  considerable  portions  are  missing, 
however,  and  we  should  not  be  able  to  reconstitute  the  ar- 

154 


FIG.  2Q2. — NORTH  WALL  OF  THE  FIRST  CHAM- 
BER IN  THE  TOMB   OF  NAKHT.      (Phot.  Beato.) 


THE  SECOND  THEBAN  AGE 


rangement  with  any  certainty,  had  not  Rameses  III.,  that  plagia- 
rist of  Rameses  II.,  imitated  them  faithfully  at  Medinet-Habu 
(Fig.  274).  By  combining  the  two,  we  are  able  to  see  what  was 
the  scheme  finally  adopted  by  the  Ramessids  for  their  mortuary 
temples.  There  was  first,  as  in  the  living  temple,  a  pylon- 
fa§ade,  then  a  court,  the  northern  portico  of  which  was  guarded 
by  Atlantes  set  against  pillars  (Fig.  231),  a  second  pylon,  a 
second  court  with 
porticoes,  arid  at  the 
end  of  this,  a  pronaos 
on  a  raised  platform, 
to  which  access  was 
obtained  by  a  flight 
of  shallow  steps.  Be- 
hind the  pronaos  was 
the  hypostyle  hall, 
hemmed  in  between 
two  rows  of  chapels 
or  store-rooms ,  and 
the  private  apart- 
ments of  the  god 
began  beyond  this 
again,  in  three  parallel 
lines  as  usual,  the 
god  in  the  centre 
with  his  hypostyle 
halls  and  his  chamber, 
the  living  Horus  on 
the  right,  the  de- 
ceased sovereign  on 
the  left.  Medinet- 
Habu  is  the  subtlest 
expression  of  the  conceptions  of  the  Theban  priesthood  as  to 
the  destiny  of  the  royal  soul,  and  the  means  by  which  its  future 
was  to  be  ensured.  The  architect  made  his  art  subservient 
to  doctrine,  and  combining  that  which  was  indispensable  to 
the  existence  of  the  gods  with  that  which  was  essential  for 
the  perpetuity  of  the  double,  he  welded  the  whole  into 
a  grandiose  and  harmonious  creation.  Would  he  have  done 
better  still  later?  Rameses  III.  was  the  last  of  the  great 
conquerors,  and  his  successors,  lacking  the  resources  provided 
by  foreign  warfare,  undertook  no  such  vast  structural  enter- 
prises as  his.  They  usurped  more  or  less  successfully  a  corner 

155 


FIG.  293. — FRAGMENT   OF  THE  BAlTLE  OF  KADESH 
IN  THE  RAMESSEUM  AT  THEBES.     (Phot.  Beato.) 


ART  IN  EGYPT 


of   the    temples    their    ancestors    had    prepared ,    and ,    graving 
their   names   upon   them,   appropriated   their   revenues. 
Just      as     the 

Memphite      Pha- 
raohs       of       an 

earlier    age    and 

the  first  Thebans 

had  lived  in  sight 

of   and    in    daily 

contact  with  their 

pyramids,  so  the 

second    Thebans 

did     not     shrink 

from       attaching 

their  dwellings  to 

their      sepulchral 

temples.        Even 

in     the     Rames- 

seum ,     we     find 

on  the  left  of  the 

first     court     the 

levelling   courses 

of  a  building,  the 
arrangement  of  the  rooms  in  which  show  it  to  have  been  a  habi- 
tation,   one  of   the   houses   of   Rameses  II.     At   Medinet-Habu, 

we  find  similar  ruins  in  the  same 
situation ;  the  south  porch  of  the 
first  court,  with  its  eight  bell- 
shaped  columns,  was,  as  it  were, 
the  religious  face  of  a  palace  of  Ra- 
meses III.,  now  destroyed  (Fig.  275). 
Three  doors  opened  into  it,  and 
from  a  sort  of  tribune  the  balcony 
of  which  projected  from  the  centre 
of  the  wall,  Pharaoh  took  part  in 
the  ceremonial  of  worship  without 
having  to  mingle  with  the  crowd. 
The  building  was  in  brick,  with 
details  of  stone  and  enamelled 
earthenware;  it  has  been  ravaged 
by  antiquity-hunters,  and  collectors 
of  manure,  but  the  remains  of  a 
villa  of  Amenophis  III.,  still  visible 
156 


FIG.  2Q4. — THEMENEP- 
TAH    WITH    TWO    EN- 
SIGNS (Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


FIG.  295. 

RAMESES  II.   BETWEEN  AMON 

AND  MUT  (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


FIG.    2Q6.— CROUCHING    FIGURE 

HOLDING   A     DIVINITY     (Museum, 

Cairo).     (Phot.  Legrain.) 


THE  SECOND  THEBAN  AGE 


about  1200  yards  further  south,  give  us  an  idea  of  what  it  was. 
Like  the  feudal  castle  of  our  own  past,  the  Theban  palace  was 


FIG.   297.— KNEELING   FIGURE 

CARRYING  A   TRIAD    (Museum, 

Cairo).     (Phot  E.  Brugsch.}. 


FIG.  298.— A  PERSON  SEATED  ON 
THE  GROUND  WITH  ONE  LEG  FLAT 
(Museum,  Cairo).  (Phot.  E.  Brugsch.) 


rectangular  on  plan ,  and  a  solid  wall, 
almost  without  doors  or  windows,  en- 
closed it.  When  this  was  passed,  one 
entered  a  labyrinth  of  little  courts,  col- 
umned rooms,  alcoves,  and  dark  cells 
leading  one  into  the  other  and  often 
ending  in  a  blind  alley.  The  main  parts 
of  the  structure  are  of  sun-dried  bricks, 
some  of  them  stamped  with  the  royal 
cartouche.  The  floor  is  of  beaten  clay, 
so  firmly  pounded  that  it  is  almost  as 
hard  as  stone.  The  walls  were  plastered 
with  a  coating  of  mud  similar  to  that 
still  used  in  Egyptian  villages.  The  ap- 
pearance of  the  various  places  does  not 
everywhere  suggest  how  domestic  life 
was  carried  on  in  these  interiors.  We 
may,  however,  surmise  that  two  oblong 
halls,  upheld  by  two  parallel  lines  of 
wooden  columns  with  limestone  bases, 

157 


FIG.   299. 
SURVEYOR  WITH 

HIS  LINE 
(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


were  guard-rooms;  courtiers  and  officers  of  the  crown  no  doubt 
thronged  them,  taking  up  their  positions  in  hierarchical  order 
on  audience -days.  A  modest  ante-room  led  thence  into  the 


FIG.   JOO.— RAMESES  PUSHING  A  BOAT   (DESTROYED)  (Museum,  Cairo). 
(Phot.  E.  Brugsch.) 

private  apartment,  where  persons  honoured  by  admission  to  the 
royal  presence  saw  before  them,  enframed  between  two  painted 
wooden  columns,  the  dais  on  which  Pharaoh  deigned  to  show 
himself.  Bath-rooms  were  numerous;  three  of  these  still  contain 
the  water -conduits  and  the  stone  slabs  on  which  the  bather 
crouched  or  lay  to  be  dried  or  massaged.  Several  bedrooms 

follow  one  another  close  by,  with  the 
platform  on  which  the  bed  was  raised. 
Other  rooms,  small  and  bare,  seem 
to  have  been  for  the  use  of  servants; 
it  is  not  known  where  the  kitchens 
and  store-rooms  were  situated.  To 
sum  up ,  we  have  here  one  of  those 
princely  residences  of  which  there  are 
so  many  not  very  comprehensible 
sketches  in  the  tombs  of  El-Amarna. 
They  were  slight  in  structure  like 
the  houses  of  modern  Egypt,  but 
covered  with  paintings  which  dis- 
guised the  poverty  of  the  material. 
Vultures  with  outspread  wings  ho- 
vered on  the  ceilings,  together  with 
flights  of  pigeons  (Fig.  276),  or  ducks 
imprisoned  in  frames  of  undulating 
lines  or  many-coloured  spirals.  On 
the  pavements  fountains  were  traced, 
158 


FIG.    3OI.— AMENOPHIS    II.     AND 

THE   SERPENT  MARITSAKRO 
(Museum,  Cairo).  (Phot.  E.  Brugsdi.) 


THE  SECOND  THEBAN  AGE 


FIG.   3O2.— TWO  PRISONERS  TIED 

BACK  TO   BACK    (Museum,    Cairo). 

(Phot.  E.  Brugsch). 


or  thickets  of  aquatic  plants, 
where  oxen  graze  and  frolic 
(Fig.  277) ;  fish  swim  under  the 
water,  ducks  cruise  among  the 
lotuses  (Fig.  278) ,  and  captives 
bound  in  constrained  attitudes 
are  ranged  in  lines  along  the 
banks. 

Civil  architecture  seems  there- 
fore   to    have    progressed,    and 
perhaps  if  we  knew  more  of  the 
palaces    of   the   Memphite    age, 
comparison    would    show    those 
of    the   Thebans    to    have   been 
richer   in    treatment,    and   more 
complex    in    arrangement;    mili- 
tary architecture   had   not  been 
modified    in   any  way,    and    the 
reason  of  its  immobility  is  obvi- 
ous.   The  conditions  of  war  on 
the  banks  of  the  Nile  had  not  changed  since  the  old  days,  and 
since    the    expulsion    of   the   Shepherds ,    conquering   Egypt   had 
never   experienced   a   reverse    serious    enough   to   cause    her   to 
reconstruct       the 
walls  of  her  towns 
on    a    new    plan. 
Not  that  the  Pha- 
raohs had  not,  in 
Syria ,       attacked 
stone  citadels  built 
on  the  most  scien- 
tific      principles ; 
only  one  of  them, 
however ,     Rame- 
ses  III.,  gave  him- 
self the  pleasure  of 
showing  his  good 
Thebans  what   he 
had  seen  in  Syria, 
by  way  of  comme- 
morating his  vic- 
tories.  Across  the 
eastern    front    of 
159 


A  JAR   (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 
FIG.  JO}. — SLAVE  BEARING 


FIG. 304. -GROTESQUE  SLAVE 
BEARING  A  JAR  (Museum, 
Cairo).    (Phot.  E.  Brugsch.J 


ART  IN  EGYPT 


FIG.    JOS.— QUEEN   TUITISHERE 
(British   Museum).     (Phot.   Perrin.J 


FIG.  306.— STATUE  OF  AMENOPHIS  I. 
(Museum,  Turin).     (Phot.  Petrie.) 


his  sepulchral  chapel  at  Medinei-Habu,  he  raised  a  battlemented 
wall  of  sandstone,  averaging  about  13  feet  in 
height.  This  was  the  equivalent  of  the  brick 
screen  of  the  old  Egyptian  fortresses,  and  the 
rampart  it  covered  was  of  brick,  but  the  first 
gateway  has  the  appearance  of  a  veritable  migdol, 
a  pair  of  pincers  in  masonry,  gripping  a  parade- 
ground  which  diminishes  sharply;  the  two  branches 
are  reunited  at  the  end  by  a  building  of  two 
storeys  where  the  real  entrance  appears  (Fig.  279). 
The  towers  are  about  70  feet  high.  The  base 
is  sloping,  to  prevent  sappers  from  approaching 
the  foot  of  the  wall ,  and  to  cause  the  pro- 
jectiles thrown  by  the  defenders  from  the  curtain 
to  ricochet  against  their  assailants.  This  is  a 
solitary  example,  and  here  again  the  Egyptians, 
having  proved  themselves  skilful  imitators  of 
the  foreigner,  returned  to  their  secular  habits. 
Thus,  during  these  centuries  of  prosperity,  all 
their  natural  talents  and  all  their  faculties  of 
invention  seem  to  have  been  concentrated  on 
a  single  object,  the  perpetual  aggrandisement 
and  embellishment  of  the  temple ,  whether  as 
the  lodging  of  the  gods,  or  the  refuge  of  dead 
160 


FIG.  307.— STATUE 

OF    AMENOPHIS  I. 

(Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


THE  SECOND  THEBAN  AGE 


FIG.  J08. 

HEAD  OF  AN  ATLAS  OF  THOTHMES  I. 
(Museum,  Cairo).  (Phot.  Legrain.) 


.ngs  whose  souls  wearied  of 
the  darkness  of  the  hypogeum. 
They  did  not  at  once  realise 
the  ideal  they  had  set  before 
themselves,  and  so  many  mo- 
numents have  perished  that  we 
are  no  longer  able  to  recon- 
stitute the  series  of  forms  through 
which  they  passed,  before  reach- 
ing the  complex  types  of  Luxor 
and  Medinet-Habu.  I  have  tried 
to  note  a  few,  but  there  are 
others  to  study.  Actual  results 
impress  us  most  strongly  with 
the  richness  of  inspiration  to 
which  these  buildings  bear  wit- 
ness, and  the  vigour  with  which 
their  builders  realised  their  boldest  inventions  in  situ;  no  archi- 
tects have  ever  rivalled  them  in  the  treatment  of  mass,  and  the 
Pharaohs  made  no  vain  boast  when  they  declared  in  their 
inscriptions  that  they  had  erected  imperishable  stones.  True, 

these    admirable    unknown    artists    did         

not  all  show  equal  talent,  and  the 
mediocre  is  not  entirely  absent  in  the 
work,  but  many  of  them  proved 
themselves  true  men  of  genius,  and 
their  names,  had  they  been  transmitted 
to  us  by  their  contemporaries,  would 
deserve  to  be  inscribed  side  by  side 
with  those  of  the  artists  to  whom  we 
owe  the  noblest  monuments  of  Greece 
and  Rome.  The  temple  as  created  by 
them  is  one  of  the  most  original  and 
mighty  conceptions  of  the  human  in- 
tellect ,  not  only  in  Egypt  under  the 
Theban  dynasties,  but  among  all  peoples 
and  in  all  ages. 

B.  PAINTING  AND  SCULPTURE. 
In  painting  and  sculpture,  as  in  archi- 
tecture,   relics  are  almost  innumerable, 
and  they  follow   in   such   strict  chrono-, 

logical    order    that   we    may   study   the 

161 


FIG.  309.— QUEEN  ISIS   (Mu- 
seum, Cairo).  (Phot.E.Brugsch.) 

M 


ART  IN  EGYPT 


FIG.  3IO.— HEAD  OF  A  STATUE 

OF     THOTHMES     III.     (Museum, 

Cairo).     (Phot.  E.  Brugsch.) 


development  of  one  of  the  great  schools, 
the  Theban,  if  not  from  year  to  year, 
at  least  from  reign  to  reign.  This  is 
not  the  case  of  the  others ,  save 
perhaps  that  of  Abydos;  at  Hermopolis 
and  Memphis,  and  in  the  Delta,  our 
lack  of  knowledge  is  such  that  we 
cannot  as  yet  discern  the  progress  of 
art.  If,  however,  we  compare  and  co- 
ordinate the  data  we  have  gained,  we 
are  enabled  to  deduce  from  them  a 
number  of  facts  and  conclusions  appli- 
cable not  only  to  Thebes,  but  to  the 
whole  of  Egypt. 

In  the  first  place,  the  principles  and 
methods  of  decorative  art  underwent 
such  serious  modification,  that  it  is 
impossible  to  confound  certain  series 
of  bas.-reliefs  and  funerary  or  martial 
pictures  of  the  second  Theban  Empire 


with  similar  pictures  of  the  Twelfth  Dynasty  and  the  Memphite 

age.    The  design  remains  just  as  pure  (Fig.  280)  and  the  sketches 
in    black    in    the    tomb    of   Seti  I.    and    certain  Theban    hypogea 

(Fig.  281)  will  bear  comparison  with 
the  best  of  those  which  abound  in 
the  mastabas  of  Sakkarah  or  Gizeh. 
But  the  boldness  of  the  artist  has 
grown  with  the  practice  of  centuries, 
and  he  attacks  compositions  and 
'  movements  which  would  have  dis- 
couraged his  ancestors.  The  rough 
sketches  drawn  upon  fragments  of 
stone  by  the  band  who  decorated 
the  syrinx  of  Rameses  IV.  bear 
witness  to  amazing  firmness  of  touch  /', 
and  an  in^xjiajjalibJe^_j/arie_ty___^P1'' 
imagination.  Whether  the  king 
charges  at  the  utmost  speed  of  his 
horses  (Fig-.  282) .  or  walking  se- 
dately, brings  a  minute  prisoner  to 
the  gods  (Fig.  283) ;  whether  the 
kneeling  priest  raises  his  hands  to 
heaven  in  prayer  (Fig.  2 
162 


FIG.    311. 

THOTHMES  IV.   AND  HIS  MOTHER 
(Museum,   Cairo).     (Phot.  E.  Brugsch.) 


THE  SECOND  THEBAN  AGE 


FIG.  312. — THE 

GODDESS  SEK- 

HET  AT  KARNAK. 

(Phot.  Legrain.) 


two    wrestlers    close   before   seizing   each   other 

(Fig.  285) ,   the   line   is   always    equally   flexible 

and   boldly   touched;    a   certain    head  of  a  man 

heightened  with  a  few  dashes  of  red  (Fig.  286) 

is  a  marvel  of  swift  precision ,    and    few  of  our 

contemporaries  could  have  given  a  more  realistic 

rendering  of   the   girl-acrobat,   throwing   herself 

back  for  a  somersault,  on  the  ostrakon  at  Turin 

(Fig.  287).    We  find  the  same  freedom,  combined 

with  a  greater  delicacy  of  line,    in   the  finished 

pictures  at  Sheikh-Abd-el-Kurnah ,   a    couple    of 

citizens  (Fig.  288),  a  pair  of  musician-priestesses 

(Fig.  289),  and,  at  El-Amarna,    the    two    little 

nude  princesses  on  their  cushion,   caressing  one 

another    with    gentle     and    ingenuous    gestures 

(Fig.  290).    The  composition,  too,  has  matured, 

and  is  almost  equal  to  the  drawing.    The  Mem- 

ptlfte  artist  had  oeen  wont  to  resolve  the  simul- 
taneous  operations   of  an   agricultural   scene   or 

of  a   battle   into   their  simple   elements,   which 

he  superposed  in  independent  rows.  The  Theban, 

under  the  Ahmessids   and  their  successors,    did 

not   discard   the    artifice  of  the    various   registers;    he    even   ad- 

KereH    to    it    strictly    in    the    treatment    of    religious    subjects ; 
-neither   in    a   temple   nor   in   a  tomb 
f  did  he  cease  to  observe  the  tradition 

bequeathed  to  him  by  antiquity.    But 

it  was  no   longer   the   same   when  he 

passed  from  pictures  ef  the  divine  to 

those  of  civil  or  military  life ,  and 
greater  liberty  is  explained  at 
in  part  by  the  constant  progress 

made  by  painting  from  the  preceding 

age  onwards.  Without    entirely  aban- 
doning  its   part   as    the    auxiliary    of 

sculpture,   it  had  learned  to  separate 

itself   from   it,    and   to    dispense  with 

its   collaboration    upon  occasion.     As 

I  have  already  indicated,    the   nature 

of  the  Theban  mountains  had  a  good 

deal    to    do  with    this.     They   consist 

of   a   very   fine  limestone,    the   strata 

r          1  •    i_  1-1  ill  FIG.  313.— STATUE  OF  A  MONKEY 

ot   which    were    dislocated    by    some      (Museum,  Cairo).  (Phot.E.Bmgsch.) 

163  M  2 


ART  IN  EGYPT 


remote   cataclysm,    in     such    a    manner    that   it    does   not    lend 
themselves  everywhere  to  the  work  of  the   chisel  with  the  same 

facility.  Though  solid 
enough  in  the  Valley  of 
the  Kings,  it  cracks  in 
every  direction  at  Sheikh- 
Abd-el-Kurnah,  and  is  fulL. 
of_jiuge  flints,  which  had 
first  to  be  removed,  and 
then  replaced  by  in- 
serted fragments.  Hence, 
inm  any  cases  the  decjur- 


FIG.    314. 

ONE   OF  THE  LIONS   OF  GEBEL-BARKAL 
(British  Museum,  London). 


rators  of  tombs  were 
content  to  cover  the 
surface  with  a  plaster 
which  hid  the  defects, 
and  to  paint  on  this  in  distemper  what  they  would  have  carved 
under  more  favourable  conditions.  Seduced  by  the  facilities  of 
the  brush,  they  became  even  more  emancipated  than  their  pre- 
cursors of  the  Twelfth  Dynasty.  They  multiplied  the  motives  on 
their  ceilings,  and  ajdej  to  the  stars  and  geometrical  designs 

which  had  hitherto  pre- 
dominated, elements  bor- 
rowed from  nature,  single 
florets ,  bouquets  of 
lotus,  bulls'  heads,  flying 
birds,  groups  of  hiero- 
glyphs of  the  happiest 
effect.  They  continued, 
from  a  lingering  respecr 
for  the  traditions  of  the 
past,  to  surround  their 
figures  with  a  line  which 
recalled  the  effects  of  the 
primitive  reliefs  (Fig.  291), 
but  they  grouped  them; 
in  attitudes  increasingly 
natural,  and  they  broke 
down  the  tyranny  oj; 
the  superposed  register^; 
Thenceforth ,  if ,  wishing 
to  represent  work  in  the 

FIG.  315.— CHAPEL  OF  THE  COW  HATHOR       f  l  l       .1         1 

(Museum,  Cairo).     (Phot.  E.  Brugsch.)  fields ,        they       ChOSC       to     , 

164 


.1  -, 


FIG.    Jl6.— BAS-RELIEF   OF  THE  COW  HATHOR 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


THE  SECOND  THEBAN  AGE 


^express  <•#",'  as  we  do,   by  the   normal   methods   of   perspective, 

they  were  free  to  do  so.     As  in  the  tomb   of   Nakht,    they  set 

upon    the    walls,    at 

various    heights  ,    ac- 

cording  to    distance, 

the  persons  who  play 

a  part  in  the  action: 

the  dead  man  super- 

intending   his    work- 

men, ploughmen  turn- 

ing  the   furrow,    the 

sower  scattering  se,ed, 

labourers       breaking 

the  sod  with  pickaxes, 

a    woodman    cutting 

down     a     tree,     the 

thirsty    toiler    taking 

a    draught    from    his 

leather  jar  (Fig.  292), 

The  experiment  is  a 
-  clumsy  one  ,   more  akin  to  the  scenes  on  a  Chinese  screen  than 

to  our   landscapes,    but  it  is  an  essay   in   perspective,   and  this 

is  no  isolated   example;   we  find   several    others   in   the   painted 

hypogea  of  Sheikh-Abd-el-Kurnah.     From 

fresco,    the    method    passed    rapidly    to 

bas-relief,  and  we  find  it  on  pylons;  here 

the  artist  gives  us,  not  offerings  and  sacri- 

fices,   but   battles,    as  at   Luxor   and  the 

Ramesseum  ,   where   the  entire   surface  is 

one  vast  composition,  in  which  the  actors 

assemble   and   disperse  without  any   sepa- 

ration of  the  planes   by   lines    (Fig.  293). 

There  is  no    unity  of  action  ,    but  a  com- 

plete narrative  is  set   forth,    some   of  the 

incidents  of   which   are  historical,   as,  for 

instance,  the  battle  of  Kadesh,  the  council 

of   war    held    by    the    Egyptian    generals, 

and  the  report  of  the  spies,   the  surprise 

of  the  camp  by  the  Hittites,  Rameses  II. 

charging,  the  arrival  of  the  reinforcements, 

the    battle  on  the  banks   of  the  Orontes, 

the  sortie  of  the  Amorrhaeans  who  saved 

the  remnant  of  the  Asiatic  army.    As  we 

165 


_J 


FIG.   317.— THE   COW   OF 

DER-EL-BAHARI 

(Museum ,  Cairo). 

(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


FIG.  318.— STATUE 

OF  AMENOPHIS  HI. 

CLOTHED     IN    THE 

ASSYRIAN   MANNER. 

(Phot.  Chassinat.) 


know  the  main  theme  from  literary  texts,  we 
do  not  find  any  great  difficulty  in  inter- 
preting the  artistic  developments  of  it,  but 
it  must  be  admitted  that  if  we  had  only  the 
picture,  it  would  be  very  difficult  for  us 
to  establish  the  chronology  of  events  and 
distinguish  their  progress  with  any  certainty. 
The  Ramesside  artist  was  as  yet  incapable 
of  discerning  the  decisive  moments  and 
seizing  the  critical  point  of  a  battle;  he  piled 
up  his  incidents  in  a  more  or  less  haphazard 
fashion,  without  troubling  about  the  time 
when  they  happened,  and  their  influence  upon 
the  final  result.  His  chief  concern  was  to 
make  the  presence  of  Pharaoh  conspicuous, 
and  to  rivet  attention  upon  him.  In  every 
crisis,  he  grouped  the  secondary  personages 
round  the  king,  and  the  better  to  draw  the 
eye  of  the  spectator  to  this  figure,  he  made 
him  of  heroic  size.  At  Luxor  as  in  the 
Ramesseum.  Rameses  II.,  standing  in  his 
chariot,  and  piercing  the  flying  Asiatics  with 
his  shafts,  is  the  centre  of  the  action.  The 
artifice  which 


consists  in  attributing  colossal  pro- 
portions to  the  prince ,  is  puerile 
in  itself;  but  in  a  huge  "machine" 
such  as  the  illustrated  record  of 
a  battle,  it  is,  after  all,  the  only 
means  of  giving  a  kind  of  unity 
to  the  decoration. 

Progress  in  less  apparent  in 
sculpture,  and  it  was  long  sup- 
posed that  here  artists  had  merely 
carried  on  the  Memphite  tradition, 
while  falsifying  and  degrading  it. 
We  so  often  recognise  the  formulae 
of  the  age  of  the  Pyramids  in  their 
works,  that  we  get  the  impression 
that  nothing  had  been  changed; 
but  as  soon  as  we  examine  their 
details,  we  find  that  novelties 
abound.  Let  us  take,  for  instance, 

166 


FIG.    JIQ. — COLOSSAL   GROUP   OF 

AMENOPHIS    III.    AND    THI    (^luseum, 

Cairo).     (Phot.  E.  Brugsch.) 


THE  SECOND  THEBAN  AGE 


FIG     320.— THE  TWO   COLOSSAL  FIGURES 
OF  MEMNON  AT  THEBES.     (Phot.  Beato.) 


the    erect    figure,    sovereign 

or  subject,  receiving  homage. 

He  stands  straight  and  firm, 

one  foot  advanced,    but  his 

hands,    which  formerly  were 

either  empty,    or   grasped  a 

fragment    of  a  sceptre    or  a 

handkerchief,  are  now  loaded 

with  gigantic  emblems.  These 

are  in  general  sacred  ensigns, 

stout  halberds  surmounted  by 

the    head    of    a    human    or 

animal    divinity ;     sometimes 

he  is  content  with  one,  some- 
times     he     demands       two 

(Fig.  294),   the  lower  extre- 
mities of  which  rest   on    the 

ground  beside  his  feet,  while 

their  faces  enframe  his  head 

right    and    left.     Groups   of 

two  seated  persons,  or  triads 

incorporated  with  a   supporting  slab   at   the   back,    like    the  Ra- 

meses  II.    between   Amon    and   Mut    (Fig.   295),    are    conceived 

entirely  in  the  ancient  taste,  but 
in  the  isolated  figures,  the  sitter 
does  not  merely  lay  his  right  hand 
on  his  breast,  like  the  Rahetep 
at  Medum ;  he  grasps  an  Osirian 
crook ,  a  scourge ,  a  scroll ,  or ,  if 
a  woman ,  a  handkerchief,  a  spray 
of  blossom,  a  sistrum.  The  kneeling 
or  crouching  figure ,  which  has 
become  frequent,  bears  in  front 
of  it  an  altar ,  a  naos ,  a  triad 
(Fig.  297) ,  a  statuette  of  a  divi- 
nity (Fig.  296);  a  roll  of  rope, 
surmounted  by  a  ram's  head,  denotes- 
the  calling  of  land-surveyor  exer- 
cised by  the  model  (Fig.  299).  Other 
types  present  themselves  which 
never  occurred  among  the  earlier 
works,  such  as  that  of  a  person 
seated  with  one  knee  drawn  up, 
167 


FIG.    321. 

AMENOPHIS,   SON   OF  HAPU 
(Museum,  Cairo).     (Phot.  Legrain.) 


ART  IN  EGYPT 


FIG.   322. 

AMENOPHIS,   THE  CROUCHING   SCRIBE 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


the  other  flat  against  the  ground 
under  him  (Fig.  298),  and  that 
of  the  kneeling  king  who  drags 
himself  along  the  ground  in 
front  of  the  god ,  pushing  an 
object  of  worship  or  an  of- 
fering, an  altar,  a  jar,  or  a 
sacred  boat  (Fig.  300).  In 
like  manner,  scenes  which  exist- 
ed only  in  bas-reliefs  on  the 
walls,  are  detached  from  it, 
and  become  stone  groups;  the 
king  standing  between  Horus 
and  Set,  and  receiving  from 
them  the  waves  of  the  water 
of  life,  the  king  escorted  by  his 
lion  and  conducting  a  chained 
barbarian  to  the  god,  a  lion 
devouring  a  captive,  a  seated 
scribe  reading  a  book  and 
carrying  a  little  monkey  on  his  neck,  (an  incarnation  of  the  god 
Thoth),  a  foster  father  squatting  on  his  haunches  and  holding  to 
his  breast  the  royal  child  whose  education  is  confided  to  him. 

The  Asiatic  or  negro  prisoners  bound 
back  to  back  are  treated  with  an 
amazing  realism,  sometimes  verging 
on  caricature  (Fig.  302).  The  beasts 
themselves  play  their  part,  and  the 
cow  Hathor  or  the  serpent  Maritsakro 
attach  themselves  to  a  Pharaoh  in 
order  to  protect  him  (Fig.  301).  All 
these  are  in  stone,  sometimes  life- 
size,  and  show  a  facility  of  invention 
and  a  flexibility  of  execution  we 
should  hardly  look  for  in  the  second 
Theban  period.  Wood  was  less  in 
favour  for  statues  than  formerly,  save 
for  those  ritual  figures  of  which  only 
fragments  remain ,  but  it  was  com- 
monly used  for  the  statuettes  which 
took  the  place  of  the  cfoi/6/e-statues 
in  the  tombs  of  the  lower  middle 
class,  and  for  certain  objects  of 
168 


FIG.  323. — SENNEFER,  HIS  WIFE 

AND  DAUGHTER  (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


THE  SECOND  THEBAN  AGE 


industrial  art  which  demanded  the  human  figure,  such,  for  in- 
stance ,  as  the  bearers  of  jars  of  kohol ;  here ,  the  number  of 
new  forms  is  considerable :  foreign 
slaves  bowed  beneath  a  sack  or  a  jar 
(Figs.  303,  304),  children  gathering 
flowers,  young  girls  swimming  and 
pushing  a  duck  or  a  goose  before 
them.  There  was  the  same  variety  in 
metal-work,  but  the  majority  of  the 
gold  and  silver  statues  have  disappear- 
ed, and  only  s  small  number  of  bronze 
examples  remain.  Statuary,  whether 
in  stone,  wood,  or  metal,  may  be  said 
to  have  developed  in  every  direction ;  far 
from  being  inferior  to  that  of  preceding 
ages,  it  surpasses  it,  as  we  have  just 
seen,  in  variety  of  motive,  and  very 
often  equals  it  in  beauty  of  handling. 
The  first  monuments  we  possess  of 
the  time  of  the  Ahmessids  are  still 

fairly  faithful  to  the  style  of  the  preceding  schools.  This  is 
notably  the  case  in  the  figure  of  Queen  Tuitishere,  in  London 
(Fig.  305),  and  in  the  two  statues  of  Amenophis  I.  at  Turin 


FIG.  324. — HEAD  OF  A  MAN 

(Museum,  Cairo). 
(Phot.  E.  Brugsch). 


FIG.  325.— TORSO  AND  HEAD 

OF  A  WOMAN  (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


169 


FIG.    326.— MOND'S    STATUETTE. 

(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


FIG.   327. — THE   EGYPTIAN  FLEET,   AT  DER-EL-BAHARI.      (Phot.   Beato.) 

and  Cairo.  That  at  Turin  (Fig.  305)  is  an  admirably  preserved 
work  in  white  limestone ;  the  king  is  seated ,  confronting  the 
spectator ,  in  the  hieratic  attitude ,  and  but  for  the  cartouches, 
we  might  well  take  it  for  a  work  of  the  Twelfth  Dynasty. 
The  Cairo  statue  (Fig.  307)  is  mutilated,  but  the  face  and  bust 
are  intact;  the  king  was  invested  with  the  insignia  of  Ta-Tenen 
and  his  flesh  was  painted.  It  has  all  the  delicacy  of  the  ancient 

Memphite  schools,  together  with  the 
firmness  of  chisel  and  the  virile 
air  which  are  characteristic  of  the 
Thebans.  The  head  of  one  of  the 
Atlantes  now  at  Cairo,  erected  by 
Thothmes  I.  in  the  court  of  the 
obelisk  at  Karnak,  establishes,  I 
think,  the  transition  from  the  an- 
cient to  the  modern  style.  As  it 
retains  the  red  colouring,  it  is 
very  life-like  in  appearance,  in 
spite  of  the  loss  of  the  head-dress 
(Fig.  308).  The  Pharaoh  himself 
seems  to  be  welcoming  the  visitor, 
and  his  round  face,  his  smiling 
eyes,  his  dimpled  cheeks  and  amiable 
mouth,  recall  the  features  of  the 
Sesostris  at  Lisht;  it  is  further 
characterised  by  a  firmness  of  touch 
170 


FIG.    328. — QUEEN  AAHMES  AT 
DER-EL-BAHARI.    (Phot.  Beato.) 


THE  SECOND  THEBAN  AGE 


FIG.  329. —QUEEN  THI,  AT  SHEIKH- 
ABD-EL-KURNAH.      (Phot.    Weigall.) 


and  an  individuality  of  expression 
lacking  in  the  others.  When  once 
the  royal  workshop  was  organised, 
the  multiplicity  of  orders  that  flowed 
in  soon  awoke  in  it  qualities  quies- 
cent since  the  invasion  of  the  Shep- 
herds, to  which  it  added  new  ele- 
ments, derived,  I  think,  from  in- 
fluences coming  from  other  parts 
of  the  valley.  The  Thebans  alone 
would  not  have  sufficed  for  the 
decoration  of  monuments,  temples, 
or  tombs.  They  received  provincial 
auxiliaries,  and  borrowed  from  them 
something  of  the  traditions  and 
temperament  these  brought  with  them 
from  their  native  cities.  Thus  rein- 
forced, the  school  subdivided  into 
several  branches,  each  of  which 
soon  assumed  its  personal  physiog- 
nomy. I  should,  for  instance,  attribute  a  good  proportion  of  the 
royal  statues  at  Turin,  and  others  recently  brought  to  light  in 
the  favissa,  the  Isis,  the  Thothmes  III.  and  the  Senmut,  to  a 
single  workshop,  probably 
established  at  Karnak.  The 
statuette  of  the  queen  Isis 
(Fig.  309)  reveals  the  initiator 
of  the  facial  type  which  pre- 
vailed under  the  Ahmessids  for 
three  generations,  the  hooked 
nose,  the  large  prominent  eyes, 
the  fleshy  mouth,  the  round 
face.  The  heavy  wig  which 
encases  the  head  was  not  calcu- 
lated to  make  the  sculptor's 
task  easier ;  he  managed,  how- 
ever, to  minimise  its  dis- 
astrous effect.  Thothmes  III. 
has  his  mother's  face,  but  the 
type  is  less  hard  (Fig.  310). 
The  statue  is  of  fine  schist, 
and  no  reproduction  could  do  FIG-  BSD—HEAD  OF  A  STATUETTE  OF 

..  *.,  ,    ,.  ,     .1  QUEEN  THI  (Museum,  Cairo). 

justice  to  the  delicacy   or  the  (Phot.  E. 


ART  IN  EGYPT 


modelling;  the  play  of  the  muscles  is  noted  discreetly,  with 
extraordinary  felicity,  and  as  the  imperceptible  shadows  it  pro- 
duces vary  as  we  pass  round  the  figure, 
the  expression  of  the  features  seems  to 
change  every  moment.  The  kneeling 
statues  of  Amenophis  II.  offering  wine 
or  water,  are  not  unworthy  of  the  series; 
although  they  show  less  individuality 
than  the  Thothmes  III.  and  the  couple, 
Thothmes  IV.  and  his  mother,  they  are 
not  wanting  in  natural  grace  (Fig.  311). 
The  touch  of  the  chisel  is  identical  in 
all,  and  reveals  a  common  origin.  I  may 
say  the  same  of  the  group  representing 
the  little  princess  Neferu-Ra  and  her 
guardian  Senmut.  Nothing  could  be  less 
conventional  than  the  free  gesture  with 
which  the  worthy  man  clasps  the  child, 
and  the  confident  self-abandonment  with 
which  it  nestles  against  his  breast.  The 
natural  movement  harmonises  well  with 
the  intellectual  benevolence  of  the  face 
and  the  smile  in  the  eyes  and  on  the 
thick  lips.  We  have  here  further  a  direct  proof  that  the  Thebans, 
like  the  Memphite  artists,  were  concerned  above  all  things  to 
get  likeness  in  their  portraits.  The  mummy  of  Thothmes  III. 

has  certainly  suffered;  the  face 
shrivelled  in  the  course  of  em- 
balming, and  the  shrinking  of 
the  flesh,  the  sinking  of  the 
eyes,  the  discoloration  of  the 
skin,  the  flattening  of  the  nose, 
make  it  very  different  to  what 
it  was  in  life.  Nevertheless,  if 
superficial  relief  has  been  lost, 
that  of  the  substructure  has  en- 
dured ;  when  we  compare  it 
with  the  modelling  of  the  statue, 
we  are  obliged  to  admit  that 
they  are  alike,  and  that  the 
sculptor  has  perpetuated  the 
expression  of  life  which  has 
passed  away  from  the  mummy. 
172 


FIG.    331. — STATUETTE  OF 

AMENOPHIS  IV. 
(The  Louvre,  Paris). 


FIG.  332.— HEAD  OF  ONE  OF  THE 
CANOPIC  JARS  OF  AMENOPHIS  IV. 
(Museum,  Cairo).  (Phot.  E.  Brugsdi.) 


THE  SECOND  THEBAN  AGE 


It  would  have  been 
strange  if  people  so  skil- 
ful in  rendering  the  human 
form  had  not  been  masters 
of  the  treatment  of  ani- 
mals. The  lion  and  its 
offspring  the  sphinx, 
the  ram,  the  monkey 
(Fig.  313),  the  falcon,  the 
vulture,  inspired  the  The- 
bans  in  admirable  works. 
Never  was  the  faculty 
of  welding  the  members 
of  different  beings  into 
a  single  body  carried 
further  than  in  such  cre- 
ations as  the  sphinx  of 
Queen  Hatshepset  at 
Rome,  the  sphinx  of 
Thothmes  III.  at  Cairo, 
the  Sekhet  with  the  lion's 
head,  standing  (Fig.  312) 
or  sitting,  the  various 

hieracocephalous  (falcon-headed)  and  criocephalous  (ram-headed) 
sphinxes,  and  rams.  The 
lions  of  Amenophis  III.  at 
Gebel-Barkal  (Fig.  314), 
have  a  nobility  of  atti- 
tude and  a  truth  of 
physiognomy  which  was 
always  lacking  in  the  lions 
of  the  Greek  and  Roman 
sculptors.  They  were  pro- 
ducts of  the  royal  work- 
shops, and  to  one  of 
these,  no  doubt,  we  must 
also  attribute  the  Amen- 
ophis II.  in  black  gran- 
ite, standing  and  leaning 
against  the  swelling  neck 
of  the  goddess  of  the 
dead,  the  serpent  Marit- 
sakro,  who  is  thus  in- 


FIG.   333. — WALL   IN  ONE   OF  THE  TOMBS   OF 
EL-AMARNA.     (Phot.  Bouriant.) 


FIG.  334.— AMENOPHIS  IV.   AND   THE  QUEEN 
(Museum,  Berlin). 

173 


ART  IN  EGYPT 


dicated  as  his  pro- 
tector. The  execu- 
tion here  is  minute 
and  trivial ;  the  work 
is  faithfully  rendered 
mythology  and  no- 
thing more.  But  the 
cow  discovered  by 
Naville  in  an  almost 
perfect  chapel  at  Der- 
el-Bahari  (Fig.  31 5),  is 
a  work  of  a  very  dif- 
ferent order.  This  is 
equal,  if  not  superior 
to  the  best  achieve- 
ments of  Greece  and 
Rome  in  this  genre, 
and  we  have  to  come 
down  through  the  ages 
to  the  greatest  animal  sculptors  of  our  own  days  before  we 
find  a  work  of  such  striking  reality.  She  is  encumbered  with 
mystic  emblems,  the  head-dress  of  discs  and  feathers  between 


FIG.  335. 

AMENOPHIS  IV.,   THE  QUEEN,   AND  THEIR  CHILDREN 
(Museum,  Berlin). 


FIG.   336. 

CAST   OF  THE  HEAD  OF  AMENO- 
PHIS IV.    (Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


174 


FIG.  337. — STUDY  WITH  THE  POINT 

FOR  THE  PORTRAIT  OF  A  PERSON  OF 

THE  TIME  OF  AMENOPHIS  IV. 

(Museum,  Berlin). 


THE  SECOND  THEBAN  AGE 


FIG.   338.— BUST   OF  AMON  AT 
KARNAK.      (Phot.  Legrain.) 


her  horns,  and  two  tufts  of  lotus,  springing  from  the  ground 
at  her  feet,  rise  to  her  shoulders  (Fig.  317).  In  her  the  faithful 
adored  Hathor,  posted  at  the  edge 
of  the  western  marsh  to  intercept 
those  who  had  lately  died,  and  ini- 
tiate them  into  the  life  beyond  the 
tomb ;  nevertheless ,  the  sculptor 
reduced  the  religious  paraphernalia 
to  their  simplest  expression.  Was 
it  he  who  created  the  theme,  or 
who,  in  other  words,  detached  it 
from  the  bas-reliefs  (Fig.  316)  to 
translate  it  into  the  round?  His 
goddess  is  no  conventional  cow 
modelled  upon  a  traditional  form ; 
she  is  an  individual  creature  chosen 
for  her  beauty  from  among  the 
sacred  flock.  In  spite  of  her  trap- 
pings and  her  Pharaoh ,  we  recog- 
nise in  her  the  kindly  maternal  beast, 
gentle,  strong,  vigorous,  and  natural. 
The  master  she  inspired  modelled 
the  relief  of  sides  and  hind-quarters  lovingly,  and  we  almost  see 
the  quivering  of  the  skin  under  the  caresses  of  the  light.  In 
the  head  he  even 
had  recourse  to 
technical  artifices 
which  appear  for 
the  first  time  in 
this  example ,  as 
far  as  I  know;  he 
treated  the  nostrils 
and  cheeks  with  a 
fine  rasp  or  file, 
and  the  furrows 
left  by  the  tool 
express  in  a  very 
curious  manner  the 
perpetual  tremor 
that  agitates  the 
face.  Life  has  been 

breathed  into  the  stone;  the  nostrils  quiver  with  the  breath  that 
passes   through  them,    and  the  eyes   are   half    closed  in  indolent 

175 


FIG.33g. — TRIUMPHAL  BAS-RELIEF  OF  SETI  I.  AT  KARNAK. 


ART  IN  EGYPT 


reverie.    The  figure  of  Pharaoh  does  not  rise  above  the  average ; 
I  am  nevertheless  inclined  to  think  that   the   group ,    taken   as  a 

whole,  is  the  finest  achievement  of. 
the  Theban  School  under  the  Ahmes- 
sids.  The  mutilated  statue  of  Amen- 
ophis  III.  in  Assyrian  dress  (Fig.  31 8) 
is  an  eccentricity.  The  gigantic 
group  of  Amenophis  III.  and  Thi  or 
Ti,  in  the  Cairo  Museum  (Fig.  319) 
is  a  marvel  of  purely  material  dex- 
terity, but  it  is  nothing  more,  and 
it  has  no  merit  save  the  immensity 
of  its  proportions.  The  colossal 
figures  in  red  sandstone  which  this 
same  Pharaoh  placed  at  the  entrance 
of  his  sepulchral  temple  on  the  left 
bank  of  Thebes,  the  two  Memnons, 
measure  65  feet  in  height  (Fig.  320). 
They  are  correct  in  style ,  and 
highly  elaborated;  in  their  present 
mutilated  condition,  they  are  chiefly 
effective  as  mass,  and  they  impress 
by  their  isolation  in  the  middle  of 
the  plain  rather  than  by  their  beauty. 

The  art  of  the  private  workshops 
is  perhaps  less  familiar  to  us  than 
that  of  the  royal  studios ,  but  it 
was  far  from  inferior  to  this.  The 
high  priest  Amenophis,  son  of 
Hapu ,  is  a  very  happy  creation, 
in  spite  of  those  retouches  of  the 
Ptolemaic  period  which  have  modi- 
fied the  expression  of  the  face 
(Fig.  321).  His  namesake  provides 
us  with  a  good  example  of  the 
type  of  the  seated  scribe  treated 
in  the  new  manner  (Fig.  322).  Let 
us  turn  to  the  trio  in  black  granite 
from  Karnak  (Fig.  323),  the  husband 
and  wife  seated  on  the  same  seat, 
the  child  standing  between  the  two. 
They  are  Theban  notables,  heavy 
of  form  and  insignificant  of  feature. 
176 


FIG.   340. — FRAGMENT  OF  A  STA- 
TUETTE   OF    PERTRIFIED    WOOD 

(Museum,  Cairo).  (Phot.  E.  Brugsch.) 


FIG.   341.— BUST   OF  KHOXSU 
(Museum,  Cairo).  (Phot.  E.  Brugsch.) 


THE  SECOND  THEBAN  AGE 


Sennefer  is  well  pleased  with  himself,   and   not  without  reason; 
he  is  commandant  of  the  Thebaid ;  he  wears  round  his  neck  the 
necklace    of    four    rows,    and    on    his 
breast  the  two  circular  ornaments,  the 
insignia   of   his  rank;    his  wife  was  the 
king's  foster-mother,  and  their  daughter 
appears  to  be  well  married.    The  artist 
has   fixed   on   the   stone,   perhaps  with 
a   touch   of   irony,    the    expression    of 
gratified     vanity    that    irradiated    their 
persons.    The  handling  is  very  searching, 
and  the  only  touch  of  convention  is  to 
be    found    in    the    torso    of    the    man, 
where   the   loose   folds  caused   by   age 
and  soft  living  are   noted  with  an  ex- 
cess   of    symmetry.      It    is   a   pity   that 
only  fragments  have  survived  of  a  couple 
contemporary  with    the  last  Ahmessids, 
who    were    buried     at    Sheikh -Abd-el- 
Kurnah.     The   head   only  (Fig.  324)   of 
the   man   has   come    down   to  us,    and 
even  this  has   lost   the  nose,   the  chin, 
and  part  of  the  mouth,  but  the  woman 
has   suffered   less  (Fig.  325).     In   spite 
nose,  the  face  is  charming,  with  the 
low   forehead   almost   concealed   by 
the   wig,    the    narrow    eyes   slanting 
upwards    towards    the    temples,    the 
slightly  prominent  cheek-bones,  and 
the  full  mouth,  the  corners  of  which 
melt    into    dimples.      The   cape   and 
the   pleated    robe    in    which    she    is 
draped  reveal  a  well  modelled   arm 
and  define  the  contours  of  the  bed) ; 
we   divine    beneath  the  veil    healihy 
hips ,    a    slender   waist ,    and    round, 
firm  breasts.    The  details  of  the  dress 
and  ornaments,   which  were  laid  on 
with    the    brush,    have    worn    away, 
but  the  material,  a  close,  crystalline 
limestone    resembling    alabaster,    is 
of  a  most  agreeable  creamy  tone.     The  unknown  woman  whose 
portrait,  half  the  size  of  life  (Fig.  326),  was  discovered  by  Mond 

177  N 


FIG.   342.— THE   SO-CAL- 
LED THI  (Museum,  Cairo). 
(Phot.  E.  Brugsch.) 

of   the  mutilation   of  the 


FIG.  343- — TUTANKHAMEN. 
(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


in  1906,  has  the  same  attitude  and  a  similar  costume,  an  almost 

transparent  drapery  from  which  the  left  hand  emerges,  holding 

a  lotus-flower  to  the  breast.  The  bust 
is  not  fully  developed,  and  the  breasts 
are  so  small  that  they  hardly  swell 
the  drapery  that  veils  them.  The  artist 
has  seized  the  characteristics  of  the 
first  dawn  of  womanhood  with  much 
truth  and  penetration,  and  the  discreet 
manner  in  which  he  suggests  the  over- 
slender  grace  of  the  model  under  the 
dress  is  masterly.  The  wig  is  so  in- 
geniously arranged  that  instead  of 
crushing  the  face,  it  forms  a  frame 
round  it,  and  gives  it  importance. 
This  face  changes  in  character,  and 
almost  seems  to  change  its  century, 
according  to  the  angle  at  which  we 
study  it.  Confronting  us,  it  is  round 
and  full,  without  superabundance  or 

"S:\SSSSSSoSSS?     looseness  of  flesh,   that   of  a   pleasant 

Cairo).  (Phot.  E.  Brugsch.)         little  Theban   girl,   pretty,   but   vulgar 

in  structure  and  expression.    In  profile, 

between  the  wings   of   her  wig,   which  fall   upon   her   shoulders 


FIG.    345.— ZAI  AND    NAI 

(Museum,  Cairo).    (Phot.  E.  Brugsch.) 


178 


FIG.   346.— AMON  AND   Mt,'T 
(Museum,Cairo).  (Phot.  E.  Brugsch-) 


THE  SECOND  THEBAN  AGE 


FIG.  347.— BAS-RELIEF  OF  SETII. 
AT  KARNAK.      (Phot.  Banville.) 


like  two  long  side-curls,  she  has  a  malicious  and  mutinous 
subtlety  very  uncommon  among  Egyptian  women ;  she  might 
well  pass  for  a  contemporary  of 
our  own ,  who  had  donned  an 
antique  headdress  and  costume  out 
of  caprice,  or  a  refinement  of  co- 
quetry. 

Great  as  was  the  activity  of  the 
statuary,  that  of  the  mural  sculptor 
was  not  inferior  to  it.  The  de- 
coration of  the  innumerable  temples 
which  were  built  at  this  period,  not 
only  at  Thebes,  but  in  the  provinces, 
admitted  of  very  little  novelty  in 
the  dogmatic  portions,  and  the  scenes 
depicting  the  bringing  of  offerings 
or  ritual  solemnities,  differ  but 
slightly  from  such  representations  in 
earlier  ages.  They  seem,  however, 
to  be  crowded  together  more  close- 
ly, and  to  be  more  numerous; 
perhaps  some  of  them,  such  as  the 

issuing  forth  in  procession  with  the  divine  bari,  are  much  later 
than  the  rest;  they  have  less  stiffness 
than  the  ancient  pictures,  and  greater 
richness  and  variety  in  the  accessories. 
In  the  majority,  however,  the  hieratic 
element  persists ,  and  they  do  not 
impress  the  spectator  by  rapidity 
of  movement,  nor  by  the  grouping 
of  the  figures,  but  by  correctness 
of  line  and  perfection  of  modelling. 
Go  to  Karnak,  or  to  Der-el-Bahari ; 
the  basreliefs  of  Thothmes,  of  Queen 
Hatshepset  and  of  Amenophis  are 
masterpieces  of  skilled  graving  and 
harmonious  colour.  Study,  at  Der- 
el-Bahari,  the  Queen  Aahmes-Nefert- 
ari,  who  has  come  to  the  end  of 
her  pregnancy,  and  is  being  con- 
ducted to  her  bed  of  labour  by  the 
divinities  who  protect  women  in  FIG-  348.— PAINTED  BAS-RELIEF 

.1,1  r       .  f  IN   THE   TOMB   OF   SETI  I. 

travail ;    the  expression   or  pain  and  (Phot.  Beatoj 

179  N  2 


ART  IN  EGYPT 


weariness  on  her  face,  and  the  languor  and  self-abandonment 
of  her  whole  person,  make  this  figure  a  most  accomplished  piece 

of  sculpture  (Fig.  328). 
Or  take  the  highly 
entertaining  expedi- 
tion to  Punt  as  a 
contrast  to  the  gentle 
sentiment  of  the  above. 
The  artist  was  not 
content  to  give  a 
general  impression  of 
the  voyage  of  an  Egyp- 
tian fleet  (Fig.  327); 
he  has  noted  in  detail 
the  local  scenery,  the 
conical  huts  perched 
on  piles  above  the 
level  of  inundation, 

FIG.   349.— BAS-RELIEF  IN   THE  MEMNONIUM  OF  .       , ,  WOHien  ,      the 

SETI  i.  AT  ABYDOS.    (Phot.  Beato.)  giraffes,  the  monkeys, 

the  oddly-shaped  fish. 

If,  as  I  imagine,  the  workshop  which  carved  this  fragment  of  a 
maritime  epic  in  the  limestone  was  the  same  to  which  we  owe 
the  cow,  we  need  feel  no  surprise  at  the  mastery  displayed. 

The  sculptors  of  Karnak  were  not 
so  successful  with  the  triumph  of 
Amenophis  II.  on  his  return  from 
Syria;  from  this  they  only  extracted 
the  material  for  a  series  of  vignettes, 
skilfully  combined,  but  without  charm 
or  originality.  The  private  workshops 
surpassed  the  royal  ones,  indeed,  in 
many  cases,  and  the  tombs  of  Sheikh- 
Abd  -  el  -  Kurnah  contain  the  finest 
paintings  and  bas-reliefs  of  the 
Eighteenth  Dynasty.  The  soft  and 
ductile  limestone  of  the  hill  lent 
itself  to  all  the  subtleties  and  even 
to  all  the  fantasies  of  the  chisel ; 
thanks  to  it,  the  Thebans  of  Amen- 
ophis III.  had  attained  a  mastery 
far  greater  than  that  of  the  Mem- 
phites.  Their  relief  is  rather  higher 
180 


FIG.  350.— SETI  I.   AT  ABYDOS. 

(Phot.  Beato.) 


THE  SECOND  THEBAN  AGE 


FIG.  351.— PRINCE  RAMESES, 
AFTERWARDS  RAMESES  II. 
AT  ABYDOS.  (Phot.  Beato) 


and  rounder,    and  consequently  less  dry 

than   that   of  the  Fifth  Dynasty;   at  the 

same  time    the  arrangement  of   the    epi- 
sodes is  richer  and  less  immutable.    The 

hypogea  of  luaa,  Kha-emhet,  and  a  score 

of  others,  shattered,   ravaged  and  muti- 
lated though  they  are ,    contain ,   to  my 

mind,  some  of  the  best,  perhaps  indeed 

the    best,    of    the    Egyptian    bas-reliefs. 

The  touch  in  these  is  fat,  long,  and  bold, 

the  drawing  free  and  flowing;  we  seem 
I  to  behold  Amenophis  III. ,  Queen  Thi 

(Fig.  329)  and  luaa  themselves.  Artists 
5  of  a  later  period  may  have  equalled 

these  works,   but   they  never  surpassed 

them. 

Then   suddenly,   at  the  moment  when 

Thebes  had  reached  its  apogee,  the 
v  semi-religious,  semi -political  madness  of 
j  Amenophis  IV.  compromised  the  existence 

,   of   its   art,    and   drawing   out   a   provincial   school   from  the  ob- 
scurity  in   which   it   was  vegetating,  sought  to  substitute  it  for 

that  of  Thebes.    When  he 

transferred  his   capital  to 

El-Amarna,  he  might  have 

taken  with  him  the  whole 

or   a   part   of   the  artistic 

staff  of  Karnak;  the  men 

who    had  worked    so  val- 
iantly for  his  father,   had 
j    lost  nothing  of  their  vigour; 

the  little  portrait  of  Thi 

at   Cairo   (Fig.  330),    the 

statuette    which     one     of 

them   made  of  him,   now 

in  the  Louvre  (Fig.  331), 

are  only   to   be   equalled 

by  the  marvellous  heads  of 

his  Canopic  jars  (Fig.  332) ; 

and     the     decorators     of 

the  tomb   of   his  minister 

Ramosis,     if     they     were 

not  the  same  who  worked 


FIG.  352.— SETI  I.  AND  THE  THREE  GODDESSES 
IN  THE  MEMNON1UM  AT  ABYDOS.  (Phot.  Beato.) 

181 


ART  IN  EGYPT 


FIG.  353. — SETI II.  IN  HIS 
TOMB.    (Phot.  Insinger.) 


on    that   of   luaa,   were    fully   equal    to   these.     At   El-Amarna, 

the  manner  of  attacking  the  stone  differs  entirely  from  the  Theban 
method,  and  is  rather  clumsy;  it  betrays 
an  old-fashioned  technique,  which  still 
persisted  at  Hermopolis,  the  largest  town 
in  the  region;  the  ancient  hypogea  of 
Beni-Hasan  revealed  its  character  in  the 
works  of  the  Twelfth  Dynasty.  If  we 
place  the  works  of  the  two  periods  side 
by  side,  it  will  sufficiently  prove  that  they 
are  the  outcome  of  the  same  teaching 
and  the  same  practice.  In  both  cases, 
the  isolated  figures  are  often  drawn  with 
a  clumsy  hand,  but  they  are  grouped 
well ;  they  act,  they  overflow  with  move- 
ment and  life.  The  episodes  are  taken 
from  the  lives  of  actual  persons,  and 
record  their  great  events  with  a  spirit 
quite  untrammelled  by  convention.  Of 
course,  the  world  had  progressed  since 
the  days  of  the  Twelfth  Dynasty,  and 
the  composition  had  become  more  scienti- 
fic; whereas  the  artist  of  Beni-Hasan  arranged  his  wrestlers  in 

symmetrical  groups,  his  confrere  at  El-Amarna  mingles  his  persons, 

and   makes  a  seething   crowd    of 

them.    The  king,  accompanied  by 

the  queen  and  one  of  his  daughters, 

goes  in  his  chariot  to  pray  in  the 

temple  of  Aton ,  or   he   summons 

one  of  his  favourites  to  the  palace, 

to  recompense  him  for  his  services 

by  the  gift  of  gold  necklaces.    He 

hands  them  to  the  favourite  from 

the  tribune ,  and  the  little  prin- 
cesses, amused  by  the  spectacle, 

throw  down  others  with  ingenuous 

gestures;     behind    the    decorated 

favourite ,    servants   indicate   their 

joy  by  bows  or  capers,  according 

to    their    social     rank    and    their 

education.      The    private    life    of 

royalty   is   treated   with   a   famil- 

.-'•',..,  ,  FIG.   354.— SEPTAH-MENEPTAH  IN 

ianty    hitherto     unknown    among  HIS  TOMB.    (Phot.  Weigaii.) 

182 


THE  SECOND  THEBAN  AGE 


the  Thebans;  Pharaoh  is 
seated  at  table  with  his 
family,  and  they  are  all 
gnawing  bones  covered 
with  meat;  or  he  lingers 
in  the  harem  to  play  with 
his  daughters  (Fig.  335) 
and  the  queen  offers 
him  a  bouquet  to  smell 
(Fig.  334),  or,  seated 
upon  his  knees,  nestles 
lovingly  against  him, 
while  his  children  caress 
each  other  on  a  cushion 
before  him ,  in  all  the 
innocence  of  their  age. 
The  equivalent  of  such 
motives  is  to  be  found 
elsewhere,  but  they  are 
arranged  and  raised  to 
the  dignity  of  sacra- 
mental themes;  the  no- 
velty here  is  the  realism 


•  w 


FIG.  355. — SCENE  FROM  THE  TOMB  OF  SEPTAH- 
MENEPTAH.  (Phot,  Weigall.) 


FIG.    356. 

SESOSTRIS   FIGHTING,   AT  ABU   SIMBEL. 
(Phot.   Oropesa.) 

183 


with  which  they  are  treat- 
ed. The  artists  of  El- 
Amarna  worked  from 
nature  even  more  closely 
than  those  of  Thebes;  we 
have  proof  of  this  in  the 
casts  they  took  (Fig.  336), 
the  studies  with  brush 
and  point  by  them  which 
have  come  down  to  us 
(Fig.  337),  and  the  manner 
in  which  they  rendered 
the  type  of  the  sovereign. 
Whereas  the  Thebans 
idealised  this,  they  trans- 
cribed it  as  they  saw 
it,  reproducing  the  low 
forehead ,  the  projecting 
face,  the  pointed  chin, 


ART  IN  EGYPT 


FIG.  357. — ONE   OF  THE  DAUGH- 
TERS  OF  RAMESES  II. 
(Museum,  Cairo).  (Phot.  E.  Brugsch.} 


the  thin  neck,  the  puffy  belly  and 
the  puny  limbs.  The  king  took  this 
in  good  part,  and  his  courtiers, 
adopting  the  type  themselves  in 
order  to  flatter,  did  their  best  to 
resemble  it.  The  result  of  these 
tendencies  was  a  very  individual 
style ,  less  independent  than  that 
which  prevailed  in  the  rest  of  Egypt, 
and,  above  all,  more  paradoxical; 
just  as  the  religion  or  Aton  was 
nothing  but  an  ancient  worship  sud- 
denly raised  to  the  first  rank  in 
order  to  check  the  disquieting  im- 
portance of  Amon,  so  the  school 
of  El-Amarna  was  but  an  ancient 
school  drawn  forth  from  its  obscurity 
by  the  will  of  the  master,  and  un- 
expectedly transformed  by  him  into 
a  royal  workshop. 

If  the  enterprise  of  Amenophis  IV.  had  succeeded,  would  the 
influence  of  these  Hermopolitans  have  supplanted  that  of  the 
Thebans?  It  is  highly  improbable;  the  Thebans  were  too  skil- 
fully organised,  and  possessed  too 
numerous  a  clientele  for  a  rapid  eclipse 
of  their  prosperity.  When,  after  some 
ten  or  fifteen  years ,  the  city  of  El- 
Amarna  was  abandoned,  and  the  artists 
who  had  shed  lustre  upon  it  relapsed 
into  obscurity,  the  Theban  School 
easily  resumed  its  place  as  the  official 
school  of  royalty.  They  did  not,  how- 
ever, take  up  things  again  at  the 
precise  point  where  they  had  been  left 
at  the  death  of  Amenophis  III.,  at  least 
as  far  as  the  bas-relief  was  concerned. 
We  have  only  to  study  the  pictures 
with  which  Tutankhamen  and  Heru-em- 
heb  adorned  the  lateral  walls  of  the 
great  colonnade  at  Luxor  to  recognise 
the  influence  of  Hermopolitan  ideas. 
Amon  comes  forth  from  Luxor  towards 
Karnak  through  the  streets  of  the  city 
184 


FIG.   358.— ALABASTER 

STATUE   OF   RAMESES   II. 

(Museum,  Turin).   (Phot.  Alinari.) 


FOUR  OF  THE  ENAMELLED  PLAQUES 

From  the  Palace  of  Rameses  III.  at  Mediiet-H»bu 

(Museum,  Cairo) 


THE  SECOND  THEBAN  AGE 


FIG.  359.— REPLICA  OF  THE  RAMESES  n. 

AT  TURIN  (Museum,  Cairo). 

(Phot.  E.  Brugsdi.) 


and  upon  the  Nile,  and  the  po- 
pulation rejoices  round  him ;  the 
composition  has  the  regularity 
and  balance  of  Theban  art,  but 
many  of  the  accessory  scenes, 
feasts  in  the  houses,  singing, 
dancing,  military  ballets,  seem 
to  have  been  borrowed  from 
that  of  El-Amarna.  Heru-em- 
heb's  master  draughtsman  had 
studied  the  work  of  his  pro- 
vincial brethren,  and  had  gleaned 
from  it  certain  ideas  for  the 
rejuvenation  of  the  traditional 
designs.  Some  touch  of  this 
inspiration  seems  to  have  passed 
into  the  triumphal  reliefs  of 
Seti  I.  (Fig.  339)  and  Rameses  II., 
but  it  vanishes  almost  immedi- 
ately after  these,  and,  in  any  case, 
it  had  no  influence  at  all  upon 
sculpture.  This  art,  hampered 
for  some  years  by  persecution ,  recovered  its  vigour  as  soon 
as  the  heresy  died  out,  and  the  royal  statuaries  of  Karnak 
produced  a  series  of  works  comparable  to  the  finest  of  their 
earlier  achievements.  They  comprise,  besides  the  bas-reliefs  of 
Heru-em-heb  on  one  of  his  pylons,  the  Amon  (Fig.  338),  which 
is  a  portrait  of  this  Pha- 
raoh ,  the  Khonsu  and 
the  Tutankhamen,  the  so- 
called  head  of  Thi  at 
Cairo,  the  group  of  Heru- 
em-heb  and  Amon  at 
Turin,  perhaps  the  Cairo 
bust  of  petrified  wood 
(Fig.  340) ,  and  a  few 

Pieces  of  less  interest, 
think  it  evident  that  the 
Khonsu  (Fig.  341)  and  the 
Tutankhamen  (Fig.  343) 
are  by  the  same  hand. 
The  two  figures  might 

1  -  °  °,  FIG.   360.— THE    TWO    COLOSSI    ON    THE    SOUTH 

almost     be     superposed ;  SIDE  AT  ABU  SIMBEL.    (Phot.  Beato.) 

185 


ART  IN  EGYPT 


FIG.   J6l. — COLOSSAL    HEAD    OF 

RABIESES  II.    (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


the  hollow  of  the  eye  is  of  the 
same  depth  in  each,  the  junction 
of  the  nose  is  identical,  as  are  the 
slight  inflation  of  the  nostrils,  the 
pout  of  the  lips,  and  the  constriction 
of  the  corners  of  the  mouth.  The 
expression  of  suffering  is  common 
to  both  faces,  but  the  indications 
of  ill-health ,  the  obliquity  and 
sunkenness  of  the  eyes ,  the 
thinness  of  the  cheeks  and  neck, 
and  the  projecting  shoulder-blades 
are  more  marked  in  the  Khonsu 
than  in  the  Tutankhamen ;  they  be- 
tray consumptive  tendencies  which 
the  artist  has  noted  with  sufficieni 
realism  to  justify,  the  diagnosis 
of  a  modern  physician.  The  group 
of  Heru-em-heb  and  Amon  is  less 
personal  in  sentiment;  but  the  two 

faces  have  a  beautiful  expression ,  and  the  technique  resembles 
that  of  the  others.  Its  affinities  with  the  so-called  Teye  are 
perhaps  less  obvious  to  those  who  know  this  only  by  drawings 
or  photographs  (Fig.  342),  but  they  become 
evident  enough  to  a  student  of  the  originals, 
and  the  peculiarities  of  the  Khonsu  and  the 
Tutankhamen  re-appear  here  in  a  modified 
form.  The  queen  is  not  a  consumptive,  but 
the  various  parts  of  her  face  indicate  great 
delicacy,  and  the  hand  which  modelled  them 
is  certainly  that  which  treated  with  so  much 
subtlety  the  contemporary  images  of  the  god 
and  the  Pharaoh.  The  Turin  group  has  the 
solemn  impress  suitable  to  the  subject,  the 
adoption  of  the  sovereign  by  his  father  Amon, 
and  his  enthronement;  the  two  heads  are 
marked,  nevertheless,  by  that  air  of  some- 
what sickly  gentleness  which  characterises 
the  others.  I  should  be  inclined  for  the 
same  reason  to  class  with  these  the  per- 

FIG.  362.— RAMESES  iv.    sonage  called    by    Mariette    Meneptah ,    who 
is  in  reality  Heru-em-heb  (Fig.  344).    Here, 
as  with   the   Khonsu    and  the  Tutankhamen, 
186 


AND  A  LIBYAN  PRISONER 

(Museum,   Cairo). 
(Phot.  E.  Brugsch.) 


THE  SECOND  THEBAN  AGE 


FIG.  363..— RUST  OF  A  STATUE 

OF  MENEPTAH  (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


the  hardness  of  the  material,  a 
close-grained  granite,  offered  serious 
difficulties  to  the  sculptor.  He 
overcame  them  with  almost  insolent 
success,  and  this  anxious,  refined 
face  is  an  unforgettable  creation. 

Seti  I.  then  received  from  the 
Eighteenth  Dynasty  the  Theban 
School  in  the  full  tide  of  its  pros- 
perity, and  he  kept  it  at  this  high 
level.  The  few  statues  of  his 
period  that  we  have,  the  group 
of  Amon  and  Mut  in  the  Cairo 
Museum  (Fig.  346),  and  that  of 
Zai  and  Na'i  (Fig.  345),  are  charming 
works,  marked  by  a  sentiment  and 
distinction  which  were  never  sur- 
passed in  the  sequel.  It  was  indeed 
characteristic  of  Theban  art  under 
this  prince,  that,  refining  still  more 

upon  the  tendency  of  Heru-em-heb,  it  sought  grace  and  elegance 
rather  than  grandeur  and  energy.  True,  the  religious  and 
triumphal  bas-reliefs  (Fig.  347)  of  the  temple  of  Karnak  show 
that  on  occasion  it  did  not  lack  breadth 
and  vigour ,  but  these  are ,  after  all, 
mere  bravura  pieces  without  any  personal 
accent,  while  the  character  of  the  period 
is  revealed  in  all  its  purity  at  the  temple 
of  Kurnah,  the  hypogeum  of  Biban-el- 
Muluk  (Fig.  348)  and  the  Memnonium 
of  Abydos.  The  pictures  in  the  hypo- 
geum are  not  all  finished,  and  entire 
halls  where  the  designer  has  finished 
his  task  while  the  sculptor  has  never 
begun  his,  are  decorated  simply  with 
sketches  in  red  and  black  inks.  These 
show  very  vividly  the  great  practical 
dexterity  of  the  ordinary  workmen,  and 
it  would  be  difficult  to  appreciate  too 
highly  the  skill  with  which  the  director, 
revising  the  labour  of  his  assistants, 
laid  the  impress  of  his  own  talent  upon  it  by  means  of  a  few 
discreet  re-touches  (Fig.  349).  The  general  effect,  however,  is 

187 


FIG.    364.— HEAD   OF  PHA- 
RAOH (Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


sad  and  solemn,  as  befits  a  tomb ;  it  is  only  at  Abydos  that  we 
get  the   full    measure   of   the   genius  of   the    school.     Study  the 

original  in  situ,  and  you  will  not  doubt 
that  the  same  master  designed  the 
decorations  both  of  temple  and  hypo- 
geum,  nor  that  some  at  least  of  the 
same  assistants  collaborated  in  the 
two.  The  relief  is  at  once  flexible 
and  precise  (Fig.  349),  a  surface  which 
the  chisel  lingered  over  lovingly,  giving 
a  kind  of  colour  to  the  epidermis  by 
a  multitude  of  almost  imperceptible 
strokes.  The  gods  and  goddesses 
have  the  features  of  the  sovereign, 
and  this  oft-repeated  profile  is  differ- 
entiated each  time  by  a  new  shade 
of  melancholy  languor.  To  have  seen 
the  Pharaoh  and  the  three  goddesses 
his  companions  (Fig.  352),  about  ten 
o'clock  on  a  fine  February  morning, 
is  to  understand  to  what  a  degree 
Egyptian  art,  so  mournful  superficially, 
may  kindle  with  life  and  exquisite 

tenderness.  The  funerary  workshop,  as  distinguished  from  that 
of  Karnak,  persist- 
ed under  the  suc- 
cessors of  Seti  I. 
We  owe  to  it  the 
Memnonium  of  Ra- 
meses  II.  at  Aby- 
dos, now  half  de- 
stroyed, and  what 
remains  of  the 
battle  -  scenes  of 
Kadesh  prove  to 
us  that  upon  oc- 
casion it  was  ca- 
pable of  boldness. 
After  the  com- 
pletion of  the  Mem- 
nonium, the  work- 
shop was  removed  to  the  left  bank,  and  devoted  itself  to  the 
decoration  of  the  royal  hypogea;  that  of  Rameses  II.  is  almost 

188 


FIG.  365.— THE  PRIEST  WITH  THE 

MONKEY  (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


FIG.   366.— A  WALL   OF  THE  HYPOGEUM    OF  PAHERI  AT 
EL-KAB.     (Phot.  Insinger.) 


THE  SECOND  THEBAN  AGE 


equal  to  that  of  Seti  I.,  and  in  those  of 
Meneptah,  Seti  II.  (Fig.  353)  and  Septan  Menep- 
tah,  we  find  isolated  figures  (Fig.  354)  and 
scenes  (Fig.  355)  which  prove  that  it  had  long 
possessed  gifted  artists.  After  Septah,  all  traces 
of  it  are  lost ,  and  it  is  probable  that  it  had 
ceased  to  exist  at  the  time  of  Rameses  III, 
at  the  beginning  of  the  Twentieth  Dynasty. 

The  period  of  Rameses  II.  has  often  been 
pronounced  the  beginning  of  the  artistic  de- 
cadence. No  opinion  could  be  less  justifiable. 
Rameses  II.  built  enormously  during  his  reign 
of  sixty-seven  years,  and  he  was  no  doubt  ob- 
liged to  employ  all  available  artists,  good  and 
bad  alike,  to  satisfy  the  exigencies  of  his 
monumental  mania:  wherever  the  traveller  goes 
in  Egypt,  he  will  almost  certainly  encounter 
a  stele  wich  bears  his  name,  a  statue,  a  votive 
bas-relief,  a  chapel ,  a  temple  of  his  period. 
For  the  most  part,  these  are  the  works  of  local 
artists,  and  are  no  more  important  than  their 
authors;  in  Nubia,  for  in- 
stance, it  is  difficult  to  imag- 
ine the  depths  of  barbaric 


FIG.   367. 

COLOSSAL    FIGURE 
OF   RAMESES  II. 
(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


FIG.    J68.— MEMPHITE 

PTAH  (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


clumsiness  into  which  the  artisans  who  worked 
on  the  hemispeos  of  Der  and  that  of  es-Sabua 
descended.  Yet  it  would  be  incorrect  to  say 
that  even  here,  we  recognise  decadence,  for 
there  can  be  no  decadence  where  there  is 
no  art,  and  such  is  the  case  in  these  two 
temples.  If  we  wish  to  form  a  just  idea  of 
art  under  Rameses  II. ,  we  must  study  it 
where  he  maintained  a  duly  organised  body 
of  craftsmen,  at  Thebes  and  in  its  de- 
pendencies, at  Abydos,  Memphis  and  Tanis; 
we  shall  see  then  that  it  compares  favourably 
with  that  of  preceding  ages.  The  triumphal/ 
bas-reliefs  of  Luxor,  Karnak,  the  Ramesseum,\ 
and  Abu  Simbel,  all  of  which  represent  the 
battle  of  Kadesh,  are  masterly  in  their  general  \ 
treatment,  and  the  artists  who  composed  them 
showed  great  fecundity  of  imagination  when 
they  had  to  adapt  this  single  subject  to  the 
189 


ART  IN  EGYPT 


FIG.  369.— THE  TRIAD  OF 
HERACLEOPOLIS  (Museum, 
Cairo).  (Phot.  E.  Brugsch.) 


varying  conditions  in  each  locality, 
resulting  from  the  shape  and  size  of 
the  panels  they  had  to  cover.  The 
religious  bas-reliefs  are  more  strictly 
conventional;  at  Thebes,  however,  they 
are  of  excellent  quality.  If  they  become 
feebler  in  certain  portions  of  the 
Memnonium  at  Abydos,  it  was  be- 
cause, after  having  re-established  the 
workshops  of  Seti  I.  in  the  capital, 
Rameses  II.  was  forced  to  make  use 
of  local  craftsmen  of  an  inferior  type. 
In  Nubia,  at  Abu  Simbel  and  Bet-el- 
Wali,  whither  he  deported  Theban 
artists,  the  snulpture  retains  its  sterling 
character.  Abu  Simbel  possesses  the 
work  which  Champollion  pronounced, 
not  without  reason,  the  masterpiece 
of  Egyptian  bas-relief,  the  Sesostris 
fighting  (Fig.  356).  It  is  somewhat 
harsh  of  aspect,  owing  to  the  coarse  texture  of  the  sandstone, 
but  the  composition  and  design  are  extraordinarily  perfect.  The 
king  has  already  struck  down  one  Libyan  chief,  and  trampling 

him  underfoot ,  he  seizes 
another  by  the  arm,  to 
thrust  him  through  with 
his  spear;  every  muscle 
is  strained  by  the  move- 
ment, and  his  whole  body 
shoots  forward  to  partici- 
pate in  the  effort  by  which 
he  throws  himself  upon  the 
enemy.  His  opponent  is 
no  longer  to  be  numbered 
among  the  living.  His 
eyelids  quiver,  his  mouth 
relaxes,  his  head  sinks  and 
droops,  his  legs  give  way 
under  him.  The  remnant 
of  life  in  him  is  concen- 
trated in  the  bust,  and 

flutters  feebly   under  the  sharp  point   that   pierces  the  flesh ;   as 
soon  as  the  conqueror  releases  him,  he  will  fall  in  a  heap,   and 

190 


FIG.   370. 

FRAGMENT   OF  A  MEMPHITE  BAS-RELIEF 
(Museum,  Cairo).     (Phot.  Quibell.) 


THE  SECOND  THEBAN  AGE 


FIG.  371.— MOURNERS  CRYING  ALOUD  IN  A  FUNERAL 

PROCESSION  IN  THE  TOMB  OF  HARMIN 

(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


will  move  no  more.    Never  has  the  drama  of  violent  death,  which 

disintegrates  the  whole  man  at  a  stroke,  and  stretches  him  inert 

on     the      ground, 

been  analysed  with 

such      knowledge, 

or     realised    with 

such  energy. 

The  statuary  is 
not  perhaps  equal 
to  the  bas-reliefs, 
although  it  has  be- 
queathed us  sever- 
al very  agreeable 
examples,  such  as 
the  portrait  of  one 
of  the  daughters  of 
Pharaoh  (Fig.  357). 
In  general  it  is  marked  by  a  paradoxical  double  tendency,  on 
the  one  hand  to  the  puerile,  on  the  other  to  the  gigantic.  The 
alabaster  Rameses  of  the  Turin  Museum  (Fig.  358),  with  its  softly 
rounded  contours,  is  still  governed  by  the  tradition  of  Seti  I., 
and  it  belongs  to  the  Theban  School,  as  is  proved  by  the  dis- 
covery of  a  replica  in  granite  (Fig.  359)  which  Legrain  found  in 
the  favissa  at  Karnak.  The  same  is  undoubtedly  the  case  of  the 
colossal  figures  of  the 
Ramesseum  and  Abu 
Simbel.  Those  of  the 
Ramesseum  have  suf- 
fered so  much  that 
the  only  sentiment 
they  evoke  is  aston- 
ishment at  their  im- 
mensity, but  those  at 
Abu  Simbel  fully  de- 
serve the  enthusiasm 
they  inspire  in  tra- 
vellers. I  have  studied 
them  by  day  and 
by  night,  from  every 
angle,  and  under  every 
play  of  light  (Fig.  360). 

At  morning,  in  the  pale  light  of  dawn,  they  seem  to  be  sounding  the 
distant  horizon  with  a  hard,  sombre  gaze;  but  soon,  when  the  sun, 

191  ^ 


FIG.  372.— CONVEYANCE  OF  OFFERINGS  IN  THE  TOMB 
OF  HARMIN  (Museum,  Cairo).  (Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


gliding  over  the  mountain-slope,  reaches  their  faces,  their  eyes  light 
up,  their  lips  tremble  and  smile,  and  for  a  moment  it  seems  as  if 
a  quiver  of  life  ran  through  their  bodies.  We  ask  ourselves  how 
the  master  who  created  them  managed  to  give  them  such  per- 
fect proportions,  on  an  incline  where  it  was  impossible  for  him 
to  stand  back,  and  where  he  could  only  begin  to  judge  of  his 

work  when  it  was  already 
far  advanced.  The  most 
remarkable  thing  about 
these  figures  is  the  manner 
in  which  they  harmonise 
with  the  landscape;  it  is 
impossible  to  conceive  of 
them  elsewhere,  or,  being 
here,  that  they  should  be 
other  than  they  are.  We 
find  it  difficult  to  imagine 
the  colossal  figures  of 
Memphis  or  Tanis ;  placed 
in  the  court  of  the  temple, 
like  those  of  the  Rames- 
seum,  they  must  have 
been  out  of  proportion 
with  the  statues  and 
buildings  that  surrounded 
them,  and  they  cannot 
have  blended  harmon- 
iously with  the  general 
structure  as  do  those  of 
Abu  Simbel.  They  were 
distinguished ,  neverthe- 
less, by  technical  qualities 
them;  the  most  famous 
and  the  figure  from 


FIG.  373-— FRAGMENT  FROM  THE  TOMB  OF 
MAIPTAH  (Museum,  Cairo).  (Phot.  E.  Brugsdi.) 


which    make   it    impossible   to    forget 
of    them,    the   Abu' I -hoi  of   Mitrahineh , 

which  the  colossal  head  in  the  Cairo  Museum  (Fig.  361)  was 
detached,  show  no  signs  of  decadence.  The  decline,  indeed,  first 
declared  itself  after  the  the  death  of  Rameses  II.,  during  the 
civil  wars  and  foreign  invasions  which  darkened  the  last  years  of 
the  Nineteenth  Dynasty.  It  was  already  manifest  under  Meneptah 
(Fig.  363),  and  more  emphatically  under  Rameses  III.,  who  copied 
his  illustrious  ancestor  heavily  and  clumsily.  The  sculptures  of 
Medinet-Habu  will  not  bear  comparison  with  those  of  Abu  Simbel 
or  Luxor,  although  some  of  the  pictures,  those  of  the  lion-hunt 

192 


THE  SECOND  THEBAN  AGE 


FIG.    374.— THE    TWO     NILE 

FIGURES  (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


and  the  aurochs-hunt,  for  instance,  are 

very  spirited;    it   is   probable   that  the 

king's  individual  taste  militated  against 

the  revival   of   the   school.     After  him, 

under  the  Twentieth  Dynasty,  some  re- 

spectable works  were  produced,  which 

do  not  rise  greatly  above  the  average; 

among  the  best  are  a  head  of  Pharaoh 

in  a  helmet,  with  thick   lips,   an  enor- 

mous nose,  and  heavy  eyes  (Fig.  364), 

and    a   little   group   in   granite  of   Ra- 

meses  IV.  conducting  a  Libyan  prisoner 

to  the  god  Amon  (Fig.  362).    There  is 

pride    in    Pharaoh's    bearing,    the    bar- 

barian's constrained  attitude  is  skilfully 

observed,    and    the    movement    of    the 

miniature     lion     which     slips     in     be- 

tween   the    two    is    rendered   with    the 

naturalism     proper     to     the     Egyptian 

artist    in     the    treatment    of    animals. 

The  priest  with  the  monkey  (Fig.  365),  or,  to  call  him  by  his  name, 

Rameses-nakht  ,  it  seated  on   the   ground,    studying  with  an  ab- 

stracted air   the  contents  of  a  scroll  spread   across  his  legs.    A 

little   hairy   monkey-headed   creature,    the 

god  Thoth,  perches  on  his  shoulder,    and 

reads   with   him.     It   was   difficult   to   co- 

ordinate   the    man    and    the    animal    in   a 

manner  not  ungraceful  ;  the  sculptor  solved 

the  problem   very  creditably.     The   priest 

bends  his  neck  a  little,   but   we  feel   that 

the  monkey  -god,  who  is  partly  concealed 

by  the  head-dress,  does  not  weigh  heavily 

upon  him.    The  style  is  Theban,  but  there 

is  greater  freedom  than  in  the  Rameses  IV.  ; 

the  latter  was  no  doubt  a  product  of  one 

of   the    royal   workshops,    while   the   man 

with   the    monkey  came   from  one   of   the 

private  workshops  at  Karnak. 

The  Theban  school,  which  had  enjoyed 
such  brilliant  opportunities  while  Thebes 
maintained  its  ascendancy  in  Egypt,  declined 


FIG.     375.— PIECE     OF     IN- 

rapidly    when    the    political    and    military      CRUSTATION    IN    GREEN 

J    -          -         -  r-  -     -  -"          ENAMEL   (Museum,    Cairo). 

(Phot.  E.  Brugsch.) 


power  devolved  on  the  cities  of  the  north. 


193 


ART  IN  EGYPT 


In  what  particular  forms  art  was  mani- 
fested in  the  provinces  during  these  long 
centuries  it  is  not  possible  to  say  with 
any  certainty,  in  view  of  the  scarcity 
of  survivals.  The  bas-reliefs  of  the  hy- 
pogea  of  El-Kab  (Fig.  366)  and  of  the 
temples  of  Elephantine,  seem  to  indicate 
that  those  of  the  south  were  under  Theban 
influences,  and  also  those  of  the  centre, 
if  we  may  judge  by  the  colossal  figure 
of  Rameses  II.  from  Hermopolis  (Fig.  367) 
and  the  triad  of  Heracleopolis  (Fig.  369). 
The  Memphite  School  prospered,  as  we 
know  by  the  inscriptions,  which  tell  us 
how  many  temples  the  Ahmessids  and 
Ramessids  built  or  restored  in  the  second 
of  their  capitals,  but,  with  the  exception 
F™n™T  m  COF"  of  certain  colossal  figures  of  Rameses  II. 

FINS     OB      THUAA     (Museum  ri-iti  11  i  i      i 

Cairo).    (Phot.  Quibeii.)         or  which  1  have  already  spoken,  and  the 
two   great  Ptahs   in   the  Cairo  Museum, 

(Fig.  368),  we  possess  hardly  anything  which  can  be  ascribed  to 
it  with  certainty.  It  has  been  thought  that  Theban  influences 
are  to  be  recognised  in  several  statues,  and  this  in  indubitable 


FIG.  377. — BUST  ON  THK  COFFIN  OF 

RAMESES  II.   (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


194 


FIG.   378. 

SANDSTONE    BUST    OF    RAMESES   IV. 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


THE  SECOND  THEBAN  AGE 


in  the  bas-reliefs  from  contemporary   tombs  preserved   at  Cairo. 

Might  we  not  suppose  that  the  personage  of  Fig.  370  came 
from  the  tomb 
of  luaa  orKha- 
emhet  ?  The 
scenes  of  do- 
mestic life 
and  of  funer- 
ary rites  are 
no  longer  ar- 
ranged in  the 
antique  fash- 
ion ,  in  de- 
corously co- 
ordinated re- 
gisters ,  with 
persons  walk- 
ing one  be- 
hind the  other. 
The  compo- 
sition and  per- 
spective are 
distinctly  The- 


FIG.    379.— STATUETTE   OF 

AMENOPHISII.  IN  BLACKENED 

WOOD  (Museum,  Caiio). 

(Phot.  E.  Brugsch.) 


FIG.   380.— STATUETTE  OF 

HERU-EM-HEB   IN  BLACKENED 

WOOD  (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


ban,  and  this  is  especially  noticeable  in  the  tomb  of  Harmin 
(Fig.  372),  in  the  picture  of  the  farewell  to  the  dead  (Fig.  371); 
the  weepers  and  the  women  of  the  family  defile  before  the 
mummy,  jumping,  dancing,  tearing  their  hair,  beating  tambourines, 
while  the  men  run  to  and  fro,  waving  long  reeds  to  keep  away 
evil  spirits ;  the 
excited  throng  has 
all  the  realism  of 
the  times  that  fol- 
lowed immediately 
after  Amenophis  IV. 
The  same  may 
be  said  of  the 
fragments  of  the 
tomb  of  Maiptah 

/C1-  OTO\  FIG-  3^1.— ARM    OF  A   CHAIR   IN   THK  FORM  OF  A  FELINE 

(rig.       616);  the  ANIMAL  (Museum,  Cairo).     (Phot.  E.  Brugsch.) 

dancer  of  the  first 

register  and  the  carpenter  of  the  second  would  not  be  out  of 
place  in  the  finest  tombs  of  Sheikh-Abd-el-Kurnah  or  El-Amarna. 
The  artists  of  the  Delta,  with  the  exception  of  the  Tanites, 

195  O  2 


ART  IN  EGYPT 


FIG.   382. 

JEWEL  CASKET  OF  AMENOPHIS  III. 
(Museum,   Cairo).     (Phot.  E.  Brugsch.) 


have  left  us  almost  nothing; 
and  here  the  perpetual  usurp- 
ation of  the  statues  of  the  first 
Theban  period  by  Rameses  II. 
is  very  perplexing,  when  we 
attempt  to  show  what  they  owed 
to  the  Nineteenth  Dynasty.  It 
is  only  after  the  accession  of 
Smendes  that  we  find  a  few 
pieces  of  a  well-marked  style, 
notably  the  twin  figures  of  Nilus 
in  the  Cairo  Museum  (Fig.  374), 
which  bring  symmetrically  ar- 
ranged offerings  of  flowers  and 
fish  to  the  gods.  They  resemble 
the  sphinx  of  Amenemhat  III.  in 
technique,  but  the  handling  is 
softer;  none  the  less,  they 
demonstrate  the  persistence  of 
the  ancient  local  art,  and  only 
some  happy  accident  is  needed  to  reveal  even  more  characteristic 
works.  Such  as  they  are,  they  justify  the  belief  that  the  north 
of  Egypt  was  not  behind  the  south  artistically,  and  that  her 
masters,  if  they  did  not  produce  so  many  remarkable  works  as 

those  of  Thebes,  were  capable 
of  carrying  on  the  tradition  of 
their  founders,  and  of  trans- 
mitting it,  with  undiminished  lustre, 
to  future  generations. 

C.  THE  MINOR  ARTS. 
The  minor  arts,  like  the  major, 
had  achieved  perfection,  and 
alone  would  furnish  materials  for 
a  history.  Pottery,  domestic  and 
funerary  furniture,  arms,  jewels 
and  goldsmiths'  work,  are  all 
the  natural  development  of  what 
had  existed  in  earlier  ages,  and 
yet  we  find  on  every  hand  com- 
binations formerly  unused,  and 
elements  many  of  which  are 
foreign.  The  Shepherds  had 
196 


FIG.   383.— THE    EMPIRE  ARM-CHAIR 
(Museum,   Cairo).     (Phot.  E.  Brugsch.) 


THE  SECOND  THEBAN  AGE 

brought  from  Asia  objects  of  common  use  and  of  martial  equip- 
ment hitherto  unknown,  among  them  the  chariot  and  the  quiver; 
conquest     and     trade     introduced     others,     and 
fashion   favouring  their  adoption,   the  Egyptians 
of  the  Eighteenth  and  Nineteenth  Dynasties  ended 
by   assimilating   table-utensils,    arms,    and    orna- 
ments,   a  certain    number   of  which  were    direct 
copies    of    Amorrhaean ,    Assyrian ,    Asiatic    and 
./Egean   models,  while   the  rest  betray   more  or 
less  markedly  the  influence  of  the  Eastern  Medi- 
terranean races.   When  we  compare  the  Mycenaean 
daggers  and  dishes  with  the  Egyptian,  and  note 
their    obvious    affinity,    it    is    very    difficult    to 
decide    which   was   modelled    on ,    or   simply   in- 
spired  by    the    other;    to    assert,    as    has    been 
done,     that    the    Egyptians    were     the    plagiar- 
ists,   is    to    make   a   statement  which   cannot   be 
proved,   while  study  tends  to  indicate  the  exact 
opposite.      We   must   further   take  into  account 
those  reactions  and  evolutions  by  means  of  which 
motives    and    forms    long    since    sent    forth    to 
foreign  lands,  often  return  to  their  original  homes 
with  new  arrangements.   If  we  consider  the  prestige 
enjoyed  by  Egypt  among  barbarous  peoples,  and 
her   supremacy   over    them ,    we   are    obliged   to    admit   a  priori 
that  she  was  likely  to  have  given  them  at  least  as  much  as  she 
borrowed  from  them;   though,  on  the    other   hand,   it  is  not  to 
be     denied     that    entire 
branches  of  her  industry, 
•such    as    ceramics,    were 
borrowed.    She  took  pos- 
session    of     the     various 
forms    of    the    Mycenaean 
vases ,    their   double   lips, 
their    twin    bodies,    their 
handles,  their  necks,   but 
[she    decorated    them    by 
processes     of     her     own, 
notably  by  covering'  them  FIG.  385.— BOX  FOR  COSMETICS 

With     enamel,      that     vivid  (Museum,  Cairo).    (Phot.  E.  Brugsch.) 

blue  enamel,  so  pure  and 

so  grateful  to  the  eye,  which  her  potters  had  recently  invented. 
Even  here,  her  inventive  spirit  did  not  fail  her,  and  her  adaptation 

197 


FIG.  384.— WOOD- 
EN SPOON  FOR 

COSMETICS 
(The  Louvre,  Paris). 


ART  IN  EGYPT 


of  foreign  forms  did  not  prevent  her  from  creating  some  that  were 
peculiarly  her  own.  I  need  only  instance  those  polished  saucers, 
and  beautiful  cups  of  blue  pottery  in  the  form  of  lotus-flowers, 
or  those  red  and  green  glazed  kohol-pots,  some  of  which  are  in 
the  forms  of  mitred  falcons,  hedgehogs,  monkeys  and  the  god 
Bes.  The-  doors  and  fagades  of  the  palaces  of  Amenophis  I. 
and  of  Rameses  III.  were  ornamented  with  polychrome  plaques 
of  pottery  incrusted  in  the  walls,  showing  Pharaoh  himself 
adoring  the  gods  (Fig.  375),  friezes  of  flowers  and  birds,  and 
rows  of  prisoners  (PI.  IV).  We  have 
further  a  whole  array  of  peculiarly 
Egyptian  ornament  in  those  necklaces 
and  bracelets  of  glazed  earth  or  coloured 
glass  pastes,  florets,  discs,  rings,  beads, 
pendants,  cartouches,  little  plaques 
covered  with  figures  or  hieroglyphs, 
which  were  the  luxury  of  the  poor,  and 
objects  of  common  use  among  the  middle 
classes.  This  again  was  the  period  when 
the  vast  family  of  amulets  began  to 
make  their  appearance ,  scarabs ,  girdle- 
knots,  little  columns,  mystic  eyes,  hawks, 
frogs,  and  twenty  other  forms  which 
fill  the  glass-cases  of  our  museums.  Many 
of  these  are  perfect  marvels.  The  fune- 
rary figurines,  the  ushebtis,  which  took 
the  place  of  the  deceased  for  the  per- 
formance of  irksome  tasks  in  the  paradise 
of  Osiris,  were  often  as  carefully  exe- 
cuted as  the  large  statues;  there  are 
some  of  enamelled  porcelain,  such  as  those  of  Thothmes  IV. 
and  Ptahmes,  which  modern  industry  down  to  the  present 
despairs  of  copying  to  perfection,  and  it  would  be  difficult 
to  say  too  much  in  praise  of  those  made  of  limestone,  painted 
wood,  green  or  blue  composition,  and  in  some  rare  cases, 
of  bronze,  which  come  from  private  tombs.  The  Theban  crafts- 
men were  neither  less  skilful  nor  less  inventive  than  the  artists 
strictly  so-called. 

This  is  most  evident  among  the  wood-carvers.  The  particular 
idea  the  Egyptians  had  of  death  gave  rise  to  various  artistic 
forms  among  them ,  which  no  longer  exist  among  us ,  or  which 
have  become  purely  utilitarian.  Among  these  were  sleighs  for 
conveying  mummies  to  the  tomb ,  boxes  and  chests  for  Canopic 

198 


FIG.    386.— BOX  FOR   COS- 
METICS   (Museum,   Cairo). 
(Phot.   E.   Brugsch.) 


THE  SECOND  THEBAN  AGE 


jars   and   figurines ,   black   sarcophagi  with   gilded   figures ,   and, 
above  all,  the  mummy-shaped  coffins.    On  these,  in  many  cases, 


FIG.    387. — PERFUME-SPOON  IX  THE  FORM   OF  A,  WOMAN    SWIMMING 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 

the  artisan  reproduced  in  the  mask  the  features  of  the  corpse 
within;  some  of  them  may  be  compared  for  truth  of  modelling 
and  richness  of  ornament  to  the  best  productions  of  the  royal 
schools;  among  the  finest  are  the  coffins  of  Thuaa  and  luaa 
(Fig.  376),  the  father  and  mother  of  Queen 
Thi,  gilded  and  incrusled  with  stones  or  glass 
paste,  and  that  of  Rameses  II.  (Fig.  377)  exe- 
cuted at  the  end  of  the  Twentieth  Dynasty 
to  replace  the  original  destroyed  by  robbers. 
These  show  all  the  qualities  of  great  sculpture, 
vigour,  expression  and  grace,  qualities  which 
persisted  among  the  Theban  funeral-furnishers 
after  they  had  begun  to  die  out  in  the  ordin- 
ary workshops;  we  cannot  but  admit  this 
when  we  compare  the  coffin  of  Rameses  II. 
with  the  stone  statue  of  Rameses  IV.  (Fig.  378), 
which  is  earlier  by  some  years.  The  images 
of  gods  and  kings  which  were  placed  in  the 
hypogea  after  serving  for  the  rites  of  sepulture, 
were  less  carefully  executed  than  the  coffins, 
but  they  retained  a  certain  grandeur  of  ap- 
pearance, if  we  may  judge  by  the  fragments 
preserved  in  the  Cairo  Museum.  Among  these 
are  figures  of  Thothmes  III.,  Amenophis  II. 
(Fig.  379)  and  Heru-em-heb  (Fig.  380)  carved 
in  cedar  or  pine-wood,  then  coated  with  pitch 
or  bitumen  in  preparation  for  the  ceremonial 
of  Opening  of  the  Mouth.  They  amaze  us, 
in  spite  of  the  mutilations  they  have  undergone.  And  the  fur- 
niture which  accompanied  them  was  produced  by  the  hands 

199 


FIG.    388. 
STATUETTE    OF  A 

NEGRESS 

(Petrie    Collection). 

(Phot.  Petrie.) 


ART  IN  EGYPT 


of  workmen;  I  do  not  think  that  the  walk  of  the  great  feline 
animals,  slow,  supple,  and  restrained,  has  ever  been  more  per- 
fectly rendered  than  in  the  figures  of  the  leopards  which 
adorned  the  funerary  chairs  of  Amenophis  II.  (Fig.  381).  Nor 
was  furniture  ever  more  elegant  or  better  adapted  to  the  require- 
ments of  daily  life.  What  could  be  more  ingenious  or  more 
charming  in  their  way  than  the  three  arm-chairs,  or  the  jewel- 
boxes  (Fig.  382),  deposited  by  Amenophis  III.  and  his  children 
in  the  hypogeum  of  the  parents  of  Queen  Thi?  The  chairs 
have  an  extraordinary  air  of  modern  com- 
fort; one  of  them,  that  in  which  the  front 
feet  are  surmounted  by  human  heads,  has 
been  christened  by  visitors  the  Empire  Chair 
(Fig.  383) ;  another  might  be  aptly  described 
as  the  Louis  XVI.  Chair. 

The  genre  in  which  technical  skill  is 
manifested  in  its  most  original  and  fertile 
form  is  that  of  the  manufacture  of  toilette 
utensils, .  and  especially  of  those  which  are 
incorrectly  called  perfume -spoons.  These 
consisted  of  a  slight  handle,  and  a 
receptacle  for  cosmetics  and  essences 
(Fig.  384);  the  variety  of  invention  dis- 
played in  design  and  proportions  is  amaz- 
ing; a  couchant  calf  whose  back  is  the 
lid  (Fig.  385);  a  fox  running  away,  car- 
rying off  a  large  fish  whose  body  forms 
the  spoon;  a  lotus  seed  hollowed  out  as 
a  bowl  on  a  bouquet  of  flowers  (Fig.  386) ; 
a  young  girl,  gathering  flowers,  or  passing 
along  the  marsh  playing  a  guitar;  a  nude  servant  bearing  offerings; 
a  grotesque  slave  bowed  beneath  the  weight  of  a  sack,  a  leather 
jar,  a  vase  (Fig.  303)  or  a  boiler  out  of  proportion  to  his  size 
(Fig.  304).  The  favourite  type,  and  also  the  most  graceful  one, 
is  that  of  the  woman  swimming  (Fig.  387),  her  outstretched  arms 
holding  up  on  the  water  a  hollow  duck,  whose  wings  fold  back 
and  form  a  lid.  Certain  statuettes,  which  look  to  our  modern 
eyes  like  drawing-room  ornaments,  are  double-statues  for  persons 
of  modest  means,  either  to  represent  the  master  of  the  house, 
or  to  ensure  his  domestic  comfort  and  the  services  of  slaves. 
As  they  were  not  very  expensive ,  they  were  much  in  request, 
and  their  manufacturers  had  acquired  inimitable  skill  in  exe- 
cuting them.  The  ethnical  type  is  rendered  with  the  utmost 

200 


FIG.  j8g. — STATUETTE 

OF  A  WOMAN 
(Phot.  Chassinat). 


THE  SECOND  THEBAN  AGE 


fidelity,  as  in  the  little  negress  of  the  Flinders 
Petrie  collection;  rarely  has  the  expression 
of  careless  gaiety  and  good  humour  proper 
to  the  black  races  been  more  happily  rendered 
(Fig.  388).  The  figure  of  the  princess  des- 
cribed by  Chassinat  (Fig.  389),  and  that  in 
the  James  Simon  collection,  emphasise,  with- 
out exaggerating,  the  characteristics  of  the 
family  of  Khu-en-Aton,  the  straight,  pointed 
face,  the  long,  thin  body,  the  curved  hip 
and  full  thigh.  The  Lady  Na'i  in  the  Louvre 
is  a  gem,  with  her  roguish  face,  her  young 
bust,  chastely  modelled  under  her  gauze  robe, 
the  lotus-bud  nestling  between  her  breasts. 
The  little  maiden  at  Turin  (Fig.  392),  adjusting 
her  earring,  is  naked,  and  quite  unabashed; 
she  is  at  that  indeterminate  age  when  the 
forms  seem  to  hesitate  between  those  of  boy 
and  woman.  Male  models  were  less  decor- 
ative as  material.  Several  who  belonged  to 
the  priesthood,  insisted  upon  being  reproduc- 
ed in  all  the  splendour  of  their  sacred  in- 
signia, and  have  gained  nothing 


FIG.   3QO. 

THE  LADY  NAI 

(The  Louvre,  Paris). 


FIG.  jgi. 

STATUETTE   OF 

A  PRIEST. 


by  the  process.  Those,  whose  statues  are  in  the 
Louvre  and  at  Cairo  (Fig.  391),  would  have  made 
a  better  impression  if  they  had  not  encumbered 
themselves  with  the  ensigns  of  their  gods,  statuettes 
of  Amon  and  of  Ptah  (Fig.  391),  and  a  great 
ram's  head  surmounted  by  a  solar  disc.  The 
three  little  fellows  at  Cairo,  with  their  deceptively 
Japanese  appearance,  and  the  statuettes  of  officers 
at  Berlin  (Fig.  394)  and  in  the  Louvre,  are  not 
unworthy  of  a  place  beside  the  Lady  Nai;  their 
short  wigs  show  the  shape  of  head  and  neck, 
their  tunics  hardly  veil  the  bust,  and  their  shapely, 
muscular  legs  emerge  robustly  from  their  turned- 
up  petticoats.  These  are  but  the  wreckage  of 
a  flourishing  industry,  and  for  the  twenty  odd 
specimens  that  have  survived,  how  many  must 
have  perished  from  antiquity  onwards,  as  fuel! 
They  show  us  that  towards  the  close  of  the 
second  Theban  age  there  was  a  semi-popular  art, 
marked  by  a  variety  of  aspect  and  a  freedom  of 
201 


ART  IN  EGYPT 


technique    very    disconcerting   to    those   who    still   hold    the    im- 
mobility of  the  Egyptian  civilisation  as  an  article  of  belief. 

It  would  seem  that  the  processes  of  cast- 
ing metal  must  have  been  perfected  in  the 
centuries  which  divide  the  two  Theban 
periods;  thenceforth  we  find  no  more  of 
those  examples  which  are  partly  cast,  partly 
hammered ,  like  the  statue  of  Pepi  I. ;  in- 
stead, we  have  life-size  cast  bronze  statues, 
like  that  to  which  the  bust  of  Rameses  IV. 
in  the  Pelizaeus  collection  must  have  be- 
longed. They  were  not  at  first  cast  all  in 
one  piece,  but  the  various  parts  were  pre- 
pared separately,  and  then  put  together 
with  tenons  imbedded  in  the  mass.  Scarce- 
ly anything  has  come  down  to  us  of  the 
metal  statuary,  bronze,  copper,  silver,  or 
gold,  and  only  statuettes  have  escaped  the 
general  destruction ;  but  the  goldsmith's  work 
and  the  jewelry  are  known  to  us  even 
better  than  those  of  the  Twelfth  Dynasty. 
Examples  of  bronze  and  copper  are  not 
lacking,  and  there  are  some  very  fine  spe- 
cimens in  our  museum,  such  as  the  two 
gilded  bowls  found  by  Newberry  in  the 
dust  of  the  tomb  of  Rakhmiriya,  with  a  little  ox  in  relief  at 
the  bottom,  or  the  dish  which  Daressy  brought  back  from  the 
hypogeum  of  Hatiyai  (Fig.  393) ; 
this  had  a  central  boss  of  gold 
or  silver  (now  lost),  and  round 
it  a  thicket  of  lotus,  among  which 
flocks  are  pasturing,  unconscious 
that  one  of  their  bulls  has  just 
been  pulled  down  by  a  lion. 
The  Louvre  possesses  some  re- 
mains of  the  plate  owned  by 
Thuti,  the  legate  of  Thothmes  HI. : 
a  perfect  gold  cup,  and  a  frag- 
ment of  a  silver  cup,  but  the 
treasure  discovered  a  few  years 
ago  at  Zakazik,  among  the  ruins 
of  the  ancient  Bubastis,  dates 

t.  il_          ,.•  £      D  TI  FIG"    393-— DISH   OF   HATIYAI 

from    the    time    of    Kameses   11.        (Museum,  Cairo).   (Phot.  E. 

202 


FIG.    392. — YOUNG    GIRL 
(Museum,  Turin). 
(Phot.   Lanzone.) 


THE  SECOND  THEBAN  AGE 


and  his  successors.  A  golden  drinking-vessel ,  in  the  form  of 
a  half-open  lotus  mounted  upon  its  stem ,  bears  the  cartouche 
of  his  granddaughter  Tuosret.  I  would  not 
propose  it  as  a  model  to  our  contemporaries, 
but  some  twenty  shallow  silver  cups,  with 
flat  bases  found  with  it  are  very  delicately 
ornamented.  At  the  bottom  of  one  of  these 
(Fig.  395) ,  is  a  lake ,  well  stocked  with  fish, 
on  which  a  little  papyrus  boat  with  a  shepherd 
and  calf  as  its  crew  floats  idly;  a  little  further 
on,  two  young  women  are  swimming  side  by 
side.  On  the  bank,  four  conventional  palm- 
trees  grow  at  equal  distances ;  winged  sphinxes 
with  female  heads  prowl  in  the  interstices,  and 
animals  run  about  distractedly:  a  wild  bull 
flying  from  a  leopard,  hares  and  gazelles 
pursued  by  foxes,  dogs  and  wolves.  The 
figures  of  the  middle  register  are  so  low  in 
relief  that  one  would  declare  them  to  be  in- 
cised; those  at  the  edge  were  repousse  more 
boldly,  then  worked  over  and  finished  with 
the  burin.  Two  golden  jugs  accompanied  the 
dishes  (Fig.  396);  one  has  a  smooth  body, 
and  a  neck  encircled  with  foliage  and  figures 
in  outline;  the  body  of  the  other  is  sym- 
metrically studded  with  ears  of  maize ,  and 


FIG.  304.— STATUETTE 
OF  AN  OFFICER 
(Museum,  Berlin). 


FIG.  395.— GOBLET  FROM  ZAKAZIK 
(Museum,  Cairo).  (Phot.  E.  Brugsch.) 


has  a  hanging  ring  fixed  to  the 
edge  of  the  neck  by  a  couchant 
calf  most  exquisitely  worked.  The 
masterpiece  of  the  collection, 
however,  is  a  ewer,  the  body  of 
which  is  covered  for  three-quarters 
of  its  height  by  longitudinal  lines 
of  ovoli ,  overlying  one  another 
like  the  scales  of  a  pine-cone. 
The  feature  that  makes  it  unique 
is  its  handle  (Fig.  397).  A  kid, 
attracted  by  the  aroma  of  the 
wine  it  contains,  has  climbed  up 
the  body,  and  looks  over  the 
brim ,  standing  boldly  upon  her 
hind  legs,  her  shins  tense,  her 
spine  rigid,  her  knees  pressed 
203 


ART  IN  EGYPT 


FIG.  396.— THE  TWO  GOLD  JARS  FROM  ZAKAZIK 
(Museum,  Cairo).  (Phot.  E.  Brugsch.) 


against  two  golden  flower-calyces,  which  spring  horizontally  from 
the  silver  surface,  her  muzzle  quivering;  a  ring  inserted  in  her 

nostrils  served  to  hang 
up  the  jug.  The  tech- 
nique is  excellent,  but 
here  the  conception  sur- 
passes the  technique ; 
nothing  could  be  better 
than  the  eagerness  of  the 
little  animal,  and  the  ex- 
pression of  greedy  desire 
expressed  in  her  whole 
body. 

Personal  ornament, 
whether  arms  or  jewels, 
was  never  treated  more 
solicitously,  and  on  the 

whole  with  greater  success  than  at  this  period.  Rings  both  for 
fingers  and  toes,  bracelets,  chains,  mirrors,  are  all  perfect  in  taste 
and  exquisitely  finished.  I  may  instance  the  mirror-handle,  the 

astounding  ivory  Bes  in  our  museum 
(Fig.  398);  even  certain  cases  for 
mirrors  are  little  short  of  master- 
pieces, such,  for  instance,  as  the 
one  found  in  the  tomb  of  Amen- 
ophis  II.  (Fig.  399) ,  on  which  the 
king's  daughter  is  seen  naked  among 
the  flowers.  And  it  is  not  only 
isolated  specimens  turned  out  of 
the  soil  by  the  accidents  of  research, 
but  whole  collections  which  show 
us  what  were  the  jewel-caskets  of 
persons  of  high  rank,  men  or 
women.  Queen  Aah-hetep,  whose 
mummy  received  a  present  from 
each  of  her  husbands  and  children, 
alone  possessed  enough  to  enable 
us  to  judge.  What  has  not  already 
been  said  about  the  dagger  and 
axe  bestowed  on  her  by  Amasis? 
(Fig.  400.)  The  dagger  in  parti- 
cular (Fig.  401)  excites  curiosity,  with  its  blade  of  dark  bronze 
set  in  massive  gold ,  on  the  surfaces  of  which  a  lion  pursues  a 

204 


FIG.  397.— THE  JUG  WITH  THE 

GOAT  FROM  ZAKAZIK 
(Museum,  Cairo).  (Phot.  E.  Brugsch.) 


THE  SECOND  THEBAN  AGE 


FIG.  398.— IVORY  HANDLE 

OF  -A  MIRROR  (Museum, 
Cairo).     (Phot.  E.  Brugsch.) 


FIG.  39g.— MIRROR-CASE 
(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


bull  in  the  presence 
of  four  great  gras- 
shoppers, and  fif- 
teen flowers  unfold 
their  petals  in  a 
delicate  damascen- 
ed gold  pattern.  It 
recalls  the  Achaean 
daggers  of  Mycenae ; 
but  did  the  dawn- 
ing civilisations  of 
Europe  borrow 
from  Egypt,  or  did 
Egypt  find  inspir- 
ation in  one  of 
their  creations?  If 
the  motive  be  for- 
eign ,  which  is  by 
no  means  proved, 
the  handling  and 

composition  are  purely  Theban.     The  same  may   be  said  of  the 

ornaments  found  with  it,   necklaces,   chains,   bracelets,   toe   and 

finger  rings.      Only    on    the   banks  of  the    Nile  would    mourning 

relatives    have   conceived  the  idea 

of  placing   among   the    plenishings 

of   her   they   bewailed,   boats  and 

their  crews  in  gold  or  silver  (Fig.  402) 

that  she  might  be  able  to  embark 

at   will   upon   the   western   sea   or 

on    the    ponds    of    her    sepulchral 

domain.    Again,  if  we  examine  the 

various  parts  of  the  large  necklace 

which  hung  round  the  queen's  neck 

(Fig.  403),  we  shall  see  that  it  had 

golden    falcons'    heads    enamelled 

with  blue,   to  fix  the  ends  to   her 

shoulders,  a  motive  in  favour  among 

the  goldsmiths  of  the  Twelfth  Dyn- 
asty, as  among  their  predecessors  of 

the  Memphite  age.    The  spirals,  the 

flowers  with  four  petals  forming  a 

Greek  cross,  the  roundels,  the  small 

bell-pendants  which  form  seven  of 

205 


FIG.  40O.— SELECTION   OF  THE 

JEWELRY  OF  AAH-HETEP 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


the  eleven  rows,    abound  on  the  scarabs  and  archaic  ornaments, 
and   what  could   be    more  distinctively  Egyptian    than   the  flying 
falcons,    the    seated    cats,    the    gazelles    turning   their   heads    as 
they  flee,  the  kids  pursued  by  lionesses? 

There  is,  however,  a  new  type,  which  comes 
from  abroad:  that  of  the  earring,  or  ear-pendant. 
We  cannot  say  whether  it  was  entirely  unknown 
to  earlier  generations  or  not,  but,  in  any  case, 
it  is  first  commonly  found  upon  the  monuments 
of  the  Eighteenth  Dynasty,  and  it  would  seem 
that  in  this,  as  in  other  goldsmiths'  work,  the 
hand  of  the  craftsman  had  become  somewhat 
heavy.  The  Pharaohs  had  acquired  a  great  deal 
of  gold  in  Ethiopia  or  in  Syria,  and  they  liked 
to  display  as  much  as  possible  of  it  on  their 
persons.  The  slender  crowns  of  the  princesses 
of  Amenemhat  or  Sesostris  were  accordingly 
superseded  by  heavy  diadems,  enormous  earrings, 
such  as  those  of  Seti  II.  (Fig.  404)  or  Rameses  XII. 
(Fig.  405) ,  the  weight  of  which  dragged  down 
the  ears,  and  bracelets  loaded  with  gems,  like 
those  Rameses  II.  gave  to  his  granddaughter 
Tuosret  (Fig.  406),  recently  added  to  the  treasures 
of  the  Cairo  Museum.  They  are  solid  and  dur- 
able, and  no  craftsman  was  ever  a  more  complete 
master  of  technique  than  he  who  executed  them, 
but  there  is  an  element  of  vulgarity  in  the  large 
lapis-lazuli  plaques  which  form  the  bodies  of  the 
ducks,  in  the  gold  reticulations  and  granulated 
lines  so  freely  lavished  upon  the  surface.  We 
get  a  similar  impression  when  at  the  Louvre  we 
examine  the  jewels  of  Prince  Khamuasit,  his 
hawk-brooch  with  the  ram's  head  (Fig.  408),  his 
pectoral  (Fig.  408),  on  which  the  vulture  and  the 
uraeus  are  enframed  as  in  the  fagade  of  a  naos, 
and  we  begin  to  think  that  perhaps  the  sober 
and  delicate  taste  of  the  old  goldsmiths  had 
been  perverted  by  the  models  which  war  or  commerce  had 
imported  from  Asia;  it  would,  however,  be  going  too  far  to 
affirm  this,  before  excavation  has  given  us  the  equivalents  in 
Mesopotamia  and  Assyria  of  the  treasures  gathered  together 
on  the  banks  of  the  Nile.  A  great  many  Asiatics  and  Europeans 
established  themselves  in  Thebes  and  other  great  cities  during 

206 
r 


FIG.  401. 
THE  DAGGER 

OF  AMASIS 
(Museum,  Cairo). 
(Phot.E.Brugsch.) 


THE  SECOND  THEBAN  AGE 


FIG.  402.— ONE  OF  THE  BOATS  OF  Qt'EEN  AAH-HETEP 
(Museum ,  Cairo).     (Phot.  E.  Brugsch.) 


the  centuries  while  the  Egyptian  hegemony  lasted.  They  modi- 
fied the  habits  of  the  nation  to  some  extent,  suggesting  new 
needs  to  them,  and 
it  certainly  seems 
that  the  passion  for 
very  elaborate  jewel- 
ry came  in  with 
them.  Here  and 
there,  however,  the 
natural  temperament 
still  prevailed ,  and 
pieces  of  exquisite 
simplicity  were  pro- 
duced ;  the  high 
priest  Pinotem  wore 

as  bracelet  a  simple  gold  reed,  ornamented  with  a  network  of 
polychrome  enamel ,  and  this  is  one  of  our  finest  specimens  in 
the  Cairo  Museum. 

In  these  narrow  domains  of  industrial  art  as  in  the  vaster  field 
of  architecture,  the  second  Theban  Age  showed  itself  capable  of 
evolving  new  forms,  although  it  adhered  for  the  most  part  to 
the  old,  and  manifested  a  vigour  of  creation  and  production 
which  equalled,  if  it  did  not  surpass,  that  of  the  Memphite  Age. 
The  ideal  of  its 
artists  was  less  pure 
and  less  serene,  but 
it  came  closer  to 
reality.  The  impulse 
which  had  carried 
Pharaoh  and  his 
armies  beyond  the 
isthmus,  had  moved 
the  whole  people, 
and  even  those  who 
took  no  part  in  the 
conquests,  theolog- 
ians, men  of  letters, 
merchants  and  artists, 
feeling  the  bound- 
aries of  their  world 


FIG.  403. — GOLD  NECKLACE  WITH  HEADS   OF 
FALCONS   (Museum,  Cairo).     (Phot.  E.  Brugsch.} 


enlarged ,     had     ex- 
tended the   circle  of  their  knowledge   and   inspiration,   each   in 
his   own   calling.     Seeing   things   on  a  great   scale,   they  sought 

207 


ART  IN  EGYPT 


to  create  greatness,  no  longer,  after  the  manner  of  their  an- 
cestors the  Pyramid-builders,  by  the  exaggerated  bulk  of  their 
material,  but  by  the  reasoned  immensity  of 
their  conceptions;  thus  architects  had  arrived 
at  the  gigantic  colon- 
nades of  Luxor  and 
Karnak ,  and  sculp- 
tors at  the  colossi 
of  the  Theban  plain 
and  Abu  Simbel. 
When  they  had 
reached  this  point, 
which  they  could  not 
surpass,  they  did  not 
long  maintain  them- 
selves at  its  level; 
exhausted  by  the 
very  effort  they  had 
made,  their  artistic 
vigour  declined  no 
less  rapidly  than 
their  military  pro- 
wess after  the  reign 
of  Rameses  III.  Their 
work  is  not  perhaps 
the  most  uniformly 
beautiful  produced  by  Egypt,  and  some  may  rank  that  of  the 
Memphite  times  above  it ;  to  me  it  seems ,  nevertheless ,  the 
most  vital ,  the  most  varied ,  the  most  complete ,  that  which  is 
most  characteristic  of  the  people,  its  defects  and  its  qualities. 
It  is  certainly  that  which  does  most  honour  to  Egypt,  and 
secures  for  her  one  of  the  highest  places  in  the  artistic  history 
of  the  world. 


FIG.  404. 

GOLD   EARRINGS   OF  SETI  I. 
(Museum ,   Cairo). 
(Phot.  E.  Brugsdi.) 


FIG.  405.— EARRING 
OF  RAMESES   XII. 
(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


208 


THE  SECOND  THEBAN  AGE 


BIBLIOGRAPHY  TO  CHAPTER  II  —  PART  II 

Second  Theban  Age.  Here  the  monuments  are  so  numerous,  aud  so  much  has  been 
written  about  them,  that  it  has  been  necessary  to  make  a  selection  of  books.  For  a  general 
survey,  we  may  recommend  Mariette,  Voyage  dans  la  Haute  Egypte,  vol.  I,  p.  55-81, 
90-98,  and  pi.  18-34,  37-38,  vol.  II,  p.  1-82,  107-109,  and  pi.  39-65,  74,  and  E.  de  Rouge, 
Album  photographique  de  la  Mission,  nos.  47-48,  125-133,  138-141,  151-155,  with  the 
portions  of  text  corresponding  to  these  numbers. 


FIG.   406.— BRACELETS   OF  RAMESES  II. 
(Museum,  Cairo)      (Phot.  E    Brugsch.) 


Architecture.  —  A.  Temples.    —   For   the   temples   of  the  Second  Theban  Empire,  see 
Perrot-Chipiez,    Histoire   de   I'Art   dans  I'Antiquite :    I,    Egypte,    1882,    LXXVI-879   p.; 

—  G.  Maspero,  Archeologie  egyptienne,  1st  ed.,  1888,  p.  66-87;  W.  Spiegelberg,  Geschichte 
der  dgyptischen  Kunst,    1903,    p.  40-51.   —   For   the   history   of  the   principal  temples   of 
Thebes  and  Nubia,  see  for  Luxor:   L.  Borchardt,  Zur  Geschichte  des  Luqsortempels,  in  the 
Zeitschrift  fiir  dgyptische  Sprache,  1896,  vol.  XXXIV,  p.  122-138;  —  Gayet,  Le  Temple  de 
Louxor  (Memoires  de  la  Mission  du  Caire,  vol.  XVIII),  1894,  Vienne  4to,  174  p.  and  LIV  pi.; 

—  G.  Daressy,  La  Procession  d'Ammon  dans  le  Temple  de  Louxor  (Memoires  de  la  Mission 
francaise,  vol.  VIII),  p.  380-391  and  XVI  pi.,  and  the  Notice  explicative  des  mines  du  Temple 
de  Louxor,  1893,  Cairo,  8vo,  IX-81  p.;  —  forKa.nak:  Mariette,  Karnak,  Paris,  1875,   Text 
4to,  88  p.,  and  Atlas  fol. ,   LV1  pi.;  —  G.  Legrain,  Rapports  sur  les  travaux  de  Karnak, 
in  the  Annales  du  Service  des  Antiquites,  I   1900,  p.  193-200,   II  1901,  p.  184-189,  265-280, 
IV  1903,  p.  1-40  and  6  pi.,  V  1904,  p.  1-43  and  6  pi.;  —  L.  Borchardt,  Zur  Baugeschichte 
des  Amonstempels  von  Karnak,  4  to,  Leipzig  1905,  47  p.  and  1  pi.;  —   for  Der-el-Bahari; 
Naville,  The  Temple  of  Deir  el  Bahari  (Egypt  Exploration  Fund,  vol.  XII— XIV,  XVI,  XIX, 
XXVI,   XXIX);    Introductory  Memoir,  4to,  London,  1894,  31   p.  and  XIV  pi.,  fol.,  I  18%, 
15  p.  and  XXXI  pi.,  II  1897,  17  p.  and  XXXII-LV  pi.,  Ill  1898,  21  p.  and  LVI-LXXXVI  pi, 


.    :i  p. 

ment  du  Ramesseum,  in  the  Annales  du  Service  des  Antiquites,  1907,  vol.  VIII,  p.  193-200; 
—  for  the  temple  of  Thothmes  III  and  other  ruined  sepulchral  temples:  Grebaut-Maspero, 
Le  Musee  egyptien,  vol.  I,  1890-1900,  p.  3-9  and  pi.  XVI-XVII;  Daressy,  La  Chapelle  d'Uazmes, 
in  the  Annales  du  Service  des  Antiquites,  1900,  vol.  I,  p.  97-108;  Flinders  Petrie,  Six  Temples 
at  Thebes,  1897,  4to,  London,  33  p.  and  XXVI  pi.;  Weigall,  A  Report  on  the  Excavation 
of  the  funeral  Temple  of  Thutmosis  III  at  Gurneh,  in  the  Annales  du  Service  des  Antiquites, 
1906,  vol.  VII,  p.  121—141  and  vol.  VIII,  p.  286;  —  for  Medinet-Habu ;  G.  Daressy,  Noticz 
explicative  des  ruines  de  Medinet-Abou,  8vo,  Cairo,  1897,  VII-120  p.;  Uvo  Holscher,  Das 
Hohe  Tor  von  Medinet-Habu,  eine  baugeschichtliche  Untersuchang ,  4 to,  Leipzig  1910, 

209  P 


ART  IN  EGYPT 

IV-68  p.  and  10  pi.;  —  for  Bet-Wali,  Garf-Husen,  Wadi-es-Sabua,  Amada,  Derr,  Abu 
Simbel,  Abahuda:  Gau,  Monuments  de  la  Nubie,  Stuttgart  -  Paris  1822,  pi.  12-14,  27-32, 
42-63;  Maspero  -  Barsanti ,  Les  Temples  immerses  de  la  Nubie.  Rapports,  p.  60-61,  87-89, 
106-168  and  pi.  LIII,  LXXXV-LXXXVII ,  CX-CXL,  CXLIV-CLXIX.  —  B.  The  tombs.  — 
For  Ethiopia:  Cailliaud,  Voyage  a  Meroe,  atlas,  fol,  Paris,  1828,  vol.  II,  pi.  VII-XV;  -"• 
for  the  royal  tombs,  see  in  addition  to  the  works  quoted  in  connection  with  the  temples 
of  Thebes:  Mariette,  Abydos,  fol.,  Paris,  I  1868,  86  p.,  and  53  pi.,  1  ol.,  II  1879,  51  p. 
and  pi.  1-21 ;  -  Caulfield-Petrie,  The  Temple  of  the  Kings  at  Abydos  (Egyptian  Research 
Account,  vol.  VIII),  4 to,  London  1902,  23  p.  and  XXVI  pi.;  —  E.  Lefebure,  Les  Hypogees 
Royaux  de  Thebes  (Memoires  de  la  Mission  du  Caire,  vol.  II-III) ,  4to,  Paris,  vol.  I  1886, 
31  p.  and  LXIV  pi.,  II  1887-1888,  VD1  and  191  p.  and  74-XLI  pi. ;  —  V.  Loret,  Les  Tombeaux 
de  Thoutmes  III  el  d'Amenophis  II  et  la  Cachette  royale  de  Biban  el-Molouk,  in  the  Bulletin 
de  I'lnstitut  egyptien,  1899,  Ill'd  series,  vol.  IX,  p.  91-112  and  15  pi.;  —  Fr.  Guilmant,  Le 
Tombeau  de  Ramses  IX  (Memoires  de  I'lnstitut  francais  du  Caire,  vol.  XV),  square  4to, 
1907,  %  1.;  —  Theodore  M.  Davis,  The  Tomb  of  Thutmes  IV,  4to,  London,  1905,  XLV-150  p. 
and  XXVIII  pi.,  The  Tomb  of  Hatshepsitou,  4 to,  London  1906,  XV-112  p.,  The  Tomb 
of  Jouiya  and  Touiyou,  4  to,  London  1907,  X-48  p.  and  XLJV  pi,  The  Tomb  of  Siphtah, 
4to,  London  1908,  45  p.  and  XXIX  pi.,  The  Tomb  of  Queen  Tiyi,  4to,  London  1910,  45  p. 
and  XXXV  lp.,  The  Tomb  of  Haremheb,  4to,  London  1911  (in  the  press);  —  for  the  private 
tombs ;  Benedite  -  Bouriant  -  Chassinat  -  Maspero  -  Scheil  -Virey ,  Tombeaux  thebains  (Memoires 
de  la  Mission  du  Caire,  vol.  V),  4 to,  Vienne  1889-1894,  657  p.;  —  Boussac,  Le  Tombeau 
d'Anna  (Memoires  de  la  Mission  du  Caire,  vol.  XVIII),  fol.,  Paris  18%,  IV  p.  and  XVI  pi.; 

—  Taylor-Griffith,   The  Tomb  of  Paheri  at  El-Kab,  fol.,  London  1894,   34  p.  and  10  pi.; 

—  N.  de  G.  Davies,  The  Rock-tombs  of  el-Amarna  (Archaeological  Survey  of  Egypt,  vol. 
XII-XVIII),  4 to,  London  I  1902,  56  p.  and  XLI  pi.,  II  1905,  48  p.  and  XLVH  pi.,  Ill  1905, 
41  p.  and  XXXK  pi.,  IV  1906,  36  p.  and  XLV  pi.,  V  1907,  37  p.  and  XLIV  pi.,  VI  1908, 
44  p.  and  XLIV  pi.  —  C.  The  Palaces.  —  In  addition  to  Flinders  Pelrie's  work  quoted  above 
on  Tell -el-Amarna,   see   G.  Daressy,   Le  Palais  d'Amenophis  HI  a  Medinet-Abou ,  in  the 
Annales  du  Service  des  Antiquites,   vol.  IV,   p.  165-170   and  1  pi.  —    Robb  de  P.  Tytus, 
A  preliminary  Report  on    the  Re-Excavation   of   the  Palace   of  Amenhetep  III,  4 to,   New 
York  1904,  25  p.  and  4  pi.;  —  Maspero,  Causeries  d'Egypte,  8vo,  Paris  1907,  p.  257-264. 


FIG.  407.— BROOCH   OF  KHAMUASIT 
(The  Louvre,  Paris). 


Painting  and  Sculpture.  —  A.  Painting.  —  In  addition  to  the  works  quoted  in  connection 
with  royal  and  private  tombs,  see  Fr  -W.  de  Bissing  and  Reach,  Bericht  fiber  die  malerische 
Technik  der  Hawata- Freshen  von  Kairo .  in  the  Annales  du  Service  des  Antiquites,  1906, 
vol.  VII,  p.  64-70;  —  Flinders  Petrie,  Egyptian  decorative  Art,  8vo,  London  1895;  — 
G.  Jequier,  La  Decoration  egyptienne,  4 to,  Paris  1910/1911,  28  pages  and  XI.  planches 
-=•  B.  Sculpture.  —  For  the  sculpture  of  the  Second  Theban  Empire,  see,  in  addition  to. 
the  general  works:  Mariette,  Album  du  Musee  de  Boulak,  fol.,  Cairo  1874,  pi.  32,  34,  37; 
—  E.  de  Rouge,  Album  photographique,  fol.,  Paris  1867,  nos.  55-56,  64,  72-74,  77,  80-85, 
125-135;  —  Fr.  W.  de  Bissing,  Denkmaler  der  agyptischen  Skulptur,  pi.  36-59,  76-97,  and 
the  corresponding  portions  of  text;  —  L,  Borchardt,  Kunstwerke  aus  dem  agyptischen 

210 


THE  SECOND  THEBAN  AGE 

Museum  zu  Kairo,  fol.,  Cairo  1908,  pi.  8-15,  24-38,  33  and  p.  6-8,  11-13,  14;  —  J.  Capart, 
L'Art  egyptien.  4 to,  Brussels  1909,  pi.  61-74,  and  the  article  by  Maspero,  La  Cachetic  de 
Karnak  et  I'Ecole  de  Sculpture  thebaine,  in  the  Revue  de  I' Art  ancien  et  moderne,  1906, 
and  XX,  p.  337-348;  the  following:  books,  pamphlets  and  articles:  G.  Leg-rain,  Statues  et 
Statuettes  de  Rois  et  de  particuliers  (Catalogue  general  du  Musee  du  Cairo),  4 to,  Cairo, 
I  1906,  p.  30-89,  and  pi.  XXVII-LXXIX,  II  1909,  40  p.  and  LIII  pi.;  —  R.  Lepsius,  Eine 
Sphinx,  in  the  Zeitschrift  fur  agyptische  Sprache,  1882,  vol.  XX,  p.  117-120;  —  G.  Maspero, 
in  O.  Rayet,  Monuments  de  I' Art  antique,  fol.,  Paris  1880-1884,  vol.  I;  la  Statue  de 
Khonsou,  in  the  Annales  du  Service  des  Antiquites,  1902,  vol.  Ill,  p.  181  and  I  pi.;  Sur 
un  fragment  de  Statuaire  Thebaine,  in  the  Revue  de  I  Art  ancien  et  moderne,  vol.  XVII, 
p.  401-404  and  1  pi.;  La  Vache  de  Deir  el-Bahari,  in  the  Revue  de  I  Art  ancien  et  moderne, 
1907,  vol.  XVII,  p.  5-18  and  3  pi.;  Les  Quatre  Tetes  de  Canapes  du  Musee  du  Caire,  in 
the  Revue  de  I' Art  ancien  et  moderne,  1910,  vol.  XXVIII,  p.  241-252  and  1  pi.;  Le  Musee 
egyptien,  vol.  I,  1890-1900,  pi.  I,  XLIV,  and  p.  3-4,  39-40,  t.  II,  1901-1907,  pi.  V-VI  and 
p.  15-20;  —  Legrain ,  Le  Musee  Egyptien,  vol.  II,  pi.  I-IX  and  p.  2-14;  —  G.  Benedite, 
A  propos  d'un  buste  egyptien  recemment  acquis  par  le  Musee  du  Louvre,  in  the  Monuments 
et  Memoires  de  la  Fondation  Plot,  1905,  vol.  XIII,  p.  3-25;  —  J.  Capart,  Tete  egyptienne 
du  Musee  de  Bruxelles,  in  the  Monuments  et  Memoires  de  la  fondation  Plot,  1906,  vol.  XIII, 
p.  27 — 34;  Une  importante  donation  d' Antiquites  egyptiennes,  in  the  Bulletin  des  Musees 
royaux  du  Cinquantenaire,  Brussels  1908,  2nd  series,  vol.  I,  p.  84-86;  —  L.  Borchardt,  Der 
Portratkopf  der  Konigin  Teje  im  Besitz  von  Dr.  James  Simon,  Leipzig  1911,  30  p.  and 
5  plates. 


FIG.   408. — PECTORAL   OF  RAMESES  H. 
(The  Louvre,   Paris). 


The  minor  Arts  —  A.  Ceramics.  —  For  household  pottery  and  objects  in  terra-cotta  or 
glazed  and  enamelled  stone  ware,  see  Fr.  W.  de  Bissing,  Fayencegefasse  (Catalogue  general 
du  Musee  du  Caire),  4 to,  Vienna  1902,  XXXI-114  p.;  —  Henry  Wallis,  Egyptian  Ceramic 
Art,  the  Macgregor  Collection,  4 to,  London  1898,  FV-85  p.  and  30  pi.;  Egyptian  Ceramic 
Art,  4to,  London  1900  p.  and  12  pi.;  —  for  the  enamelled  earthenware  decoration  of 
palaces  and  temples:  G.  Daressy,  Plaquettes  emaillees  de  Medinet-Habou ,  in  the  Annales 
du  Service  des  Antiquites,  1910,  vol.  XI,  p.  49-63.  —  B.  Wood-work.  For  funerery  objects, 
see  Theodore  M.  Davis,  The  Tomb  of  Jouiya  and  Touiyou ,  4 to,  London  1907,  pi.  VI-X> 
XII-XVI,  XXXIII-XLI;  —  E.  Quibell,  Tomb  of  Yuaa  and  Thuiu  (Catalogue  general  du 
Musee  du  Caire),  4to,  Cairo,  1908,  80  p.  and  LX  pi.;  —  for  toilet  utensils,  see:  G.  Maspero, 
in  O.  Rayet,  Monuments  de  I' Art  antique,  vol.  I;  —  J.  Capart,  L'Art  et  la  Parure  feminine 
dans  I'ancienne  Egypte,  in  the  Annales  de  la  Societe  d'archeologie  de  Bruxelles,  1907,  vol.  XXI, 

211  p  2 


ART  IN  EGYPT 

p.  303-334;  Figurine  egyptienne  en  bois  au  Musee  de  Liverpool,  in  the  Revue  archeologique, 

1907,  vol.  II,  p.  369-372  and  1  pi.;  —  for  wooden  rfoui/e-statuettes:  G.  Maspero,  in  A.  Rayet, 
Monuments  de  I  Art  antique,  vol.  I ;  —  Chassinat,  Une  Tombe  inviolee  de  la  XVIII''-  Dynastie, 
in  the  Bulletin   de  I'lnstitut  francais  d'archeologie  orientale,  1901,   vol.  I,   p.  225-234;    — 
Flinders   Petrie,   An  Egyptian  Ebony  Statuette  of  a  Negress,   in  Man,  1901,    no.  157;    *— 
G.  Benedite,  La  Statuette  de  la  dame  Tout,  in  the  Monuments  et  Memoires  de  la  Fondation 
Plot,  1895,  vol.  II,  p.  29-37  and  pi.  H-IV,  —  C.  Goldsmith's  work  and  jewelry :  In  addition 
to  the  two  works  by  Vernier   quoted   above,   in   connection   with   the   jewels   of  the  first 
Theban  period,   consult   his   article:   Notes   sur  les   boucles  d'oreilles   egyptiennes,    in   the 
Bulletin  de  I'lnstitut  francais  d'Archeologie  Orientale,  1911,  vol.  VIII,  p.  15-41  and  pi.  I-VII, 
also:  Mark  Rosenberg1,  ^Egyptische  Einlage  in  Gold  und  Silber,  4 to,  Frankfort-on-the-Main 
1905,  12  p.;  —  Fr.  W.  de  Bissing-,   Eine  Bronzeschale  Mykenischer  Zeit,    in  the  Jahrbuch 
des  Deutschen  Archceologischen  Instituts,  1898,  vol.  XIII,  p.  28  —  56,  Ein   Thebanischer  Grab- 
fund  aus  dem  Anfange  des  Neuen  Reiches,  fol.,  Berlin  1890 — 1909  (unfinished) ;  —  Daressy, 
Un  Poignard  du  temps  des  Rois  Pasteurs,  in  the  Annales  du  Service  des  Antiquites,  1906, 
vol.  VII,  p.  115-120  and  1  pi.;  —  C.  C.  Edgar,  The  Treasure  of  Tell  Basta,  in  the  Musee 
egyptien,  vol.  II,  p.  93-108   and   pi.  XLIII-LV;   —   G.  Maspero,    Causeries  d'Egypte,   8vo, 
Paris  1907,  p.  335-341,  Le   Tresor  de  Zagazig,    in    the   Revue  de  I' Art  ancien  et  moderne, 

1908,  vol.  XXIII,  p.  401-412  and  vol.  XXIV,  p.  29  38;  —  Mariette,  Le  Serapeum  de  Memphis, 
fol.,  Paris  1857,  3'"'1  part.,  pi.  9,  11-12,  20;  —  L.  Borchardt,  Kunstwerke  aus  dem  dgyptischen 
Museum  zu  Kairo,  pi.  43-44  and  p.  18. 


212 


F1U.  409. — THE  ISLAND   OF  PHILAE  AND  ITS  MONUMENTS  SEEN  FROM  THE   SOUTH- 
WEST   BEFORE    THE  COMPLETION    OF    THE    BARRAGE    AT    ASSUAN.     (Phot.  Beato.) 


PART  III 


Architecture  among  the  Sa'ites:  the  Theban  temple  is  transformed,  and  resolved  into  the 
Ptolemaic  and  Roman  type  —  Painting  and  Sculpture  —  The  Minor  Arts:  their 
Development  under  the  Influence  of  Greek  Conceptions  —  The  Death  of  Egyptian  Art. 

THE  political  decadence  of  Thebes  and  the  fall  of  the  Egyptian 
Empire  hampered  the  progress  of  art,  but  did  not  interrupt 
it  altogether,  as  some  have  supposed.  The  schools  of  archi- 
tecture, sculpture  and  painting  which  existed  under  the  Rames- 
sids  continued  to  produce,  some  even  brilliantly,  and  the  entry 
into  public  life  of  certain  cities  of  the  Delta,  which  had  hitherto 
dragged  out  an  obscure  existence,  led  to  the  tardy  development 
of  new  schools.  It  must  indeed  be  admitted  that  none  of  these, 
even  in  their  best  moments ,  displayed  that  sovereign  activity 
which  had  characterised  the  Theban  and  Memphite  Schools. 
There  are,  further ,  some  among  them,  the  Saite ,  for  instance, 
which  are  hardly  more  than  names  for  us  as  yet.  Sais  has 

213 


ART  IN  EGYPT 


FIG.  410.— THE  MOST  ANCIENT  TEMPLE  OF  NAPATA. 


disappeared,  her  temples  are  laid  low,  and  their  ruins  have 
perished  so  completely  that  we  can  no  longer  even  trace  their 
plan  on  the  ground;  while  such  of  her  statues  and  bas-reliefs 

as  have  escaped 
destruction ,  are 
scattered  indiscri- 
minately in  muse- 
ums, and  we  are 
unable  to  distin- 
guish them  from 
other  relics  of  the 
same  period.  Hero- 
dotus speaks  of  her 
buildings  and  their  colossi  in  terms  which  show  that  he  considered 
them  equal  to  those  of  Memphis.  I  may,  however,  be  allowed  to 
say,  after  having  visited  the  site  on  which  they  stood,  that  all 
the  temples  put  together  cannot  have  made  up  a  whole  compar- 
able in  extent  to  that  formed  by  the  buildings,  I  will  not  say 
of  Karnak ,  but  of  Luxor.  It  must  be  remembered  that  the 
Pharaohs  who  devised  them  did  not  command  the  almost  inexhaust- 
ible resources  of  the  conquering  dynasties,  'Asia  no  longer 
poured  a  steady  stream  of  gold  and  silver  into  their  treasury; 
the  fortune  they  spent  in  building  was  drawn  from  the  valley 
itself,  or  the  regions  nearest  to  the  African  desert.  They  had 

no  lack  of  precious  me- 
tals, as  we  know  from 
the  enumeration  of  the 
sums  which  Osorkon  II. 
devoted  to  the  restoration 
of  one  of  the  sanctuaries 
of  Bubastis;  but  what 
did  these  represent  when 
compared  with  those  of 
which  Thothmes  III.  and 
Rameses  II.  had  formerly 
disposed?  The  grandiose 
enterprises  which  their 
predecessors  had  carried 
on  for  centuries  were 
denied  to  them,  but,  re- 
duced though  their  wealth 

was  in  comparison  with  the  riches  of  the  past,  it  was  sufficient 
to  make  them  the  boldest  architects  in  the  world  of  their  day. 

214 


FIG.  411. — SANCTUARY  OF  THE  TEMPLE  OF 

TIHHAKAH  AT  NAPATA 

(After  Cailliaud). 


THE  SAITE  AGE 


FIG.  412. — RUINS   OF  THE  PROPYLAEA   OF  TIRHAKAH 
AT  KARNAK.     (Phot.  Beato.) 


They  worked  so  assiduously,  from  the  Cataracts  to  the  Mediterranean, 
that  they  not  only  preserved  the  tradition  of  great  art,  but  trans- 
mitted it  intact  to 
their  foreign  suc- 
cessors, Greek  or 
Roman.  The  major- 
ity of  the  temples 
we  admire  in  the 
Said  are  the  work 
of  the  Ptolemies 
or  the  Caesars. 

A.  ARCHITECTURE. 

The  various  types 
of  the  house ,  the 
palace  and  the  tomb, 
had  been  so  ingeni- 
ously perfected  in 
the  course  of  the 
second  Theban  age, 
that  they  changed 
very  little  subsequently,  and  it  is  extremely  difficult  to  distinguish 
the  Grseeco-Roman  house  from  that  of  the  Nineteenth  or  Twen- 
tieth Dynasty  among  the  ruins.  The  former  was,  in  fact,  very 
often  nothing  but  the  latter,  rebuilt  on  the  same  plan,  and  partly 
with  the  same  materials.  Mastabas,  whether  of  the  Theban  age 
with  their  pyramidal 
crowns,  or  of  the 
Memphite  age  with 
their  flat  roofs,  were 
no  longer  in  favour; 
they  had  been  super- 
seded everywhere 
by  hypogea  with  or 
without  external  chap- 
els. At  Thebes  it 
was  rarely  thought 
worthwhile  to  cut  new 
ones ,  so  numerous 
were  those  which  had 
been  rifled  and  aban- 
doned after  the  extinction  of  the  families  who  had  founded 
them.  These  were  requisitioned,  or  bought  cheaply  from  the 

215 


FIG.  413. — EAST  FACADE  OF  THE  TKMPI-E  OF  HEBT. 
(Phot.  Baraize.) 


ART  IN  EGYPT 


FIG.  414.— THE  TEMPLE  OF  DAKHLEH.      (Phot.  Lythgoe.) 


corporations    of   priests,    and    one   or   two   of    the   rooms   were 
adapted,  together  with  the  well;  after  two  or  three  generations, 

the  second  -  hand 
owners  suffered  the 
fate  they  had  in- 
flicted on  their 
predecessors,  and 
usurpers  in  the 
third  degree  were 
in  like  manner  de- 
spoiled by  new  in- 
truders. It  was  dif- 
ferent at  Memphis, 
where  the  sand, 
covering  up  the 
mastabas  or  pre- 
vious centuries,  generally  preserved  them  from  attack  by  those 
who  coveted  ready-made  sepulchres.  Violations  were,  however, 
sufficiently  frequent  to  induce  the  invention  of  a  type  which 
had  some  chance  of  escaping  it.  At  Sakkarah  and  at  Gizeh, 
accordingly,  the  following  system  was 
adopted:  a  cavity,  some  40  to  50  feet 
wide,  and  from  70  to  90  feet  deep,  was 
dug  out  in  the  plateau,  and  beside  it,  on 
the  south,  a  little  square  well,  from  4  to 
6  feet  wide,  which  communicated  with  it 
at  the  bottom.  A  huge  compact  block 
of  limestone  was  then  lowered  into  it  to 
serve  for  a  sarcophagus,  covered  with 
mystical  scenes  and  inscriptions;  in  this 

I  a   basalt   coffin   of   anthropoid   form   was 

imbedded,  and  round  it  a  vaulted  chamber 
was  raised,  built  of  small  dressed  limestone 
blocks,  the  inner  walls  decorated  with 
written  prayers,  and  sometimes  with  figures 
borrowed  for  the  most  part  from  the  Book 
of  the  Dead,  or  the  Ritual  of  the  Pyramids; 
a  narrow  rectangular  window  was  left  in 
the  middle  of  the  vault.  On  the  day  of 
burial,  when  the  mummy  had  been  laid 
in  its  bed,  the  two  covers  of  basalt  and 
limestone  were  adjusted,  then  the  passage  which  led  to  the 
little  well  was  walled  up,  and  the  two  cavities  were  filled  in 

216 


FIG.  415. — PLAN  OF  THE 

TEMPLE  OF  HORUS  AT 

EDFU. 


THE  SAITE  AGE 


FIG.  416.— PYLON  OF  THE  TEMPLE  OF  HORUS  AT  EDFU. 
(Phot.  Beato.) 


with  sand  and  fragments  of   stone.     Thieves   attacked   the   little 

well  at  all  hazards;  when  they  got  into  the  passage,  they  began 

to  clear  it,  but  the 

sand      which      the 

masonry  had  kept 

in  place  fell  in  upon 

them,  and  prevent- 

ed them  from  ad- 

vancing.   I  myself, 

every  time  I  have 

attempted  to  enter 

by  this  way,  have 

been   stopped  just 

as    were    the    ma- 

rauders of  old,  and 

have  had  to  resign 

myself   to  digging 

out    the    principal 

cavity.    As  it  con- 

tains from  6000  to 

16000    cubic   feet   of  soil,   this  is  generally  the  work  of  about 

three  months,  with  gangs  of  100  workmen.    We  can  understand 

that   persons    obliged    to    work   furtively   and   in   small   numbers, 

in   perpetual   fear  of  surprise,   should  have   respected  such  well- 

defended       sepul- 

chres; nearly  all  of 

those  we  discover 

at    Sakkarah     are 

virgin  tombs. 

The  plan  of  the 
temples,  like  that  of 
the  tombs,  was  very 
gradually  modi- 
fied. The  Tanite 
and  Bubastite  Pha- 
raohs built  a  great 
many  in  Lower 
Egypt,  only  shape- 
less fragments  of 
which  remain  to  us 
at  Tanis  and  Buba- 


stis. 


In     Upper 

,  Tr 

the   priest- 


FIG.  417.—  COURT  AND  PORTICOES  OF  THE  TEMPLE  OF 

HORUS  AT  EDFU.    (Phot.  Beato.) 
217 


ART  IN  EGYPT 


FIG.  418. — PRONAOS  AND  TERRACES  OF  THE  TEMPLE 

OF  HORUS  AT  EDFU  FROM  THE  TOP  OF  THE  PYLON. 

(Phot.  Insinger.) 


kings  of  the  Theban  Amon,  less  wealthy  than  their  rivals  "of  the 
north,    confined  themselves  to  repairing  or  completing  the  work 

of  the  Ahmessids 
and  Ramessids;  it 
was  all  they  could 
do  to  save  enough 
money  to  complete 
the  decoration  of 
the  temple  of 
Khonsu ,  or  to 
build  chapels  in  a 
vicious  style,  such 
as  that  of  Osiris, 
Master  of  Eternity, 
in  the  eastern 
quarter  of  Karnak. 
The  foundation  of 
the  Kingdom  of 
Napata,  which  se- 
parated Ethiopia 
from  the  Said 
shortly  after  the  accession  of  Shashank  I.,  had  cut  off  one  of  their 
chief  sources  of  revenue,  the  gold  they  had  obtained  from  the 
rivers  and  placers  of  the  Upper  Nile.  Some  of  the  artists 

employed  by  the  successors  of  Her- 
Heru  followed  their  descendants 
into  their  new  country,  and  the 
Theban  School  was  thus  divided  into 
two  branches ;  one  of  these  dragged 
out  a  languid  existence  in  its  ancient 
workshops,  while  the  other  started 
on  a  new  career  in  the  far  south. 
Ethiopia  has  not  been  sufficiently 
explored  to  enable  us  to  judge  its 
art  equitably.  The  temples  of  Napata 
do  not  seem  to  have  differed  much 
from  those  of  Thebes  in  plan  and 
decoration.  The  most  ancient  of 
them  (Fig.  410),  that  which  the 
victorious  Piankhi  erected  about  the 
middle  of  the  eighth  century  before 
Christ  at  the  foot  of  the  Holy  Moun- 
tain, recalls  the  plan  adopted  by 
218 


FIG.  4ig.— THE  NAOS  OF  NECTA- 
NEBUS  AT  EDFU.  (Phot.  Beato.) 


THE  SAlTE  AGE 


FIG.  420.— vEDICULA   ON   THE   TERRACE   OF  THE 
TEMPLE  OF  HATHOR  AT  DENDERAH.  (Phot.  Beato.) 


Amenophis  III.  at  Luxor.  It  is  entered  by  a  court  some  150  feet 
long  bordered  by  porticoes,  the  twenty-six  columns  of  which  are 
about  6  feet  in  diameter; 
next  comes  a  kind  of 
pronaos  with  forty-six 
columns,  then  a  hypostyle 
hall  with  ten  columns, 
and  behind  it,  the  three 
chapels  of  the  Theban 
triad,  that  of  Amon  in 
the  centre,  those  of  Mut 
and  Khonsu  right  and 
left.  The  temple  of  Tir- 
hakah,  which  is  later  by 
about  forty  years,  belongs 
to  the  category  of  the 
hemi-speos,  and  we  might 
suppose  it  to  have  been 
copied  from  one  built  by 

Rameses  II.,  Derr  or  Wadi-es-Sabu'a ,  were  it  not  for  certain 
original  features.  A  portico  of  eight  columns  gives  access  to  a 
court  of  greater  length  than 
breadth,  with  porticoes  of  eight 
square  pillars,  on  the  outer  faces 
of  which  were  colossal  figures 
of  the  god  Bes.  The  sanctuary 
is  completely  imbedded  in  the 
rock,  and  the  approach  to  it  is 
guarded  by  two  pillars  of  Bes 
(Fig.  411).  It  is  improbable  that 
this  type  was  invented  by  the 
Ethiopian  architects,  and  indeed 
it  is  derived  from  the  Osirian 
pillar  used  at  Medinet-Habu ;  we 
should  no  doubt  find  it  in  Egypt, 
if  all  the  monuments  built  after 
the  disappearance  of  the  Rames- 
sids  still  existed.  The  distin- 
guishing peculiarity  of  the  exam- 
ples at  Napata  is  a  certain  fury 
of  execution  which  gives  this 
essentially  savage  god  a  character 
even  more  savage  than  that 


219 


FIG.  421.— HATHOR  COLUMNS  OF  THE 

PRONAOS  AT  DENDERAH. 

(Phot.  Beato.) 


ART  IN  EGYPT 


FIG.  422.— DOUBLE  COURT  AND  PRONAOS 
AT  KOM-OMBO.  (Phot.  Beato.) 


bestowed  on  him  by  the  pure  Thebans;  a  strain  of  Sudanese 
barbarism  was  already  manifesting  itself  in  the  exiled  branch  of 

the  Theban  school. 
In  Egypt,  on  the  other 
hand ,  the  Ethiopian 
Pharaohs,  transform- 
ed into  the  masters 
of  their  former  mas- 
ters by  an  unimagin- 
able turn  of  fortune, 
made  no  changes  in 
local  methods.  The 
chapel  commemorat- 
ing his  accession 
which  Tirhakah  inter- 
polated more  or  less 
successfully  to  the 
north  west  of  the 
holy  lake  at  Karnak, 

between  the  wall  of  Meneptah  and  the  rampart  of  Rameses  II., 
is  built  and  decorated  in  a  deplorable  manner,  but  wholly  with- 
out any  traces  of  exoticism.  Though  the  huge  column  of  the 
great  court  (Fig.  412),  the  only  fragment  still  erect  of  the 
triumphal  propylaea  he  had  designed  to  build  in  front  of  the 
pylon  of  Rameses  in  place  of  the  avenue  of  rams,  belongs  to  a 
new  order  of  conception,  that  conception  is  purely  Egyptian, 
and  marks  the  terminal  point  of  that  slow  evolution  which  had 
been  accomplished  in  the  course  of  several  centuries  in  the 

workshops  of  the  Thebaid 
and  the  Delta. 

For  indeed  this  little 
known  period  which  lasted 
from  the  Twenty  -  first 
Dynasty  to  the  end  of 
the  Persian  Period,  coin- 
cided with  a  rich  after- 
math of  developments 
and  transformations  ,in 
architecture.  Elements  it 
had  previously  neglected, 
and  combinations  of  which 
it  had  not  thought,  were  suddenly  manifested  to  it,  and  the  uses 
it  learned  to  make  of  them,  though  they  did  not  modify  the 

220 


FIG.  423.— THE  KASR-KARUN  (After  Lepsius). 


THE  SAITE  AGE 


HBB 


FIG.  424.— ROMAN  TEMPLE  AT  MEDiNET-HABU  (After  Beato). 

main  principles  of  its  doctrine ,  led  it  to  regulate  its  practice 
more  strictly,  and  to  vary  it.  Thus  the  open  portico  which  formed 
the  fagade  of  the  hypostyle  hall  in  many  of  the  Theban  temples, 
at  Luxor,  in  the  temple  of  Khonsu,  in  the  Ramesseum,  and  at 
Medinet-Habu ,  sometimes  on  a  level  with  the  court  of  honour, 
sometimes  from  6  to  10 
feet  higher,  with  a  low 
balustrade  right  and  left 
of  the  central  staircase 
leading  up  to  it,  grew  in 
depth  and  breadth  to  the 
proportions  of  a  monu- 
mental vestibule,  a  pronaos 
of  three  or  four  rows, 
closed  in  front  by  a  party 
wall  half-way  up  the  co-  ':~m&>  -m 

lumns  of  the  outer  row, 
pierced  by  a  single  door 
on  the  longitudinal  axis 
of  the  building.  Again, 
architects  conceived  the 
idea  of  placing  normally 

in  front  of  the  pylon,  propylaea  more  or  less  imposing,  conducting 
the    faithful   to    the    true    entrance   of   the   house    of   the   god, 

221 


FIG.  425. — FACADE  OF  THE  TEMPLE  OF 
DER-EL-MEDINET.  (Phot.  Baraize.) 


ART  IN  EGYPT 


FIG.  426.— PRONAOS  OF  THE  TEMPLE  OF 
DER-EL-MEDINET.   (Phot.  Beato.) 


by  a  kind  of  triumphal  way  ana- 
logous on  a  small  scale  to  that 
which  connected  Luxor  and  Kar- 
nak,  several  doors  rather  near 
one  another;  these  had  two 
massive  jambs  surmounted  by 
the  usual  cavetto,  and  then  an 
oblong  hall  open  to  the  sky, 
enclosed  by  a  variable  number 
of  columns  in  pairs,  connected 
by  low  walls  with  the  uraeus 
cornice;  instead  of  an  archi- 
trave there  was  a  series  of  divine 
emblems,  probably  the  falcons 
or  hawks  of  Horus,  for  which 
the  cubes  on  the  columns  served 
as  bases.  As  I  have  just  said, 
the  columns  of  Tirhakah  at 
Karnak  belonged  to  propylaea 
of  this  kind,  the  most  ancient 

known  to  me ;   it  had  nine  sisters,  of  which  only  the  lower  courses 

now    exist.      In    the    interior, 

the  arrangement  of  the  chapels 

had    been    regularised,    and 

the  result  was  a  very  sensible 

amelioration  in  the  disposition 

of  the   sanctuary.     The  Hall 

of  the  Sacred  Boat,  formerly 

a   room   open  at  either  end, 

lost  its  four  colums,  and  was 

closed  up  at  the  back,  so  that 

it  could  only  be  entered  by 

a  single  door  facing  the  pylon. 

The  other  rooms  were  ranged 

hierarchically   round    it,    the 

Hall  of  the  Statues,  which  was 

the  actual  sanctuary,  immedia- 
tely behind  it,  and  the  rooms 

of  the  paredri  right  and  left 

of   the   Hall   of  Statues,   on 

three    sides    of   the    passage 

which  isolated  the  Hall  of  the 

Boat.     This   was    no   abrupt 


222 


FIG.  427. — THE  MAMMISI  AT   EDFU. 
(Phot.  Oropesa.) 


THE  SAITE  AGE 


metamorphosis;  we  note  its  gradual  evolution  in  several  temples 
of  the  Saite  period  which  escaped  destruction  or  restoration 
under  the  Ptolemies.  The  earliest  example  of  a  sanctuary  whit 
a  single  door,  encir- 
cled by  an  ambulatory, 
is  furnished  by  the 
chapel  of  Amenartas 
at  Medinet-Habu,  but 
350  years  later,  Ale- 
xander II.  and  Philip 
Arrhidseus  built  at 
Luxor  and  Karnak, 
sanctuaries  with  two 
doors  in  the  long  axis 
of  the  room.  We  are 
rather  better  informed 
as  to  these  matters 
since  the  temple  of 
Hebt  (Fig.  413)  in  the 
Great  Oasis,  has  been 
more  closely  studied. 
Founded  by  Darius  I. 
and  finished  by  Necta- 
nebus  II.,  it  has  both 
the  successive  doors 
and  the  preliminary 
hall,  but  the  three 
sanctuaries  are  still 
side  by  side  as  in 
temples  of  the  purely 
Theban  order  (Fig. 
414),  and  the  central 
shrine  is  a  veritable 
the  walls  of 


naos, 


FIG.  428. — A  WALL  IN  THE  TEMPLE   OF  ISIS 
AT  PHIL^E.     (Phot.  Beato.) 


which     are      covered 
with    sacred    pictures 

on  a  small  scale.  It  is,  to  tell  the  truth,  a  second-rate  building, 
relegated  to  one  of  the  poorest  and  most  distant  provinces, 
and  it  can  give  us  but  little  idea  of  the  true  temple  of  the 
Said;  nevertheless,  when  we  compare  it  with  the  debris  of  the 
same  period  which  has  survived,  and  then  with  the  complete 
temples  of  the  Ptolemies,  we  are  easily  convinced  that  from 
the  time  of  Psammetihus  I.  to  the  Macedonian  conquest,  archi- 

223 


ART  IN  EGYPT 


tects  produced  far  more   than  was  supposed 
was    in    its    infancy.     We    can    hardly    assert 


FIG.  429. — THE  LONG  WESTERN  PORTICO  AT  PHILjE. 
(Phot.  Beato.) 


Ombo;  if  they  did  not  create  masterpieces, 
difficult  to  form  an  opinion,  at  least  they 
taught  their  pupils  how  to  produce  them. 
Let  us  take  the  best  preserved  of  their 
temples  now  remaining,  the  only  complete 
one  indeed,  that  of  Horus  at  Edfu  (Fig.  415). 
It  is  in  the  form  of  a  much  elongated  rect- 
angle ,  running  from  north  to  south  on  its 
main  axis,  and  with  its  fagade  to  the  south 
(Fig.  416).  Roughly  speaking,  the  arrange- 
ment is  the  same  as  in  the  temple  of  Khonsu, 
but  the  differences  in  detail  are  very  great. 
In  the  first  place,  in  the  temple  of  Khonsu, 
as  in  that  of  Rameses  III.,  and  all  the  known 
temples  of  the  Theban  age ,  the  exterior 
wall,  which  starts  from  one  of  the  towers 
of  the  pylon,  turns  twice  at  a  right  angle, 
and  comes  back  symmetrically  to  the  other 
tower,  is  not  a  free  wall,  forming  an  enclo- 
sure for  the  body  of  the  building;  it  is 
bound  up  indissolubly  with  it,  in  such  a 

224 


when  Egyptology 
as  yet  that  they 
showed  any  strong 
inventive  faculty, 
but  it  is  clear  that 
they  strove  to  get 
new  results  from 
the  elements  they 
had  inherited  from 
their  predecessors, 
and  that  they  em- 
ployed them  more 
rigorously  and  lo- 
gically, if  in  a  less 
grandiose  manner. 
Egypt  owes  them 
the  magnificent  and 
harmonious  plan 
she  applied  in  her 
closing  centuries 
at  Philae,  Dende- 
rah,  Edfu  and  Kom- 
as  to  which  it  is 


FIG.  43O. 

MIXED  HATHOR 

CAPITAL  AT  PHIL.S. 


THE  SAITE  AGE 


manner  that  the  rooms  which 
surround  the  sanctuary  abut 
upon  it,  and  forms  their 
back  wall.  In  the  temple 
of  Horus,  on  the  other  hand, 
it  is  independent,  save  at 
the  extremities,  where  it 
joins  the  pylon,  and  it  enve- 
lopes the  temple  without 
touching  it  at  any  point; 
it  interposed  like  a  screen 
between  the  external  world 
and  the  domain  of  the  god, 
and  prevented  the  profane 

from    seeing   anything   that 

•     -j       c 
was  going  on  inside,   oome- 

times  it  was  dispensed  with, 
when  time  or  money  was 
lacking,  and  this  was  the 
case  at  Denderah,  and  pro- 
bably also  at  Esneh.  The  FIG. 
Saite  and  Ptolemaic  plan 
had  then,  as  an  addition  to 
the  main  features  of  the  Ramesside 


FIG.  432.— EAST  FACADE   OF  THE   MAMMISI 
AT  PH1L.E.      (Phot.  Fiorclli.) 

225 


4JI.— THE  KIOSK    OF  NECTANEBUS 
AT  PHILvE.      (Phot.  Fiorelli) 


plan,  what  I  may  call  an  iso- 
lating barrier,  and 
as  soon  as  the 
gateway  is  passed, 
the  differences  be- 
come more  mark- 
ed. The  Rames- 
side colonnade 
ran  along  the 
two  lateral  walls, 
and  joined  the 
facade  or  the  se- 
cond pylon,  as  in 
the  Ramesseum, 
or,  if  the  facade 
had  its  own  raised 
portico  with  a  bal- 
ustrade between 
the  columns ,  as 
in  the  temple  of 
Q 


ART  IN  EGYPT 


FIG.  433.  — CAPITALS  OF  THE  COLUMNS  OF  THE 
PRONAOS  AT  PHIL^E.  (Phot.  Beato.) 


Khonsu   and    at   Medinet-Habu,    it  was    merged   more    or   less 
adroitly  into  the  colonnade.    In  the  temple  of  Horus  the  portico 

is  set  into  the  back'  of 
the  pylon  on  either  side 
of  the  door  (Fig.  417);  it 
then  follows  the  lateral 
isolating  walls,  and  re- 
mains as  distinct  from  the 
main  building  as  these ; 
it  stops  a  short  distance 
from  their  two  angles,  and 
leaves  a  narrow  space 
between  them  and  itself. 
The  pronaos  here  has 
three  rows  of  columns, 
but  the  balustrade  of  the 
Ramessids  was  raised  be- 
tween the  columns  of  the  first  row,  and  its  panels  acted  as 
an  effectual  screen  for  the  more  intimate  rites  (Fig.  418).  The 
court  was  the  only  part  of  the  divine  house  which  was  really 
public;  only  those  privileged  by  their  religious  calling  or  social 
rank  penetrated.. beyond  it,  and,  the  better  to  emphasise  the 
division ,  a  partition-wall ,  the  postern  of  which  was  carefully 
closed  in  general,  barred  the  ambulatory  between  the  enclosing 

wall  and  the  sanctuary 
to  the  height  of  the 
pronaos ;  the  vulgar  were 
never,  it  would  seem, 
admitted  to  view  the 
scenes  from  the  life  and 
wars  of  Horus  with  which 
the  inner  walls  were 
adorned.  When  the  pro- 
naos had  been  passed 
through,  the  sort  of  fore- 
court into  which  one 
entered,  consisting  of  a 
hypostyle  hall,  a  vestibule, 
and  its  adjacent  chambers, 
was  more  or  less  on  the 
ancient  plan ,  but  the 

private  apartments  of  the  god  were  arranged  in  the  new  manner. 
Passing  through  an  anteroom  common   to  both  staircases  of  the 

226 


FIG.   434.      PTOLEMAIC   DOORWAY    BEFORE   THE 

TEMPLE"  OF  KHONSU.    (Phot.  Beato.) 


THE  SATTE  AGE 


FIG.  435.— THE  PROPYLAEA  AND  FACADE  OF 

THE  ORATORY  OF  THE  THEBAN  PTAH. 

(Phot.  Legrain.) 


terrace,  the  little  court  giving  access  to  the  Chapel  of  the  New 
Year,  several  dark  chapels,  and  finally  the  tabernacle  of  the 
sacred  boat  were  reached. 
In  this  the  Bubastites  had 
replaced  the  wooden  naos 
of  antiquity,  by  a  naos 
of  chased  and  polished 
stone;  under  the  Saites, 
each  temple  contained  at 
least  one  of  these,  some- 
times of  considerable  size, 
which  was  much  more  the 
real  home  of  the  god  than 
the  boat.  That  at  Edfu 
(Fig.  419)  has  only  three 
lines  of  inscriptions  on 
the  door-jambs,  but  it  is 
the  best  of  those  we 
know  as  regards  beauty  of  material  and  finish  of  workmanship. 
The  passage  which  encloses  the  naos-chamber  on  three  sides  is 
bordered  by  chapels  assigned  to  the  gods  of  the  Enneas;  that 
at  the  end,  where  the  rites  of  worship  were  carried  on  while 
the  chamber  of  the  boai  remained  closed,  was  Kere  called  the 
Forge ,  in  memory  of  the  blacksmiths  who  helped  Horus  to 
conquer  Egypt,  and  who  were  supposed  to  have  been  his  first 
priests.  I  may  add  in 
conclusion  that  -the  ter- 
races, formerly  forsaken, 
or  given  up  to  the  sacred 
astronomers,  had  acquired 
considerable  importance, 
thanks  to  the  expansion 
of  the  Osirian  myths.  On 
appointed  days  the  mys- 
teries of  the  passion  and 
resurrection  or  Osiris  were 
celebrated  here  in  shrines 
(Fig.  420),  or  before 

stelae    designed    for    this        FIG  436._THE  KIOSK  OF  TRAJAN  AT  pHILyK> 
purpose;     the    parapets,  (Phot.  Beato.) 

which    were    made    very 

high,  prevented  the  people  from  seeing  anything  of  the  spectacle 
offered  to  the  priests  and  their  associates. 

227  Q2 


ART  IN  EGYPT 


FIG.  437.— PLAN   OF  THE  ISLAND  OF  PHIL^E 
(After  Cne.  Lyons). 


Such  was  the  final  type  of  the  Egyptian  temple.    It  had  already 
taken   form   under  Alexander,    and  it   persisted    throughout    the 

last  centuries  of  paganism, 
subject  to  the  slight  vari- 
ations which  local  worship 
imposed  on  its  architects. 
As  Denderah,  for  instance, 
was  the  palace  of  Hathor, 
the  columns  of  the  pronaos 
had  the  Hathor  -  capital 
with  four  faces,  and  the 
ceiling  rested  on  a  forest 
of  immense  sistra  (Fig. 
421).  At  Kom-Ombo, 
where  the  worship  of  a 
single  divinity  had  given 
way,  under  the  Theban  Pharaohs,  to  that  of  the  twin  gods  Ho- 
rus  and  Set,  and  afterwards,  when  Set  was  proscribed  for  the 
murder  of  Osiris,  that  of  Horus  and  Sobek,  a  single  dwelling 
would  not  have  sufficed  for  the  habitation  of  two  beings  so 
diverse  as  the  falcon  and  the  crocodile;  the  temple  was  accord- 
ingly divided  into  two  parallel  aisles  joined  together  in  the 
longer  axis,  but  it  cannot  be  described  as  two  adjoining  temples; 

though  there  are 
two  separate  rooms 
for  the  boats,  the 
whole  forms  but  a 
single  block  en- 
compassed by  a 
corridor ,  along 
which  are  the  rooms 
of  theEnneas,  those 
on  the  south  for 
Sobek ,  those  on 
the  north  for  Ho- 
rus. In  the  same 
manner  the  other 
constituents  are  not 
doubled ,  but  are 

FIG.    438. — SOUTH   FRONT   OF    THE    TEMPLE    OF    ISIS  AT  !•     •   1       1  1        • 

PHIL*  DURING  THE  INUNDATION.     (Phot.  Fiorelli.)  divided     each     into 

two  sectors  belong- 
ing respectively  to  the  hawk  and  the  crocodile  according  as  to 
whether  they  are  situated  north  or  south  of  the  diameter  drawn 

228 


THE  SAITE  AGE 


from  east  to  west.  Neither  are  there  two  pylons,  but  between 
the  towers  of  the  usual  pylon  are  twin  doors,  the  openings  of  which 
correspond  to  those  giving  access  to  the  two  chambers  of  the 
sacred  boats.  We  have  thus  a  pronaos,  a  hypostyle  hall ,  three 
bi-lobate  rooms,  if  I  may  so  express  it,  and  at  their  sides  the 
usual  closets  and  staircases.  Two  stone  walls  with  two  ambu- 
latories ran  round  these ;  the  exterior  wall  abutted  on  the  pylon 
as  at  Edfu,  but  the  in- 
terior wall  formed  a 
prolongation  of  the 
pronaos  north  and  south; 
it  was  furnished  on  the 
east  with  a  row  of  cells, 
probably  to  lodge  the 
officiating  priests,  three 
for  Horus  on  the  north- 
east, three  for  Sobek  on 
the  south-east;  in  the 
centre,  on  the  line  of 
junction,  a  staircase  led 
to  upper  storeys  now 
demolished.  The  plan, 
drawn  on  paper,  shows 
us  a  monument  too  wide 
for  its  length,  the  dis- 
proportion of  which  must 
have  emphasised  the  hea- 
viness for  which  Egyptian 
architects  are  blamed.  We 
ask,  indeed,  with  some 
uneasiness  if  the  effect 
of  that  gaping  double 

void  between  the  towers  of  the  pylon  can  have  failed  to  be 
disastrous;  perhaps,  however,  there  was  no  pylon,  but  merely  a 
wall  like  that  which  encloses  the  court  of  Amenophis  III.  at  Luxor. 
Apart  from  this  point,  as  to  which  there  is  no  certainty,  the 
scholars  and  travellers  who  visit  Kom-Ombo  find  nothing  to 
offend  the  eye  (Fig.  422) ;  the  master  who  conceived  and  executed 
it  under  the  Ptolemies  reconciled  the  general  traditions  of  the 
school  and  the  exigencies  of  religion  with  a  great  deal  of  tact. 
Temples  on  a  smaller  scale,  such  as  I  have  called  the  temple 
of  the  small  town,  lent  themselves  less  readily  to  changes  than 
the  others.  Although  they  remained  faithful  to  the  Ramesside 

229 


FIG.  439.— EAST  BLOCK  OF  THE  SECOND 
PYLON  AT  PHIL^E.  (Phot.  Beato.) 


tradition,  the  examples  at  Sanhur,   at  El-Kalaa,  near  Coptos,  of 
Assuan,  of  Dakhlah  in  the  Great  Oasis,  of  Kasr-Karun  (Fig.  423) 

and  even  that  at  Medinet-Habu 
(Fig.  424),  all  dating  from  the 
Roman  period,  have  not  proved 
sufficiently  attractive  to  engage  the 
attention  of  archaeologists.  Others 
are  but  the  accessory  features  of 
a  greater  building,  such  as  the 
delicate  chapel  of  Hathor  at  Philae 
with  its  dainty  court  and  single 
chamber.  The  best  specimen  of 
these  is  the  charming  temple  of 
Hathor  and  Amenophis  at  Der-el- 
Medinet,  which  fascinated  the  artists 
of  the  French  Commission,  but  has 
been  unduly  neglected  since  their 
time.  It  stands  like  a  sentry  towards 
.the  entrance  of  the  savage  gorge 
which  leads  from  the  base  of  Kurnet- 
Murrai  to  the  Valley  of  the  Queens, 
and  the  Coptic  monks  who  occupied 
it  for  a  long  time  rebuilt  the  brick 
rampart  bestowed  on  it  by  the  Ptolemies  almost  on  the  origin- 
al foundations.  It 
forms  a  rectang- 
le, the  facade  of 
which  looks  to  the 
south-east,  and  it 
is  quite  bare  on 
the  outside ,  the 
only  ornaments 
being  the  torus 
and  cavetto  which 
crown  the  walls 
(Fig.  425).  Inside 
is  a  vestibule  sup- 
ported by  two  col- 
umns with  floriated 
capitals,  separated 
from  the  pronaos 
hv  HIP  iisnfll  naH-i  FIG-  441.— SOUTH-WESTERN  ANGLE  OF  THE  MAMMISI 

/  ,,/,-..      V*' "  AND   GREAT  DOOR   OF   THE  TEMPLE   OF  ISIS  AT  PH1L.E. 

tion- wall  (r  ig.  426);  (Phot.  Fioreiu.) 

230 


FIG44O.—  PRONAOSAND  HYPOSTYLE 

HALL   OF   THE   TEMPLE    OF    ISIS    AT 

PHIL.K.     (Phot.  Beato.) 


FIG.  442.— TEMPLE  OF  DABUD  BEFORE  RESTORATION. 
(Phot.  E.  Brugsch.) 


THE  SAITE  AGE 

the  three  sanctuaries  adjoin  the  pronaos,  side  by  side,  and  a 
staircase  against  the  western  wall  led  to  the  terraces.  The  pro- 
portions ,  the  co- 
lour, the  illumina- 
tion ,  are  all  of 
extraordinary  re- 
finement, and  even 
the  sculpture  is 
good  for  the  peri- 
od; the  seven  masks 
of  Hathor  in  a  line 
over  the  door  on 
the  front  wall  of 
the  pronaos,  form 
an  unexpectedly 
agreeable  frieze. 
The  two  other  chapels  in  the  plain  on  the  left  bank,  south  of 
Medinet-Habu,  are  far  from  possessing  the  same  attraction  for 
the  traveller.  The  nearer  of  the  two,  called  by  the  natives  Kasr- 
el-Aguz,  which  was  dedicated  to  a  local  saint  by  the  Ptolemies, 
was  preceded  by  a  portico,  but  it  consists  only  of  a  wide  vesti- 
bule, and  three  dark  chambers  in  a  line  one  behind  the  other. 
The  other,  the  Kasr-el-Shalauit,  was  built  for  the  Isis  of  Her- 
monthis  under  Otho,  Vespasian  and  Domitian,  and  finished  by 

Hadrian   and    Antoninus 

Pius;  it  contains  an  iso- 
lated sanctuary  in  the 
middle,  two  little  rooms 
on  the  north-west,  three 
on  the  south-west,  and 
in  the  last  of  these,  the 
narrow  staircase  leading 
to  the  roof-terraces.  These 
little  temples  must  not  be 
confused  with  the  Mam- 
misi ,  the  house  of  re- 
treat to  which  the  god- 
dess retired  every  year 

FIG.   443.— TEMPLE   OF  TAFEH  BEFORE  tO      Spend      in      Tltual      SOU- 

RESTORATION.    (Phot.  E.  Brugsch.)  tude    the    unclean  weeks 

of  travail  and  convales- 
cence. We  know  by  the  scenes  depicted  at  Luxor  and  Der- 
el-Bahari,  that  the  queens  or  mother-divinities  were  placed  in 

231 


ART  IN  EGYPT 


a  kind  of  quarantine  at  the  birth  of  their  children,  but  their 
usual  prison  was  concealed  in  the  interior  of  their  houses.  Com- 
plete isolation  and  retirement  became 
obligatory  under  the  Ethiopians  or  the 
Saites.  and  the  first  indubitable  examples 
of  it  are  contemporary  with  the  Ptole- 
mies. The  Mammisi  stood  at  a  short 
distance  from  the  principal  temple,  in 
front  of  the  facade  and  to  the  right 
on  coming  out  of  the  pylon  at  Edfu 
(Fig.  427),  at  Kom-Ombo,  at  Philae, 
and  to  the  left  at  Denderah;  only  in 
one  instance,  at  the  temple  of  Khonsu 
at  Karnak,  was  it  placed  behind  the 
facade,  almost  in  the  shadow  of  the 
western  wall  of  the  hypostyle  hall.  In 
general ,  it  was  built  in  the  form  of 
the  hypaethral  temple,  which  had  been 
abandoned  since  the  Eighteenth  Dyna- 
sty, and  it  consisted  almost  everywhere 
of  a  single  room,  surrounded  by  a 
colonnade,  and  preceded  sometimes  by  a 
court  and  a  monumental  doorway;  the 
columns  are  of  the  Hathoric  order  at  Edfu,  images  of  the  god 
Bes  being  applied  to  them  at  Denderah,  with  the  complication 
of  a  floriated  capital  at  Philse.  The  Mammisi  of  Opet  at  Karnak 
is  conceived  in  a  totally  different  spirit,  which  I  attribute  to  a 
purely  local  cause.  Popular  piety  had  enshrined  the  most  deeply 

venerated  of  Theban  relics, 
the  head  of  Osiris,  in  the 
very  building  where  it  was 
supposed  the  godwasborn 
of  Opet  and  the  shrine 
had  been  placed  against 
the  ordinary  Mammisi. 
New  arrangements  were 
necessitated  by  a  new  use ; 
the  chamber  of  travail  with 


FIG.   444.— PLAN   OF   THE 

TEMPLE    OF    KALABSHAH 

(After  Barsanti). 


FIG.  445.— TEMPLE   OF   DAKKEH 
(After  a  drawing  by  Burton). 


its  Hathor- columns  was 
accompanied  by  corridors 
and  rooms,  the  most  re- 
mote of  which,  a  sort  of  sculptured  crypt,  was  the  tabernacle, 
or  rather  the  cenotaph  of  the  god. 

232 


THE  SAITE  AGE 


Like  the  plan,  the  elements  of  construction  and  decoration 
were  directly  derived  from  those  of  the  Ramesside  age,  but  the 
use  of  them  had  been  systematised  and  defined.  The  idea  that 
not  only  the  whole  temple,  but  each  of  its  chambers,  was  an 
image  of  the  world,  predominates  more  and  more,  and  manifests 
itself  even  in  minute  details  of  ornamentation.  Thus  the  lower 
part  of  the  walls  was  covered  with  a  flora  and  a  fauna  the  variety  of 
which  increased  from  century  to  century,  from  Edfu  to  Kom-Ombo, 
Denderah  and  Philae.  Bouquets  of  field  and  river  flowers,  buds 
and  blossoms  of  lotus,  thickets 
and  marshes  in  the  midst  of 
which  oxen  wander  and  birds 
nestle,  processions  of  Niles  and 
Nomes  bearing  the  customary 
tribute  in  great  profusion,  all 
that  land  and  river  produce  was 
figured  by  the  Ptolemaic  and 
Roman  architects,  in  places  where 
the  Ramessids  had  prescribed 
a  plain,  smooth  band,  painted 
black,  and  separated  from  the 
carved  panels  of  the  wall  by 
superposed  lines  of  red,  yellow 
and  white.  This  rustic  deco- 
ration admitted,  in  addition 
to  natural  forms,  conventional 


FIG.  446.— KIOSK   OF  KARTASSI. 
(Phot.  E.  Brugsdi.) 


motives,  flowers  united  into  arti- 
ficial bouquets  with  interlace- 
ments, divine  emblems  such  as 
the  lily  crowned  with  feathers 
of  Nefer-temu,  even  prisoners  bound  together  in  couples  at  the 
stake.  The  decoration  of  ceilings  was  no  longer  restricted  to  a 
sprinkling  of  yellow  stars  on  a  dark  blue  ground,  or  a  flight  of 
vultures;  they  exhibit  the  constellations  of  the  Egyptian  heavens 
with  their  protecting  divinities  or  their  conventional  animal  sil- 
houettes ;  in  certain  chambers  in  the  Chapels  of  the  New  Year  and 
the  Osirian  processional  altars  on  the  terraces  at  Denderah,  zodi- 
acs composed  in  imitation  of  those  of  the  Greeks,  replace  the 
stellar  imagery  of  purely  Egyptian  origin.  Between  plinth  and 
cornice,  that  is  to  say,  according  to  doctrine,  between  heaven 
and  earth,  there  was  a  like  profusion  of  pictures.  There  are  no 
longer  only  two  or  three  registers  upon  the  walls,  but  five  or 
six  or  more,  and  their  multiplication  was  the  natural  consequence 

233 


ART  IN  EGYPT 


FIG.  447. — CHAPEL  IN  THE  QUARRIES 
OF  KARTASSI.     (Phot.   Thedenat.) 


of  the  auxiliary  virtue  attributed  to  divine  figures  and  their  actions.  If 
two  or  three  scattered  divinities  preserved  a  chapel  from  destruction, 

twenty  or  more  would  necessarily 
increase  the  security.  Moreover, 
the  empty  spaces  between  the 
figures  were  like  so  many 
breaches,  through  which  evil  in- 
fluences might  penetrate;  artists 
did  their  best  to  reduce  them 
by  increasing  the  number  and 
length  of  formulae  and  explana- 
tory texts.  Thus  the  Ptolemaic 
and  Roman  temples  are  literally 
peppered  with  scenes  and  in- 
scriptions, the  closer  and  more 
minute  as  the  date  of  construction 
advances ;  a  hall  of  medium  size 
at  Edfu ,  Denderah ,  Ombo  or 
Philae  (Fig.  428)  contains  more 
than  the  great  hypostyle  hall 
of  Karnak.  And  in  many  cases, 
the  magic  of  this  external  defence 
was  doubled  by  an  internal  defence  known  only  to  the  initiated. 
In  Pharaonic  buildings  there  are  few  examples  of  secret  chambers, 

corridors  or  cabinets  conceal- 
ed in  the  thickness  of  the  walls. 
I  know  of  one  at  Abydos  in 
the  Memnonium  of  Seti  I., 
another  at  Medinet-Habu  in 
the  cenotaph  of  Rameses  III., 
and  finally,  one  in  the  temple 
of  Khonsu  to  the  left  of  the 
sanctuary.  The  last  is  a  veri- 
table crvpt  near  the  ceiling, 
13  feet  from  the  ground,  in 
the  space  between  two  acces- 
sible chapels,  and  was  perhaps 
the  hiding-place  of  the  priest 
who  pronounced  the  oracles. 
On  the  other  hand,  the  Ptole- 
maic and  Roman  temples  al- 
ways contain  a  number  more 
or  less  great,  which  are  group- 


FIG.  448.— THE  EAST  GATE  OF 
KASR-IBRIM.     (Phot.  Oropesa.) 


THE  SAlTE  AGE 


ed  in   general    round  the  'sanctuary   of  the  boat.    They  seem  to 
have  been  built  after  the  downfall   of  the   national  dynasties  to 


FIG.  449. — FIELD   OF  THE  PYRAMIDS   OF  MERGE 
(After  Lepsius). 

receive  the  sacred  treasures,  the  precious  utensils,  the  jewels,  statues, 
and  mystic  emblems,  all  the  material  used  by  Pharaoh  in  the 
rites  he  alone  could  celebrate,  which  naturally  fell  into  disuse 
when  foreign  monarchs  ascended  the  throne;  it  was  stored  in 
subterranean  cells,  in  the  hope  that  the  day  would  come  when 
the  ancient  sovereignty  would  be  restored,  and  the  antique  rites 
would  reappear  in  all  their  splendour.  Edfu  possesses  at  least 
two,  and  Denderah  more  than  a  dozen,  some 
of  which  are  bare,  while  others  are  orna- 
mented with  bas-reliefs  which  indicate  the 
uses  to  which  they  were  destined.  Although 
they  were  entered  only  by  a  few  priests, 
their  sculptures  are  as  carefully  executed  and 
as  delicate  in  style  as  those  of  the  public 
rooms ;  the  god ,  who  could  see  everything, 
would  not  have  tolerated  mediocrity  in  his 
secret  places,  any  more  than  in  his  public 
domain.  Such  a  room  was  entered  by  an  aper- 
ture ,  sometimes  on  a  level  with  the  floor, 
sometimes  placed  so  high  that  it  could  not 
be  entered  without  a  ladder;  a  movable  stone, 
concealed  in  the  decoration,  masked  this  in- 
gress, known  only  to  the  priests,  who  trans- 
mitted the  secret  from  generation  to  genera- 
tion. 

Pillars  and  columns  were  modified  by  the 
same  influences  as  mural  decoration,  not  in 
the  base  and  the  shaft,  which  remained  essen- 

235 


FIG.    450.— PLAN    AND 

ELEVATION       OF       AN 

ETHIOPIAN      PYRAMID 

(After  Lepsius). 


FIG.  451. — THE  TEMPLE   OF  MESAURAT 
(After  Lcpsiu.s). 


ART  IN  EGYPT 

tially  identical  with  those  of  the  Ramesside  age,  but  in  the  capi- 
tal.   The  pillar  was  used  much   less  than   in  earlier  times;   or  at 

least  we  rarely  see  it  in 
the  temples  which  have 
survived.  On  the  other 
hand,  the  evolution  of 
columns  and  their  capitals, 
once  initiated,  continued 
rapidly,  and  a  hasty  survey 
of  some  of  the  most  fa- 
mous temples,  those  of 
Edfu  or  Philse,  show  what 
a  variety  of  forms  it  en- 
gendered; the  number  was 
so  great  as  to  suggest  to 
architects  a  highly  ingeni- 
ous combination.  Whereas  their  Ramesside  predecessors  had,  as 
a  rule,  allowed  only  a  single  kind  in  any  one  part  of  a  building, 
and  even  when  they  introduced  two,  as  in  their  hypostyle  halls, 
rigorously  restricted  each  to  the  special  place  assigned  to  it, 

these  later  architects  mingled  them  all, 
but  in  such  a  manner  that  each  kind 
was  represented  by  two  symmetrical 
examples  in  the  general  scheme;  thus 
at  Edfu,  the  dactyliform  or  papyriform 
capitals  which  succeed  each  other  in 
the  west  portico  of  the  great  court 
have  their  counterpart  at  the  same 
height  and  in  the  corresponding  row 
in  the  east  portico.  When  once  this 
principle  had  been  laid  down,  it  was 
applied  assiduously,  and  its  effect  was 
to  co-ordinate  the  somewhat  vague 
sensations  evoked  in  the  spectator's 
mind  by  the  infinite  diversity  of  types. 
When  we  enter  the  court  at  Edfu, 
or  the  long  porticoes  of  Tiberius  at 
Philae  (Fig.  429),  we  are  struck  instinc- 
tively by  the  impression  of  unity  they 
give,  and  on  reflection  we  perceive 
that  this  unity  arises  from  the  symmetry 
observed  between  one  side  and  the  other  in  the  use  of  capitals. 
In  the  beginning,  the  Saiites  were  content  to  utilise  the  styles 

236 


FIG.  452. — STATUE   OF  A 

SQUATTING    FIGURE    (Museum, 

Cairo).     (Phot.  Legrain.) 


THE  SAlTE  AGE 


FIG.  453.— PETUBASTIS 
(Museum,  Cairo).     (Phot,  E.  Brugsch.) 


for  which  the  monuments  of  the  past  furnished  models,  even 
those  which  might  have  been  supposed  to  have  been  entirely  dis- 
credited. Thus  at  Elephantine  we 
find  polygonal  columns  with  six- 
teen faces  erected  by  Amasis 
and  bearing  his  name.  This,  how- 
ever ,  was  exceptional ,  and  the 
so-called  proto-Doric  order  was 
never  used  after  his  time,  save 
in  those  cases  where  the  Ptolemies 
deliberately  imitated  the  fashions 
of  the  Eighteenth  Dynasty,  as  in 
the  chapel  of  the  Theban  Ptah, 
which  they  aspired  to  pass  off  as 
a  work  of  Thothmes  III.  Every- 
where else,  they  adhered  to  the 
styles  preferred  by  the  Ramessids, 
the  lotus,  papyrus,  and  dactyliform 
columns,  and  in  the  places  where 
religion  demanded  them,  Hathor- 
columns,  but  so  modified  as  to 
be  almost  unrecognisable.  It  even 
happened  sometimes  that  their  new 

conception  of  an  order  so  far  triumphed  over  the  old  that  the 
Hathor  capital  no  longer  consisted  merely  of  two  female  masks 
welded  together  at  the  nape  of  the  neck,  not  always  with  a  very 
happy  effect,  even  at  Der-el-Bahari ;  it  was  expanded  into  four 
faces ,  united  by 
the  folds  of  the 
headdress,  and  the 
square  cube  which 
crowned  them 
showed  over  each 
a  naos  between  two 
volutes  (Fig.  430). 
The  column  thus 
terminated  has  the 
form  of  a  gigantic 
sistrum ,  and  in 
spite  of  the  want 
of  proportion  be- 
tween the  shaft  and  the  head,  it  has  a  grand  appearance ;  there 
is  nothing  more  impressive  in  Egypt  than  the  Hathor-portico  at 

237 


FIG.  454.— OSORKON  II.   PUSHING  A  BOAT 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


FIG.   455. — AMENARTAS 
(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


Denderah.  At  Philae,  in  the  kiosk  of  Necta- 
nebus  (Fig.  431)  and  in  the  colonnade  of  the 
Mammisi  (Fig.  432),  the  architect  has  intro- 
duced the  Hathorian  motive  as  a  sort  o! 
abacus  or  cube  between  the  architrave  and 
other  capitals  with  floriated  decorations. 
The  combination  has  a  certain  bizarre  grace 
not  unpleasing;  the  art  with  which  the 
sistrum  is  incorporated  with  the  bouquet 
from  which  it  rises  is  truly  admirable.  The 
modifications  to  which  the  papyriform  column 
was  subjected  are  no  less  happy.  The  first 
step  was  to  apply  four  or  more  bunches  of 
flowers  to  the  edges  of  the  open  corolla, 
which  clothed  its  nudity;  the  details  of  this 
addition,  conventionalised  more  and  more, 
soon  gave  the  whole  the  appearance  of  a 
heraldic  lily.  This  same  ornament  laid  upon 
the  lotus  or  palm-leaf  capitals,  reduced  them 
by  degrees  to  identity  with  the  papyriform 


capital;  towards  the  end  of  the  Ptolemaic 
age  and  under  the  Caesars,  the  original 
forms  of  the  three  orders  had  become 
hardly  more  than  an  almost  invisible 
support  on  which  motives  borrowed  from 
the  flora  of  the  country,  leaves,  flowers, 
buds,  grasses,  clusters  of  dates  and 
bunches  of  grapes,  rose  in  vigorous  profu- 
sion. All  these  were  not  heaped  up  or 
applied  haphazard ;  the  various  elements 
were  gradually  combined  with  an  exqui- 
site sense  of  arrangement  and  proportion; 
designers  did  not  create  them  '  all  at 
once,  and  we  can  follow  the  development 
of  their  ideas  from  the  relatively  simple 
essays  of  Edfu  to  the  more  cunning 
complexities  of  Philae  and  Esneh.  The 
hypostyle  hall  of  the  great  temple  of 
Isis  at  Philae  undoubtedly  contains  their 
most  perfect  achievement  (Fig.  433). 
Not  only  are  the  motives  they  carved 
on  each  of  the  capitals  inconceivably  rich 
and  tasteful ;  they  heightened  the  effect 

238 


FIG.  456. 

ANKHNASNUFIABRI  (Museum, 
Cairo).     (Phot.  E.  .Brugsch.) 


THE  SAITE  AGE 


of  these  by  shades  of  green,  blue,  red,  and  yellow  incomparably 
soft  and  harmonious;  those  painted  columns,  which  the  enlarge- 
ment of  the  barrage  at  Assuan  will 
soon  destroy,  are  among  the  purest 
and  most  delicate  creations  not  only 
of  Egyptian,  but  of  any  art.  It  would 
seem  that  in  Upper  Egypt,  at  any 
rate,  the  first  Ptolemies  merely 
cleaned  and  restored  the  buildings  of 
the  Theban  Pharaohs.  Thus  Ptolemy 
Soter,  under  the  name  of  Philip 
Arrhidaeus  and  Alexander  Aigos, 
transformed  the  columned  chapels 
in  which  Thothmes  III.  and  Ameno- 
phis  III.  had  kept  the  sacred  boat  at 
Karnak  and  Luxor  respectively,  into 
dark  sanctuaries;  he  retained  the 
two  doors  of  the  original  plan,  one 
of  which,  however,  that  at  the  back, 
soon  disappeared.  The  great  temple 
of  Amon  was  encumbered  with  ex- 
voto  offerings  and  statues,  and  the 
ruins  which  the  Ethiopian  and  As- 


FIG.  457. — HEAD  OF  A  STATUE 

OF  TJRHAKAH  (Museum,  Cairo). 

(Phot.  E.  Bmgsdi.) 


Syrian  invasions  had  accumulated 
here,  had  never  been  restored  com- 
pletely. All  this  rubbish,  which  was 
of  no  interest  to  contemporaries,  was 
buried  in  an  immense  favissa  dug 
out  between  the  hypostyle  hall  and 
the  seventh  pylon;  then  the  eastern 
walls  of  the  hypostyle  hall  were 
consolidated,  the  level  of  the  floor 
was  raised  a  little,  the  pavement  was 
replaced,  and  the  bases  of  the  col- 
umns, already  eaten  away  by  salt- 
petre, were  restored.  The  monumen- 
tal doorway  in  front  of  the  temple 
of  Khonsu  was  entirely  rebuilt  (Fig. 
434).  The  oratory  of  the  Theban 
Ptah  (Fig.  435)  was  restored,  and 
a  last  pylon,  more  immense  than 
any  of  the  others,  was  being  built  in  front  of  the  pylon  of  Rameses  II., 
when,  the  town  having  revolted,  Ptolemy  Soter  II.  destroyed  it 

239 


FIG.  458.— PSAMMETICHUS  I. 
(British  Museum ,   London). 


ART  IN  EGYPT 


FIG.45g.-PSAMMETICHUSIII. 
(The  Louvre,  Paris). 


in  the  year  87  B.  C.  On  the  other  bank 
of  the  river  Ptolemy  IV.  Philopator  and 
Ptolemy  Euergetes  founded,  in  the  place- 
of  a  ruined  shrine  of  Amenophis  III.,  the 
little  temple  of  Der-el-Medinet,  dedicated, 
as  I  have  already  said,  to  the  goddess 
Hathor  and  the  magician  Amenophis,  son 
of  Hapu,  whose  worship  was  then  so 
popular.  Their  activity  was  not  confined 
to  the  limits  of  the  ancient  capital ; 
wherever  we  go  in  Middle  Egypt  and  in 
the  Delta,  we  find  grandiose  traces  of 
it,  at  Sebennythos,  at  Xois,  at  Bubastis, 
at  Heracleopolis ,  at  Oxyrrhynchos ,  at 
Hermopolis,  at  Kau-el-Kebir,  at  Akhmim, 
and  chaotic  though  the  remains  of  their 
work  are,  we  are  obliged  to  admit 
that  in  dimensions  at  least  they,  bear 
comparison  with  the  most  important  monu- 


ments of  the  Theban  age.  We  should,  however, 
have  some  difficulty  in  determining  what  was 
its  general  character,  if  the  Said  had  not  pre- 
served certain  temples,  intact,  or  nearly  so, 
at  Esneh,  Edfu,  Kom-Ombo,  Assuan,  and  Philae 
(Fig.  409).  The  Romans  continued  the  work  of 
the  Ptolemies;  the  few  buildings  or  fragments 
of  buildings  which  date  from  their  domination, 
such  as  the  pronaos  of  Esneh  and  the  kiosk  at 
Philae  (Fig.  436) ,  show  that  the  schools  of 
architecture  had  not  degenerated  under  their 
rule.  Philae  was  to  the  later  centuries  what 
Karnak  had  been  to  the  earlier  ages,  the  fa- 
voured spot  where  all  the  sovereigns  had  work- 
ed uninterruptedly,  and  consequently  where 
the  successive  phases  of  the  evolution  of  art 
may  be  most  clearly  observed  (Fig.  437),  Amasis 
had  built  a  chapel  there ,  and  in  front  of  this 
Nectanebus  had  put  up  first  a  gatewav  and  then 
a  pavilion  which  marked  the  southern  landing- 
stage  for  travellers  coming  from  the  south. 
Ptolemy  Philadelphus  demolished  the  building 
of  Amasis,  and  devised  the  present  temple 
(Fig.  438),  pylons  (Fig.  439),  courts,  hypostyle 

240 


FIG.  460. 
STATUETTE    OF  A 

QUEEN 

(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


THE  SAITE  AGE 


hall  (Fig.  440),  sanctuary,  and  mammisi  (Fig.  441),  which  where 
decorated  by  his  Macedonian  or  Ethiopian  successors.  Towards 
the  northern  end  of  the 
island,  Augustus  erected 
a  chapel  in  the  Roman 
style,  as  if  to  emphasise 
the  rights  of  the  hmpire, 
but  having  accomplished 
this  act  of  annexation, 
he  returned  to  the  native 
tradition  in  the  great 
southern  colonnade  and 
the  unfinished  kiosk,  in- 
correctly called  the  Kiosk 
of  Trajan.  It  was  also 
purely  on  the  Egyptian 
system  that  Claudius 
built  -the  sanctuary  and 
Hadrian  the  propylaea  of 

the   west.     The   town  occupied  all  the   space   not  filled   by  the 
temple ;  towards  the  close  of  the  third  century  of  our  era,  it  was 


FIG.  461.— MENTEMHET  (Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


FIG.  462.— NSIPTAH  (Museum,  Cairo). 
(Phot.  Legrain.) 


FIG.  463. — MENTEMHET 
(Museum,  Cairo).  (Phot.  E.  Brugsch.) 


a  kind  of  museum,  in  which  the  later  manifestations  of  Egyptian 
art  could  be  studied  almost  reign  by  reign. 

241  R 


ART  IN  EGYPT 


FIG.   464.— THOUERIS 

(Museum,   Cairo). 
(Phot.  E.  Brugsch.) 


Beyond  lay  what  was  still  Egypt  for  a 
time,  but  an  Egypt  which  had  relapsed 
into  a  semi-barbarous  state,  and  in  which, 
artistic  life  was  only  maintained  by  the  effort 
of  a  foreign  will.  The  kings  of  Ethiopia, 
who  disputed  the  possession  of  the  Nubian 
marches  first  with  the  Ptolemies  and  then 
with  the  Caesars,  have  left  few  traces  of 
their  supremacy  there.  They  founded  two 
temples,  however,  at  least,  one  near  Dabud 
(Fig.  442),  some  half  dozen  miles  south  of 
Philae.  the  other  atDakkeh,  on  what  was  prob- 
ably the  site  of  an  oratory  of  Thothmes  III. ; 
the  Ptolemies  completed  the  first,  the  Ro- 
mans enlarged  the  second,  and  when  they 
had  thoroughly  colonised  the  country,  they 
built  small  chapels  here  and  there-,  at  Tafeh 
(Fig.  443),  at  Kartassi,  at  Dendur,  at  Mahar- 
raka,  and  a  great  temple  at  Kalabshah.  The 
plan  (Fig.  444)  is  that  used  in  Egypt  proper 
at  the  same  period,  but  the  execution  is  less  careful.  Kalabshah 
is  not  unlike  Edfu,  and  is  imposing  in  spite  of  the  obliquity 
of  its  pylon  and  the  disproportion  between  its  depth  on  the 

main  axis  and  the  breadth  of 
its  facade;  but  Dakkeh  (Fig. 
445)  when  it  was  intact,  made 
a  far  more  favourable  impres- 
sion. The  Ethiopian  king  Er- 
gamenes  first  raised  a  small 
shrine  to  Thoth-Paotnuphis,  to 
commemorate  the  taking  of 
Dodecaschoenus,  (i.  e.  the  terri- 
tory of  the  Cataract  between 
Philae  and  Syene),  from  the 
Macedonians.  Ptolemy  Philo- 
pator  and  Ptolemy  Physcon 
successively  added  to  it  a  sekos, 
a  pronaos  with  two  columns 
in  the  fagade,  and  a  pylon; 
less  than  a  century  later,  Au- 
gustus surrounded  the  primi- 
tive chapel  with  a  number  of 
new  buildings ;  on  the  east,  an 


FIG.  4&5- 

HEAD   OF  A   SAITE  KING 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


242 


THE  SAITE  AGE 


aedicula  dedicated  to  the  two  lions  of  the  Heliopolitan  Enneas, 
Shu   and   Tefnut,    on   the   south  a   sanctuary   with   its   naos    in 


FIG.  466.— STATUE 

OF  A   PRIEST 

(Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


FIG.  467. 
STATUE   OF  PSAM- 

METICHUS  I. 
(Museum,  Turin). 


FIG.  468.— STATUE 
OF  HORUS  AS  A 
CHlLD(Mus.,  Cairo). 
(Phot.  E.  Brugsch.) 


granite,  on  the  west  a  flight  of  steps  leading  to  the  terrace  of 
the  sanctuary,  and  finally,  a  double  stone  wall,  which  made  the 
whole  a  kind  of  donjon  destined  to  protect  the  village  of  Pselchis 
against  the  barbarians. 
Dendur  is  set  against  the 
mountain,  in  front  of  the 
grotto  in  which  lies  the 
local  saint  who  was  wor- 
shipped here,  and  its 
quay,  the  triumphal  gate 
which  precedes  the  pro- 
naos,  and  the  pronaos  itself 
are  agreeable  in  design. 
The  chapel  hewn  in  the 
sandstone  of  the  quarries 
of  Kartassi  in  the  first 
century  of  our  era  is  happy 

in   its   effect  (Fig.  447).    The  kiosk  (Fig.  446),    which   is  all  that 
remains  of  the  local  temple,  recalls  the  kiosk  of  Trajan  at  Phila? 

243  R  2 


FIG.   469. — STATUETTE   OF  RECUMBENT  OSIRIS 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


FIG.  47O.— STATUE   OF 

ISIS   (Museum,  Cairo). 

(Phot.  E.  Brugsdl.) 


FIG.  471.— STATUE   OF 

OSIRIS  (Museum,   Cairo). 

(Phot.  E.  Brugsch.j 


from  a  distance,  but  it  owes  its  more  picturesque  charm  to  its 
position  on  a  kind  of  rocky  spur  on  the  river-bank.  Mahar- 
raka,  the  southernmost  temple  built  by  the  Emperors,  is  also 
the  latest.  Seen  from  outside, 
it  is  a  rectangular  box,  dull  and 
heavy  in  appearance;  inside,  it 
is  a  court  bordered  by  colonnades 
on  the  south,  west,  and  north, 
and  towards  the  north-east  angle, 
a  block  of  masonry  in  the  thick- 
ness of  which  is  a  spiral  staircase, 
the  only  one  of  its  kind  in  the 
southern  regions.  Beyond  Mahar- 
raka,  there  are  no  memorials  of 
the  Latin  domination  save  at  Ib- 
rim ,  where  the  eastern  door  of 
the  wall  still  exists  (Fig.  448), 
while  towards  the  northern  ex- 
tremity of  the  plateau  there  is  a 
nondescript  building,  half  temple, 
half  barrack,  which  certainly  dates 
from  the  time  of  Septimius  Sever- 
us.  This  outpost  of  the  Empire 

244 


FIG.  472. — THE  COW  HATHOR   OF 

SAKKARAH  (Museum,  Cairo). 

(Phot.  E.  Brugsch.j 


THE  SAITE  AGE 


evidently  remained  its  artistic,  as  well  as  its 
political  dependency,  but  did  its  influence  ex- 
tend further  south,  to  the  territories  connected 
with  the  kingdom  of  Meroe?  We  know  so 
little  of  the  history  of  Ethiopia  after  the  time 
of  Tirhakah  and  his  immediate  successors,  that 
it  is  difficult  to  say  anything  definite  on  this 
head.  The  tombs  of  the  Ethiopian  Pharaohs, 
the  plans  and  designs  of  which  have  survived, 
are  obviously  derived  from  the  pyramid-mastabas 
of  the  Theban  age.  (Fig.  449).  They  include  a 
pyramid,  in  fact,  but  it  rests  on  a  low  base, 
has  a  sharper  point  than  the  Memphite  pyramids, 
and  tori  dividing  its  faces ;  a  monumental  door, 
or  even  a  pylon,  marks  the  entrance  of  the 
chambers,  which  are  frequently  decorated  with 
bas-reliefs  (Fig.  450).  Long  after  Tirhakah,  the 
temples  continued  to  be  purely  Egyptian  in  type, 
though  they  show  a  slovenliness  of  execution 
which  betrays  the  unskilfulness  of  their  archi- 
tects and  stone-cutters,  but 
towards  the  close  of  the  first 
century  after  Christ,  Greek 


FIG.  473. 

STATUE  ROUGHLY 

BLOCKED  OUT 

(Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


FIG.  474.— STATUE 
FINISHED  BUT  FOR 

THE   HEAD 
(Museum,  Cairo). 
(Phot.    E.  Brugsch.) 


elements  began  to  mingle  with  the  Pharaonic, 
and  a  hybrid  style  was  formed,  of  which  the 
temple  of  Messaurat  is  the  best  known  example 
so  far  (Fig.  451).  It  dates  from  the  third 
century  of  our  era,  that  is  to  say,  from  the 
last  days  of  the  monarchy.  Meroitic  art  may 
have  long  survived  the  downfall  of  the  dynasties 
of  Egyptian  origin;  I  am  inclined  to  recog- 
nise its  influence  in  what  we  know  of  Axu- 
mitan  art. 

Architecture  then  had  been  dead  for  some 
time  in  Egypt  proper,  when  the  last  and  most 
obscure  of  its  progeny  perished  on  foreign 
soil.  It  was  still  so  vital  and  so  magnificent 
in  the  third  century,  when  it  produced  Ombos 
and  Esneh,  that  we  cannot  explain  its  sudden 
downfall  by  inherent  causes.  Up  to  its  supreme 
moment,  its  masters  knew  their  craft  thoroughly, 
and  applied  its  principles  with  incontestable 
superiority ;  it  would  surely  have  subsisted  for 
245 


ART  IN  EGYPT 


centuries  longer,  if  it  had  not  been 
so  thoroughly  impregnated  with 
the  religious  conceptions  to  which 
it  owed  its  inspirations ,  that  it 
could  not  survive  them.  The  Ro- 
man basilica,  and  even  the  Greek 
temple,  had  hardly  any  constituent 
elements  which  were  exclusively 
pagan ;  at  the  price  of  certain 
modifications  of  no  great  impor- 
tance, they  lent  themselves  to  the 
requirements  of  triumphant  Chris- 
tianity, and  furnished  it  with  sui- 
table churches.  But  the  arrange- 
ments which  made  the  Egyptian 
temple  the  mysterious  abode  of 
Amon,  Horus,  or  Isis  and  their 
court,  had  no  longer  any  reason 
for  existence  in  relation  to  the 
new  God.  What  need  was  there 

for  the  chambers  of  the  boat,  the  shrine  of  statues,  the  chambers 
of  the  Enneas,  the  hypostyle  halls,  the  pronaos,  in  a  religion 
the  whole  ritual  of  which  was  performed  in  public,  and  |n  which 


FIG.   475.— HEAD  OF  A   STATUE 
(Museum,  Berlin). 


FIG.  476.— HEAP    OF    A   STATUE 
(The  Louvre,  Paris). 


246 


FIG.  477.— BUST  OF  A   STATUE 
(Museum, Cairo).  (Phot.E.Brugsch.) 


THE  SAITE  AGE 


the  paraphernalia  of  images  and  vases  was  reduced  almost  to 
nothing?  Christ  and  his  saints  did  not  hide  in  dim  recesses, 
and  did  not  rigorously  exclude  certain  categories  of  the  faithful; 
they  flung  wide  their  doors,  and  allowed  all  who  joined  their 
community  to  approach  them  freely.  Those  of  the  Egyptians 
who  rallied  to  them  could  not  help  renouncing  their  national 
architecture,  as  they  had  renounced  their  national  ' — 

writing;  both  were  too  close  a  reflection  of  the 
antique  religion  to  survive  it. 

B.    SCULPTURE. 

Sculpture  did  not  offer  such  a  vigorous  resis- 
tance to  the  action  of  centuries  as  architecture ; 
yet  it  had  its  happy  seasons,  and  at  various 
times  produced  works  which  deserve  to  rank 
very  highly  in  the  estimation  of  our  contempor- 
aries. 

Many  of  these  works  were  still  of  the  Theban 
school.  We  know  how,  rather  more  than  a 
century  after  the  death  of  Rameses  III.,  the 
priests  of  Amon  usurped  authority  and  ruled 
over  the  Thebaid,  sometimes  under  the  title  of 
High  Priest,  sometimes  under  that  of  King. 
They  allowed  their  relatives  to  set  up  statues 
in  the  temples,  and  this  privilege,  continuing 
throughout  five  or  six  centuries,  ended  by  en- 
cumbering the  halls  with  works,  many  of  which, 
if  not  actually  bad,  lack  originality.  The  favour- 
ite attitude  was  inelegant.  It  represents  the 
sitter  swathed  in  the  closely  fitting  mantle  from 
which  only  the  hands  emerge ,  squatting  on 
the  ground,  the  thighs  drawn  up  to  the  chest, 
the  arms  crossed  on  the  knees  (Fig.  452).  How  was  it  possible 
to  get  any  effect  out  of  an  attitude  which  condemned  the  model 
to  be  nothing  but  a  bundle  with  a  head  ?  It  had  been  popular 
since  the  Twelfth  Dynasty,  and  there  are  good  examples  of  it 
of  the  time  of  Thothmes  HI. ,  in  the  Senmut  of  Cairo  and  that 
of  Berlin ;  but  under  the  High  Priests  and  the  SaYtes,  it  became 
almost  obligatory  for  the  ex-votoes  of  the  temples.  All  the  merit 
of  these  figures  lies  in  the  head,  which  is  often  very  delicate, 
as  in  the  Pedishashi  at  Berlin,  in  which  an  expression  of  joyous 
youth  and  good-nature  atones  for  the  clumsiness  of  the  whole, 
and  the  Petubastis  at  Cairo  (Fig.  453).  In  cases  where  the  model 

247 


FIG.  478.— COLOS- 
SAL STATUE  OF 
AMENOPHIS,  SON 

OF  HAPU 

(Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


FIG.  47g. 

TANITE  STATUE 
(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


consented  to  adopt  a  less  irksome  position, 
our  admiration  is  no  longer  confined  to 
the  head,  and  the  fine  qualities  of  the  school 
are  revealed.  They  proclaim  themselves  in 
the  limestone  statuette  of  Osorkon  II.,  drag- 
ging himself  along  the  ground,  and  pushing 
before  him  a  boat  (now  mutilated),  as  an 
offering  to  his  god  (Fig.  454);  and  again, 
though  in  a  lesser  degree,  in  the  alabaster 
Amenartas  so  much  admired  by  Mariette 
(Fig.  455);  the  face  is  doleful  and  lifeless, 
but  the  modelling  of  the  bust  and  abdomen 
is  chaste  and  delicate.  The  Ankhnasnufiabri 
(Fig.  456)  of  Cairo  might  almost  bear  com- 
parison with  the  Amenartas,  but  for  the 
headdress  which  weighs  down  her  head,  and 
the  heavy  pillar  against  which  she  is  set; 
the  features  have  a  somewhat  affected 
delicacy,  but  the  rest  is  poor  in  design. 
This,  indeed,  is  characteristic  of  the  works 
of  this  period :  the  limbs  are  often  neglected, 
while  the  face  is  very 


carefully  treated.  For 
every  one  or  two  passable  bodies,  we 
shall  find  some  twenty  fine  heads,  such 
as  the  energetic  head  of  the  Ethiopian 
Tirhakah  (Fig.  457),  with  his  almost 
negroid  face,  the  intelligent  head  of 
that  crafty  old  peasant,  Psammetichus 
I.  (Fig.  458),  the  melancholy  head  of 
Psammetichus  III.  (Fig.  459).  The  muti- 
lated queen  at  Cairo,  which  dates  from 
the  Ethiopian  period  (Fig.  460)  is  a  little 
rough,  perhaps  because  of  the  material 
in  which  it  is  carved,  the  pink  granite 
of  the  cataract,  but  it  does  not  lack 
decision  and  nobility.  The  portrait  of 
Mentemhet  (Fig.  461),  who  ruled  at 
Thebes  at  the  end  of  the  Twenty -fifth 
Dynasty,  is  the  most  vigorous  example 
known  to  us  of  this  last  Theban  school. 
The  man  was  common,  even  brutal  in 
appearance,  and  the  singular  wig  with 

248 


FIG.  480.— ALEXANDER 

AIGOS    (Museum,    Cairo). 

(Phot.  E.  Brugsch.} 


which  he  thought  well  to  crown  himself 
on  this  occasion,  was  not  calculated  to 
temper  the  mulish  vulgarity  of  his  counten- 
ance; all  the  more  credit  is  due  to  the 
artist  for  having  built  up  a  work  of  such 
power  that  it  remains  superb  in  spite  of 
mutilation,  on  such  an  unpromising  foun- 
dation. The  statues  of  Nsiptah  (Fig.  462 
and  463)  son  of  Mentemhet  and  his  heir 
in  the  administration  of  the  principality 
of  Amen,  are  not  marked  by  the  same 
almost  excessive  realism ;  nevertheless,  the 
sculptor  has  faithfully  reproduced  the 
expression  of  self-sufficiency  and  aristo- 
cratic inanity  which  differentiates  this  per- 
sonage from  his  father.  I  might  enumerate 
some  nine  or  ten  examples  which,  though 
not  equal  to  this,  will  bear  comparison 
with  it;  one  among  them,  the  Thoueris  at 
Cairo,demands  speci- 


FIG.  481.— STATUETTE 

OF  A  WOMAN 

(Museum,  Alexandria). 

(Phot.  Breccia.) 


FIG.  482.— THE   SAME 

IN   PROFILE 

(Museum,  Alexandria). 

(Phot.  Breccia.) 


al  mention  for  its 
monstrosity  of  con- 
ception. It  repre- 
sents neither  a  hu- 
man being  nor  a  normal  animal  (Fig.  464), 
but  a  hippopotamus  with  a  huge  muzzle, 
smiling  jaws,  flabby  breasts,  and  swollen 
abdomen,  the  form  in  which  the  Egyptians 
incarnated  one  of  the  divine  protectresses 
of  maternity.  Rising  elegantly  upon  her 
hind  legs,  her  two  forepaws  resting  upon 
symbolical  knots  of  rope,  she  is  cut  out 
of  a  block  of  green  breccia  with  a 
precision  which  somewhat  redeems  the 
strangeness  of  her  appearance,  but  all 
the  skill  of  the  technique  fails  to  mask 
her  hideousness;  one  cannot  but  pity  the 
master-  craftsman  whose  religion  obliged 
him  to  treat  a  motive  so  unfavourable 
to  art  seriously. 

All  the  schools  of  Northern  Egypt  put 
together  have   not   given   us   one   half   of 
that    which   has   been   bequeathed   us   by 
249 


ART  IN  EGYPT 


FIG.  483.— THE  ALEXAN- 
DRIAN HORUS  (Museum, 
Cairo).  (Phot.  E.  Brugsch.) 


the  Theban  School.  The  Memphite  School 
has  left  us  a  few  busts  of  kings,  Sai'te  or 
Ptolemaic,  (Fig.  465)  and  certain  rare  statues, 
distinguished  by  that  characteristically  soft 
and  delicate  execution  which  has  made  it 
possible  for  us  to  recognise  its  existence: 
these  are ,  a  full-length  statue  of  Psammet- 
ichus  I.  (Fig.  467) ,  a  statue  of  a  priest 
holding  an  Osiris  before  him  (Fig.  466),  a 
statue  of  a  youthful  Horus,  naked,  his  finger 
to  his  lip,  the  plait  over  his  ear,  the  uraeus 
on  his  forehead  (Fig.  468),  a  statuette  of  an 
Osiris  mummy,  lying  face  downwards  on  his 
base,  and  raising  his  head  in  the  first  spasm 
of  resurrection.  The  four  monuments  of 
Psammetichus  in  green  breccia  preserved  in 
the  Cairo  Museum,  which  belong  to  the  be- 
ginning of  the  Persian  period,  are  the  most 
remarkable.  I  deliberately  pass  over  the  table 
of  offerings,  which  is  merely  a  good  piece  of 

work  by  the  marble-cutter  of  some  necropolis ;  the  Isis  (Fig.  470), 

and  theOsiris  (Fig. 471),  at  whose  feet  it  originally  stood,  are  mark- 
ed ,    it   must    be    admitted ,    by   a   flatness    of 

inspiration  in  painful    contrast  to  the  supreme 

skill  of  their  technique.  The  modelling  is  correct, 

but   soft   and   nerveless,    the    eyes  empty,    the 

smiles  inane,  the  faces  inanimate ;  they  are,  in 

fact,    a  perfect   anticipation  of  those  religious 

figures  which  abound  in  our  modern  eccle- 
siastical warehouses.  The  cow  which  accom- 
panied them  (Fig.  472)  is  posed  in  the  same 

manner  as  that  of  Naville,  and  also  wears  the 

two  huge  feathers  of  Hathor;  Psammetichus  is 

standing   in    the   shadow   of   her  head,    in  the 

attitude  of  Amenophis  II.     The  SaTte  sculptor, 

like    his   Theban    predecessor,    was    unable    to 

disengage  the  legs  of  his  beast,  and  has  retained 

a  stone  partition    between   her   belly   and   the 

ground;   nevertheless,   he   was   determined    to 

show   her   complete  on  either  side,   so  to  her 

one   head,   she  has  two  chests  in  profile,  two 

bodies  and  two  sets  of  legs.  The  contours  have 

an  unpleasant  dryness  due  to  the  hardness  of 

250 


FIG.  484. 
MEMPHITE  STATUE 

(Museum,   Cairo). 
(Phot.  E.  Brugsch.) 


THE  SAITE  AGE 


the  stone,  but  the  modelling  is  extraordinarily 
fine  and  the  faces  both  of  man  and  goddess 
are  marked  by  a  serenity  touched  with  melan- 
choly. It  is,  in  fact,  an  excellent  example, 
and  one  we  greatly  admired  before  seeing  the 
group  of  Der-el-Bahari.  The  mythological  con- 
vention is  perhaps  less  embarrassing  than  in 
the  latter,  but  the  formula  of  the  workshop 
manifests  itself  more  aggressively.  Hathor  is 
a  conventionalised  heifer  which  has  lost  the- 
natural  grace  and  freedom  of  the  good  Egyptian 
milch-cow;  she  has  all  the  elegance  and  all  the 
insipidity  of  the  Isis  and  Osiris.  Beyond  these 
pieces,  I  know  of  none  which  deserve  mention, 
but  there  are  so  many  statues  of  the  Sa'ite  age 
scattered  in  museums  with  no  indication  of  their 
origin,  that  we  may  be  sure  some  of  these  will 
have  to  be  assigned  to  the  Memphite  School, 
when  we  are  better  acquainted  with  its  charac- 
teristics at  this  period.  The 
other  schools  of  the  Delta 


FIG.  485.— MEM- 
PHITE STATUETTE 

(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


FIG.  486. 
MEMPHITE  STATUE 

(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


are  no  more  familiar  to  us,  and  for  the  same 
re.ason;  we  must  be  content  for  the  present 
to  recognise  the  general  features  of  the  period, 
without  attempting  to  distinguish  local  pecu- 
liarities. 

Saite  artists  did  not  forsake  either  granite 
or  the  softer  materials,  limestone  and  sand- 
stone ;  they  showed,  however,  a  marked  prefer- 
ence for  hard,  close-grained  stones  such  as 
basalt,  breccia,  and  serpentine,  and  excelled 
in  the  art  of  rendering  them  supple.  Their 
fine  works  may  therefore  be  recognised 
generally  at  a  glance  by  the  beauty  of  the 
substance,  and  the  pellucid  polish  with  which 
they  clothed  it;  but  in  addition  to  these 
material  indications,  there  are  others  more 
subtle  which  result  from  the  manner  in  which 
they  interpreted  the  human  form.  On  the 
one  hand,  they  tended  to  conventionalise 
it  more  and  more,  and  they  modelled  it 
from  the  drawings  of  masters,  and  the  pattern- 
books  of  the  workshop  rather  than  from 
251 


ART  IN  EGYPT 


FIG.  487.— HEAD  AND  BUST  OF  A 

WOMAN  (Museum,  Cairo). 

(Phot.  E.  Brugsch) 


nature;  excavators  have  found, 
and  are  still  finding  in  the  ruins, 
what  may  be  described  as  ready- 
made  statues,  some  entirely 
blocked  out  in  the  rough  (Fig.  473), 
others  with  the  bodies  finished, 
and  a  shapeless  block  of  stone 
left  for  the  head  (Fig.  474),  to 
await  the  client  whose  likeness 
it  was  to  receive ;  also  feet,  hands, 
arms  and  heads  in  different  stages 
of  preparation,  which  were  used 
for  the  instruction  of  pupils,  or 
were  the  products  of  their  ex- 
periments. Under  the  influence  of 
this  method,  the  science  of  ana- 
tomy languished,  contours  became 
soft,  the  muscles  relaxed  and  were 
incorrectly  placed,  the  planes  of 
the  flesh  were  merged  one  into 

the   other  and   became   perfectly   smooth.      On  the  other  hand, 

great  pains  were  taken  to  make  the  head  as  exact  a  reproduction 

of   the  original    as  possible,   and    in  order 

to  succeed  in  this,  sculptors  were  no  longer 

content  to  render  the  features  of  the  face 

very  faithfully  in  the  stone;  they  gave  much 

attention  to  the  modelling  of  the  neck  'and 

skull,    which   had  hitherto  been  neglected. 

Our   museums   contain    examples   of  these 

disconcerting  statues,  in  which  the  feebleness 

of  the  body  is  in  such  striking  contrast  to 

the  truth  of  the  face.    The  wrinkles  of  the 

forehead   are    emphasised    with  scrupulous 

insistence,  the  sunken  eyes  and  the  crows' 

feet    at    the    corners,    the    muscles    which 

encircle  the  nostrils,  the   laughing  lines  of 

the  mouth,  the  curve  or  the  flatness  of  the 

nostrils;    the    double- statues  no  longer  re- 
present   their  master   as  uniformly  young; 

if    he     is    mature    or    old    they    show    the 

stigmata    of    age.       The    head    at    Berlin 

(Fig.  475)  like  that  in  the  Louvre  (Fig.  476), 

is   the  unflattered  portrait   of  a  Memphite 

252 


FIG.  488. — HEAD  AND 

BUST   OF  A   WOMAN 

(Barracco  Collection). 

(Phot.  Bissing.) 


THE  SAITE  AGE 


citizen  ,    whose    ugliness   has    been    transferred    to    green    schist 

or   serpentine  with   the   mechanical    precision    of  a  photographic 

plate.   The  skull  of  the  shaven 

priest   at   Cairo    (Fig.  477)    is 

as  minutely  modelled  as  if  the 

sculptor    had    been     commis- 

sioned to  make  an  anatomical 

model   for   a  medical   school; 

it   shows    every  bump  ,    every 

depression,  all  the  asymmetries, 

and  a   doctor  could  tell    at  a 

glance  if  there  were  congenital 

defects  in  the  original. 

This  recrudescence  of  realism 

is  not  to  be  attributed  to  the 

appearance  of  the  Greeks  on 

the  scene;  the  Greeks  of  the 

fifth  and  fourth  centuries  B.  C. 

did  not  carry  this  almost  painful 

striving   after   resemblance  so 

far.  It  was  the  natural  develop- 

ment, and  as  it  were  the  con- 

sequence of  the  ancient  theory 

of   the    double,    and   was    produced    under   the   influence   of   the 

changes  introduced  into 
costume  by  the  fashion  of 
the  period.  After  the 
Twenty  -sixth  Dynasty  it 
would  seem  that  the  use 
of  the  wig  gradually  be- 
came less  general  ,  and 
that  it  disappeared  entirely 
under  the  Persians.  The 
priests,  who  kept  their 
skulls  bare  for  reasons  of 
professional  cleanliness, 
lived,  it  may  be  said,  with 
uncovered  heads  ,  and 
tnose  members  or  the 

Other     Upper     claSSCS    who 

were     not     affiliated     in 

some    way    to    the    priesthood,    acquired    the    habit    of   wearing 
short  hair.     Once  again  it  was  religious  dogma  which  drew  art 

253    . 


FIG.  489.— HEAD  AND  BUST   OF  A  MAN 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


FIG.  490.  —  SPHINX  OF  THE  ROMAN  PERIOD 

(Museum,  Alexandria).     (Phot.  Breccia.) 


ART  IN  EGYPT 


into   a  new  path,  and  when  artists  copied   the  head  so  exactly, 
it  was   in  the   hope   of  securing   for   the  double   all  the  benefits 


FIG.  401.— CROWNED  COLOSSUS  IN  THE  ISLAND  OF  ARGO. 

he  had  enjoyed  on  earth.  It  is  not  surprising,  therefore,  that 
the  example  thus  set  by  the  Memphite  School  should  have 
been  followed  by  all  the  others.  The  few  Theban  statues  of 
Ptolemies  which  have  survived  betray  a 
like  interest  in  the  accidents  of  the  face. 
Those  of  the  princes  of  Asyut,  which  seem 
to  be  related  to  the  Hermopolitan  School, 
show  traces  of  the  same  influence,  and 
those  of  the  Tanite  School  which  we  find 
at  Cairo  (Fig.  479)  have  not  escaped  it, 
a  fact  which  will  surprise  no  one,  seeing 
that  an  analogous  tendency  makes  itself 
felt  in  the  earliest  of  its  productions,  the 
sphinxes  of  Amenemhat  III.  Nevertheless, 
in  the  cities  colonised  by  the  Macedonians, 
at  Alexandria,  at  Memphis,  and  even  at 
Thebes,  the  sight  of  Greek  statues,  and 
perhaps  contact  with  the  masters  who  exe- 
cuted them,  had  finally  made  some  im- 
pression on  the  natives,  and  though  they 
never  entirely  abandoned  their  ancestral 
traditions,  they  hellenised  them  to  some 
extent.  The  Theban  Colossus  of  Ameno- 
phis,  son  of  Hapu  (Fig.  478),  is  purely 
254 


FIG.  492. 

THE  GOD  AMON  AND  AN 
ETHIOPIAN  CANDACE 

(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


THE  SAlTE  AGE 


FIG.  493. — ETHIOPIAN   STATUES  OF  SOULS 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


Egyptian,  and  that  which  represents  an  Alexander  Aigos  (Fig.  480) 
has  no  exotic  elements  save  the  arrangement  of  the  hair  and 
the  cast  of  the  face.  In 
the  Alexandrian  statuette 
(Figs.  481,  482),  the  atti- 
tude and  costume  are,  on 
the  other  hand,  purely 
Egyptian,  but  the  Greek 
afflatus  has  passed  over 
the  body,  animating  every 
part  of  it,  the  rounded 
bust,  the  small,  firm 
breasts,  the  closely  mo- 
delled belly,  the  well- 
developed  hip,  the  slen- 
der, nervous  leg.  The 
priest  Horus  (Fig.  483), 
less  delicate  in  handling,  is  much  more  advanced  in  evolution; 
it  looks  like  a  Greek  work  executed  by  an  Egyptian  rather  than 
a  purely  Egyptian  creation.  Here  again  the  body  is  open  to 
criticism;  the  shoulders  are  not  broad  enough,  the  chest  is  too 
narrow,  and  the  artist  had  a  difficulty  in  rendering  both  the 
fall  of  the  arms  and  the  folds  of  the  chlamys. 
The  head  is  not  bad;  the  nose  is  thin 
and  straight,  the  chin  square,  the  jaw  ob- 
stinate, and  the  whole  has  a  certain  general 
resemblance  to  the  portraits  of  the  young 
Augustus.  These  hybrid  statues  were  also 
produced  at  Memphis  at  this  period.  One, 
which  is  of  basalt  (Fig.  484),  is  not  unlike 
our  Horus  in  costume  and  attitude,  while 
the  other  (Fig.  485),  in  limestone,  represents 
a  priest  walking  and  holding  a  naos  in  both 
hands;  the  eyes  are  inserted,  and  the  eye- 
brows have  been  blackened  with  kohol;  the 
whole  work  is  uninteresting.  There  is  the 
same  unskilfulness  in  the  large  limestone 
figure  of  our  museum  (Fig.  486),  and  al- 
though the  heads,  male  (Fig.  489)  and  female 
(Figs.  487,  488),  near  it  give  the  impression 
of  faithful  likenesses,  the  dryness  of  the 
chisel  has  played  the  excellent  intentions 
of  the  sculptor  false. 

255 


X 


FIG.    494.— STATUE    RE- 
CARVED  AS  A   CJESAR 

(Museum,  Cairo). 
(Phot.  E.  Brugsdi.) 


ART  IN  EGYPT 


FIG.  495- 
COLOSSUS  OF  THE 

ROMAN  PERIOD 

(Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


Certain  ancient  types  were  modified  in  this 
last  stage  of  sculpture;  that,  for  instance,  of 
the  female  sphinx,  who  began  to  lean  her  head 
to  one  side  and  to  cross  her  forepaws  (Fig.  490). 
No  original  type  was  born  of  this  belated 
alliance  between  the  Greek  and  the  Egyptian 
spirit;  the  remnant  of  creative  vigour  which 
the  old  tradition  kept  alive,  had  taken  refuge 
in  Ethiopia.  It  is  not  manifested  in  the  crowned 
Colossus  of  the  island  of  Argo  (Fig.  491),  nor 
in  the  few  royal  statues  which  have  come  down 
to  us  from  Argo,  Napata,  and  Meroe;  the 
Cairo  group  (Fig.  492) ,  in  which  a  Queen 
Candace  stands  beside  an  Amon ,  has  some 
pride  and  spirit,  in  spite  of  the  imperfection 
of  the  execution,  but  it  has  no  elements  which 
are  not  purely  Egyptian.  And  yet  recent  exca- 
vations have  revealed  a  new  conception  of  the 
soul  among  this  people  which  was  relapsing 
into  barbarism  (Fig.  493).  Taking  as  their  point 
of  departure  the  bird  with  a  human  head  which 
had  in  all  times  served  to  express  it,  they 

substituted  a  human  body  for  that  of  the  falcon,  at  first  without 

altering  the  proportions;  but  soon,  enlar- 
ging the  miniature  body  to  normal  dimen- 
sions, they  produced  what  I  have  called 

the  soul-statue   in  contradistinction   to  the 

double-statue,  the  figure  of  a  man  or  woman 

over   whose    shoulders    the    falcon's    skin 

hangs  like  a  cloak.    It  does  not  seem  that 

this  type  ever  penetrated  into  Egypt  proper. 

Indeed,  soon  after  its  appearance,  the  purely 

indigenous  schools  of  sculpture  were  either 

closed,    or,   all   along   the   valley,    merely 

produced  artisans  incapable  of  a  passable 

work;    when,    about   the  beginning  of  the 

third  century,  certain  towns  of  the  Fayum 

or   the  Delta  wished   to  erect  monuments 

in   honour  of   the  Caesars,    Commodus  or 

Caracalla ,  they  were  reduced  to  borrowing 

an  antique  statue  and  re-carving  the  face 

(Fig.  494),  or,  if  they  demanded  an  original, 

the   artist  gave  them  the  caricature  of  a 

256 


FIG.  4Q6. — STATUE  OF  A 
MONKEY  (Museum  of  the 
Vatican).  (Phot.  Petrie.) 


THE  SAITE  AGE 


FIG.   497.— LION  PASSANT 
(Museum  of  the  Vatican).     (Phot.  Petrie.) 


Colossus  in  the  antique  style  (Fig.  495).  The  hybrid  art  of 
Alexandria  was  on  a  higher  level,  as  we  learn  from  the  two 
statues  of  a  man  and 
woman  of  the  tomb  of 
Kom-es-Shugafa,  and  the 
animal  -  sculptors  were 
able  for  some  time  yet  to 
produce  figures  of  monk- 
eys (Fig.  496),  sphinxes 
(Fig.  498),  and  lions 
passant  or  seated  (Figs. 
497,  499),  which  may  be 
taken  for  living  animals 
at  a  first  glance.  The 
pacific  lion  of  Fig.  500, 
and  the  Kom-Ombo  statue,  of  which  only  the  crowned  head  has 
survived  (Fig.  501),  were  probably  among  the  last  efforts  of 
Egyptian  or  Egyptianistic  art;  none  of  the  monuments  I  have 
met  with  so  far  seem  to  me  later  than  the  second  half  of  the 
third  century  after  Christ. 

The  history  of  the  Saite  and  Graeco-Roman  bas-relief  is  anal- 
ogous to  that  of  the  statues.  It  had  its  glorious  moments  from 
the  eighth  to  the  third  century  before  Christ;  it  then  passed 
through  a  long  period  of  decadence,  and  closed  towards  the 
middle  of  the  third  century  of  our  era.  A  few  stelae ,  the 
rescript  of  Nectanebus  on  Naucratis,  and  the  Horus  with  the 

crocodiles  (Fig.  502)  in 
the  Cairo  Museum,  will 
bear  comparison  with  the 
best  works  of  the  Ah- 
messide  age,  and  the  frag- 
ment of  a  decree  of 
Ptolemy  Euergetes  has  a 
motive  unknown  in  earlier 
periods,  that  of  a  Pharaoh 
on  horseback  charging  the 
enemy,  sarissa  in  hand 
(Fig.  503).  The  bas-reliefs 
of  buildings  anterior  to 
the  Greek  conquest  are  of 
a  very  pure  style  in  the 
Delta,  at  Beh-bet,  for  instance,  where  Nectanebus  I.  restored 
the  temple  of  Isis ;  even  certain  fragments  of  a  temple  of  Akoris, 

257  S 


FIG.  498.— SPHINX  OF  THE  PTOLEMAIC   PERIOD 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


FIG.  4QQ. — SEATED  LION 

(Museum,   Cairo). 
(Phot.  E.  Brugscli.) 


found  at  Sakkarah,  would  be  almost  equal 
to  those  of  Seti  I.  at  Abydos,  if  their  deli- 
cacy were  not  marred  by  a  certain  insipidity. 
Over-roundness  of  modelling,  and  softness 
of  contours  are  indeed  the  defects  we  note 
in  the  decoration,  defects  which  became 
more  and  more  marked  towards  the  end. 
They  make  themselves  felt  even  in  the  ex- 
perimental pieces,  and  the  collection  of 
models  which  every  studio  possessed,  relics 
which  are  found  in  quantities  in  the  ruins 
of  the  great  towns,  from  Tanis  to  Edfu  and 
Philae.  Figures  of  kings  and  queens  are 
the  most  frequent  (Fig.  504) ,  and  this  is 
natural,  for  the  reigning  Pharaoh  and  the 
women  of  his  family  had  an  immemorial 
right  to  be  represented  on  the  walls  of  the 
temples,  and  it  was  also  customary  to  lend 
their  features  to  the  gods  and  goddesses 
with  whom  they  consorted;  animals  and 
hybrid  forms  of  godhead,  half  man,  half 
beast,  also  abound,  and  this  again  is  not 
surprising,  for  sculptors  were  perpetually  called  upon  to  execute 
them ,  both  in  pictures  and  inscriptions.  One  of  the  masters 
would  trace  on  thin  slabs  of  limestone,  sometimes  squared 
on  the  surface,  the  better  to  instruct  the  tyro  in  the  proper  relation 

of  parts,  the  portrait  of  a  Pto- 
lemy or  a  Cleopatra  (Fig.  505), 
in  various  skilfully  graduated 
stages,  from  the  moment  of 
sketching  in  the  silhouette  and 
the  relief  (Fig.  506),  until  that 
when  they  are  finished  in  their 
slightest  details  (Fig.  507). 
Several  of  these  examples  are 
masterpieces,  and  there  are 
things  in  the  Cairo  Museum, 
such  as  the  head  of  a  lioness 
and  the  image  of  a  bull ,  no 
whit  inferior  in  delicacy  of 
touch  to  the  best  in  the  temple 
of  Abydos  or  the  tomb  of  Seti  I. 
Yet  even  in  these  we  discern 


FIG.   5OO. — THE  I.TON  OF  KOM-OMBO 
(Museum,    Cairo).     (Phot.  E.  Brugsch.) 


258 


THE  SAITE  AGE 


the  tendencies  which,  becoming  more  and  more  pronounced, 
brought  about  irremediable  decline.  It  is  not  good  for  an  ap- 
prentice to  be  kept  exclusively  at  the  reproduction  of  models, 
however  perfect.  He  loses  touch  with  reality,  he  becomes  a 
machine,  and  soon  he  prides  himself  on  being  nothing  but  the 
servile  copyist  of  antique  forms.  Plato,  no  doubt,  reflected  the 
state  of  mind  of  his  Egyptian  contemporaries  when  he  praised 
as  admirable  the  persistence  with  which  they  had  produced  the 
same  types  without  change  for  thousands  of  years.  If,  thanks  to 
the  beauty  of  its  stereotyp- 
ed designs,  late  Saite  art 
retained  a  certain  elegance, 
it  soon  had  nothing  of  its 
primitive  originality  and  crea- 
tive vigour  left.  Its  figures 
had  become  mere  puppets 
without  any  anatomical  basis; 
the  nose  became  rounder, 
the  lips  more  pouting,  the 
chin  thicker,  the  cheeks  hea- 
vier, the  mouth  was  set  in 
a  smile  which  lifts  it  at  the 
corners  and  draws  up  the 
nostrils  towards  the  eyes. 
This  contraction  of  the  whole 
countenance,  slight  under  the 
first  Ptolemies,  degenerates 
under  their  successors  and 
the  Caesars  into  a  grimace, 
which  gives  the  person  re- 
presented a  distressingly  silly  expression. 

Thebes  and  Philse  are  almost  the  sole  places  where  we  may 
trace  the  progress  of  this  decadence.  The  bas-reliefs  of  the  gate 
of  Nectanebus  on  the  east  of  the  great  temple  of  Amon  are  dignified 
and  agreeable,  if  they  have  no  claim  to  supreme  excellence,  and 
the  same  qualities  reappear  at  Luxor  in  the  sanctuary  of 
Alexander  II.,  at  Karnak,  on  the  walls  of  the  granite  chamber 
constructed  by  Philip  Arrhidseus  in  the  shrine  of  the  sacred  boat 
of  the  time  of  Thothmes  III.  They  begin  to  die  out  in  the  little 
temple  of  Ptah  (Fig.  508),  and  the  decoration  of  the  large  door 
built  by  Ptolemy  Physcon  for  the  hypostyle  hall  is  frankly 
detestable;  above  all,  in  the  places  where  sculptors  presumed 
to  fabricate  bas-reliefs  in  the  names  of  Thothmes  (Fie.  509)  or 

259  s    2 


FIG.   501.— HEAD   OF  A   STATUE 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


FIG.  5O2.— STELE  OF  HORUS 
ON  THE  CROCODILES 

(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


Rameses  (Fig.  511),  the  handling  is  so 
feeble  that  the  imposture  is  obvious  to 
the  most  ignorant  eye.  The  material 
in  which  they  are  carved  is  in  part 
responsible  for  their  shortcomings. 
Whereas  the  architects  of  the  Saite 
period  used  a  close,  durable  sandstone, 
capable  of  keeping  the  play  of  the 
chisel  firm,  those  of  the  Ptolemaic  age 
were  content,  no  doubt  for  economical 
reasons,  with  a  soft,  coarse-grained 
sandstone  which  did  not  lend  itself  to 
precise  lines  or  delicate  transitions  in 
relief.  Sculptors  were  accordingly  ob- 
liged to  suppress  in  the  contours  and 
the  modelling  of  their  figures  minutiae 
for  which  the  stone  was  unsuitable, 
and  as,  even  by  avoiding  these  as 
much  as  possible,  they  did  not  alto- 
gether escape  such  accidents  as  the 
splitting  or  crumbling  of  the  work, 
they  substituted  for  relief  on  a  sunk  surface  (relief  en  creux), 
which  had  been  almost  obligatory  in  former  centuries  relief  on 

the  normal  surface  of  the 
wall ,  which  they  applied 
even  to  inscriptions.  It  thus 
became  easier  for  them  to 
complete  their  decoration 
without  endangering  their 
figures  and  hieroglyphics 
unduly;  but,  on  the  other 
hand,  they  secured  integrity 
at  the  cost  of  flexibility; 
their  works,  reduced  tb  the 
utmost  simplicity  of  model- 
ling for  fear  of  accidents 
during  execution ,  look  as 
if  they  had  lost  their  skin, 
and  were  presenting  hastily 
flayed  figures  to  view.  At 
Philae  the  decadence,  though 
slower,  was  not  less  sure, 
and  was  brought  about  by 


*<--. 


FIG.   503. 

PTOLEMY  EUERGETES    CHARGING 
(Museum,   Cairo).     (Phot.  E.  Brugsch.) 

260 


THE  SAITE  AGE 


the  same  causes  as  at  Karnak;  we  follow  it  step  by  step,  from 
the  pavilion  of  the  doorway  of  Nectanebus  II.  to  the  shrine  of 
Euergetes  II. ,  the  porticoes  of  Augustus  and  Tiberius,  the  doors 
of  Hadrian  and  Trajan.  In  such  of  the  temples  as  were  decor- 
ated on  a  homogeneous  plan,  without  too  long  a  period  between 
the  dates  of  beginning  and  completing 
the  work,  Edfu,  for  instance,  there  is 
no  distressing  contrast  between  room 
and  room,  and  the  unity  of  the  deco- 
ration conceals  its  feebleness  to  a 
certain  extent.  No  very  keen  study 
is  needed,  however,  to  perceive  that 
the  sculptor  here  was  a  workman,  who 
mechanically  transferred  a  stereotyped 
design  to  the  wall,  and  was  no  longer 
capable  of  giving  it  a  personal  impress 
(Fig.  510);  his  main  preoccupation  was 
the  correct  reproduction  of  costume 
and  liturgical  accessories  (Fig.  512). 
For  the  great  Theban  School,  which 
had  inspired  all  the  provincial  schools 
of  the  Said  for  centuries,  was  dying, 
if  indeed  it  was  not  already  dead  at 
the  moment  when  Edfu  and  Denderah 
were  decorated,  and  the  local  work- 
shops could  only  reproduce  mechani- 
cally the  motives  the  architect  ordered, 
as  in  those  pictures  where  the  Emperor 
Domitian  comes  to  worship  the  gods 
of  Thebes;  where  architecture,  a  mathe- 
matical art,  had  retained  its  vitality 
for  centuries,  a  few  years  had  sufficed 
to  cause  sculpture  to  degenerate  and 
bring  it  to  the  point  of  death.  The 

stelae  of  the  Fayum  (Fig.  514)  and  the  bas-reliefs  of  Macrinus 
and  Diadumenianus  at  Kom-Obos  (Fig.  515),  the  last  which  bear 
a  date,  have  neither  life  nor  style,  nor  anything  else;  they  are 
mere  hieratic  lumber,  timidly  manufactured  by  an  ignorant  work- 
man. Ethiopia  was  no  better  equipped  than  Egypt  proper.  The 
heads  of  some  of  her  kings  (Figs.  516 — 517)  look  like  caricatures, 
so  clumsy  is  their  execution.  Even  in  Nubia,  where ,  it  might 
have  been  thought,  the  domination  first  of  the  Ptolemies  and 
then  of  the  Caesars  would  have  tended  to  preserve  the  Theban 

261 


FIG.   504. — MODELS  OF  HEADS 
AT  DIFFERENT    STAGES 

(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


tradition,  the  bad  taste  characteristic  of  Meroe  prevailed.  The 
sculptures  at  Dakkeh  (Fig.  518)  and  Kalabshah,  which  are  for  the 
most  part  of  the  time  of  Augustus ,  are  worse  than  those  of 
Ombo,  later  by  over  200  years.  The  temple  of  Dakkeh,  as 
we  know  (cf.  p.  242)  received  its  final  form  under  Augustus, 
who  added  the  sanctuary  and  pylon.  The  pylon-towers  have 
a  few  reliefs  on  the  outside.  The  sanctuary  reliefs  show  an 
unknown  Emperor  doing  homage  to  various  deities.  Among  the 
reliefs  of  a  chamber  built  by  the  Ethiopian  Ergamenes  (con- 
temporary with  Ptolemy  II.), 
is  one  of  the  king  pouring 
a  libation.  The  reliefs  in  the 
temple  of  Kalabshah  were 
never  completed,  and  the 
crudity  of  those  which  were 
executed  is  such  as  to  obviate 
any  regrets  on  this  score.  They 
decorate  the  pronaos  and  the 
three  following  rooms,  and 
represent  Augustus  sacrificing 
to  the  gods  of  Egypt. 
At  Maharrakah,  the  ancient 
Hierasycaminos,  side  by  side 
with  certain  execrable  bas- 
reliefs  purely  Egyptian  in 
style,  I  must  instance  a  last 
attempt  to  fuse  the  Alexand- 
rian and  Pharaonic  styles,  Isis 
and  Horus  in  Greek  draperies, 
sitting  pr  moving  after  the 
manner  of  Roman  deities  (Fig.  51 9).  The  intention  was  praiseworthy, 
but  no  more  barbaric  work  ever  disgraced  an  Egyptian  chisel. 

The  funerary  bas-relief  of  the  SaYte  age  is  considered  by 
archaeologists  an  evidence  of  the  renascence  of  ancient  Memphite 
art,  and  the  purity  and  tenderness  which  characterise  it  are 
readily  attributed  to  direct  imitation  of  the  Ancient  Empire. 
It  cannot  be  denied  that  the  Egyptians  of  this  period  must  often 
have  penetrated  into  the  earlier  hypogea,  but  if,  after  pillaging 
and  demolishing  them  to  make  fresh  use  of  their  materials,  they 
derived  texts  and  magic  formulae  from  them,  they  had  no  need 
to  borrow  school  motives.  The  designs  prescribed  for  use  in 
the  mastabas  had  never  been  completely  abandoned,  although, 
under  the  Ahmessids  and  the  Ramessids,  the  imitation  of  Theban 

262 


FIG.   505.— FIGURE   OF  A  QUEEN  OR 

GODDESS   AT   TWO   DIFFERENT   STAGES 

(Museum,   Cairo).     (Phot.  E.  Brugsch.) 


THE  SAlTE  AGE 


FIG.   5O6. — SKETCH 

FIGURE  OF  A  KING 

(Museum,   Cairo). 

(Phot.  E.  Brugsch.) 


processes  had  loaded  them  with  details 
alien  to  the  primitive  conception.  When 
the  political  hegemony  of  Thebes  was  over- 
thrown, its  artistic  influence  vanished  rapid- 
ly, and  soon,  towards  the  end  of  the 
Bubastite  age,  or  under  the  Ethiopians, 
local  inspiration,  casting  off  these  adven- 
titious elements,  returned  to  an  almost 
primitive  simplicity.  Thus,  in  the  tomb  of 
Patanafi  or  that  of  Psammetiknufisashmu, 
the  general  scheme  is  that  of  the  bas-reliefs 
of  the  Fifth  and  Sixth  Dynasties,  for  the 
bearers  of  offerings  walk  in  single  file, 
separated  and  juxtaposed  (Fig.  520),  not 
superposed  in  profile;  but  as  soon  as  we 
give  more  than  a  superficial  attention  to 
the  work,  we  note  combinations  which 
we  had  not  perceived  at  first.  The  primi- 
tive fiction  of  the  wakf,  indicating  a 
number  of  domains,  each  with  its  geo- 
graphical individuality,  had  been  replaced 
by  the  idea  of  a  tribute  levied  once  for  all  upon  the  entire 
patrimony;  the  procession  of  the  domains  is  therefore  merely 
an  artistic  survival,  a  motive  of  earlier 
times,  used  to  symbolise  the  rendering 
of  dues  to  the  dead  man,  but  no  longer 
corresponding  to  rites  performed  round 
the  tomb.  An  indifferent  staff  of  priests, 
professional  mourners,  musicians  and 
dancers  was  now  employed  for  the 
ceremonies,  in  place  of  the  deceased's 
own  servants.  It  is  not  surprising  that 
this  perversion  of  the  original  idea  should 
have  entailed  modifications  in  the  treat- 
ment of  the  scenes.  On  the  bas-relief 
of  Zanufi  in  the  Cairo  Museum,  the 
women  are  not  drawn  from  a  uniform 
pattern.  The  artist  has  made  them 
young,  in  accordance  with  the  ancient 
tradition,  but  they  have  no  longer  the 
dainty  roundness  and  the  virginal  appear- 
ance which  pleased  in  earlier  ages; 
he  has  given  them  the  heavy  breasts, 

263 


FIG.  507.— FINISHED  FIGURE 

OF  A  KING  (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


thick  waist,  full  abdomen,  wide  hips,  and  firm  gait  of  married 
women,  and  often  of  mothers  (Fig.  521).  As  was  natural,  the 
men  are  little  changed,  but  their  attitudes  are  less  conventional 
than  formerly,  and  one  of  them,  the  man  in  the  middle,  brings 
his  left  shoulder  forward  with  an  attempt  at  perspective  of  which 
no  one  seems  to  have  thought  before  the  Theban  period.  The 
desire  for  variety  is  even  more  strongly  marked  in  the  animals; 

not  a  single  one  of  the  eight 
bullocks  the  men  are  leading 
is  in  the  same  attitude  as 
its  fellows.  It  is,  however, 
in  a  bas-relief  in  the  Museum 
of  Alexandria  that  the  dif- 
ference between  the  old  style 
and  the  new  is  most  pron-< 
ounced  (Fig.  522).  Here, 
every  head  of  a  man  or  a 
woman  has  been  studied 
separately,  as  in  a  modern 
bas-relief,  and  the  rendering 
has  been  carried  very  far; 
I  doubt  if  we  shall  ever  find 
in  any  example  of  the  time 
of  the  Pyramids  an.  equi- 
valent for  the  player  of  the 
triangle  or  the  female  mu- 
sician who  is  striking  the 
drum.  Not  only  have  the 
FIG.  508.— A  BAS-RELIEF  m  THE  TEMPLE  fjaiirps  parh  a  diffprpnt  nhv- 

OF  THE  THEBAN  PTAH.  iigurcs  eacn  a  cmrerent  pny- 

(Phot.  Legrain.)  siognomy,     suggesting     the 

probability    of     portraiture, 

but  their  gestures,  their  costumes,  their  draperies  and  accessories 
are  all  diversified.  The  painter,  too,  came  to  the  aid  of  the 
sculptor  with  a  daring  impossible  to  an  artist  of  the  Fifth  or 
Sixth  Dynasty.  His  brush  supplied  the  portions  of  the  dresses 
which  the  chisel  could  not  have  expressed  clearly  enough;  he 
even  added  shades  which  his  colleague  was  accustomed  to  neglect 
entirely,  so  that  where  the  colour  has  disappeared  the  work  is 
incomplete.  It  is  for  this  reason  that  we  are  puzzled  at  first  by 
the  cut  of  the  wrapper  worn  by  one  of  the  wives  of  Psamme- 
tiknusashmu,  a  curious  garment,  which  though  drawn  up  round 
the  neck  and  fitting  closely  to  the  limbs,  seems  to  leave  the 
bust,  abdomen  and  thighs  bare,  while  draping  the  shoulders, 

264 


THE  SAlTE  AGE 


back,  and  loins.  This  peculiarity  did  not  exist  originally;  the 
painter  had  supplied  the  sculptor's  omissions  by  a  few  discreet 
touches,  and,  thanks  to  him,  the  body,  revealing  its  contours 
under  a  wash  of  red  or  blue,  was  clothed  in  a  semi-transparent  robe. 
At  this  period  there  was  no  longer  much  question  of  local 
schools,  but  this  particular  style  spread  throughout  Egypt,  and 
even  penetrated  to  Thebes;  we  find  it  in  several  of  the  hypogea 
of  Asasif,  notably  that  of  Abai,  where  it  has  the  same  character- 


FIG.  5OQ. — PASTICCIO   OF  EIGH- 
TEENTH DYNASTY  RELIEF ;  TEMPLE 
OF  THE  THEBAN  PTAH.  (Phot.Legrain.) 


FIG.   510.— BAS-RELIEF  IN  THE 

TEMPLE   OF   EDFU. 

(Phot.  Beato.) 


istics  as  at  Memphis,  and  this  is  not  surprising,  since  the  "divine 
spouses  of  Amon"  of  the  Twenty-sixth  Dynasty  were  Saite  prin- 
cesses who  brought  their  households  with  them ;  the  persons  of 
their  suite  either  prepared  their  tombs  in  accordance  with  the 
usages  of  their  own  country,  or  imposed  its  fashions  on  the 
Thebans.  Here,  at  least,  no  foreign  influence  intervened  at  first, 
and  we  have  to  come  down  to  the  first  Caesars  to  find  hybrid 
works,  in  which  the  Egyptian  manner  is  wedded  to  the  Greek, 
as  in  the  figures  of  Antaeus  and  Isis  in  the  Cairo  Museum 
(Fig.  523),  At  Memphis,  the  case  was  very  different,  and  as  we 
gradually  advance,  under  the  Persians,  under  the  last  Sa'ites, 
under  the  Ptolemies,  we  feel  that  by  living  in  contact  with  the 
Greeks,  first  those  of  Naucratis,  and  then  those  of  Alexandria, 
the  people  of  the  Delta  had  ended  by  drawing  inspiration  from 

265 


them.  The  ample  pallium  cut  into  battlements  below  in  which 
Psammetiknufisashmu ,  his  scribe,  and  the  musician  with  the 
triangle,  are  draped,  was  a  garment  borrowed  from  neighbouring 
Greece.  The  mantle  of  the  woman  playing  the  drum  is  also 
Greek;  she  herself  bears  a  strong  likeness  to  her  companions 

on  painted  Alexandrian 
stelae.  Such ,  however, 
was  the  strength  of  tradi- 
tion among  this  singular 
people,  that,  although  they 
consented  to  disguise  a 
sculptured  relief  by  pain- 
ting over  it,  they  declined 
to  modify  the  foundation 
itself;  the  silhouette  of  the 
figures  remained  archaic, 
even  when  colour  had 
modernised  their  super- 
ficial detail.  The  foreign 
influence,  was  moreover, 
very  slow  in  its  action; 
the  fragments  of  the  tomb 
of  Psammetiknufisashmu, 
in  which  it  is  incontes- 
table, are  not  earlier  than 
the  Twenty-ninth  or  Thir- 
tieth Dynasty,  and  I  do 
not  hesitate  to  assign  the 
two  bas-reliefs  of  the 
tomb  of  Zanufi  to  the 
reign  of  one  of  the  first 
Ptolemies.  This  would 
make  them  the  latest,  as 
they  are  certainly  the  fin- 
est that  we  know.  With  these  may  be  classed  those  admirable 
Memphite  sarcophagi  of  the  families  of  Tchaho  and  Ankh-Hapi 
at  Cairo,  on  which  every  little  figure  and  even  every  hieroglyph, 
is  carved  as  conscientiously  as  if  it  were  a  motive  on  a  cameo; 
their  artistic  merit  is  very  slight,  but  they  are  the  perfection  of 
craftsmanship.  After  them  not  only  art  but  craftsmanship  de- 
clined suddenly,  save  perhaps  in  the  neighbourhood  of  Alexandria, 
where  bas-reliefs  in  a  mixed  style  analogous  to  that  of  the  statues 
of  Horus  are  occasionally  excavated.  At  Kom-es-Shugafa,  the 

266 


FIG.   5IT. — PTOLEMAIC  PASTICCIO   OF  THE 

STYLE   OF  THE   EIGHTEENTH  DYNASTY. 

(Phot.  Legrain.) 


THE  SAITE  AGE 


piety  of  a  great  family  suggested  the  construction  of  a  Pharaonic 
hypogeum,   at   the   beginning   of   the   third   century   of   our  era 
(Fig.  524).    The  principal 
scenes  of   ancient   times 
were    reconstituted ,    the 
mummy  on  its  lion-footed 
couch  given  over  to  the 
care    of   Horu>,    Anubis 
and  Thoth  (Fig.  525),  the 
sovereign  before  an  Apis 
bull  which  an  Isis  shelters 
under  her  wings,  priests 
reciting  the  office  of  the 
dead,    or   offering  sacri- 
fice to  IsiS,  but  the  whole  FIG.  512. — A  BAS-RELIEF  OF  THE  TEMPLE 
is   bedaubed  with   crude  OF  OMBOS.    (Phot.  Beato.) 
colours  and  the  technique 

shows  how  utterly  the  Egyptian  style  was  forgotten;  even  our 
own  artists  of  the  seventeenth  and  eighteenth  centuries  were 
not  more  remote  from  their  originals,  when  they  published  so- 
called  facsimiles  of  the  Ramesside  stelae  and  bas-reliefs. 

C.   PAINTING  AND  THE  MINOR  ARTS. 

Independent  painting  played  an  increasingly  important  part  as 
the  antique  artistic 
conception  of  the 
Egyptians  declined. 
This  could  hardly 
have  been  other- 
wise ,  seeing  that 
they  had  been 
brought  into  daily 
contact  with  a 
people  like  the  Hel- 
lenes, among  whom 
painting  had  eman- 
cipated itself  from 
sculpture  to  become 
an  art  in  itself;  but 
their  works,  execu- 
ted upon  non-dur- 
able materials,  wood  or  canvas,  have  perished  for  the  most  part. 
The  remnant  is  chiefly  made  up  of  papyri,  the  panels  of  funerary 

267 


FIG.    513. — THEBAN   BAS-RELIEF   OF   THE   TIME 

OF  DOMITIAN.     (Phot.  E.  Brugsch.) 


FIG.   514. — STELE    OF  THE  FAYUM 
(Museum,  Cairo).    (Phot.  E.  Brugsch.) 


caskets,  stelae,  coffins  or 
cartonnages  of  mummies, 
the  most  ancient  of  which, 
those  of  the  priests  of 
Amon  or  Mentu,  carry  on 
the  tradition  of  the  Ra- 
messide  workshops.  The 
vignettes  of  the  papyri  are 
often  veritable  miniatures, 
perfect  in  line  and  fresh 
in  colour;  the  best  of  them 
are  unquestionably  those 
of  the  Ritual  of  Queen 
Mat-ka-Ra,  wife  of  one 
of  the  king-priests  of  the 
Twenty-first  Dynasty  (PI. 
III).  The  panels  are  less 
carefully  executed ,  but 
they  lack  neither  vigour  nor  harmony,  as  is  shown  by  that  of 
Pakheri  in  the  Cairo  Museum  (Fig.  526),  where  we  see  the  deceased 

bringing  offerings  to  Hathor 
the  cow,  who  is  emerging  from 
the  western  mountain.  The 
stelae  generally  represent  the 
dead  man  or  woman  adoring 
Amon  or  Mut,  or,  from  the 
Twenty  -  first  and  Twenty- 
second  Dynasties  onward, 
more  frequently  Harmachis. 
They  are  generally  speaking 
the  work  of  some  scrupu- 
lous artisan  who  executed 
them  conscientiously  accord- 
ing to  the  principles  of  the 
school,  but  several  were  paint- 
ed by  artists  of  talent,  and 
bear  comparison  with  the  best 
miniatures  of  the  Theban 
School.  I  may  instance  that 
of  the  priestess  Zadamonefo- 
nukhu  (PL  II)  in  the  Cairo 
Museum;  half  naked  under 
her  flowing  robe,  she  raises 


TIG.   515. — BAS-RELIEF  OF  MACRINUS 

AND  DIADUMENIANUS  AT  KOM-OMBO. 

(Phot.  Thedenat.) 


268 


THE  SAITE  AGE 


FIG.   5l6. 

HEAD   OF    AN    ETHIOPIAN    KING 
(Museum,  Cairo).  (Phot.  E.  Brugsch.) 


her  hands  to  the  god,  claiming  her  share  of  the  offerings  heaped 
before  him.  She  is  charming  in  the  deliberate  rigidity  of  her 
hieratic  pose,  and  never  did  a  more 
supple  line  envelope  a  young  body 
more  gracefully  slender.  The  colour, 
at  once  rich  and  tender,  relieves  the 
melancholy  proper  to  a  funerary 
monument  by  its  gaiety,  and  the  gray- 
blue  background  against  which  the 
figures  and  inscriptions  are  set  mod- 
ifies their  vivacity.  Under  the  princi- 
pal scene,  in  the  midst  of  a  land- 
scape almost  unique  so  far,  the  corner 
of  the  burial  ground  where  the  de- 
ceased rests  is  indicated  very  effec- 
tively. A  sandy  mountain,  yellow 
streaked  with  red,  descends  from 
left  to  right.  The  facades  of  three 
tombs  are  set  against  it,  and  before 
them  a  kneeling  woman  smites  her 

forehead  with  her  hands,  bewailing  the  priestess;  a  nabeca  and 
three  date-trees  loaded  with  fruit  rise  on  the  right,  and  between 
the  two  latter  is  the  table  of  offerings  on  which  the  soul  will 

Eresently  alight  to  take  its  meal.  The  composition  is  not  badly 
alanced,  and  the  solitary  mourner  in  the  middle  commands  the 
spectator's  attention.  If  we  compare 
it  with  other  stelae  of  the  same 
category,  we  shall  recognise  in  it 
a  desire  to  renew  the  expression 
of  the  religious  idea  they  set  forth, 
and  to  bring  it  closer  to  reality 
than  had  hitherto  been  done.  In 
the  earlier  conception ,  the  lowest 
register  represented  the  scenes  that 
passed  on  earth,  the  rites  performed 
by  survivors  in  honour  or  the  rela- 
tive they  lamented,  while  in  the 
upper  register  the  deceased  was  seen 
arriving  in  the  other  world  and  re- 
ceiving from  the  gods  his  share  of 
the  sacrifice.  The  artist  commission- 
ed to  execute  the  stelae  of  Zadamonefonukhu ,  conscious  of  the 
absurdity  of  associating  as  in  contemporary  life  persons  some 

269 


FIG.  517.— HEAD  OF  AN  ETHIOP- 
IAN KING   (After  Lepsius). 


ART  IN  EGYPT 


of  whom  were  still  living  on  the  earth,  while  others  had  left  it, 
and   thinking  further  that  the  distinction  between  the  terrestrial 

ceremony   and   the    apo- 
theosis  ir 
the   gods 


the 


FIG.   5l8. — BAS-RELIEF  AT  DAKKEH. 
(Phot.   Oropesa.) 


abode  of 
not  suffi- 
ciently marked,  took  upon 
himself  to  accentuate  the 
contrast  to  the  eye  as 
strongly  as  to  the  spirit. 
Retaining  the  upper  pic- 
ture unaltered,  he  replac- 
ed the  fictitious  tete-a- 
tete  of  the  lower  register 
by  an  episode  of  real  life, 
one  that  might  have  been 
seen  any  day  in  cemete- 
ries. The  dead  woman  is 
invisible,  but  one  of  the 

women  of  the  family  has  come  to  bring  her  the  homage  of  her  rela- 
tives, and  to  weep  for  her;  the  antique  idea  persisted  under  a  reju- 
venated form,  better  adapted  to  the  tendencies  of  contemporary  art. 
Like   all    the   other    attacks    upon    tradition,    this    one   failed. 

Three  stelae  of  the 
same  period  and 
origin  show  that 
the  Cairo  example 
was  no  solitary 
caprice ;  in  these 
the  landscape  is 
less  elaborate  and 
the  lower  register 
contains  only  a  con- 
ventional sketch, 
either  of  the  moun- 
tain (Fig.  527)  or 
of  the  garden 
where  the  soul  was 
supposed  to  shel- 
ter during  the  heat 
of  the  day,  a  group 
of  date-palms  and 

sycamores  very  hastily  painted.    It  was  the  same  with  the  coffins 
and  cartonnages.   For  these  the  Theban  artists  continued  to  invent, 

270 


FIG.   gig.— BAS-RELIEF  OF  MAHARRAKAH 
(Museum,  Cairo).     (Phot.  E.  Brugsch.J 


THE  SAlTE  AGE 


if  not  new  forms,   at  least  new  beauties.     In  some  cases  the  lid 
reproduced  not  only  the  mask,   but  the  contours  and  relief  of 


FIG.  520.-BAS-RELIEF  OF  PSAMMETIKNUFISASHMU  (Museum,  Cairo).  (Phot.E.Brugsch.) 

the  whole  body,  with  all  the  details  of  the  costume,  the  wig, 
the  plain  or  pleated  skirt,  the  gauze  robe,  the  sandals,  so  that 
it  has  the  appearance  of  a  statue  of  the  defunct,  lying  in  state. 
Towards  the  end  of  the  Twentieth  Dynasty  and  under  those 
that  followed,  it  was  sometimes  as  carefully  executed  as  a  real 
statue;  the  work  is  comparable  to  work  in  limestone,  but  it  is 
more  sincere  and  less  in  bondage  to  a  school  convention.  The 
subject,  of  course,  admitted  of  no  fancy  in  the  attitude;  the 
model  had  to  lie  flat  on  his  back,  his  head  straight,  his  chest 
expanded,  his  legs  and  his  feet  joined;  the  only  variation  was 
in  the  position  of  the  arms,  which  sometimes  lie  along  the  sides, 
sometimes  are  crossed  on  the  breast.  The  whole  is,  however, 
so  true  in  its  proportion,  the  colours  which  heighten  each  part 
are  so  happily  har- 
monised, that  we 
get  almost  an  illu- 
sion of  life.  These 
cartonnages,  espe- 
cially those  of  wo- 
men, almost  attain- 
ed to  perfection  to- 
wards the  Twenty- 
second  Dynasty. 
Thus  the  Princess 
Tantkalashiri,  who 
died  in  the  reign, 
of  Osorkon  II.,  is 
wrapped  in  a  dra- 
pery of  pale  pink  which  defines  the  contours  without  indiscreet 
emphasis.  Her  arms  are  free,  and  one  of  them  is  laid  upon 

271 


FIG.  521. — BAS-RELIEF  OF  ZANUFI  (Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


her  bosom  under  the  breasts;  some  half  dozen  amulets,  the  sign 
of  life,    the   buckle,    the   altar  with  four  tablets,    hang  from  her 


FIG.  522. — BAS-RELIEF  OF  PSAMMETIKNUFISASHMU  (Museum,  Alexandria). 
(Phot.  E..  Brugsch.) 

wrist  and  protect  her.  The  face,  enframed  by  the  puffed  wig, 
is  a  rather  flat  oval,  with  small  but  merry  eyes,  a  thick  mouth, 
a  short  nose,  and  an  expression  of  gaiety  and  good  humour. 
Even  on  inferior  coffins,  the  carved  wooden  mask  is  often  ex- 
cellent down  to  the  beginning  of  the  Persian  era,  after  which 

the  decadence  of  Thebes 
was  complete,  and  her 
funeral  -  furnishers  produ- 
ced nothing  but  com- 
mercial articles,  barbarous 
both  in  colour  and  mod- 
elling. The  workshops 
of  Lower  Egypt  held  their 
own  better,  and  partici- 
pated in  the  revival  of 
Sa'ite  art.  Under  the  first 
Ptolemies,  when  they  were 
relieved  from  the  necessity 
of  copying  in  wood  the 
uncouth  masses  of  green 
schist  or  granite  coffins, 
they  executed  works  comparable  to  the  best  Theban  examples; 
such  are  the  coffin  of  Psammetichus,  discovered  at  Wardan  (Fig.  528) 

272 


FIG.  523.— ANTAEUS  AND  ISIS  (Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


THE  SAITE  AGE 


FIG.   524- 
A  ROOM  IN  THE  HYPOGEUM   OF  KOM-ES-SHUGAFA. 


and  that  of  the  lady  Tataharsiasi ,  which  belongs  to  the  Berlin 
Museum  (Fig.  529).  The  Greek  influence  is  supreme  here  in 
the  arrangement  of 
the  headdress,  but 
the  technique  is 
Egyptian;  the  sculp- 
tor belonged  to  one 
of  those  mixed 
schools  whose  exis- 
tence at  Memphis 
and  Alexandria  to- 
wards the  end  of 
the  Ptolemaic  age  I 
have  already  noted. 

The  Egyptian  pa- 
lette had  enriched 
itself;  it  included 
at  least  two  kinds 
of  pink,  five  or  six 
shades  of  green, 
blue,  and  yellow, 

violets,  lilacs,  and  mauves,  notably  in  the  regions  colonised  by 
the  Greeks,  round  Alexandria,  in  the  Fayum,  at  Ptolemais  in  the 
Thebaid.  The  Egyp- 
tian painters  promptly 
yielded  to  the  temp- 
tation to  paint  por- 
traits or  decorate 
walls  in  the  manner 
of  the  works  they  saw 
in  the  Hellenic  villas. 
We  must  go  to  the 
Oases  to  find  in  the 
temples  or  the  hypo- 
gea  some  fragments 
of  these  mural  paint- 
ings, in  which  the 
drawing  is  incorrect 
and  the  touch  unskil- 
ful (Fig.  530). 

Portraits ,    on    the 

other   hand ,    have    come    down   to   us   in    considerable    numbers, 
thanks  to  a  caprice  of  fashion.     Towards  the  middle  of  the  last 

273 


FIG.   525. 
DECORATION  ON  THE  TOMB    OF    KOM-ES-SHUGAFA. 


ART  IN  EGYPT 


century  before  our  era,  the  rich  families  of  the  Fayum  and  Upper 
Egypt,  disgusted  by  the  heaviness  and  coarseness  of  the  wooden 

coffins,  conceived  the  idea 
of  substituting  for  carv- 
ed masks  some  equivalent 
in  which  their  artistic 
tastes  would  be  more 
respected ,  and  likeness 
better  observed.  They 
had  recourse  to  two  differ- 
ent methods.  The  first, 
that  of  which  we  have 
the  oldest  examples,  con- 

;  FIG.  526.-PAINTED  PANEL  OF  pAKHERi  sisted    in    replacing    the 

(Museum,  Kairo).    (Phot.  E.  Brugsch.)  face  in  relief  by  a  panel 

painted    with    wax     and 

slightly  re -touched  with  distemper,  or  even  painted  entirely  in 
distemper;  it  was  set  into  the  cartonnage,  above  the  mummified 
head,  and  the  deceased  seemed  to  be  looking  through  a  window 
to  see  what  was  going  on  in  the  world  of  the  living,  just  as 
his  ancestors  had  done  by  means  of  the  eyes  formerlv  painted 

on  the  sides  of  the  coffin.  The 
majority  of  these  portrait -coffins 
were  found  in  the  burial  grounds 
of  Dimeh  or  Hawara,  and  date 
from  the  century  before  or  the 
first  century  after  Christ;  several, 
now  in  the  Louvre,  come  from 
Thebes  and  represent  members  of 
the  noble  clan  of  the  Soters,  who 
flourished  under  the  Antonines. 
Some  of  them  are  so  excellent  in 
design  and  colour  that  they  might 
almost  be  ascribed  to  a  good  Ital- 
ian master  of  the  fifteenth  century, 
but  though  painted  by  native  Egyp- 
tians, they  have  nothing  in  common 
with  Pharaonic  art;  their  source 
of  inspiration  is  Alexandrian.  The 
second  method  is  less  completely 
exotic  (Fig.  531);  it  required,  in- 
stead of  the  wooden  mask,  a  plaster  bust,  in  the  hollow  of  which 
the  veiled  head  of  the  mummy  was  encased,  and  the  lower  edges 

274 


FIG.   527.— PAINTED   STELE 
(Museum,   Cairo).     (Phot.  E.  Brugsch.) 


THE  SAITE  AGE 


FIG.  528. — COFFIN  FROM  WARDAN 
(Museum,  Cairo).  (Phot.  E.  Brugsch.) 


of  which  disappeared  beneath  the  bandages  of  the  chest.     The 

design    as    a   whole    was   Greek;    but    certain    details    remained 

specially  Egyptian,  such  as  the  custom  of 

inserting  the  eyes,  and  imitating  their 

natural   effect  by  incrustations  of  talc 

or  glass.  This  fashion  obtained  towards 

the  second  century  of  our  era  among 

the     higher     classes     of     Heptanomis 

(Middle  Egypt)  from  the  Fayum  to  the 

confines  of  the  Thebaid.    Even  in  the 

Fayum  it  did  not  exclude  the  use  of 

cartonnage    masks    modelled    in    the 

semblance   of   the    dead,    but  treated 

in  the  Greek  manner,  with  indications 

of  the  upper  part  of  the  clothing,  faces 

enlivened  by  paint,  jewels,  curled  locks 

and  wreaths  of  flowers;   seeing  them 

in  our  glass  cases,  we  might  take  them 

for  busts  of  coloured  wax  (Fig.  532). 

At  Akhmim  and  in  its  outskirts,    the 

custom  of  laying  the  dead  in  their  out- 
door costumes  upon  their  tombs  came 

into  favour  again  towards  the  times  of  Severus,  but  instead  of 

giving  them  the  loin-cloth  and  archaic 
wig,  they  were  dressed  in  their  modern 
costume,  tunics,  peplums,  mantles  made 
by  some  fashionable  modiste,  whose 
inventions  were  probably  some  months 
behind  the  latest  creations  of  Alexandria 
or  Rome  (Fig.  533).  The  box  is  made 
of  Nile  mud,  plastered  on  a  framework 
of  cardboard  or  stuccoed  cloth ;  it  was 
given  a  form  approximating  to  that 
of  the  person  for  whom  it  was  destined, 
and  over  the  modelling  was  painted, 
in  distemper  or  tempera,  the  face,  the 
flower-crowned  head,  the  hands,  jewels, 
and  multi-coloured  stuffs. 

This  was  the  last  effort  towards 
originality  of  the  artists  who  worked 
for  the  burial-grounds.  Mummy-paint- 
ing and  sculpture  ceased  towards  the 
middle  of  the  third  century,  at  the 
275  T  2 


FIG.   529. 

COFFIN   OF   TATAHARSIASI 
(Museum.  Berlin). 


ART  IN  EGYPT 


same  time  as  ordinary  sculpture  and  painting,  and  for  the  same 
reasons;  the  civil  wars,  the  effects  of  which  were  felt  in  Egypt, 
and  then  the  invasion  of  Christianity,  destroyed  all  the  arts  which 
depended  for  their  existence  on  the  maintenance  of  the  antique 
religions.  Sculpture  in  metal  and  ceramics  were  also  affected. 
As  they  overcame  the  technical  difficulties  of  casting,  the  Saite 
masters  were  emboldened  to  increase  the  size  of  their  works  until 
at  last  they  succeeded  in  casting  figures  larger  than  life  in  a 
single  piece.  Not  one  of  these  metal  colossi  has  come  down  to 

us  intact,  but  we 
possess  fragments 
which  enable  us  to 
reconstitute  their 
appearance ,  such 
as  the  hand  grasp- 
ing the  hydra,  now 
in  the  Cairo  Mu- 
seum, which  Dani- 
nos  found  among 
the  ruins  of  Mem- 
phis: it  is  termin- 
ated at  the  wrist 
by  a  rectangular 
tenon  which  held 
it  to  the  arm,  and 
the  effigy  of  the 
kneeling  king  to 
which  it  belonged 
must  have  been 
about  6l/2  feet  in 

height.  But  this  is  exceptional.  The  statue  of  Petukhanu,  the 
torso  of  which  was  in  the  Stroganoff  collection,  was  barely  life- 
size,  and  the  most  important  pieces  we  have  of  the  Bubastite 
or  Saite  ages  are  rarely  as  much  as  3  feet  in  height.  Several 
of  them  are  finer  than  the  best  contemporary  examples  in  lime- 
stone or  granite,  notably  the  little  sphinx  of  Apries  in  the  Louvre 
(Fig.  534)  and  the  kneeling  Tirhakah  at  Cairo  (Fig.  535).  The 
Karomama  in  the  Louvre  (Fig.  537)  bought  by  Champollion  of 
a  dealer  who  had  himself  bought  it  at  Luxor,  is  Theban  in  hand- 
ling. The  queen  is  standing,  dressed  in  a  long,  closely  -  fitting 
gown  with  flowing  sleeves,  her  head  crowned  with  a  ceremonial 
wig,  the  forelocks  of  which  overhang  her  brow;  the  eyes  are 
inserted,  and  the  divisions  of  the  wig,  as  well  as  the  folds  of 

276 


FIG.   530.— PAINTING  OF  A  HYPOGEUM  IN  THE   OASIS 
OF  BAHRIYAH.      (Phot.  Moritz  Bey.) 


THE  SAITE  AGE 


FIG.   531. — PLASTER  MASK   OF 

A  MUMMY  (Museum,   Cairo). 

(Phot.  E.  Brugsch.) 


the  dress,  were  incrusted  with  gold. 
The  body  is  finely  modelled  under  the 
stuff,  but  the  head  is  above  praise. 
Karomama  was  certainly  no  beauty, 
with  her  long ,  beak  -  like  nose ,  her 
sunken  nostrils,  her  dry  mouth  and 
bony  chin.  But  as  she  had  a  lofty 
bearing,  the  artist,  unable  to  make  her 
attractive,  concentrated  all  his  powers 
on  the  pride  and  energy  of  the  face; 
his  Karomama  is  the  incarnation  of 
what  he  conceived  the  wife  of  Pharaoh 
and  Queen  of  Egypt  should  be.  The 
Takushit  of  the  Athens  Museum  (Fig. 
536)  on  the  other  hand  is  a  woman 
of  the  middle  class,  a  worthy  lady  of 
Bubastis,  and  her  statuette,  probably 
the  product  of  a  local  workshop,  is  a 
contrast  in  the  somewhat  flaccid  round- 
ness of  its  contours  to  the  nervous 
spareness  of  the  Karomama.  The  good  dame  has  started  off 
with  the  left  foot,  and  she  walks  without  haste,  her  right  arm 
hanging,  her  left  held  against  her  breast;  her  drapery  moulds 
while  affecting  to  conceal  her  full  hips 
and  abdomen,  and  her  round,  heavy 
breasts.  The  face  is  broad  and  fat 
under  the  wig  of  short  locks  rising 
in  tiers  one  above  the  other,  and  her 
narrow  eyes,  her  short  nose,  her  fleshy 
lips  and  rounded  cheeks  are  those  of 
a  fellah  woman  without  any  touch  of 
race.  The  bronze  with  its  mixture  of 
gold  and  silver,  is  irradiated  by  soft 
reflections  which  seem  to  animate  the 
forms;  the  dress  is  covered  as  with 
an  embroidery  by  religious  scenes  and 
inscriptions  incised  and  filled  in  with 
a  silver  line. 

We  can  hardly  say  whether  the 
activity  of  the  Theban  foundries  relax- 
ed from  the  beginning  of  the  Saite 
period,  or  whether  the  lack  of  Theban 
bronzes  is  due  merely  to  the  perversity 

277 


FIG.   532.— MASK   OF  A  MUMMY 

IN  PAINTED   CARTON 
(Museum,  Cairo).  (Phot.  E.  Brugsdi.) 


ART  IN  EGYPT 


of  chance.  But  with  the  exception  of  an  enormous  Osiris  found 
in  the  favissa  at  Karnak,  the  finest  and  most  important  bronzes 
of  our  collections  all  come  from  Memphis  or  the  Delta.  Bubastjs 
was  the  home  of  the  four  bronzes  bought  by  the  Louvre  at  the 
Posno  sale.  The  first,  one  Masu,  whose  name  is  tattooed  on  his 
breast  (Fig.  538)  near  the  heart,  advances  towards  the  spectator 
with  a  proud,  confident  movement;  the  face,  somewhat  disfigured 
by  the  loss  of  the  eyes,  which  were  of  enamel 
encircled  with  silver,  breathes  energy  and 
arrogance.  The  second  is  less  vigorous  in 
bearing,  but  the  third  (Fig.  539),  a  Horus  who 
originally  lifted  up  a  jar  from  which  he  poured 
water  over  a  king  kneeling  before  him,  is 
harsher  and  drier,  and  was  perhaps  cast  in 
the  same  workshop  as  the  kneeling  Horus  at 
Cairo;  the  composition  of  the  metal  seems  to 
be  identical  in  each,  the  handling  is  similar, 
and  the  manner  in  which  the  bird's  head  is 
adjusted  to  the  human  bust  is  marked  by  the 
same  exactness.  It  must,  indeed,  be  allowed 
that  these  divine  monsters,  in  whom  the  hu- 
man and  animal  natures  are  allied,  inspired 
the  makers  of  bronzes  more  happily  than  the 
statuaries.  The  Basts  and  Sekhets  discovered 
by  Barsanti  at  Sais  (Fig.  540)  are  not  only 
comparable  to  the  black  granite  Sekhets  of 
Amenophis  III.,  but  superior  to  them  in  dignity 
of  attitude,  and  the  suggestion  of  restrained 
vitality.  Their  cat  or  lion  heads  rest  more 
easily  on  their  feminine  shoulders,  and  they 
are  less  in  the  nature  of  a  defiance  to  the 
laws  which  rule  the  division  of  species.  The 
lions  of  Thmuis  and  Tell -es- Sab  are  no  earlier  than  the  first 
Ptolemies.  Those  of  Horbet  were  cast  under  Apries  (Fig.  541). 
They  were  part  of  a  mechanical  contrivance  for  closing  the  doors 
of  a  temple,  and  they  had  a  wooden  beam  prosaically  inserted 
in  their  hind-quarters,  but  the  artist  turned  the  conditions  imposed 
by  their  functions  to  excellent  decorative  account;  he  imagined 
them  lying  flat  on  the  ground ,  in  an  oblong  cage ,  the  lateral 
walls  of  which  were  pierced  to  show  their  bodies,  while  their 
heads  and  forepaws  emerged  from  the  open  trap  in  front.  He 
simplified  the  lines,  but  in  the  manner  of  which  the  Egyptians 
were  masters,  neither  suppressing  nor  weakening  any  of  those 

278 


FIG.   533.— COFFIN 

OF  AKHMIM 
(Museum,   Cairo). 
(Phot.  E.  Brugsch.) 


THE  SATTE  AGE 


FIG.    534. — LITTLE  BRONZE   SPHINX  OF 
APRIES  (The  Louvre,  Paris). 


which  give  the  animal  its  character;  the  face  is  calm  and  soberly 
majestic.  The  cat  is  treated  no  less  happily  than  the  lion,  and 
it  may  be  said  without 
exaggeration  that  among 
the  thousands ,  either 
whole  cats  or  heads  of 
cats,  brought  out  in  1878 
from  the  favissa  of  Bubas- 
tis,  very  few  were  bad, 
or  even  mediocre  (Fig. 
542);  no  people  ever 
showed  more  skill  in  seiz- 
ing the  undulating  grace 
of  the  beast,  the  treach- 
erous softness  of  its  atti- 
tudes, and  the  expression 

of  its  mask,  now  dreamy,  now  mutinous.  The  other  animals  — 
rams,  Apis  or  Mnevis  bulls  (Fig.  543),  crocodiles,  cynocephali, 
the  innumerable  figurines  of  Amon  (Fig.  544),  Osiris,  Isis,  Horus, 
Nit,  Anubis  with  a  dog's  muzzle  (Fig. 
545),  Sekhet  with  a  lion's  face  (Fig.  546), 
Thoth  with  the  head  of  a  monkey  or 
an  ibis,  do  not  bear  comparision  with 
the  cats  and  lions;  though  many  of  them 
are  remarkable  for  the  perfection  of 
their  casting,  or  the  delicacy  of  their 
chasing,  the  majority  are  the  prosaic 
reproductions  of  non-artistic  types  de- 
vised for  the  edification  of  the  faithful. 
They  bear  the  same  relation  to  the 
splendid  bronzes  at  Cairo  and  in  the 
Louvre  as  do  the  gilded  and  painted 
saints  of  the  St.  Sulpice  quarter  to  the 
works  of  the  great  Christian  sculptors 
of  France  and  Italy. 

And  here  we  are  confronted  by  a 
problem  the  solution  of  which  we  can 
only  divine  at  present.  Among  the 
innumerable  bronzes  found  in  the  same 
places,  where  they  seem  to  have  been 
deposited  at  the  same  time,  we  find 
some  so  different  from  the  rest  in  style,  that  were  we  not  certain 
of  their  origin,  we  should  be  disposed  to  attribute  them  to  very 

279 


FIG.   535-— BRONZE 

STATUETTE    OF    TIRHAKAH 

(Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


FIG.  536.— THE  LADY 

TAKUSHIT 
(Museum,  Athens). 
(Phot.  E.  Brugsch.) 


diverse  periods  and  localities.  It  is  in  the 
cats  above  all  that  these  divergences  are  most 
strongly  marked.  Some  are  vigorous  and 
realistic  after  the  manner  of  the  best  Theban 
sculptors;  their  silhouettes  have  a  certain 
harsh  abruptness  of  contour  which  is  not  the 
result  of  any  lack  of  skill  in  the  artist,  but 
the  effect  of  a  determination  to  express  the 
energy  and  strength  rather  than  the  grace 
and  ease  of  the  animal's  movements.  With 
others,  however,  the  desire  for  elegance  gets 
the  mastery,  and  the  contours  are  softened 
to  the  verge  of  flaccidity;  we  recognise  the 
Memphite  technique  in  its  most  trivial  aspect. 
Noting  these  contrasts,  we  are  inclined  to 
ask  whether  the  fact  that  such  dissimilar 
works  were  all  buried  together  in  the  favissa 
of  the  temple  of  Bast  is  sufficient  evidence 
that  they  were  all  manufactured  at  Bubastis. 
May  not  the  pilgrims  who  dedicated  them  to 
the  goddess  have  brought 


them  from  their  native  towns?  Their  dissi- 
milarity would  be  comprehensible  enough  in 
this  case ;  those  in  which  we  seem  to  distinguish 
the  impress  of  Theban  or  Memphite  art  would 
then  have  made  their  way  hither  from  Thebes 
or  Memphis.  But  even  if  we  accept  this 
hypothesis,  we  shall  not  have  resolved  the 
difficulties  entirely.  Carefully  examined,  these 
groups  do  not  present  a  homogeneous  appear- 
ance, for  whereas  some  of  the  examples 
really  reveal  the  characteristics  of  the  Saite 
age,  many  others  would  seem  from  their 
treatment  to  be  earlier  by  several  centuries: 
and  yet  the  circumstances  of  the  find  and  the 
nature  of  the  bronze  hardly  permit  us  to 
doubt  that  they  were  all  cast  within  the  space 
of  a  few  years.  An  observation  I  made  in  the 
ruins  of  a  pottery  workshop  discovered  last 
winter  behind  one  of  the  mounds  of  Eshmunen, 
may  help  to  explain  this  anomaly.  The  majority 
of  the  moulds  for  lamps  and  of  the  kiln-refuse 
it  still  contained,  belonged  to  the  Christian 

280 


FIG.   537-— QUEEN 

KAROMAMA 
(The  Louvre,  Paris). 


JUDGMENT  BEFORE  OSIRIS 

(Vignette  from  the  funerary  Papyrus  of  Queen  Mal-ka-Rct, 
Twenty-first  Dynasty) 


I 


THE  SAITE  AGE 


FIG.   538.— BRONZE 

STATUETTE   OF  MASU 

(The  Louvre,  Paris). 


era,  as  we  learn  from 
the  crosses  and  in- 
scriptions, but  others 
are  decorated  with 
pagan  figures  and 
legends,  and  cannot 
be  later  than  the 
second  or  third  cen- 
tury of  our  era ;  the 
potter  must  have 
had  in  the  back  of 
his  shop  old  models 
which  came  from 
his  distant  predeces- 
sors ,  and  these, 
slightly  altered  to 
suit  the  require- 
ments of  the  new 
religion ,  were  still 
sold  occasionally. 


FIG.  539. 

BRONZE    HORUS 

(The  Louvre,  Paris). 


It  is  probable  that  founders  also  preserved  old-fashioned  moulds, 
and  continued  to  cast  with  them  from  time  to  time  for  their 
clients.  Thus  some  Theban  devotee  of  Bast  might,  before  starting 

for  Bubastis,  have  provided  himself 
with  ex-votoes,  cats,  or  cat -headed 
statuettes,  or  other  figures  of  divinities 
which,  though  of  new  metal  and  fresh 
from  the  furnace ,  were  none  the  less 
the  work  of  older  generations  by  virtue 
of  the  moulds  used. 

The  same  may  be  said  of  the 
countless  divinities  made  of  different 
compositions  or  of  terra-cotta,  which 
swarm  in  the  tombs  and  cities  of 
the  Saite  period  and  of  the  Graeco- 
Roman  epoch.  The  last  centuries  of 
paganism  were  above  all  centuries  of 
pious  imagery  for  the  use  of  dead 
and  living,  at  least  in  the  Delta  and 
the  northern  part  of  Middle  Egypt, 
for  the  Said  never  fell  into  these  ex- 
cesses, and  the  use  of  amulets  was 
not  much  more  general  here  than  in 
281 


FIG.   540. — BRONZE    SEKHET 

(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


in  the  glorious  -days  of  the  second  Theban  Empire.  It  was 
inevitable  that  manufacturers  and  dealers  should  spare  them- 
selves the  trouble  of  in-* 
venting  new  types  and 
sacrificing  their  old 
models,  so  long  as  these 
could  be  made  to  suffice 
for  the  demand  and 
content  their  customers. 
And ,  naturally ,  objects 
prepared  by  the  hundred, 
and  even  by  the  thou- 
sand, for  daily  sale,  could 
not  fail  to  be  mediocre 


FIG.   541. — BRONZE  LION  OF  APRIES 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


and  lacking  in  originality. 
There  are  many  of  which 
we  can  only  say  that 
they  faithfully  express  the  hieratic  attitude,  the  gesture,  costume, 
head-dress,  and  exterior  attributes  of  the  god  they  represented; 
this  was  all  the  devout  asked,  and  it  was  the  same  with  the 
Ushabtiu  (substitutes).  Provided  they  vaguely  suggested  the 
mummy  by  their  forms,  and  the  name  of  their  master  had 
been  traced  on  them,  together  with  the  opening  words  of  the 

consecrated  prayer,  they  served 
for  the  rite,  and  this  was  all- 
sufficient;  at  the  beginning  of 
the  Roman  period  many  were 
sold  which  are  hardly  more  than 
pieces  of  clay  or  paste  length- 
ened out,  with  a  vague  indi- 
cation of  the  head  and  the  feet, 
things  more  barbarous  than  the 
most  barbarous  Polynesian  idols. 
Here  and  there,  however,  we 
meet  with  examples  which  stand 
out  from  the  general  level  of 
ugliness,  and  are  almost  finer 
than  those  of  the  great  period. 
They  come  generally  from  the 
wells  of  Sakkarah,  and  belong 
to  the  time  of  the  Persian  do- 
mination, or  to  the  early  reigns 
of  the  Macedonian  dynasty.  The 
282 


FIG.   542. 

BRONZE  CAT  (Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


THE  SAlTE  AGE 


FIG.  543- 

STATUETTE  OF  APIS 
(Museum,   Cairo). 
(Phot.  E.  Brugsch.) 


best,    those    of   a   certain  Admiral   Patanesis, 
varied  in  size  from  4  to  10  inches.    Modelled 
in  a  very   pure   paste,    and   fired  with  extra- 
ordinary skill,    they  were  glazed  with  a  non- 
lustrous   clear,  vivid   blue,  the   freshness   and 
evenness  of  which   are  unimaginable;    I   have 
seen    nothing   to    approach    them   in    modern 
porcelain.    The  head  is  a  gentle,   melancholy 
portrait;    the    only   thing  comparable  to  it  in 
its  own  genre  is  the  little  blue  porcelain  head 
at   Cairo   (Fig.  547) ,    perhaps    an   Apries    or 
Necho  II.    Others,  though  not  so  beautiful  as 
these,    show   a   laudable    effort    to    produce 
something  new;  I  may  instance  the  little  group 
of  green  enamelled  frit,  which,  inspired  perhaps 
by  a  motive   of  the  time  of  Amenophis  IV., 
represents  queen  Amenartas  seated  on  Amon's 
lap  and  passing  her  arms  lovingly  round  his 
neck  (Fig.  548);    the  kohol  jar,    the  body  of 
which   is   formed   by  a  head   of  Apries   in   a 
Greek    helmet    (Fig.  549),     and    the    votive 
statue  in  green  paste  of  Nufiabres,  standing  on  a  high  pedestal 
and  holding  the  naos  of  the  Osiris- 
mummy  in  front  of  him  with  both 
hands.     Some   twenty  of  the  Nits, 
Ras,    Horuses,    Ptahs    and    Nefer- 
Atmus    in    porcelain    preserved    in 
the  Cairo  Museum,   were  executed 
by   workmen    brought    up    in    the 
good  school.    Whereas  their  neigh- 
bours  in   the  glass  cases   show  the 
rounded,  flaccid  forms  which  pleased 
the  Ptolemaic   sculptors,   we   note 
in    them   the    nervous,    and    some- 
times   rather    dry    handling   of   an 
earlier  age.    Of  course  it  was  not 
very    easy    to    mark    the    play    of 
muscles    in    works    barely    ten    or 
twelve  inches  high,  sometimes  con- 
siderably less.     Artists   accordingly 
adopted  the   plan  of  enclosing  the 
limbs    in    a    series    of    frankly    cut 
planes  with  sharp  angles,  and  exag- 

283 


FIG.  544- 

HARPOCRATES,   OSIRIS,   AND 
AMON.     BRONZE   STATUETTES 

(Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


gerating  the  proportions  of  the  anatomical  details  which  they 
preserved  in  the  knees,  feet,  arms  and  face,  but  with  such  an 
intelligent  sense  of  effect 
that  we  have  to  examine 
them  a  second  time,  if 
we  have  not  been  already 
informed  of  the  device, 
before  we  notice  the 
exaggeration.  If  they  had 
respected  the  true  di- 
mensions ,  certain  ele- 
ments of  the  human 
body  would  have  been 
so  attenuated  as  to  be- 
come almost  invisible, 
and  the  general  impres- 
sion of  truth  would  have 
suffered.  Several  of  these 
figurines  are  treated  so 
skilfully  that  instead  of 
appearing  what  they,  are, 
miniatures  of  men  or 
animals,  we  feel  when  we 
examine  them  as  if  we 

were  looking  at  colossal  figures  from  the  wrong  end  of  a  field-glass. 
The  Egyptians  of  the  Pharaonic  age  had  used   plain   earthen- 

ware ,   neither   glazed  nor  coloured, 

only  for  the  manufacture  of  coarse 
domestic  utensils  and  amulets, 
chiefly  articles  intended  for  the  poor, 
ubshabti,  beads,  figures  of  the 
gods,  more  especially  Bes;  it  is 
only  exceptionally  that  we  find,  to- 
wards the  close  of  the  second  Theban 
age,  heads  of  Canopic  vases  in  clay 
as  delicately  executed  as  if  they  had 
been  in  stone  or  enamelled  ware. 
From  the  accession  of  the  Ptolemies, 
and  probably  under  the  influence  of 
Greece,  taste  developed.  We  know 
what  masterpieces  were  bequeathed 
to  us  by  the  potters  of  Alexandria; 
several  of  the  statuettes  found  in 
284 


FIG.    545.— BRONZE 

STATUETTE    OF   ANUBIS 

(Museum,   Cairo). 

(Phot.  E.  Brugsdi.) 


FIG.  540. — BRONZE 

SEKHET 

(Museum,   Cairo). 
(Phot.  E.  Brugsch.) 


FIG.   547.— ROYAL  HEAD  IN 

BLUE  ENAMEL  (Museum,    Cairo). 

(Phot.  E.  Brugsdi.) 


THE  SAlTE  AGE 


the  burial  grounds  of  Meks  equal  those  of  Tanagra.  The  natives 
imitated  their  foreign  comrades,  and  gradually  the  use  of 
earthenware ,  baked  or  unbaked ,  but 
always  painted  in  bright  colours,  be- 
came general  from  one  end  of  the 
valley  to  the  other.  It  found  favour 
more  especially  in  localities  where 
there  were  colonies  of  Hellenes,  in  the 
Delta,  at  Memphis,  in  the  Fayum,  at 
Hermopolis,  at  Akhmim,  at  Syene,  but 
it  also  made  its  way  into  places  that 
had  remained  purely  Egyptian.  Its 
manifestations  are  innumerable,  from 
the  decorative  plaques  in  temples  and 
public  buildings  to  household  utensils, 
lamps,  domestic  lares,  groups  represent- 
ing episodes  in  private  life,  grotesque 
and  sometimes  obscene  figurines,  camels 
(Fig.  551),  elephants  (Fig.  550),  birds, 
and  the  majority  are  industrial  rather 
than  artistic  creations  (Fig.  552).  Nev- 
ertheless, some  of  the  subjects  are 
treated  with  a  most  amusing  dexterity 
(Fig.  553),  and  bronze  was  even  used 
in  some  cases  (Fig.  555).  A  study  of  Perichon  Bey's  collection 
is  particularly  instructive  for  this  genre.  It  all  came  from  the 
tells  of  Eshmunen,  the  ancient  city 
of  Thoth,  and  the  majority  of  the 
pieces  composing  it  do  not  go  back 
further  than  the  second  century  of 
our  era.  Yet  at  Cairo  there  are 
heads  of  dwarfs  and  idiots  of  sur- 
prising truth  (Fig.  554).  Sugar-loaf 
skulls,  narrow  retreating  foreheads, 
eyes  overhung  with  bushy  eyebrows, 
crooked  noses,  bony  cheeks,  hanging 
lips,  minute  chins,  enormous  ears  set 
on  each  side  of  the  head  like  the 
handles  of  an  ill-made  pitcher  —  no 
feature  is  lacking  of  all  that  makes 
up  well-observed  human  deformity; 
two  or  three  strokes  of  the  thumb 
lengthened  and  kneaded  the  paste 

285 


FIG.  548. — AMON 

AND  QUEEN  AMENARTAS 

(Museum,  Cairo). 

(Phot.  Legrain.) 


FIG.   549.— VASE  IN  THE  FORM 

OF  A  HEAD  IN  A  HELMET 

(The  Louvre,  Paris). 


ART  IN  EGYPT 


to  the  desired  module;  then  a  pinch  here  and  another  there 
to  bring  out  the  protuberances  of  the  face,  a  stroke  of  the 
graver  for  the  mouth,  two  pellets  for  the  eyes,  and  there  it 

was,  as  ugly  as  nature,  but 
more  amusing.  Animals 
are  treated  with  no  less 
spirit ,  dogs  especially, 
not  the  thin  greyhound, 
the  prototype  of  the  so- 
called  jackal  Anubis,  but 
the  pug,  with  the  angry 
muzzle ,  pointed  ears, 
long  waving  hair  and 
curly  tail ,  or  the  good 
fellow  of  no  particular 
breed  (Fig.  556) ,  who 
thinks  his  constant  bark- 
ing protects  the  house, 
but  whose  true  function 

FIG.   55O. — TERRA-COTTA   ELEPHANT 

(Museum,  Cairo).  (Phot.  E.  Brugsch.)  is   to    be    tormented    by 

the  children  in  it.    Here 

and  there  are  feminine  heads  so  graceful  that  they  would  not 
disgrace  the  Alexandrian  series;  they  are  purely  Greek.  The 
only  persons  who  have  not  entirely  forsworn  their  Egyptian 
character  are  the  fashionable  divinities.  Harpocrates  chubby  as 
a  Pompeian  Cupid,  but  adorned  with  a  minute  pschent,  Agatho- 
demons  with  a  uraeus  body  and  an  Isis  head  (Fig.  557),  Isis  cha- 
stely draped  (Fig.  558),  and  others  destined  to  serve  as  wives 

to  the  dead,  their  tunics  rolled 
up  on  their  breasts;  these  replac- 
ed the  statuettes  of  blue  and 
green  porcelain  towards  the 
close  of  the  first  century  after 
Christ,  and  until  the  definitive 
triumph  of  Christianity  they 
sufficed  for  popular  devotion. 
The  same  transformation  took 
place  in  the  other  minor  arts, 
though  we  are  not  yet  in  a  po- 
sition to  note  its  successive 
stages.  Furniture  retained  the 
ancient  forms  in  its  essentials, 

FIG.   551.— TERRA-COTTA   CAMEL 

(Museum,  Cairo).   (Phot.  E.  Brugsch.)         at   least   among  the   poor  and 

286 


THE  SAITE  AGE 


the  lower  middle  classes;  the  domination  of  the  foreigner  had, 
in  fact,  altered  nothing,  or  almost  nothing,  in  the  habits  of  the 
fellahin  and  the  artisan ,  and  even 
the  introduction  of  a  current  coinage 
had  not  affected  the  conditions  of 
their  domestic  life  as  might  have 
been  expected.  They  did  not  want 
a  single  piece  of  furniture  more  than 
their  ancestors  had  used  under  the 
Pharaohs,  and  the  little  they  required 
they  continued  to  make  on  the  con- 
secrated models,  beds  and  arm-chairs 
with  lion's  feet  incrusted  with  ivory, 
bone  or  ebony,  stools  and  benches 
with  leather  seats  and  many-coloured 
cushions ,  linen  -  chests ,  bread  -  bins, 
jewel  -  caskets ,  kohol  -  pots ,  perfume- 
boxes;  they  admitted  innovations 
only  in  certain  funerary  articles.  The 
catafalque  in  which  the  mummy 
journeyed  to  the  tomb,  under  the 
Tanites  of  the  Twenty-first  Dynasty, 
an  enormous  rectangular  case  laid 
upon  a  sleigh,  became  under  the  Ptolemies  a  carved  wooden 
bed  with  a  canopy.  The  one  in  the  Edinburgh  Museum,  which 
Rhind  got  at  Sheik  -Abd-  el  -Kurnah,  simulates  a  kiosk  with  a 


FIG.  552. 

TERRA-COTTA  GROTESQUE 
(Museum,  Cairo).  (Phot.E.  Brugsch.) 


FIG.  553-— GROTESQUE  HEAD  IN 

TERRA-COTTA  (Museum,  Cairo). 

(Phot.  E.  Brugsch.) 


287 


FIG.  554. — GROTESQUE  HEAD 
IN  TERRA-COTTA  (Museum, 
Cairo).  (Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


barrel -vault,  three  sides  of  which  are  upheld  by  little  columns 
of  coloured  wood.  The  fourth,  that  of  the  head,  has  a  facade 
with  three  superposed  cornices,  each  decorated  with  its  winged 
disc,  the  whole  bordered  by  a  row  of  rampant  uraei;  a  door 
between  two  columns,  guarded  by  serpents,  was  supposed  to 
give  access  to  the  interior.  The  mummy  within  was,  as  it  were, 
in  a  peripteral  temple  the  sanctuary  of  which  was  his  coffin. 
The  catafalque  in  our  Museum  (Fig.  559)  which  I  found  at 
Akhmim  in  1885,  is  conceived  in  a  spirit 
more  attuned  to  its  funereal  function.  Its 
lateral  columns  are  replaced  by  cut  out 
pieces  of  painted  wood  representing  the 
goddess  Maat,  the  Truth  who  protected  the 
doubles  at  the  tribunal  of  Osiris ;  she  crouches 
on  her  haunches ,  her  pen  on  her  lap ,  and 
beside  her  the  winged  Isis  and  Nephthys  of 
the  ordinary  sarcophagi  fill  up  the  space 
at  the  short  ends.  The  vault  is  of  open- 
work, and  on  each  of  the  seven  curves 
which  compose  it  are  painted  vultures,  spread- 
ing out  their  wings  above  the  mummy; 
two  statuettes  of  Isis  and  Nephthys,  posted 
at  the  two  extremities,  lament  as  prescribed 
by  ritual.  The  work  is  agreeable  to  the 
eye,  and  if  provincial  artisans  were  capable 
of  productions  so  tasteful,  we  may  imagine 
what  those  of  Memphis  could  do;  here 
again,  the  cult  of  the  dead  prevented  art 
from  falling  too  low,  when  it  sank  into 
decadence  in  civil  life.  There  is  reason  to 
believe  that  Hellenism  made  way  among  the  rich,  and  that  the 
same  class  who  under  the  first  Caesars  substituted  their  wax  portraits 
for  wooden  coffin-masks,  furnished  their  houses  in  the  western 
fashion,  like  modern  Egyptians,  who  buy  the  furniture  for  theii 
dining,  reception,  and  bed-rooms  in  Venice,  Paris,  and  London. 
None  of  these  Hellenistic  pieces  of  furniture  have  come  down 
to  us,  but  in  1901  Daninos  found  at  Memphis  fragments  of 
several  carrying-chairs  which  had  belonged  to  one  of  the  last 
Sa'ite  Pharaohs.  The  wood,  which  was  in  bad  condition,  was 
profusely  decorated  with  small  bronze  plaques,  some  in  very 
low  relief,  others  cut  out  flat  in  the  metal  and  incised  the 
designs  being  Niles  (Fig.  560)  and  Osirises  bringing  offerings 
(Fig.  562),  or  helmeted  kings  (Fig.  561),  Thothmes  III.,  Osorkon  III. 

288 


FIG.  555- 

GROTESQUE  FIGUR- 
INE IN  BRONZE 
(Museum,   Cairo). 
(Phot.  E.  Brugsch.) 


THE  SAlTE  AGE 


Psammetichus  II,  Amasis.  It  is  possible  that  they  came  from 
Thebes,  in  the  trousseau  of  some  princess  married  at  Memphis; 
whatever  their  origin,  they  are  mediocre  in  design  and  even 
more  so  in  execution. 

Goldsmith's  work  and  jewelry  alone  flourished  to  the  end,  and 
were  transmitted,  by  a  complete  cycle  of  transformations,  to  the 
Byzantines  and  then  to  the  Arabs,  thus  escaping  to  some  extent 
the  destruction  of  the  Pharaonic  civilisation.  In  the  beginning, 
under  the  Twenty-second 
and  Twenty -sixth  Dy- 
nasties, these  productions 
differed  only  by  almost 
imperceptible  shades  from 
those  of  the  Theban  age. 
The  shallow  goblets,  some 
Egyptian,  others  Cypriot, 
but  in  the  Egyptian  style, 
discovered  in  the  palaces 
of  theSargonids  in  Assyria, 
resemble  those  in  the 
treasure  of  Bubastis;  ne- 
vertheless, martial  scenes 
occur  frequently  in  them, 
and  the  progress  of  military 
art  complicates  them  with 
incidents  unknown  to  the 
strategy  of  earlier  gener- 
ations ,  cavalry  charges 
side  by  side  with  chariot  charges.  But  for  this,  the  composition 
is  very  little  changed;  as  formerly,  it  is  arranged  in  concentric 
bands,  in  which  the  incidents  are  separated  by  florets  or  trees. 
The  influence  of  Greece  began  to  make  itself  felt  towards  the 
end  of  the  Sa'i'te  period,  and  several  of  the  pieces  from  Tukh- 
el-Karamus  are  importations  from  Ionia,  as,  for  instance,  the 
bracelet  with  the  Eros,  the  rhyton,  the  two  perfume-burners  in  the 
form  of  altars ;  but  others  were  manufactured  in  Egypt  by  Egyp- 
tians, and  these  are  not  the  least  remarkable.  The  oxide  from 
which  we  have  not  been  able  to  free  them  mars  their  purity  of 
contour  and  delicacy  of  ornamentation;  but  it  does  not  prevent 
us  from  recognising  that  they  are  covered  with  true  Egyptian 
motives,  treated  in  the  Egyptian  manner,  lotus-flowers  or  buds, 
running  ornament,  foliage,  and  clusters  of  aquatic  plants.  We 
distinguish  these  still  more  plainly  on  the  silver  vessels  of  Thmuis, 

289  U 


FIG.   556. 

TERRA-COTTA  DOG  (Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


ART  IN  EGYPT 


where  no  extraneous  matter  clogs  the  surfaces  (Fig.  563).  They 
are  deep  libation-cups,  rounded  at  the  bases,  the  bodies  of  which, 

slightly     compressed 

towards  the  top,  open 

out  widely  at  the  lip. 

A    rosette    enclosed 

in     a     circle    marks 

the  centre  of  gravity, 

and      focusses      the 

external    decoration, 

lotus  -  blossoms      al- 
ternately   in    bloom 

and    in    bud,     then 

narrow    leaflets   laid 

closely  together,  their 

points    separated  by 

ovae   in    relief.     The 

handle  of  the  cover 

is    formed    by    two 

lotus-flowers  laid  flat 

upon  the  surface  and 

united  by  the  stems. 

Some   of  the  pieces 

were   beaten    out   in 

a  mould  of  hard  stone,  or  repousse,  and  then  retouched  with 
the  point;  others  were  chased  solidly  upon  the  silver;  in  several 
cases  the  most  salient  parts  of  the  decoration,  the  ovae,  for  instance, 
were  cast  and  worked  separately,  and  then  soldered  to  the  sur- 
face. It  is  hard  to  say  which  is  more  admirable  in  the  majority, 


FIG.  557- 

SERPEXT-ISIS 

(Museum,   Cairo). 

(Phot.  E.  Brugsch.) 


FIG.   558. 
TEHRA-COTTA  ISIS 

(Museum,   Cairo). 
(Phot.  E.  Brugsch.) 


FIG.  559.— CATAFALQUE  OF  AKHMIM  (Museum,  Cairo).     (Phot.  E.  Brugsch.) 

290 


THE  SAlTE  AGE 


the  mastery  of  the  technique  or  the  perfection  of  the  taste  dis- 
played in  the  composition.  When  we  compare  them  with  the 
treasure  of  Bosco  Reale,  we  cannot  but  think  that  the  Egyptian 
goldsmiths'  work  of  the  SaYte  and  Ptolemaic  periods  must  have 
sometimes  furnished  models  for  the  metal-workers  of  imperial 
Rome. 

The  jewelry  of   the  Bubastite  and  Tanite   dynasties  carries  on 
the   tradition  of  preceding   ages  almost  without  a  break,   in  the 


FIG.   560. 

NILE  IN  PIERCED 
BRONZE 


FIG.   56l.— FIGURE 

OF  PHARAOH  IN 

PIERCED    BRONZE 

(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


FIG.  562.— FIGURE 
OF  AN  OSIRIS-NILE 
IN  PIERCED  BRONZE 


form  of  bracelets,  rings,  earrings,  broad  necklaces  and  slender 
chains.  We  have  to  come  down  to  the  Psammetichan  kings  to  find 
new  designs  among  them.  I  have  said  elsewhere  that  the  ves- 
sels made  by  the  lonians  for  Necho  II.  probably  inspired  the 
craftsman  who  chased  the  clasps  of  the  necklace  in  the  Louvre 
(Fig.  564) ;  these  are  imitations  of  galleys,  with  their  flat  chamber, 
their  spur,  their  swan  or  goose-necked  poop.  The  little  amulets 
which  served  as  a  kind  of  magic  cuirass  to  the  mummies  of  the 
great  Sa'ite  dignitaries  entombed  at  Sakkarah,  owe  nothing  to 
the  foreigner,  and  the  original  types  to  be  met  with  among  them 
are  exclusively  national  —  the  tiny  gold  palm-trees,  with  scaly 
trunks  and  heavy  clusters  of  dates,  and  the  cynocephalous  figures 
worshipping  before  a  cartouche  crowned  with  feathers  (Fig.  565). 


ART  IN  EGYPT 


Some  are  composed  of  thin  flakes  of  gold  hammered  out  and 
soldered  together;  others  are  worked  upon  miniature  ingots  to 
which  the  accessories  have  been  added,  and  we  admire  the  dex- 
terity of  the  chasing.  The  seated  cat,  the  two  cynocephalous 
figures  standing  on  each  side  of  the  Osirian  fetich,  the  Isis  suck- 
ling Horus,  the  boat  of  Sokaris  resting  upon  its  cradle,  with  its 
crew  of  tiny  fish  and  falcons,  lose  nothing  on  close  examination 
through  a  microscope.  And  the  most  surprising  thing  is,  that 
the  patient  work  of  the  tool  has  not  produced  dryness  or  awk- 
wardness :  the  proportions  of  the  parts  are  calculated  as  skilfully 
as  those  of  the  faience  or  lapis-lazuli  figurines,  and  nowhere  do 
we  see  better  how  completely  the  Egyptians  had  mastered  human 

and  animal  forms  than 
in  these  infinitesimal  ob- 
jects. Some  of  the  figures 
of  Osiris,  of  Isis,  of  Thoth 
and  of  Amon  discovered 
by  Edgar  with  the  treasure 
of  Tukh-el-Karamus,  sug- 
gest decadence  only  by 
a  touch  of  affectation  and 
over -refinement.  Soon, 
however,  the  Greek  mod- 


FIG.   563. — SILVER  PLATE   OF  THMTTIS 
(Museum,  Cairo).     (Phot.  E.  Brugsch.) 


els,  so  free  and  so 
various  in  conception, 
spread  throughout  the 
country,  to  the  detriment  of  the  Pharaonic  types.  From  the 
accession  of  the  first  Caesars,  only  jewels  and  amulets  in 
the  Italian  or  Hellenic  style  were  sold  in  the  towns,  or 
bestowed  on  mummies.  These  were  twisted  serpents  with 
emerald  or  garnet  eyes,  for  bracelets,  keepsake-rings  for  hair, 
the  bezel  composed  of  a  massive  gold  plaque  or  a  cameo,  chains 
with  heavy  links,  earrings  in  the  form  of  bunches  of  grapes, 
crescents,  shells,  diadems  of  Gorgons'  heads  with  crinkled  hair, 
the  entire  jewel-case  of  the  Roman  or  Byzantine  lady.  The 
mania  for  western  jewelry  obtained  even  at  the  court  of  the  last 
Meroitic  Pharaohs,  and  the  jewels  Ferlini  stripped  from  the  mummy 
of  a  Candace  a  century  ago,  came  from  a  workshop  more 
than  half  Greek.  The  exotic  designs  executed  in  gold  for  the 
rich,  were  transferred  to  silver  for  the  benefit  of  the  poor,  and, 
interpreted  by  rustic  goldsmiths,  they  regained  something  of  the 
ancient  technique  under  their  hands.  In  the  more  modern  exam- 
ples of  the  treasure  of  Ben-ha  there  are  bracelets  with  checkered 

292 


THE  SAlTE  AGE 

ends  or  twists  which  recall  the  old  types  of  the  Ramessids :  if  we 
compare  them  with  contemporary  Egyptian  ornaments,  we  shall 
find  that  they  differ  from  them  only  in  insignificant  details,  and 
if  they  were  offered  for  sale  in  a  village  shop,  the  fellahin  would 
buy  them  without  suspecting  their  antiquity. 


FIG.  564.— CLASP  FOR  A  NECKLACE  IN  THE 
FORM  OF  A  BOAT  (The  Louvre). 


BIBLIOGRAPHY  TO  PART  III 

Sa'ite  Age.  —  The  Saiite  and  Graeco-Roman  ages  have  been  neglected  by  most  of  the 
writers  who  have  treated  the  history  of  art  in  Egypt  They  have  been  content  for  the 
most  part  to  describe  a  few  of  the  monuments,  without  attempting  to  sum  up  their  general 
character. 

Architecture.  —  A.  Tombs  —  For  the  tombs  of  the  Sa'ite  and  Ptolemaic  period,  see 
Barsanti-Maspero,  Fouilles  autour  de  la  pyramide  d'Ounas  (Extract  from  Annales  du  Service 
des  Antiquites,  vol.  I,  99),  p.  36-174,  8vo,  Cairo  1900-1909  (unfinished);  —  Rhind,  Thebes, 
its  Tombs  and  their  Tenants,  8vo,  London  1862.  —  B,  Temples.  —  For  the  manner  in 
which  the  style  of  columns  developed  from  the  time  of  the  Ramessids  to  thatiof  the  Saites, 
see  A.  Koester,  Die  agyptische  Pflanzensaule  der  Spatzeit,  in  the  Recueil  de  Travaux,  1901, 


The  Egyptian  Sudan,  1907,  8vo,  London,  vol.  I,  p.  337-435,  vol.  II,  p.  1—184;  the  same 
may  be  said  of  those  of  the  Sa'ite  period:  Cailliaud,  Voyage  a  I'Oasis  de  Thebes,  fol., 
Paris  182-186;  —  H.  Brugsch,  Reise  nach  der  grossen  Oase  El-Khargeh,  4to,  Leipzig  1878, 
VI-93  p.  and  XXVII  pi.;  —  Honroth-Rubensohn-Zucker,  Bericht  fiber  die  Ausgrabungen  aujF 
Elephantine,  in  the  Zeitschrift  fiir  agyptische  Sprache,  1907,  p.  14-61  and  9  pi.  —  The 
principal  temples  of  the  Graeco-Roman  age  have  been  the  subject  of  some  fairly  exhaustive 
monographs :  Philae :  G.  Benedite,  Le  Temple  de  Philce  (Memoires  de  la  Mission  permanente 
da  Catre,  vol.  XIII,  XVII),  4 to,  1  1895,  388  p.  and  vignettes,  II  1909,  356  p.;  Kom-Ombo: 
J.  de  Morgan,  Kom-Ombo  4 to  ,  Vienne,  2  vols.  1895-1909;  Edfu;  J.  Dumichen,  Bau- 
urkunde  der  Tempelanlagen  von  Edfu,  in  the  Zeitschrift  fiir  agyptische  Sprache,  1870,  p.  1-14, 
1871,  p.  25-32,  88-89,  105-111,  1872,  p.  33-42,  1873,  p.  109-130;  —  M.  de  Rochemonteix, 
Le  Temple  d'Edfou,  4 to  (Memoires  de  la  Mission  francaise,  vol.  X-XI),  Vienne,  1892-1899 

293 


(unfinished);  —  E.  Chassinat  et  Pieron,  Le Mammisi  d'Edfou  (Memoires  de  I'lnstitut  francais 
d'Archeologie.  vol.  XVI),  4 to,  Cairo,  1909,  208  p.  and  LII  pi.;  le  Kasr-el-Agouz,  a  Thebes: 
D.  Mallet,  Le  Kasr-el-Agouz  (Memoires  de  I'lnstitut  francais  d'Archeologie,  vol.  XI),  4to, 
Cairo  1909,  103  p.  and  I  pi.;  the  temple  of  Opet  at  Karnak:  —  M.  de  Rochemontei;, 
CEuvres  diverses.  8vo,  Paris  1894,  p.  167-318,  and  pi.  I-XVI;  the  temple  of  the  Theban  Ptah, 
at  Karnak:  G.  Legraiu,  Le  Temple  de  Ptah-Ris-anbouf ,  a  Thebes,  in  the  Annales  du 
Service  des  Antiquites,  vol.  Ill,  p.  38-56,  97-115;  the  temple  of  Der-el-Medinet ,  in  the 
Commission  d'Egypte,  Ant.  vol.  II,  p.  317-340  and  plates,  vol.  II,  pi.  34-37;  —  the  temple 
of  Denderah :  J.  Dumichen ,  Baageschichte  der  Tempelanlagen  von  Denderah ,  4  to ,  Leip- 
zig, 1865,  46  p.  and  XK  pi.;  —  Mariette,  Denderah,  text  4 to,  Paris  1875,  VI-347  p., 
atlas  fol.,  Paris  1873,  I-IV,  Supplement  1874,  9  pi.  —  For  the  temples  of  Nubia,  see 
Gau,  Les  Monuments  de  la  Nubie,  fol.,  Paris,  Stuttgart,  1823,  pi.  1-11-15-26,  33-41;  — 
Maspero-Barsanti ,  Les  Temples  immerges  de  la  Nubie,  Rapports,  4to,  Cairo,  1909-1911, 
XX 111-21 5  p.  and  CLIX  plates,  with  plans  in  addition. 

Sculpture.  —  For  tlie  general  character  of  sculpture  in  SaTte  and  Grace-Roman  times 
see  Maspero,  Le  Musee  egyptien,  4 to,  Cairo,  1906,  vol.  II,  p.  74-92  and  pi.  XXXII-XLII, 
—  Mariette,  Album  du  Musee  de  Boulag,  fol,  Cairo,  1872,  pi.  33-35;  —  Fr.  W.  de  Hissing, 
Denkmdler  dgyptischer  Skulptur,  fol.,  Munich,  1906-1911,  pi.  60-75,  98-119,  and  the  corres- 
ponding portions  of  the  text;  —  L.  Borchardt,  Kunstwerke  aus  dem  agyptischen  Museum 
zu  Kairo,  fol.,  Cairo,  1908,  pi.  16,  24-30,  48  and  p.  8-9,  13-14,  19;  —  C.  C.  Edgar, 
Sculptor's  Studies  and  unfinished  Works  (Catalogue  general  du  Musee  du  Caire),  4to, 
Cairo,  1906,  XII-91  p.  and  43  pi.  Various  questions  of  detail  and  several  isolated  monu- 
ments have  been  studied  by  Gourlay-Newberry,  Mentu-em-hat,  in  the  Recueil  de  Travaux, 
18%,  vol.  XX.  p.  188-192;  —  G.  Benedite,  Une  Tete  de  Statue  royale,  in  the  Gazette  des 
Beaux-Arts,  1897,  vol.  XVIII,  p.  35-42;  —  W.  Golenischeff ,  Eine  neue  Darstellung  des 


1895,  vol.  XVII,  p.  105-113,  with  three  plates;  —  C.  C.  Edgar,  Remarks  on  Egyptian 
Sculpture  Models,  in  the  Recueil  de  Travaux,  1905,  vol.  XXVII,  p.  137-150;  —  Maspero, 
la  Vache  de  Deir-el-Bahari ,  in  the  Revue  de  I  Art  ancien  et  moderne,  1907,  vol.  XXII, 
p.  5-18,  with  a  photogravure  plate. 

Painting  and  the  minor  Arts.  —  This  is  the  section  in  art-history  in  which  the  biblio- 
graphy is  most  scanty.  A.  Painting.  The  only  general  work  is:  C.  C.  Edgar,  Grceco- 
Egyptian  Coffins,  Masks  and  Portraits  (Catalogue  general  du  Musee  du  Caire,  4 to, 
Cairo,  1895,  XIX-136  p.  and  XL VIII  pi.;  for  details  see  Maspero,  Melanges  de  Mythologie 
et  d'Archeologie  egyptiennes,  8vo,  Paris,  vol.  IV,  p.  241-248  and  1  pi.;  —  C.  Watzinger, 
Griechische  Holzsarkophagen  aus  der  Zeit  Alexanders  des  Grossen ,  Abusier  III  (Wissen- 
schaftliche  Veroffentlichungen  der  Deutschen  Orient-Gesellschaft ,  vol.  VI),  4  to  Leipzig, 
1905,  96  p.  and  4  pi.;  —  H.  Schafer,  Priestergraber  und  andere  Grabfunde  vom  Ende  des 
alien  Reiches  bis  zur  griechischen  Zeit  vom  Totentempel  des  Ne-user-Re  (Wissenschaftliche 
Veroffentlichungen  der  Deutschen  Orient-Gesellschaft,  vol.  VIII),  4 to,  Leipzig,  1908,  VIII-185 
p.  and  13  pi.  —  B.  Work  in  bronze.  —  Only  one  general  treatise:  Edgar,  Greek  Bronzes 
(Catalogue  general  du  Musee  du  Caire),  4 to,  Cairo,  1904,  XI-99  p.  and  XIX  pi.;  various 
studies  on  points  of  detail:  G.  Daressy,  Une  Trouvaille  de  bronzes  a  Mitrahineh,  in  the 
Annales  du- Service  des  Antiquites.  1902,  vol.  Ill,  p.  139-150  and  3  pi.;  Statuette  grotesque 
egyptienne,  in  the  Annales,  1903,  vol.  IV,  p.  124-125;  —  G.  Benedite,  Une  Statuette  de  reine 
de  la  Dynastic  Bubastite  au  Musee  du  Louvre,  in  the  Gazette  des  Beaux-Arts,  1896,  vol.  XV. 
p.  447-485  and  1  pi.;  Un  Guerrier  Libyen,  figure  egyptienne  en  bronze  incruste  d'argent, 
conservee  au  Musee  du  Louvre,  in  the  Memoires  et  Monuments  de  la  fondation  Plot, 
1902,  vol.  IX,  p.  123-133;  —  t.  Chassinat,  Une  Statuette  en  bronze  de  la  reine  Karomama, 
in  the  Memoires  et  Monuments  de  la  fondation  Plot,  1897,  vol.  IV,  p.  15-25  and  pi.  Ill;  — 
G.  Maspero,  Melanges  de  Mythologie  et  d'Archeologie  egyptiennes,  8vo  Paris,  vol.  IV, 
p.  259 — 266  and  2  pi.  and  Sur  une  chatte  de  bronze  egyptienne,  in  the  Revue  de  I' Art 
ancien  et  moderne,  1902,  vol.  XI,  p.  377-380;  —  Mariette,  Monuments  divers,  folio,  Paris, 
pi.  41 ;  Album  du  Musee  de  Boulaq,  pi.  5,  9.  —  C.  Ceramics.  —  Two  collections  of  Egypto- 
Grecian  figurines,  that  of  C.  C.  Edgar,  Greek  Moulds  (Catalogue  general  du  Musee  du 
Caire),  4 to,  Cairo  1903  XVII-89  p.  and  XXXIII  pi.,  and  that  of  Valdemar  Schmidt,  De 
Grcesk-Agyptiske  Terrakote  i  Ny  Carberg  Glyptothek,  1911,  Copenhagen  94,  p.  and 
LXIII-VII  pi.;  —  For  certain  isolated  monuments  cfr.  G.  Legrain,  Sur  un  groupe  d'Amon 
et  d'Ameniritis  I.,  in  the  Recueil  de  Travaux,  1909,  vol.  XXXI,  p.  139-142;  —  On  porcelain 
and  pottery  in  general  down  to  the  Graco-Roman  period  cf.  Fr.  W.  de  Bissing,  Fayence- 
gefasse  (Catalogue  general  du  Musee  du  Caire),  4 to,  Vienna,  1902,  XXI-114  p.  — 

294 


THE  SAlTE  AGE 


D.  Joinery.  —  For  carved  and  painted  wooden  catafalques,  see  Rhind,  Thebes,  its  Tombs 
and  their  Tenants,  8vo,  London  1862,  p.  111-112  and  frontispiece.  —  Maspero,  Archeologie 
egyptienne,  1st  ed.,  1888,  p.  278-279  and  fig.  256-257.  —  E.  Goldsmith's  Work  and  Jewelry. 


and  their  Tenants,  8vo,  London  1862,  p.  111-112  and  frontispiece.  —  Maspero,  Archeologie 
egyptienne,  l»t  ed.,  1888,  p.  278-279  and  fig.  256-257.  —  E.  Goldsmith's  Work  and  Jewelry. 
In  addition  to  the  two  works  by  Vernier  quoted  above,  see,  for  the  treasure  of  Tukh-el- 


Karamus:  Maspero,  Causeries  d'Egypte,  8vo,  Paris,  1907,  p.  305-310;  —  C.  C.  Edgar, 
Report  on  an  Excavation  at  Toukh  el-Garamous,  in  the  Annales  du  Service  des  Antiquites, 
1906,  vol.  VII,  p.  205-212 ;  for  Saite  goldsmith's  work ,  Maspero ,  Lettre  sur  une  trouvaille 
de  bijoux  egyptiens  faite  a  Sakkarah,  in  the  Revue  de  I' Art  ancien  et  moderne,  1900, 
vol.  VIII,  p.  353-358 ;  —  for  Meroitie  goldsmith's  work,  Schaf er-M611er-Schubart,  j&gyptische 
Goldschmiedearbeiten  (in  the  Mitteilungen  aus  der  agyptischen  Sammlung) ,  II,  8vo  Berlin, 
1910,  243  p.  and  36  pi.,  where  also  we  shall  find  the  jewels  and  goldsmith's  work  of 
earlier  ages  owned  by  the  Berlin  Museum. 


FIG.  565. — GOLDEN  AMULETS   OF  THE  SAITE  PERIOD   (Museum,  Cairo). 
(Phot.  E.  Brugsch.) 


295 


CONCLUSION 

SUCH,  in  its  main  lines,  is  the  history  of  Egyptian  Art.  Its 
first  pages,  those  which  deal  with  its  origin  are  completely 
lacking ;  and  if  several  chapters  of  its  more  recent  ages  have  been 
reconstituted,  others  are  still  full  of  lacunae,  or  break  off  ab- 
ruptly. We  know  not  how  long  it  lasted,  but  more  than  forty 
centuries  intervene  between  the  moment  when  it  begins  for  us 
and  that  at  which  it  ends.  Up  to  the  present,  this  is  the  longest 
period  through  which  it  has  been  given  to  us  to  follow  more  or 
less  continuously  the  evolution  of  one  of  the  great  artistic  nations 
of  antiquity. 

Is  it  possible  for  us  to  discern  already  some  of  the  causes 
which  made  Egyptian  art  what  it  was,  and  preserved  its  character? 
I  have  shown  more  than  once  in  these  pages  that  it  did  not  seek 
to  create  or  to  record  beauty  for  its  own  sake.  It  was  originally 
one  of  the  means  employed  by  religion  to  secure  eternal  life 
and  happiness  for  the  dwellers  upon  earth.  This  end  attained, 
if  beauty  resulted  in  addition,  it  was  received  with  joy,  though 
it  was  by  no  means  looked  upon  as  indispensable  to  the  per- 
fection of  the  work  undertaken;  no  effort  was  made  to  ensure 
it,  if  such  effort  could  be  held  in  any  way  to  interfere  with  the 
desired  consummation.  As  art  owed  to  the  gods  and  the  dead 
dwellings  secure  against  destruction,  it  was  concerned  from  the 
beginning  to  choose  among  materials  and  forms  those  which  seem- 
ed to  it  best  fitted  to  secure  the  longevity  of  temples  and 
tombs.  It  therefore  invented,  from  purely  utilitarian  motives, 
that  prodigious  architecture,  the  colossal  masses  and  powerful 
lines  of  which  leave  on  the  mind  of  the  spectator  the  strongest 
impression  of  indestructibility  that  any  works  by  human  hands 
have  ever  produced.  And  since  bas-reliefs  and  statues  had  at 
first  no  reason  for  existence  beyond  that  of  affording  an  imper- 
ishable lodging  for  the  souls  of  divinities,  and  the  doubles  of 
the  incarnate,  art  in  the  beginning  aspired  only  to  express  faith- 
fully the  idea  of  divine  types  formed  by  the  people,  and  to  per- 
petuate in  stone,  in  metal,  or  in  wood,  the  features  of  persons 

296 


CONCLUSION 

whose  rank  or  fortune  had  secured  for  them  the  privilege  of 
immortality.  Soon,  however,  the  same  interested  motive  which 
had  induced  it  to  carve  faithful  portraits,  led  it  to  disregard 
this  exactitude  in  certain  points.  It  was,  of  course,  necessary 
that  the  doubles  should  find  their  fictitious  bodies  sufficiently 
like  their  actual  ones  to  feel  at  ease  in  them;  but  their  second 
existence  would  hardly  have  seemed  a  blessing  to  them,  had  they 
been  condemned  to  spend  it  with  limbs  weakened  by  all  the 
infirmities  of  age.  By  substituting  for  the  sickly  or  decrepit 
reality  the  figure  of  the  individual  as  he  was  in  his  youth  or  in 
the  vigour  of  his  maturity,  the  artist  conferred  on  him  more 
certainly  the  full  enjoyment  of  his  strength  and  faculties.  This 
is  why  there  are  so  few  statues  of  old  men  before  the  Sa'ite 
period;  even  when  a  centenarian  was  represented,  Amenophis, 
son  of  Hapu,  or  Rameses  II,  their  portraits  are  not  very  different 
to  what  they  must  have  been  in  their  youth.  And  if  we  go 
further  into  details,  was  it  not  an  analogous  scruple  which  caused 
the  rarity  of  nude  statues  ?  Nudity  was  a  sign  of  low  condition  •— • 
for  all  but  children,  and  persons  of  good  family,  inflicting  it  on 
their  statues,  would  have  risked  finding  themselves  confounded 
with  the  proletariat,  and  losing  caste  in  the  next  world.  If 
there  are  some  few  exceptions  to  this  rule,  it  was  because  some 
superior  interest  made  the  model's  singularity  of  advantage  to 
him.  The  Anisakha  at  Cairo  showed,  by  laying  aside  his  loin- 
cloth,  that  he  was  circumcised,  and  thus  gained  the  advantages  KK 
due  to  the  faithful  who  bore  this  mark  of  ritual  initiation ;  other- 
wise would  he  have  given  his  double  a  body  naked  as  that  of 
a  labourer,  with  loins  and  thighs  exposed? 

It  was  then  the  desire  for  utility  which  gave  Egyptian  sculpture 
that  combination  of  hieratic  idealism  and  realism  to  which  it 
owes  its  most  personal  charm.  This  was  not  without  disadvan- 
tages to  it,  since  it  robbed  it  of  some  part  of  its  liberty,  but 
what  it  gained,  if  it  did  not  entirely  compensate  for  this  loss, 
at  least  went  far  to  minimise  it.  The  case  was  very  different 
with  painting.  I  have  said  that  it  played  an  honourable  part  in 
the  civilisation  of  archaic  times,  whether  it  was  charged  with  the 
ornamentation  of  the  house  or  the  tomb,  or  whether  —  though 
of  this  we  have  no  certain  proof  as  yet  —  it  had  its  allotted 
task  in  palaces  and  temples.  It  had,  in  short,  this  advantage 
over  the  other  arts,  that  its  apparatus  was  simpler,  and  its  pro- 
cesses less  expensive.  Nevertheless,  as  the  fragility  of  its  methods 
left  the  gods  and  the  dead  ill  protected  against  supreme  disso- 
lution, it  yielded  to  sculpture  as  soon  as  the  latter  had  acquired 

297 


ART  IN  EGYPT 

facility  in  its  technique,  and  became  a  secondary  art  almost 
everywhere;  where  a  certainty  of  immortality  wajs  insisted  on,  it 
was  no  more  than  the  humble  servant  of  its  comrade.  When  it  was- 
laid  down  that  the  gods  could  only  become  imperishable  in  a  medium 
itself  imperishable,  it  could  not  serve  them  independently,  and  was 
reduced  to  being  a  mere  play  of  artificial  tones  without  any  form 
but  those  of  the  sculptured  reliefs  over  which  it  was  spread.  It 
succeeded  nevertheless  in  emancipating  itself  by  degrees  in  the 
tombs,  but  this  was  not  the  result  of  a  spontaneous  effort  of 
development ;  it  was  simply  because  the  conceptions  of  the  after- 
life were  modified  and  enlarged.  As  long  as  universal  belief 
tied  the  double  to  the  spot  where  the  corpse  rested,  care  for  its 
well-being  demanded  that  it  should  be  surrounded  by  indestruct- 
ible scenes ;  colour  remained  an  accessory,  and  it  was  only  toler- 
ated by  itself  in  places  where  the  nature  of  the  rock  was  recal- 
citrant to  the  chisel.  From  the  day  when  it  began  to  be  ima- 
gined that  the  soul,  no  longer  inhabiting  the  vault,  could  dispense 
with  an  eternal  decoration,  the  number  of  painted  hypogea  in- 
creased. It  might  have  been  supposed  that  the  painter,  left  to 
himself,  would  speedily  have  discovered  the  resources  of  his 
craft,  and  would  have  developed  them  in  directions  where  the 
sculptor  was  powerless  to  follow  him.  But  this  was  by  no  means 
the  case;  the  traditions  and  routines  to  which  it  had  been 
subjected  for  centuries  had  obtained  such  mastery  over  him,  that 
he  had  no  longer  the  energy  to  cast  them  off.  He  discarded 
some  of  the  stereotyped  forms  bequeathed  to  him  by  his  prede- 
cessors, he  treated  others  with  greater  breadth  and  liberty,  he 
intermingled  his  lines  more  harmoniously,  he  gained  greater  flexi- 
bility in  his  processes,  but  on  the  whole  he  remained  what  ages 
of  subjection  had  made  him.  He  utilised  all  the  ingenuity  and 
the  experience  of  centuries  to  copy  with  his  brush  the  silhouettes 
the  sculptor  had  cut  out  with  his  chisel,  and  persisted  in  filling 
them  in  artlessly  with  uniform  planes,  without  any  effort  to  model 
their  reliefs  by  combinations  of  shade  and  half-tones  scientifically 
opposed  or  graduated.  The  fatality  of  utilitarianism  continued 
to  enslave  him,  when  it  had  long  ceased  to  have  any  raison 
d'etre. 

It  will  be  readily  understood  that  a  people  among  whom  the 
manifestations  of  art  were  so  strictly  subordinated  to  the  material 
advantage  of  those  who  evoked  them,  was  very  little  concerned 
to  preserve  the  names  of  their  authors.  Thus  many  of  the 
masterpieces  which  are  anonymous  for  us  were  equally  so  for 
their  contemporaries.  The  temple  raised  at  Luxor  by  Amen- 


CONCLUSION 

ophis  III.  in  honour  of  his  father  Amon,  the  statue  which 
the  Sheikh-el-beled  had  hidden  in  his  tomb,  the  bas-relief  on 
which  Sesostris  struck  down  the  Libyan  heroes  —  in  all  these 
the  interest  of  the  person  they  immortalised  would  not  have 
tolerated  that  the  merit  of  their  execution  should  have  been  claimed 
in  any  degree  by  their  creators.  If  the  name  of  the  artist  had  been 
associated  with  that  of  the  master,  the  former  would  have  parti- 
cipated in  the  benefits  they  conferred,  and  by  so  much  would 
the  bliss  to  which  the  double  or  the  god  had  a  right  have  been 
diminished.  It  happened  sometimes,  when  the  artist  belonged  to 
the  household  of  a  great  personage,  that  the  master  would  asso- 
ciate him  by  special  favour  with  his  posthumous  destiny,  and 
we  owe  our  knowledge  of  one  or  two  sculptors  to  this  unusual 
condescension;  but  the  exception  is  one  of  those  which  prove 
the  rule.  In  a  general  way  we  shall  not  be  mistaken  if  we 
assume  that  the  principle  of  utility  forbade  all  those  who  prac- 
tised an  art  to  sign  their  works,  and  consequently  condemned 
them  to  oblivion.  We  should  like,  indeed,  to  know  what  they 
were  called,  what  was  their  native  city  or  their  condition  of  life, 
who  had  been  their  first  teachers,  and  by  what  efforts  those 
geniuses  who  made  the  plans  of  Der-el-Bahari  or  the  Hypostyle 
Hall ,  raised  the  Pyramid  of  Chephren ,  and  carved  the  Seated 
Scribe  in  the  Louvre,  the  Thothmes  III.  and  the  Amenartas  at 
Cairo,  the  Seti  I.  and  the  goddesses  of  Abydos,  outstripped  the 
crowd  of  their  competitors.  The  choice  that  fell  on  them  to 
undertake  these  great  tasks  proves  sufficiently  that  they  did  not 
pass  unnoticed  among  their  immediate  circle,  and  that  they 
enjoyed  in  their  day  the  reputation  of  being  the  most  skilful 
and  the  most  gifted  in  their  craft.  Fame  was  not  lacking  to  them, 
at  least  in  their  lifetime,  and  among  those  who  surrounded  them, 
but  when  their  generation  had  passed  away,  the  admiration  of 
the  new  races  was  poured  out  on  the  Pharaohs  or  the  rich  men 
who  had  employed  them ;  the  memory  of  the  bold  craftsman  who 
dared  to  design  and  execute  the  specs  of  Abu  Simbel  was  not 
handed  down  with  his  work  as  was  that  of  Ictinus  with  the 
Parthenon.  It  was  thus  that,  ignorant  of  the  ambition  of  immor- 
tality by  fame,  the  action  of  which  is  so  powerful  among  the 
moderns,  the  Egyptian  masters  were  for  the  most  part  content  to 
observe  conscientiously,  as  they  would  have  done  in  any  ordinary 
calling,  the  rules  which  the  teachings  of  their  predecessors  had 
assured  them  were  necessary  to  the  well-being  of  souls  human 
and  divine.  When  by  chance  any  were  born  whose  inventive 
minds  rebelled  against  the  half  technical,  half  religious  education 

299 


of  the  workshop,  their  efforts  towards  progress  or  reform  had 
no  serious  results.  Might  it  not  well  be  that  by  changing  some- 
thing in  the  recognised  processes,  they  would  compromise  the 
salvation  of  their  models?  In  doubt,  the  crowd  stood  aloof  from 
them,  and  held  prudently  to  the  old  customs;  I  have  described 
above  how  an  attempt  at  picturesque  painting  failed  in  the 
Theban  necropolis  under  the  Twenty-first  or  Twenty-second 
Dynasty,  and  yet  painting  was  of  all  the  arts  that  which  tended 
at  the  time  to  dissociate  itself  most  from  the  useful.  By  refusing 
thus  to  modify  the  themes  and  types  of  an  earlier  age  save  in 
details,  Egypt  gave  her  art  that  character  of  uniformity  which 
strikes  us.  The  personal  temperament  of  the  individual  is  reveal- 
ed only  by  almost  imperceptible  shades  of  handling,  and  the 
majority  of  visitors  carry  away  with  them  from  museums  and 
ruins  the  sense  of  a  collective  impersonality,  slightly  varied  here 
and  there  according  to  time  and  place  by  the  greater  or  lesser 
degree  of  skill  in  the  executant.  They  do  not  understand  what 
an  amount  of  natural  talent  and  acquired  science  the  unknown 
authors  of  great  temples  and  fine  sculpture  expended,  to  make 
themselves  more  than  mere  skilful  craftsmen. 

I  am  far  from  asserting  that  religious  utility  was  the  sole 
consideration  here;  it  was  the  main  one,  that  which  after  having 
inspired  the  dawning  arts,  governed  their  developments  to  the 
last,  and  had  I  leisure,  I  should  like  to  show  how  its  influence 
appears  in  every  direction,  not  only  in  the  major  arts  but  in 
industry.  Not  that  the  Egyptians  were  conscious  of  it  each  time 
that  it  determined  a  momentary  progress  or  decadence  among 
them,  but  that,  by  instinct  and  by  routine,  they  followed  the 
incline  on  which  they  had  been  launched  at  the  beginning  of 
their  history.  Other  antique  nations  were  influenced  by  it  as 
well  as  they,  and  throughout  the  world,  in  Assyria,  in  Chaldaea, 
in  Asia  Minor,  in  Syria,  architecture,  sculpture,  and  painting 
were,  as  in  Egypt,  means  for  ensuring  to  gods  and  men  together 
with  eternal  life,  prosperity  before  and  after  death ;  but  whereas 
in  Greece  the  desire  for  pure  beauty  soon  triumphed ,  Egypt, 
falling  behind  more  and  more  in  her  archaic  methods  of  thought, 
ended  by  becoming  incapable,  I  will  not  say  of  adopting  the 
nobler  conceptions  that  were  growing  up  around  her,  but  of 
realising  their  value.  The  divorce  between  her  inveterate  routine 
and  the  enterprising  spirit  of  the  new  world  was  so  profound 
when  Christianity  rose  before  her,  that  she  could  offer  it  nothing 
it  could  adapt  to  its  needs,  even  with  many  alterations  of  artistic 
expression.  The  western  arts  lent  it  their  basilicas,  their  statues, 

300 


CONCLUSION 

their  bas-reliefs,  their  frescoes,  and  it  readily  evolved  from  these 
forms  suitable  to  its  ideas  and  its  beliefs.  But  as  for  Egypt, 
how  could  the  Christ  find  an  abode  in  those  dark  and  massive 
temples,  where  every  line,  every  chamber,  every  decorative  motive, 
every  accessory  of  furniture,  would  have  recalled  dogmas  and 
practices  he  reprobated?  How  could  his  priests  and  people 
have  metamorphosed  into  images  of  their  saints  and  prophets, 
or  reconciled  with  their  hopes  of  immortality  those  bestial  or 
half-human  statues,  and  those  pictures,  the  elements  and  compo- 
sition of  which  they  declared  to  have  been  governed  throughout 
antiquity  by  the  most  impure  of  demons?  There  was  in  a  Ju- 
piter enough  beauty,  independent  of  all  religious  faith,  to  enable 
an  artist,  stripping  it  of  its  heathen  trappings,  to  make  it  the 
Christian  conception  of  the  one  God;  but  to  what  person  or 
what  incarnation  would  it  have  been  possible  to  assimilate  the 
cold,  inanimate  figures  of  an  Amon,  a  Ptah,  an  Osiris,  even 
when  their  characteristic  insignia  and  attitudes  were  taken  from 
them  ?  The  demands  of  utility,  which  had  made  them  what  they 
were,  had  riveted  them  by  bonds  so  numerous  and  so  solid  to 
the  dying  creed,  that  they  had  no  choice  but  to  depart  with  it. 
The  art  of  Egypt,  like  its  literature,  its  science,  its  current  civi- 
lisation, was  one  with  its  religion;  the  blow  which  struck  at 
the  one  struck  at  the  others  and  crushed  them. 

It  died  completely,  and  the  world  lost  sight  of  it;  for  some 
fifteen  centuries,  nothing  was  known  of  it,  save  that  classical 
writers  described  it  as  marvellous,  and  that  rare  travellers  had 
seen  some  gigantic  remains  of  it  in  the  neighbourhood  of  Cairo, 
or  in  the  deserts  of  the  Thebaid.  The  draughtsmen  and  scholars 
of  the  French  expedition  having  brought  it  to  light  again  some 
hundred  years  ago,  it  has  reconquered  that  place  in  the  esteem 
of  the  present  generation  from  which  the  neglect  of  former  ages 
had  dethroned  it.  To  tell  the  truth,  it  is  not,  and  I  fear  it  never 
will  be,  one  of  those  arts  which  fire  the  student  at  first  sight. 
Some  of  the  works  it  has  bequeathed  us  command  instant  admir- 
ation, and  a  first  glance  suffices  to  enable  us  to  understand 
them  as  easily  as  the  finest  works  of  Greece  or  Rome.  The 
merits  of  the  rest  are  not  at  once  apparent;  we  can  only  grasp 
them  after  patient  study,  and  they  must  be  pointed  out  to  people 
who  have  not  time  to  discover  them  for  themselves.  Is  not  this 
the  same  in  literature,  and  are  there  not  poets,  Pindar,  for 
instance,  whose  verses  are  the  delight  only  of  a  chosen  few? 
Their  beauty  is  as  real  as  when  thev  were  first  composed,  but 
the  long  commentary  they  require  before  yielding  their  charm 

301 


ART  IN  EGYPT 

has  obscured  it  to  the  eyes  of  the  multitude.  Artists  and  men 
of  letters,  who  were  disconcerted  at  first  by  Egyptian  sculpture 
and  painting,  have  recovered  from  their  stupefaction;  they  take 
an  extreme  pleasure  in  appreciating  them,  and  it  is  primarily  of 
them  I  have  been  thinking  in  writing  these  pages.  It  is  not  to 
be  supposed  that  professional  Egyptologists  and  critics  will  endorse 
without  considerable  reserve  all  the  opinions  and  judgments  they 
will  have  read  in  them;  but  should  they  reject  them  all,  they 
will  nevertheless,  I  hope,  have  gained  something.  For  is  not 
this,  in  fact,  the  first  time  that  an  attempt  has  been  made  to 
relate  in  a  consecutive  fashion  the  history  of  an  art  as  extinct 
as  the  races  of  monsters  we  find  imbedded  in  the  lower  strata 
of  our  globe?  By  studying  closely  the  vicissitudes  of  its  existence, 
its  hesitations,  its  progress,  its  failures,  its  recoveries,  and  its 
long  agony,  they  will  learn  to  recognise  more  precisely  than 
they  have  hitherto  done  some  of  the  principles  which  regulate 
the  incubation,  the  birth,  the  efflorescence  and  the  death  of  the 
arts  of  other  nations. 


302 


INDEX 


INDEX 


References  to  Illustrations  are  indicated  by  an  ' 


A. 

Akoris,  Temple  of,  257. 

Alexander  the  Great,  228. 

Aah-hetep,  204. 

Alexander  II.,  223,  259. 

Aahmes,  148,  170*. 

Alexander  Aigos,  239,  248*.  255. 

Aahmes-Nefert-ari,  179. 

Alexandria,  254,  257,  265,  266, 

Aat-tcha-Munt.    see    Medinet- 

273.  275,  284. 

Habu. 

Alexandria  Museum,  264. 

Abai,  265. 

Amada,     132-135;     Facade    of 

Abahuda,  143. 

Temple  of,  131*. 

Abu'I-hol,  192. 

Amasis,  129,  204,  237,  240,  289. 

Abuni,  the  soldier,  20,  20*. 

Amelineau,  7. 

Abu-Roash,  38,  83. 

Amenemhat  III.,  116,  118,  120; 

Abu-Simbel,  144,  189,  190,  191, 

Statue  of,    115*;  Statuette 

208,  299;   Esplanade  before 

of,  115*. 

great  Speos  at,    139*;    Fa- 

Amenartas,  223,  238*,  248,  299; 

jade  of  great  Speos  at,  1  39*; 

and  Amon,  283,  285*. 

Fajade  of   little   Speos   at. 

Amenemhat,  100,  110.  124,  133, 

138*. 

206,  254. 

Abusir,  45,  48,  57,  62. 

Ameni,  101. 

Abydos,  3,  4,  6,  7,  19,60,77,91, 

Amenophis  the  Scribe,  168*.  176. 

99,  101.  102,  110,  112,  114, 

Amenophis,  son  of  Hapu,  167*, 

153,154.  162,  187-190,299; 

176,  240,  254.  297;  Colos- 

School of,  24,  105. 

sal  Statue  of,  247*. 

j£dicula  on  Terrace  of  Temple  of 

Amenophis  I.,  139,  150;   Statue 

Hathor,  219*. 

of,  160*. 

/Egean  peoples,  125. 

Amenophis  II.,   150,   172,   173, 

Afai,  Hypogeum  of,  37. 

180,  199,  204;  and  the  Ser- 

African  mud    architecture,    28, 

pent      Maritsakro,       158*; 

28*. 

Wooden  Statuette  of,  195*. 

Agathodemons,  286. 

Amenophis  III..   130,   131,   134, 

Aha,  Tablet  of,  3. 

140,    150,    156,    173.    176, 

Ahmessids,  139,  163,  169,   171, 

180,  181,  184,200,219,239, 

176,  177.  194,  218,  262. 

240,    278,  299;   Statue    of 

Akhmim,  37,  60,  240,  275,  285. 

Amenophis  III.  in  Assyrian 

305 

Dress,  166*;  and  Queen 
Thi,  166*. 

Amenophis  IV.,  139,  181,  184. 
195,  283;  Cast  of  Head  of, 
174*;  Head  of  Canopic  Jar 
of,  172*,  181;  and  his  Queen, 
173*;  with  the  Queen  and 
their  Children,  174*.  183; 
Statuette  of,  172*. 

Amentit,  57. 

Amon  or  Amen,  138,  151,  167, 
184-187,201,218,219.239, 
246,  247,  249,  256,  265.  268, 
279,  292,  299,  301 ;  Bust  of. 
175*;  Temple  of,  259;  with 
Amenartas,  283,  285*;  with 
Candace,  254*;  with  Mut, 
178*. 

Amon-Ra,  114. 

Amorrhaeans,  165. 

Amulets,  198.  281,  291,  292; 
golden,  295*. 

Animals  in  Egyptian  Art,  173- 
175,  278. 

Anisakha,  84,  304. 

Ankh-Hapi,  266. 

Ankhmara,  7 1 ;  Dancers  in 
Tomb  of,  69*. 

Ankhnasnufiabri,  238*,  248. 

Anonymity  of  Egyptian  artists, 
66,  298-300. 

Anonymous  Statue,  1 12*. 

Ant  feus  and  Isis,  265,  272*. 

Antef  dynasty,  113.  114. 


INDEX 


Antef,  Prince.  113. 

Antonines,  the,  55,  274. 

Antoninus  Pius,  231. 

Anubis,  152.  267,  279,  286; 
Bronze  Statuette  of,  284*. 

Apis,  267,  279;  Statuette  of, 
283*. 

Apries,  278,  283;  Sphinx  of,  276. 

Arabs,  289.  • 

Archaic  Statuette,  78*. 

Architecture,  125. 

Argo,  Isle  of,  256. 

Arm  of  a  Chair  in  form  of  a  fe- 
line Animal,  195*. 

Asasif,  148,  2JS5. 

Ashu,  57. 

Asia,  206;  Minor,  125,  300. 

Assuan.  27.  230,  239,  240. 

Assyria,  206.  289,  300. 

Asyut,  254. 

Athens  Museum,  277. 

Atlantes,  155;  at  Medinet-Habu, 

124*;    of   the  Ramesseum, 

127*. 
Atlas  of  Thothmes  I..  Head  of, 

161*,  170. 

Atonor  Aten,  182,  184. 
Augustus.  241,  242.  261,  262. 


B. 

Bakers,  88*. 

Barsanti,  278. 

Bas-relief.  179-181;  at  Abusir, 
57;  in  Alexandria  Museum, 
264;  from  Chapel  of  Sahu- 
Ra.  59*;  of  the  Cow  Ha- 
thor,  165*;  at  Dakkeh,  270; 
at  Edfu.  265*;  of  Macrinus 
and  Diadumenianus,  261, 
268*;  of  Maharrakah,  270*; 
Memphite,  190*;  in  the 
Memnonium  of  Seti  I.  at 
Abydos,  180*;  of  Psamme- 
tiknufisashmu,  263,  266, 
271*.  272*;  of  Seneferu. 
58*;  of  Seti  I.,  179*;  at 
Sinai,  22*;  in  the  Temple  of 
Ombos,  267*;  in  the  Tem- 
ple of  the  Theban  Ptah, 
264*;  in  the  Tomb  of  Mek- 
hu,  61*;  from  Tomb  of 
Menthuhetep,  111*;  of  the 
time  of  Domitian,  267*;  of 
Zanufi.  271*. 

Bast  (goddess),  278;  temple  of, 
280. 


Battle  of  Kadesh,  in  Ramesseum, 

155*.  165. 
Beh-bet,  257. 
Beket,  105,  109. 
Ben-ha.  292. 
Beni-hasan,   101,  103.  106,   108, 

110,  182. 

Beni-Mohammed,  60. 
Berlin  Museum,   87,   201,   247, 

252. 

Bersheh.  101,  106. 
Bes(god),  202,  219,  232. 
Betchau,  see  Neter-baiu. 
Bet-el-  Wali,    113,    190;     Hemi- 

speos  of,  137*. 
Bet-Khallaf,  7,  38,  39;   Plan  of 

Tomb  at,  8*;  Tomb.  8*. 
Birket-el-Kurun,  96. 
Biban-el-Muluk,  149,  153. 
Blue  enamel  Head,  284*. 
Blunted  Pyramid,  100. 
Boat,  sacred  or  solar,  46. 
Boat  of  Queen  Aah-hetep,  207. 
Book  of  the  Dead,  66,  216. 
Bosco  Reale,  291. 
Box  for  Cosmetics,  197*.  198*. 
Bracelets  from  Abydos,  2*. 
Bracelets  of  Rameses  II.,  209*. 
Brawl  on  the  Water,  68*. 
Brewer,  71*. 
Brick  Boat  of  Ra-en-user.  46, 

50*. 

Bronze  Statues,  Two,  75*. 
Brooch  of  Khamuasit,  210*. 
Brussels  Museum,  80. 
Bubastis,  118, 129,202,214,217, 

240,278-281. 

Bull  lassoed  by  the  King,  56*. 
Bust  of  a  Statue,  246. 
Byzantines,  289. 


c. 

Csesar,  215,  238,  242.  256,  259, 
261,  265,  288;  Statue  re- 
carved  as  a,  255*. 

Cairo,  Museum  of,  22,  51,  71,  79, 
81,  86,  87,  88,  90,  92,  118, 
119,170,  173,  176,  181,185, 
187,  192,  194,  199,201,206, 
247-250,  254,  256,  258,  265, 
266,  268,  276,  278,  279,  283, 
285,  288,  297.  299. 

Camel,  terra-cotta,  286*. 

Candace,  Queen,  256,  292. 

Canopic  jars,  181. 

306 


Capitals,  97.  103.  232;  bell- 
shaped,  129;  lotus,  129; 
palm-leaf.  129;  Hathor,  97. 
224*.  228,  237;  irregular, 
129. 

Caracalla,  256. 

Caramania,  77,  273. 

Cartonnages,  270-273. 

Cat,  Bronze,  282*. 

Cat  watching  for  Prey,  106*. 

Catafalque,  287,  288;  of  Akh- 
mim,  290. 

Cataracts,  143,  215. 

Chairs,  200;  "Empire"  Chair, 
196*. 

Chaldaea,  300. 

Champollion,  103.  190,  276. 

Chapel  of  Apis,  146;  of  Cheops, 
44;  of  Chephren,  plan,  48*; 
of  the  Cow  Hathor,  164*; 
of  Mekhu,  38,  43*;  of  Ra- 
en-user,  49*;  of  Sabni,  38, 
43*;  of  Sahu-Ra,  plan,  49*. 

Chapels,  sepulchral,  146,  147. 

Chassinat,  83,  201. 

Cheops,  26,  41,  62,  77;  in  ivory, 
74*. 

Chephren.  47;  the  aged,  70*; 
alabaster,  79*;  the  Great, 
79*;  Pyramid  of,  41.  299; 
Statue  of,  81-84. 

Chinese  artists,  69. 

Christianity  destructive  to  Egyp- 
tian Art,  246,  276,  286,  301. 

Clandius,  241. 

Clasp  for  Necklace  in  form  of  a 
Boat,  293*. 

Cleopatra,  258. 

Coffin  of  Akhmim,  278*;  of  Ta- 
taharsiasi,  275;  of  Wardan, 

275. 

Coffins,  199,271,272. 
Column  of  the  Ambulatory  at 

Karnak,    129*;     of    Ra-en- 
user,  54*;   of  Shepses-Ptah, 

53*. 
Colossi,  191,  192,256,257,276; 

at  Abu  Simbel,    185*.   191; 

of  the  Ramesseum,  192;  of 

Luxor,  132*. 
Colossus     of     Argo,     Crowned, 

254*;  of  the  Roman  period, 

256-*. 
Columns,    127,    235-238;     loti- 

form,  49;    palm  if  or  in,  53. 
Commodus,  256. 
Concert,  90*. 
Cook,  87. 


INDEX 


Coptic  Monastery,  151. 
Coptos,  113. 
Copts.  230. 
Cow  of  Der-et-Bahari,  165*,  174, 
175;     of    Sakkarah,    244*. 
250,   251;    turning   to   her 
Calf,  66*.     See  also  Hathor. 
Crouching  Figure  holding  a  Di- 

Diadumenianus,  261,  268*. 
Didufriya,  Statue  of,  81*,  83. 
Dimeh,  274. 
Dish  of  Hatiyai,  202*. 
Dodecaschcenus,  242. 
Dog,  terra-cotta,  289*. 
Domitian,  231.261. 
Double  or  Ka,  10,  17,  20,  31,  32. 
37.  38,  41,  73,  77,  86.  89, 

Ethiopian    Pyramid,    plan    and 
elevation,  235. 
Ethiopian  Statues  of  Souls,  255. 
Euergetes  II.,  261. 

F. 

D 

253.  254. 
Doui/e-Statues,  168,  200,  201, 
250. 
Drah-Abu'1-Nekkah,  148. 

False   Doors,   31-33;    of   Atoti, 
37*;    of  Mereruka,  37*;    of 
Neferseshem-Ptah,  38*. 
False  Pyramid  at  MedOm,  44*. 
Fay  urn  the,  36,   100,   110,  124, 

Dabud,  Temple  of  ,  231*.  242. 
Dagger  of  Amasis,  206*. 
Dahshur,  26,  33,  38,  41,  62.  100, 

Dwarf  of  Gizeh,  82*. 
Dwarf  Khnemu-hetep,  82*. 

261,273-275,285. 
Female  Figure,  58*. 
Ferlini,  292. 
Figure  of  a  Queen  or  Goddess  at 

111,  120,  122. 
Dakhleh,  Temple  of,  216*.  230. 
Dakkeh,  Temple  of,  232*.  242, 
262. 
Daninos,  276,  288. 
Daressy,  202. 
Darius  I.,  223. 
Daughter  of  Rameses  II.,  184*. 
Daughters    of   Amenophis    IV., 
153*.  163. 
Decadence  of  Egyptian  Art,  259. 
Decoration  of  Interiors.  50-56; 
in  Palace  at  Medinet-Habu, 

E. 

Earring  of  Rameses  XII.,  208*. 
Earrings  of  Seti  I.,  208*. 
Edfu,  15.  224,  227.  228,  233,  236, 
238.   240,   242,   258,    261; 
Nomes    bringing    offerings, 
14*. 
Edgar,  292. 
Edinburgh  Museum,  287. 
Egypt,  Lower,  53,  272;   Middle, 
96,    100,    101,    240.    281; 
Upper,  36,  38.  53,  96,  239, 

Figurines  of  Animals,  3*. 
Finished  Figure  of  a  King,  263*. 
Fortress      of     Kom  -  el  -  Ahmar 
(plan),  4*. 
Fortresses,  159,  160. 
France,  2,  279. 
French  Expedition,  230,  301. 
Frontality,  Law  of,  73. 
Furniture,  286,  287. 

G. 

51-73;     in    Tomb-chamber 
of   Rameses   V.,    141*;     in 
Tomb  of  Kom-es-Shugafa, 
273*. 
Decorative  Art  in  Egypt,  162- 
166. 
Delta,  19,  24,  36.  162,  195,  213, 
220,240  256,251,265,278, 

274. 
Egyptian   Fleet,   at  Der-el-Ba- 
hari, 170. 
El-Amarna,  158.  163,  181-185. 
El-Armarna,  195;  Wall  in  Tomb 
at,  173*. 
Elephant,  terra-cotta,  286*. 
Elephantine,  60,   129,   130,   132, 

Garf-Husen,  138*.  144. 
Garstang,  7;   Hemi-speos. 
Gebel-Barkal,  173. 
Gebel-Silsileh,  Chapel  at,   135*. 
142. 
Geese  of  Medum,  51,55*. 
German  excavations,  45. 
Gizeh,  33,  38,  45,  62,  80,  114. 

281,285;  School  of  the,  54. 
Denderah,  36,  60,  1  13,  224,  225, 
228,229.232-234,238,261. 

194,  237;    Chapels   at.   38; 
East  Temple   at  (section), 
130*;  a  Prince  of,  61*. 

216;    corner  of  Necropolis 
at,  30*. 
Goblet  of  Zakazik,  203*. 

Dendur,  242,  243. 
Der-el-Bahari,  99,  113,  114,  129, 
148,  149,  151-153,  174,  179, 
231,    237,    299;    Esplanade 
at,  142*;   General  View  of, 
142*. 
Der-el-Medinet,    148.  230,   240; 
Facade  of  Temple  at,  221*; 
Pronaos  of  Temple  at,  222*. 
DSr-el-Melak,  37. 
Derr,  144,  189,  219. 

El-Kab,  36.  131,  194. 
EI-Kalaa,  230. 
Ellahun,  100. 
Enamelled    Chamber    of    King 
Zoser,  92*. 
Enneas,  227,  228,  243,  246. 
Ergamenes,  242. 
Erment,  113. 
Eshmunen,  280,  285. 
Esneh.  225,  238,  240,  245. 
Ethiopia,  24,  126,  206,  218,  242. 

Gold  Jars  of  Zakazik,  204*. 
Graeco-Egyptian  Sculpture,  255, 
262,  265. 
Greece,  161,  174,301. 
Greeks,  253,  273,  289. 
Grooved  Facade  of  Thinite  For- 
tresses, 4*. 
Grotesque    Figures    in    Bronze, 
288*. 
Grotesque    Head,    Terra-cotta, 
287*. 

Deshasheh,  Scene  in  a  Siege  at, 
107*.  108. 
Diadem  of  Khnemit,  123*. 

245.261. 
Ethiopian    King,    Head    of    an, 
269*. 

Grotesque,  Terra-cotta,  287. 
Grotesque  Slave  bearing  a  Jar, 
159. 

307 


INDEX 


Grotesques,  285. 
Group   of   Persons  standing  or 
sitting,  17*. 


H. 

Habi  Sadu  (festival),  79. 

Hadrian,  231.241.261. 

Hapu,  240,  297;  the  dwarf,  19*. 
20. 

Hapsefai.  101. 

Harmachis.  268. 

Harmin,  195;  Mourners  in 
Tomb  of,  191*;  Offering- 
bearers  in  Tomb  of,  191*. 

Harpocrates,  286;  with  Osiris 
and  Amon,  283*. 

Hathor,  22,  83,  152.  168,  174- 
175.  228,  240;  and  Ameno- 
phis, Temple  of,  221*,  230; 
Chapel  of,  230;  Columns, 
219*;  Cow,  250,  251,  268; 
Cow  at  Sakkarah,  244*; 
heads,  145,  231;  Pillars. 
131.  See  also  under  Capi- 
ta's  and  Cow. 

Hatiyai,  202. 

Hatshepset,  Queen,  130,  150, 
151,  173,  179. 

Hawara,  100,  116,274. 

Hawk's  Head,  Golden,  94*. 

Head  in  the  Louvre,  85*. 

Head  of  a  Colossal  royal  Statue, 
117*. 

Head  of  a  Man,  169*. 

Head  of  a  Statue,  246*.  259*. 

Hebt,  223;  East  Facade  of  Tem- 
ple, 215. 

Heliopolis,  19,  44,  54,  61,  97. 

Heptanomis,  275. 

Heracleopolis,  60,  104,  110,  194, 
240. 

Hermonthis,  231. 

Hermopolis,  60,  96,  162,  182, 
194, 240,  285;  School  of,  24, 
106-108. 

Heru-em-heb,  139,  143,  150, 
184-187,  199;  Head  of  a 
Statue  of,  178*;  Wooden 
Statuette  of,  1 95*. 

Her-Heru,  218. 

Hesi,  Wooden  panels  of,  59,  60*. 

Hieraconpolis,  4,  6,  7,  78. 

Hierasycaminos,  262. 

Hittites,  165. 

Holy  Mountain,  218. 


Horbet,  278. 

Ivory,  use  of,  77;  bas-relief,  74*; 

Horus,  21,  32,  1  12,  153,  155,  168, 

Feet  of  Bed  and  Stool,  2*; 

222,  227-229,  246,  250,  266. 

Mirror-handle,  205*. 

267.   278.   279,   283.   292; 

* 

Alexandrian,  250*;   Bronze, 

281*;  as  a  Child,  243*;  with 

the  Crocodiles,  257*;   Tem- 

J- 

ple  of  ,  at  EdfO  (plan).  2  1  6*. 

224,  225;  Court  and  porti- 

Japanese Artists,  69. 

coes   of   Temple    of,    217*; 

Jewel-casket  of  Amenophis  III.. 

Pronaos    and    Terraces    of 

196*. 

Temple  of,  218*;  Pylon  of 

Jewelry.  92,  120-122,  200,  204- 

Temple  of,  217. 

206,   289-293;    of  Aah-he- 

Humorous  Episode,  65*. 

tep,   205*;    from  Dahshur, 

Husband  and  Wife  seated,  72*; 

121*. 

standing,   72*;    of   unequal 

Jug   with   Goat   from   Zakazik, 

Height.  73*. 

204*. 

Hyksos  or  Shepherd  Kings,  1  1  7, 

124,  125,  159.  171,  196. 

Hypogea,  36,  37,  147-150,  215- 
217. 

K. 

Hypogeum  of  Ameni,   102*;    of 

Amenophis    II.,    141*;     of 

Ka-apiru,  32;    see  also  Sheikh- 

Khnemu-hetep,     101*;      of 

el-beled. 

Kom-es-Shugafa  (room  in). 

Kadesh,  Battle  of.  165,  166;  re- 

273*;    of.  Paheri.    wall    in. 

liefs,  155*.  189. 

188*;   of  Seti  I.  Sketch  in. 

Kalabshah,    242,   262;    plan   of 

148*. 

temple,  232*. 

Hypostyle  Hall,  th?,   127,   128; 

Kames,  148. 

Karnak,  97,  1  13,  1  16,  127.  129, 

and     Transverse     Section, 

130.  133,  136,  139-141,  145. 

1  7R*  •     '      f-K     ]WI 

171.  176,  179,  180,  187.  188, 

of  Seti  I  *    143'   in  the  Ra- 

208,214,218,220,222,223, 

messeum,  127*. 

232,    234,    239,    240,    260; 

Avenue  of  Rams  at,   133*; 

Court  of  the  Temple  of  Ra- 

meses    III.    at,    131*;     Fa- 

I. 

vissa,    191;    Great  Temple 

of    Amon,    95*;     Ruins   of 

Propy  laea  of  Tirhakah,  215; 

I  brim.  244. 

Workshops  at,  193. 

ctinus,  299. 

Karomama,    Queen,    276,    277, 

ncrustation   in   green   Enamel, 

280*. 

193*. 

Kartassi,   242,  243;    Chapel  in 

nfantry  from  Meir,  91*. 

Quarries  of,  234*  ;  Kiosk  of. 

nfluences,  foreign,  in  Egyptian 

233*. 

minor  arts,  197,  289. 

Kasr-el-Aguz,  231. 

onia,  289. 

Kasr-el-Shalauit,  231. 

'  sis  (goddess),   231,   246,   251, 

Kasr-es-Sayad.  37,  60. 

267.    279,    288,    292;     and 

Kasr-Ibrim,  East  Gate,  234*. 

Horus    (relief),    262,    270; 

Kasr-Karun,  220*.  230. 

statue   of,   244*;     in   terra- 

Kau-el-Kebir, 37,  60,  240. 

cotta,  290*;  Temple  of,  14*. 

Kauit,  Princess,  1  1  3. 

257*. 

Kha-emhet,  181,  195. 

Isis,  Queen,  161*.  171. 

Khamuasit,  Prince,  200. 

Italy,  279. 

Khasakhmui,  76*. 

luaa,   181.   182,   195,   199;    and 

Khayanu,  124. 

Thuaa,  200*. 

Kheti.  105,  108.  109. 

308 


INDEX 


Khnemit,  121;   Crown  of,  122*; 

Court  of  Amenophis  III.  at. 

Diadem  of,  123*. 

133*;  Court  of  Rameses  II. 

Khnemu,  Chapel  of,  132. 

at,  135*;   Temple  of  Amen- 

Khnemu-hetep, 101;  hypogeum 

ophis  III.  at,  137*. 

of,  101*,  106,  107. 

Khonsu(god),  15,  185,  186,  218, 

219;    Bust  of,  176*;   Tem- 

M. 

ple  of.  11*.  221,  224,  226, 

232;  Temple  at  Thebes  (sec- 

tion), 132*,  135-137. 

Maat  (goddess),  288. 

Khonsu,  Temple  of,  239. 

Macrinus,    261;     Bas-relief   of. 

Khu-en-Aton,  201. 

268*. 

Khufu-enekh,  27;  Sarcophagus, 
of,  27*. 

Maharrakah,  242,  244,  262. 
Maiptah,  195;    Fragment  from 

Kiblah,  52,  53. 

Tomb  of,  192*. 

Kitchen,  89*. 

Mammisi,  or  Birth-house,  "231, 

Kneeling    Figure     carrying     a 

232,  238,   241;    at  Edfu, 

Triad,  157*. 

222*. 

Kneeling  Man,  77*. 
Kokome,  40. 

Man,  Head  and  Bust  of  a,  253*. 
Mariette.  81,  117,  124,  180,248. 

Kom-el-Ahmar,  1  1*.  37,  79.  92; 

Maritsakro.  168,  173. 

Mastaba,  the,  29,  30,  31,  36,  37, 

Door-jamb   in    1  emple   or, 
7*. 

38,41,45,62.98-100,  102. 

Kom-es-Sagha,  96;    Temple  of, 
96*;    Interior  of  Temple  of 

215.     216;      with     hollow 
Chamber   and    discharging 

96*. 

Arch,  42*;    at  Gizeh,  30*. 

Kom-es-Shugafa,  257,  266. 

and  Shaft  in  same,  29*;    of 

Kom-Ombo,  55,  224,  228,  229, 

Ka-apiru,    Cell    and    Fore- 

232. 233,  240.  257;  Double 

court,  34*;  of  Khabeuptah, 

Court  and  Pronaos  at,  220*. 

35*;    of   Menefer,   Facade, 

Koseir-el-Amarna,  37. 

34*;   of  Mereruka,  33,  36*. 

Kurnah,  187;    Facade  of  sepul- 

37, 65,  73;  of  Ne  er-hetep. 

chral  Temple,  1  44*. 

Forecourt,  35*;   of  Ti,  Por- 

Kurnet-Murrai, 148,  230. 

tico  of,  36*;  of  Zazamenekh, 

35*. 

Masu,  278;  Bronze  Statuette  of, 

281*. 

L. 

Mat-ka-Ra,  Queen,  268. 

Medinet-Habu,    127,    130,    131, 

Legrain.  116,  191. 

134,  149,  155,  160,  161,  192, 

Libyan  Chief   struck   down  by 

219,   221,   223,   226,   230; 

Sahu-Ra,  56*. 

Fagade  of  Temple   of  Ra- 

Libyan Desert,  38. 

meses  III.  at,  146*;  Migdol 

Lion     passant,     257*;      seated, 

of,    148*;    Roman   Temple 

258*. 

at,  221*;    Second  Court  in 

Lion    at    Gebel-Barkal,     164*; 

Temple  of,  145*. 

Bronze,     time     of    Apries, 

Mediterranean,  215. 

282*;   of  Kom-Ombo,  258*; 

Medum,  26,  33,  38,  45,  62,  80, 

of  Sakkarah,  52*. 

167;   False  Pyramid  at,  40; 

Lisht,  97,  100,  111,  170. 

Geese  of,  51,  72;   Tomb  at. 

London  (British  Museum),  22, 

51;   Statues  of,  81. 

169,  288. 

Meir,   37;     decorations   at,   61; 

Louvre,  The,  57,  75,  80,  87,  1  19, 

Fat  Men  of,  62*;  Lean  Men 

181,201,202,206,252,274, 

of,  62*;  Soldiers  of,  90,91*. 

276,  279. 

Mekhu.  38. 

Luxor.  138-140,  161,  165.  166, 

Meks,  285. 

192,208,214,219,220,221, 

Memnon-Colossi      at      Thebes, 

223,   231.   239,  259.   276; 

167*.  176. 

309 

Memnonium,  153,  187,  188,  190; 
of  Seti  I.  at  Abydos,  plan, 
143*. 

Memphis,  19,  29,  37,  38.  95,  106. 
114,126,148,  162,189,192, 
216,  254,  265,  273,  276,  278. 
280.  285,  288,  289. 

Memphite  Dynasties,  23. 

Memphite  School,  24,  26,  61.  79. 
81,  105,  107,  108,  110,  113, 
213.  251,  254;  contrasted 
with  Theban  School,  115. 

Memphite  Statue,  250*;  Stat- 
uette, 251. 

Meneptah,  186,  189,  192,  220; 
Bust  of  a  Statue  of,  187*; 
with  two  Ensigns,  156*. 

Menes,  6,  23. 

Menkhau-Heru,  57;  Statue  of, 
57*;  Statuette  of.  82. 

Mentemhet,  241*,  248. 

Menthu-hetep,  99,  1 14.  1 15,  148, 
152;  in  the  Costume  of 
Apotheosis,  112*;  Tomb  of, 
113. 

Mentu  (god),  120,  268;  Temple 
of,  134. 

Mereruka,  33.  37.  65.  73.  See 
also  under  Tomb. 

Meroe,  245,  256.  262. 

Mesaurat,  Temple  of,  245;  236*. 

Mesopotamia,   125,  206. 

Metal,  casting,  202,  276;  use  of. 
77,78;  utensils,  92. 

Migdol,  160. 

Mirmashau,  1 18;  Colossal  Statue 
of,  117*. 

Mirror-case,  205*. 

Mitrahineh.  192. 

Mnevis,  279. 

Models  of  Heads  at  different 
Stages,  261*. 

Mond's  Statuette.  169*.  177. 
178. 

Monkey,  Statue  of  a,  163*,  256*. 

Mummy,  Mask  of  a,  277*. 

Munich  Museum.  80.  110. 

Mural  Decorations,  18. 

Musician-priestesses,  1 52*. 

Mut(god),  167,  187,  219,  268. 

Mut-emua,  Queen,  138. 

Mycenae,  205. 

Mycerinus,  27;  Pyramid  of,  41; 
Sarcophagus  of,  27*;  Statue 
of.  16*.  83;  Statuette  of, 
82;  Triad  of.  81*;  and  his 
Wife.  80*. 

Myers*  Bust,  85;  Statuette,  83*. 


INDEX 


N. 

Nai.  187;  the  Lady,  201*. 
Nakadah,  7,  26,  27.  113;  Plan 

of  Tomb,  7*. 
Nakht,  165. 

Name  Horus  of  Chephreu,  32*. 
Napata,   218.   219,   256;    most 

ancient    Temple    at,    214*; 

Sanctuary  of    I  irhakah  at, 

214*. 

Naples  Museum,  80. 
Napriti,  58. 

Nasi.  The  Lady,  77*.  80. 
Naucratis.  257,  265. 
Naville.  174.  250. 
Necklace  with  Heads  of  Falcons, 

Gold.  207*. 
Necho  II..  291. 
Nectanebus,  238,  240,  257,  259; 

Naosof.  218*. 
Nectanebus  II.,  223.  261. 
Nefer-Atmu,  283. 
Nefer-hetep.  32. 
Nefer-ka-Ra,  43,  48. 
Nefer-ka-Ra-Huni,  26. 
Neferseshemptah,  34,  68*. 
Nefert,  80;  Queen,  116M18. 
Nefert-ari,  Queen,  144. 
Nefer-temu,  233. 
Neferu,  85*;   Statuette  of,  86. 
Neferu-RI,  172. 

Negress,  Statuette  of,  199*.  200. 
Nephthys,  288. 

Neter-baiu,  22;    Tablet  of  (ob- 
verse), 23*;   (reverse),  24. 
Newberry,  202. 
Nibu,  the  Dog.  18*.  20. 
Nile,  38,  90,  139.  159.  185.  218; 

Figure  of  the,  1%,  233,  288; 

in   pierced    Bronze,    291*; 

Twin  Figures  of,  193*. 
Nit  (goddess).  279.  283. 
No.  I.  Statue  at  Cairo,  77*.  79. 
Nomes  (divisions  of  Egypt),  8, 

16.  63,  233. 
Nsiptah,  241,  249. 
Nubia.  24,  143,  189,  190,261. 
Nufiabres,  283. 
Nutir,  the  Lady,  19*,  20. 


0. 

Oases,  273. 

Obelisks,  97;  at  Karnak,  125*; 
of  Ptah-hetep  and  Maiti, 
41*;  of  Heliopolis.  98*. 


Obelisk-worship,  46. 

Offering-bearers,  89*. 

Officers,  Statuettes  of,  201,  203*. 

Ombos,  234,  245.  262. 

Opening  of  the  Mouth,  199*. 

Opet  (goddess),  232. 

Orontes,  165. 

Osiris,  79,  115.  134.  145.  154, 
218,227.228,232,250,251, 
279,288,292.301;  Bronze, 
278;  Osiris-Nile  in  pierced 
bronze,  291*;  Recumbent, 
243*;  Statue  of,  244*. 

Osorkon  II.,  214,  248;  pushing 
a  boat.  237. 

Osorkon  III.,  288. 

Ostraka  at  Cairo  and  Turin, 
149*-151*.  163. 

Otho,  231. 

Oxford.  79. 

Oxyrrhynchos,  240. 


Painted  bas-relief  in  Tomb  of 
Seti  I..  179*. 

Painted  Decoration  of  a  Pri- 
vate House,  103*. 

Painted  Stele.  274*. 

Painting,  Egyptian,  50,  51,  267- 
270. 

Painting  of  a  Hypogeum  at  Bah- 
riyeh,  276*. 

Painting  and  Sculpture,  com- 
bination of,  103. 

Painting  without  sculptured 
Background,  103*. 

Pakheri.  268;  Painted  Panel  of, 
274*. 

Paris,  22.  288. 

Parthenon,  152,  299. 

Pasticcio  Ptolemaic,  266*;  in 
Temple  of  Theban  Ptah, 
265*. 

Patanafi,  263. 

Patanesis,  283. 

Pavement  in  Palace  at  El 
Amarna,  147. 

Peasant  Squatting.  106*. 

Pectoral  of  Amenemhat,  III., 
121*;  of  Rameses  II.;  of 
Sesostris  III.,  120*. 

Pedishashi,  Statue  of,  247*. 

310 


Pelizaeus  Collection,  202. 

Pepi  I.,  202;  Bust  of,  76*;  Head 
of  Statuette  of,  76*;  Statue 
of,  78*. 

Pepi  II.,  95. 

Perfume-spoon  in  Form  of 
Woman  swimming,  199*. 

Perichon  Bey,  285. 

Persians,  253. 

Person  seated  with  one  Leg  flat, 
157*. 

Person  of  the  time  of  Ameno- 
phis  IV  .  174*. 

Petrie,  Flinders,  3,  77,  91,  96, 
201. 

Petubastis,  237*;  Statue  of, 
247*. 

Petukhanu,  276. 

Pharaoh  in  a  Helmet,  Head  of, 
187*.  193;  Horus.  110*; 
Khasakhmui,  76*,  79;  in 
pierced  Bronze,  291*. 

Pharaohs.  9,  10,  16,  17.  18,  21. 
27,  52.  55.  57,  62,  95,  96, 
100.112,113,114,119.130, 
135,  151,  156,  158,  159,  175. 
176.183,207.217,228.239. 
257,  288;  Ethiopian,  220. 
245;  Palaces  of.  156-159; 
Tombs  of,  38. 

Phil*.  224,  230.  232,  233.  236- 
240.  242.  243,  258-260; 
General  View  of,  213*; 
Capitals  of  Prohaos,  226*; 
East  Block  of  second  Pylon. 
229*;  Kiosk  of  Nectanebus. 
225*.  238;  Kiosk  of  Trajan, 
227*.  240;  Mammisi,  225*; 
Mammisi  and  Great  Door, 
230*;  Plan  of  Island  of, 
228*;  Pronaos  and  Hypo- 
style  Hall,  230*;  Temple 
of  Isis  during  Inundation, 
228*;  Wall  in  Temple  of 
Isis,  223*;  Western  Portico. 
224*. 

Philip  Arrhidsus.  223.  239. 
259. 

Piankhi,  218. 

Pillar  at  Karnak,  97*. 

Pillars,  235-238. 

Pinotchem,  207. 

Plaques,  polychrome,  198. 

Plate,  gold  and  silver.  202,  203, 
289,  290. 

Plato,  259. 

Portraits,  273.  275. 

Posno,  Sale.  278. 


INDEX 

Pottery,  91,  284. 

Q- 

Priest  and  Monkey,  188*,  193. 

Priest,  Statuette  of  a,  201*. 

Qa-au  Stele,  20*.  21. 

Private  Person,  Statue  of  a,  119, 

Queen,  Statuette  of  a,  240*. 

120. 

Quibell,  9.  92. 

Pronaos,  the,  221. 

Proto-Doric  or  primitive  Order, 

103,  129,  144,237;  at  Kar- 

R. 

nak,  129*. 

Psammetichus    I..    223,    239*, 
243*.  248,  250. 

Rameses  I..  140,  154. 
Rameses  II.,  139,  140.  143,  150, 

Psammetichus  II.,  289. 
Psammetichus  III.,  240*. 

154,155,165-167.185,188- 
191,  194,214,219,220,239, 

Psammetiknufisashmu,  263,  266; 
Bas-relief  of,  271.272. 
Pselchis,  243. 
Ptah,  113.  194,  201,  239,  283, 

260,  297;  Alabaster  Statue 
of.    184*.    185*;    Bust   on 
Coffin  of,   194*;    Coffin  of, 

301;Memphite,  189*;  The- 
ban,    129;    Propylaea    and 
Facade  of  Oratory  of  The- 

199;     Colossal    Figure    of, 
189*;     Colossal    Head    of, 
186*;  pushing  a  Boat,  158. 
Rameses  III.,  133,  145,  151,  155, 

ban,  227*;  Temple  of,  133, 

*>^7 

159,189,192,198.208.224, 

Ljl  . 

Ptah-enekh,  65*.  66. 
Ptah-hetep,    66,    71;     and    his 
Wife,  67*. 

247. 
Rameses  IV.,  162.  199,  202;  and 
a    Libyan    Prisoner,    186*, 

Ptahmes,  198. 

193;  Sandstone  Bust  of.  194. 

Ptolemais,  273. 
Ptolemaic  Doorway,  226*. 
Ptolemies,   215,   223,  230,   231, 
232,237,239,240,242,254. 
259,261,266,272,278,287. 
Ptolemy,   258;    Euergetes,   240, 

Rameses  XII.,  206. 
Rameses  of  the  Twentieth  Dy- 
nasty, the,  140,  151. 
Rameses,     Prince,     afterwards 
Rameses  II.,  181*. 
Ramesesnakht,  193. 
Ramesseum,  127,  144,  154,  156, 

257;     Euergetes    charging, 
260*;     Philadelphus,    240; 
Philopator,   242;    Philopa- 
tor  IV.,  240;  Physcon,  242. 
259;  Soter,  239. 

165,    166,   221.  225;    Gen- 
eral View  of,  144*. 
Ramessids,    the,   194,   213.  218, 
219,  233,  262.  293. 

Punt,  180. 

Ramosis,  181. 

Pylon,  the,  48,  97,  126,  127;  of 
Heru-em-Heb   at    Karnak, 
126*. 

Ra  (the  sun-god),  44,  46,  283. 
Ra-hetep,  80,  84,  87,  167. 
Ra-en-user,  44,  45,  49;    Chape 

Pyramid,    of    Cheops,    42;     of 
Chephren,  41.  299;    False, 
40,    44*;     Great    Pyramid 
and  Sphinx,  45*;  with  Obe- 
lisk, 44;  origin  of  the,  40; 
of  Pepi  II.,  35;    of  Ra-en- 
user,  45*;    of  Unas,  Door 

of,  49*,  56;  Statuette  of,  82. 
Ra-nefer,  71,  82. 
Reindeer  Period,  2. 
Reisner's  Mycerinus,  80*. 
Rekaknah,  7. 
Relief  en  creux,  12,  260. 
Renascence    of    Egyptian    Art, 

in,  47*;  of  Unas,  plan,  46*. 
Pyramidium,  the,   146,   147;    of 
Dahshur,  100*. 

262. 
Rhind,  287. 
Ritual  of  the  Pyramids,  216. 

Pyramid-mastaba,  the,  100,  101, 

Rome,  161,  174,275.301. 

146,  147;   of  an  Apis,  140*; 

of  Drah-Abu'1-Nekkah,  98*: 

section  of  a,  98*. 

S. 

Pyramids,  26,  40,  42,  48,  62,  81, 

83,  98.  113;    of  Gizeh,  45; 

Sabni,  Chapel  of,  38. 

of  Meroe,  245. 

es-Sabua,  189. 

311 

Sadunimet,  87*. 
Saft-el-Henneh,  Naos  of,  6. 
Sahu-Ra,  56;    adopted  by  the 

goddess,    57*;     Chapel    of. 

44,  49*;  Temple  of,  49. 
Said,  the,  20,  83,  215,  218,  223. 

240. 

Sals,  213,  278;  School  of,  24. 
Saite  period,  15,  48,  242,  265. 
Sakkarah,  32-34,  38,  44,  62.  63. 

77,  80,  91,  114.  216.  217. 

258,  282,  291 
Sanhur,  230. 
Sapui,  80. 
Sarcophagus     of     Khefu-enekh, 

27*;  of  Mycerinus,  27*;  of 

Princess  Kauit.   111*.  113. 
Sa-Renput,    Portico    of.     100*; 

Tomb  of,  101. 
Sargonids,  289. 
Scribes,  Statues  of,  86-88,  119; 

crouching,    86*;     kneeling, 

87*;  seated,  87,  299. 
Sebek-emsef,  118*.  119. 
Sebek-hetep.     118*.     119,    124. 

240. 

Secret  chambers,  234,  235. 
Sekhet,     173,    278,    279,    281; 

Bronze,  284*;    at  Karnak. 

163*. 
Seneferu,  20.  40.  41.  44.  51.  62. 

80. 

Senmut,  171.  247. 
Sennefer,    177;    with  his  Wife 

and  Daughter,  168*. 
Septah  Meneptah,    189;    Scene 

in  Tomb  of,   183*;    in  his 

Tomb,  182*. 
Septimius  Severus,  244. 
SerJab,  34.  37.  42. 
Serpent-Isis,  290*. 
Serpent,  King,  21.  23;  Stele  of. 

21*. 
Servant   carrying  his   Master's 

Baggage,  88*. 
Servant  tarring  a  Jar,  10*. 
Service  of  Antiquities,  96. 
Sesostris  I.,   97,    100,   111-113. 

116,   119;    Head  of,   113*; 

Statues  of,  108*.  109*. 
Sesostris  II..  118.  120. 
Sesostris  III..   114.  120;    Head 

of,  114*. 
Sesostris    IV.,    116;     Head    of. 

114*. 
Sesostris  fighting,  183*.  190,  206. 

299. 
Set  (god),  168,  228. 


INDEX 


Seti  I..  21,  140.  151.  153,  154. 

"ready-made,"  252;  roughly 

162,  185,  187-191,  258;   at 

blocked   out,   245;    seated. 

Abydos,     180*;      and     the 

74*;    standing,  bearing  en- 

three Goddesses,  181*.  188, 

signs,  17*. 

299. 

Statuette    of    petrified    Wood, 

Seti  II..  189,  206;  in  his  Tomb. 

176*. 

182*. 

Statues,  17. 

Severus,  275. 

Stete,  20,  21,  23.  33,  52,  54,  62, 

Sheikh  -Abd-  el  -Kurnah.     148, 

148,  261,  266,  268,  270. 

149,    153*.    163-165,    177, 

Stele  Door  of  Nibera,  39*;    of 

180,  287. 

Usiru,  33*;    of  the  Fayum, 

Sheikh-el-beled,  83*,  84,  85,  1  12, 

268*;  in  form  of  False  Door, 

195,  299;  supposed  Wife  of 

31*;    of  Horus  and  Croco- 

the, 83*. 

diles,   260*;    of   King   Ser- 

Shepherd Kings,  see  Hyksos. 

pent,    21*;     of   Nikhafitka, 

Shepsestah,  49. 

40;    of  Prince  Antef,    113; 

Shoulder   Movement,   Tomb  of 

of  Ta-au,  20*;    of  Sesostris 

Khnemu-hetep,  105*. 

III.,  112;   of  Seti,  28*. 

Shu  (god),  243. 

Step  Pyramid,  39,  40,  43*.  91; 

Shunet-ez-Zebib,  1*. 

Section  of,  44*. 

Siege   of   a   Fortress,    104*;     at 

Storehouses  with  Brick  vaults  in 

Deshasheh,  107*. 

Ramesseum,  145*. 

Siesis,  111;   Portrait  of,  108*. 

Stroganoff,  276. 

Silsileh,    143;     Speos   at;    Plan 

Surveyor  with  Line,  157*. 

and  Fajade,  136*. 

Syene,  110,242,285. 

Silver  Plate  of  Thmuis,  292*. 

Syria,  77,  159,  206,  300. 

Simon,  James,  201. 

Sinai,  21,  26,  96;  bas-reliefs  at. 

21,22*,  59. 

Siut,  108,  110. 

T. 

Sketch  Figure  of  a  King,  263*. 

Sketch  in  a  Tomb,  153*. 

Slave  bearing  a  Jar,  159*. 

Table  of  offerings,  32,  38,  46. 

Sloping  Passage  in  Tomb  of  Ti, 

Tafeh,  Temple  of,  231*.  242. 

29*. 

Takushit,  277,  280*. 

Smcndes,  196. 

Tanagra,  285. 

Sobek,  228,  229. 

Tanis,  117,217,258;  School  of, 

Sokaris.  292. 

24,  195. 

Soldiers  at  Siut,  105*. 

Tanite  Statue,  248*. 

Solar  Temple  of  Ra-en-user,  50*. 

Tantkalashiri,  271. 

Soul  Statues,  256*. 

Tataharsias  .    273;    Coffin    of, 

Spain,  2. 

275*. 

Speos,  the,  141-146. 

Tchaho,  266. 

Sphinx,  80,  81.  117,  118,  173; 

Tefnut,  243. 

of  Amenemhaal  III.,  116*, 

Tefyeh,  105. 

124,    196;    of   Apries,   276, 

Tell-es-Sab,  278. 

279*;  of  Gizeh,  78*;  of  the 

Temenos,  the,  41. 

Ptolemaic  Period,  257*;    of 

Temples,  130;  decoration  of  the, 

the   Roman   Period,    253"; 

18;   small,  229,  230. 

Temple  of  the,  47,48,51*; 

Teye,  see  Thi. 

Interior    of    Temple,    51*. 

Thebaid,  220,  247,  273,  275,  300. 

52*. 

Theban  Citizens,  1  52*. 

Squatting  Figure,  236*. 

Theban  School,  24,   113.  213. 

Statue,   with   body   in   Sheath. 

Thebes,  48,  101,  104,  114,  116, 

18*;    finished   but   for   the 

117,  125,  127,  135,  147,  162, 

Head,  245*;   full  face,  15*; 

179,  183,  189,  190,  193,  196, 

full  face,  walking,   16*;    of 

206,213,218,254,259,261, 

the  Hyksos   Period,    125*; 

263,  265,  274,  280,  289. 

312 

Thi,  Queen.  171,  176,  181,  185, 
186,  200;  so-called  Head 
of,  177*;  Head  of  Statuette 
of,  171*. 

Thinite  Tablet,  25*;  Period,  6. 
7;  Pharaohs,  7;  School,  26, 
61,79.81. 

This,  80. 

Thmuis,  278,  289. 

Thoth,  60,  168,  193,  267,  285, 
292. 

Thoth  Paotnuphis,  242. 

Thothmes  I.,  148,  149,  170. 

Thothmes  II.,  130. 

Thothmes  III.,  113,  129,  130. 
132,  133,  140,  150,  151,  171, 
172,  173,  179,  199,202,214, 
237,  239,  242,  247,  259,  288. 
299;  Head  of  a  Statue  of, 
162*. 

Thothmes  IV.,  140,  150,  151, 
172,  198;  and  his  Mother, 
162*. 

Thoueris,  242,  249. 

Thuaa,  199;  Coffin  of,  194*. 

Thuti,  202. 

Ti,  see  under  Tomb. 

Ti,  Statue  of,  84. 

Tiberius,  236. 

Tirhakah,  219,  220,  222,  245, 
248,  276;  Head  of,  239*; 
Statuette  of,  279*. 

Toilet  utensils,  200,  204. 

Tomb  of  Adu,  Denderah,  42*. 

Tomb  of  Ankhmara,  Dancers 
in,  69*. 

Tomb  of  Heru-hetep,  Painted 
Interior  in,  102*. 

Tomb  of  Kha-em-het,  Sketch  in, 
149. 

Tomb  of  Mekhu,  38,  43*. 

Tomb  of  Menthu-hetep,  Plan, 
99*. 

Tomb  of  Nakht,  wall  in,  154*. 

Tomb  of  Ptah-hetep,  Brawl 
between  Boatmen,  69*; 
Herdsman  driving  Bulls, 
67*;  Perspective  of  Regis- 
ters, 66*;  Sacrifice,  63*. 

Tomb  of  Ra-hetep,  wall,  63*. 

Tomb  of  Sabni,  43*. 

Tomb  of  Sa-Renput  II.,  101. 

Tomb  of  Ti,  34,  65,  72,  73; 
Bringing  Corn,  13*;  Cram- 
ming Geese,  70*;  Harvest 
Scenes,  12*;  Hippopotamus 
Hunt,  55*;  Plan  of,  41*; 
Servants  of  Ti  bringing  of- 


INDEX 


fcrings,  12*;  Slaughter  of 
Cattle,  13*;  Sloping  Pas- 
sage in,  29*. 

Tomb  with  pyramidal  summit, 
Theban,  140*. 

Tombs,  27,  28,  29,  97,  98;  dec- 
oration of,  18. 

Torso  and  Head  of  a  Woman, 
169*. 

Trajan,  261;  Kiosk  of,  241,  243. 

Triad  of  Heracleopolis,   190*. 

Triumphal  bas-relief  of  Seti  I., 
175. 

Tuitishere,  Queen,  160*,  169. 

Tukh-el-Karamus,  289,  292. 

Tuosret,  203,  206. 

Turah  limestone,  47,  67,  75,  84. 

Turin  Museum.  80,  169,  170, 
171,  186,  191,201. 

Tut-ankhamen,  177*.  184-186. 

Twin  Statues  at  Cairo,  1 1 8*. 

Two  Prisoners  tied  Back  to 
Back,  159*. 

Types  of  archaic  Chapels  and 
Temples,  5*. 


u. 

Unas,  41,  56;    Pyramid  of,  43; 

Temple  of.  49. 
Ushabtia.  198,  282.  284. 


Utilitarian  element  in  Egyptian 

art,  300. 
Uzueri,  58. 


V. 

Valley  of  the  Kings,  164;  of  the 

Queens,  230. 
Vase  in  the  form  of  a  Head  in  a 

Helmet,  285*. 
Vassalli,  51. 
Venice,  288. 
Vespasian,  231. 
Vienna  Museum,  1 1 9. 


W. 

Wadi-es-Sabua,  144.  219;  Av- 
enue of  Sphinxes  at,  137*. 

War-Dance,  104*. 

Wood,  use  of,  77. 

Wood-carving,  198. 

Wooden  Head  from  Lisht,  109*. 

Wooden  Naos,  5*. 

Wooden  Spoon  for  Cosmetics, 
197*. 

Wooden  Statue,  Bust  of,  75*. 

Wooden  Statue,  84*,  85. 

Wooden  Statuettes,  168. 


Woman  grinding  Corn,  9*,  72*. 
Woman,  Head  and  Bust  of  a, 

252*;  Torso  and  Head  of  a, 

169*. 
Woman   in   sculptured    groups, 

74. 
Woman,   Statuette  of   a,  200*. 

249*. 
Wrestling,  Beni  Hasan,  107*. 


Xois,  240. 


X. 


Y. 


Young  Girl,  202*;  Statuette  of. 
201*. 


Zadamonefonukhu.  268. 

Zai,  187;   and  Nai,  178*. 

Zakazik,  202. 

Zanufi,  263,  266. 

Zauti   37. 

Zawyet-el-Aryan,  26,  43;  Floor 

and   Libation-trough,   48*; 

Inclined  Way,  47*. 
Zoser.  38.  39,40,91. 


313 


ERRATA. 

P.  44.  1.  18.  Substitute  for  lines  18-22  the  following:  "Sahu-Ra  and  Ra-en-user.  They  were 
approached  by  propylaea  built  at  the  foot  of  the  plateau  (Fig.  81),  beyond  which  a  long 
incline  rose  to  the  body  of  the  building,  the  arrangements  of  which  varied.  In  the  chapel 
of  Sahu-Ra  (Fig.  82),  for  instance,  there  was  a  dark  passage,  then  a  colonnaded  court,  then 
a  complicated  series  of  cells  and  storehouses,  and  in  the  obscurity  of  the  background,  the 
stele,  in  the  form  of  a  closed  door,  where  the  office  of  the  dead  king  was  celebrated.  It  was 
a  temple  ",  etc. 

1s.  48,  inscription  of  Fig.  79,  for  "  Foor  "  read  "  Floor." 

P.  49,  inscription  of  Fig.  81,  for  "Plan  of  the  Chapel"  read  "Plan  of  the  Propylaea." 

P.  268,  inscription  of  Fig.  515,  for  "  Diadumenian"  read  "  Diadumenianus." 


AU6  2  6  1988 


/ 


PLEASE  DO  NOT  REMOVE 
CARDS  OR  SLIPS  FROM  THIS  POCKET 

UNIVERSITY  OF  TORONTO  LIBRARY