Skip to main content

Full text of "A history of art in ancient Egypt"

See other formats


^SSSSSSSH 


gBW  ■»iii*ri"ii<» 


n^ttm$^mamm-!im9u!Mn 


^\w.  '■  I 


f' 


/ 


V 


A  HISTORY 


OF 


ART    IN    ANCIENT    EGYPT. 


^0229 


A  HISTOR 


OF 


^rt  in  indent 


LEGISLATIVE* 


^^^^S 


FROM        THE        FRENCH 


OF 


GEORGES    PERROT, 

TROFFSSOR    IN    THE    FACULTY    OF    LETTERS,    PARIS  ;    MEMBER    OF   THE   INSTITUTE 

AND 

CHARLES  CHIPIEZ. 


ILLUSTRATED  WITH    FIVE   HUNDRED   AND   NINETY-EIGHT   ENGRAVINGS  IN   THE  TEXT, 
AND   FOURTEEN   STEEL  AND   COLOURED   PLATES. 


/JV   TWO    VOLUAfES.—  VOL.  I. 
TRANSLATED  AND  EDITED  BY 

WALTER    ARMSTRONG,    B.A.,    Oxon., 

AUTHOR    OF    "ALFRED    STEVENS,"     ETC. 


alon^On:    chapman    and     hall,    Limited. 
^fU)  §oife:  A.   C.  ARMSTRONG  AND  SON. 

1883. 


HonHon : 

Iv.  Clay,  Soxs,  and  Taylor, 

BREAD   STKEET   HIM.. 


PREFACE. 

M.  Perrot's  name  as  a  classical  scholar  and  archaeologist, 
and  M.  Chiplez's  as  a  penetrating  critic  of  architecture,  stand  so 
high  that  any  work  from  their  pens  is  sure  of  a  warm  welcome 
from  all  students  of  the  material  remains  of  antiquity.  These 
volumes  are  the  first  instalment  of  an  undertaking  which  has 
for  its  aim  the  history  and  critical  analysis  of  that  great  organic 
growth  which,  beginning  with  the  Pharaohs  and  ending  with 
the  Roman  Emperors,  forms  what  is  called  Antique  Art.  The 
reception  accorded  to  this  instalment  in  its  original  form  is 
sufficient  proof  that  the  eulogium  prefixed  to  the  German  trans- 
lation by  an  eminent  living  Egyptologist,  Professor  Georg  Ebers, 
is  well  deserved  ;  "  The  first  section,"  he  says,  "  of  this  work,  is 
broad  and  comprehensive  in  conception,  and  delicate  in  execution  ; 
it  treats  Egyptian  art  in  a  fashion  which  has  never  previously 
been  approached."  In  clothing  it  in  a  language  which  will,  I 
hope,  enable  it  to  reach  a  still  wider  public,  my  one  endeavour  has 
been  that  it  should  lose  as  little  as  possible,  either  in  substance 
or   form. 

A  certain  amount  of  repetition  is  inevitable  in  a  work  of 
this  kind  when  issued,  as  this  was,  in  parts,  and  in  one  place  ^ 
I  have  ventured  to  omit  matter  which  had  already  been  given 
at  some  length,  but  with  that  exception  I  have  followed  M. 
Perrot's  words  as  closely  as  the  difference  of  idiom  would  allow. 
Another  kind  of  repetition,  with  which,  perhaps,  some  readers 
may  be  inclined  to  quarrel,   forced  itself  upon  the  author  as  the 

^  Page  92,  Vol.   I. 


vi  Preface. 

lesser  of  two  evils.  He  was  compelled  either  to  sacrifice  detail 
and  precision  in  attempting  to  carry  on  at  once  the  history  of 
all  the  Egyptian  arts  and  of  their  connection  with  the  national 
religion  and  civilization,  or  to  go  back  upon  his  footsteps  now  and 
again  in  tracing  each  art  successively  from  its  birth  to  its  decay. 
The  latter  alternative  was  chosen  as  the  only  one  consistent  with 
the  final  aim  of  his  work. 

Stated  in  a  few  words,  that  aim  is  to  trace  the  course  of  the 
great  plastic  evolution  which  culminated  in  the  age  of  Pericles 
and  came  to  an  end  in  that  of  Marcus  Aurelius.  That  evolution 
forms  a  complete  organic  whole,  with  a  birthday,  a  deathday, 
and  an  unbroken  chain  of  cause  and  effect  unitinor  the  two. 
To  objectors  who  may  say  that  the  art  of  India,  of  China,  of 
Japan,  should  have  been  included  in  the  scheme,  it  may  be 
answered  :  this  is  the  life,  not  of  two,  or  three,  but  of  one. 
M.  Perrot  has  been  careful,  therefore,  to  discriminate  between 
those  characteristics  of  Egyptian  art  which  may  be  referred 
either  to  the  national  beliefs  and  modes  of  thought,  or  to 
undeveloped  material  conditions,  such  as  the  want  or  supersti- 
tious disuse  of  iron,  and  those  which,  being  determined  by  the 
very  nature  of  the  problems  which  art  has  to  solve,  formed  a 
starting  point  for  the  arts  of  all  later  civilizations.  By  means  of 
well-chosen  examples  he  shows  that  the  art  of  the  Egyptians  went 
through  the  same  process  of  development  as  those  of  other  and 
later  nationalities,  and  that  the  real  distinguishing  characteristic  of 
the  sculptures  and  paintings  of  the  Nile  Valley  was  a  continual 
tendency  to  simplification  and  generalization,  arising  partly  from 
the  habit  of  mind  and  hand  created  by  the  hieroglyphic  writing, 
partly  from  the  stubborn  nature  of  the  chief  materials  employed. 

To  this  characteristic  he  might,  perhaps,  have  added  another, 
which  is  sufficiently  remarkable  in  an  art  which  had  at  least  three 
thousand  years  of  vitality,  namely,  its  freedom  from  individual 
expression.  The  realism  of  the  Egyptians  was  a  broad  realism. 
There  is  in  it  no  sign  of  that  research  into  detail  which  dis- 
tinguishes  most   imitative  art  and   is    to  be   found  even    in    that 


Preface. 


Vll 


of  their  immediate  successors ;  and  yet,  during  all  those  long 
centuries  of  alternate  renascence  and  decay,  we  find  no  vestige 
of  an  attempt  to  raise  art  above  imitation.  No  suspicion  of 
its  expressive  power  seems  to  have  dawned  on  the  Egyptian 
mind,  which,  so  far  as  the  plastic  arts  were  concerned,  never 
produced  anything  that  in  the  language  of  modern  criticism  could 
be  called  a  creation.  In  this  particular  Egypt  is  more  closely 
allied  to  those  nations  of  the  far  east  whose  art  does  not  come 
within  the  scope  of  M.  Perrot's  inquiry,  than  to  the  great 
civilizations  which  formed   its  own  posterity. 

Before  the  late  troubles  intervened  to  draw  attention  of  a 
different  kind  to  the  Nile  Valley,  the  finding  of  a  pit  full  of  royal 
mummies  and  sepulchral  objects  in  the  western  mountain  at 
Thebes  had  occurred  to  give  a  fresh  stimulus  to  the  interest  in 
Egyptian  history,  and  to  encourage  those  who  were  doing  their 
best  to  lead  England  to  take  her  proper  share  in  the  work 
of  exploration.  A  short  account  of  this  discovery,  which  took 
place  after  M.  Perrot's  book  was  complete,  and  of  some  of  the 
numerous  art  objects  with  which  it  has  enriched  the  Boulak 
Museum,  will  be  found  in  an  Appendix  to   the  second  volume. 

My  acknowledgments  for  generous  assistance  are  due  to 
Dr.  Birch,  Mr.  Reginald  Stuart  Poole,  and  Miss  A.  B.  Edwards. 

W.  A. 


t 


CONTENTS. 


PAGE 

INTRODUCTION i— Ixi 

TO  THE  READER Ixiii— Ixiv 


CHAPTER   I. 

THE    GENERAL    CHARACTER    OF    EGYPTIAN    CIVILIZATION. 

§  I.  Egypt's  place  in  the  History  of  the  World i — 2 

§  2.  The  Valley  of  the  Nile  and  its  Inhabitants 2 — 16 

§  3.  The  Great  Divisions  of  Egyptian  History 16 — 21 

§  4.  The  Constitution  of  Egyptian  Society — Influence  of  that  Constitu- 
tion upon  Monuments  of  Art 21 — 44 

§  5.  The  Egyptian  Religion  and  its  Influence  upon  the  Plastic  Arts  .    .  44 — 69 
§  6.  That  Egyptian  Art  did  not  escape  the  Law  of  Change,  and  that 

its  History  may  therefore  be  written 70 — 89 

§  7.  Of  the  place  held  in  this  work  by  the  Monuments  of  the  Memphite 

Period,  and  of  the  Limits  of  our  Inquiry 89 — 93 


CHAPTER    II. 

PRINCIPLES    AND    GENERAL   CHARACTERISTICS    OF    EGYPTIAN    ARCHITECTURE. 

§   I.  Method  to  be  Employed  by  us  in  our  Study  of  this  Architecture    .  94 — 96 

§  2.  General  Principles  of  Form 96 — 102 

§  3.  General  Principles  of  Construction. — Materials    ,             103 — 106 

§  4.  Dressed  Construction 106 — 113 

§  5.  Compact  Construction 113— 114 

§  6.  Construction  by  Assemblage 114 — 119 

§  7,  Decoration 119 — 125 

VOL.    I.  b 


Contents. 


CHAPTER  III. 

SEPULCHRAL  ARCHITECTURE. 

PAGE 

§   I.  The  Egyptian  Belief  as  to  a  Future  Life  and  its  Influence  upon 

their  Sepulchral  Architecture 126 — 163 

§  2.  The  Tomb  under  the  Ancient  Empire 163 — 241 

The  Mastabas  of  the  Necropolis  of  Memphis 165 — 189 

The  Pyramids 189—241 

§  3.  The  Tomb  under  the  Middle  Empire 241 — 254 

§  4.  The  Tomb  under  the  New  Empire 255 — 317 


CHAPTER  IV. 

THE    SACRED    ARCHITECTURE    OF    EGYPT. 

§   I.  The  Temple  under  the  Ancient  Empire 318 — ;^^;^ 

§  2.  The  Temple  under  the  Middle  Empire 333 — 335 

§  3.  The  Temple  under  the  New  Empire 335 — 433 

§  4.  General  Characteristics  of  the  Egyptian  Temple 434 — 444 


LIST    OF    ILLUSTRATIONS. 


COLOURED   PLATES. 

The  Arab  Chain,  from  near  Keneh To  face  page  102 

The  Pyramids,  from  old  Cairo ,,  102 

Karnak,  bas-reliefs  in  the  Granite  Chambers „  124 

Seti  I.,  bas-relief  at  Abydos „  126 

General  view  of  Karnak „  360 

Perspective  view  of  the  Hypostyle  Hall,  Karnak „  368 

Thebes,  the  plain,  with  the  Colossi  of  Memnon „  376 


FIG.  PAGE 

1.  During  the  Inundation  of  the  Nile 3 

2.  Hoeing 4 

3.  Ploughing 4 

4.  Harvest  scene 5 

5.  The  Bastinado 6 

6.  Statue  from  the  Ancient  Empire 10 

7.  The  Sheikh-el-Beled 11 

8.  Hunting  in  the  INLarshes 14 

9.  Shadouf 15 

10.  The  White  Crown 16 

11.  The  Red  Crown 16 

12.  The  Pschent 16 

13.  Seti  L  in  his  War-Chariot 2t, 

14.  Rameses  IL  in  adoration  before  Seti 25 

15.  Homage  to  Amenophis  HI 26 

16.  Construction  of  a  Temple  at  Thebes 27 

17.  Columns  in  the  Hypostyle  Hall,  Karnak 28 

18.  19.  Scribes  registering  the  yield  of  the  harvest 29 

20.  Colossi  of  Amenophis  IH 3° 

21.  Scribe  registering  merchandize 3^ 

22.  Boatmen 32 

23.  Cattle  Drovers ZZ 


xii  List  of  Illustrations. 

FIG.  PAGE 

24.  Bakers 33 

25.  Women  at  a  loom 34 

26.  Netting  birds 35 

27.  Sliepherds  in  the  fields 36 

28.  Winnowing  corn 36 

29.  Herdsmen 37 

30.  From  the  tomb  of  Menofre 39 

31.  Water  Tournament 42 

32.  Mariette's  House 43 

33.  Amenhotep,  or  Anienophis  HI.,  presented  by  Phre  to  Amen-Ka    ....  45 

34.  Amen  (or  Amnion) 51 

35-  Ptah 52 

36.  Osiris      53 

37.  The  goddess  Bast 54 

38.  Painted  bas-relief 58 

39.  Sekhet 59 

40.  Isis-Hathor 60 

41.  A  Sphinx 61 

42.  Touaris , 63 

43.  Rannu 64 

44.  Horus 65 

45.  Thoth 66 

46.  Sacrifice  to  Apis 67 

47.  Statue  from  the  Ancient  Empire 73 

48.  Woman  kneading  dough 74 

49.  The  Scribe  Chaphre 75 

50.  The  Lady  Na'i 76 

51.  Ouah-ab-ra 79 

52.  Sculptor  at  work  upon  an  arm 81 

53.  Sculptor  carving  a  statue 83 

54.  Artist  painting  a  statue 85 

55.  Isis  nursing  Horus 87 

56.  Chephren 90 

57.  Ti,  with  his  wife  and  son      91 

58.  Square  building 97 

59.  Rectangular  and  oblong  building 97 

60.  The  Libyan  chain,  above  the  Necropolis  of  Thebes 98 

61.  General  appearance  of  an  Egyptian  Temple 99 

62.  Temple  of  Khons,  at  Thebes       100 

63.  Temple  of  Khons,  Thebes 100 

64.  Temple  of  Khons,  Thebes 100 

65.  From  the  second  court  of  Medinet-Abou,  Thebes 10  r 

66.  Ramesseum,  Thebes loi 

67.  The  Egyptian  Gorge  or  Cornice 102 

68.  Capital  and  Entablature  of  the  Temple  of  the  Deus  Rediculus  at  Rome  .  104 

69.  The  Egyptian  "bond" 107 

70.  Double-faced  wall loS 


List  of  Illustrations.  xiii 


FIG.  PAGE 

71,  72.  Elements  of  the  portico 108 

73.  Egyptian  construction 109 

74.  Element  of  an  off-set  arch       iii 

75.  Arrangement  of  the  courses  in  an  off-set  arch iii 

76.  Off-set  semicircular  arch iii 

77.  Voussoir 112 

78.  Arrangement  of  voussoirs 112 

79.  Semicircular  vault 112 

80.  Granaries,  from  a  bas-relief      .    .  " 113 

81.  Modern  pigeon  house,  Thebes 114 

82.  Elements  of  wooden  construction 116 

83.  Wooden  building  (first  system) 117 

84.  Wooden  building  (second  system) 118 

85.  Seti  I.  striking  prisoners  of  war  with  his  mace 124 

86.  Stele  of  the  eleventh  dynasty 131 

87.  Mummy  case  from  the  eighteenth  dynasty 137 

88.  Man  and  his  wife  in  the  style  of  the  fifth  dynasty 138 

89.  Sekhem-ka,  his  wife  Ata,  and  his  son  Khnem,  in   the  style  of  the  fifth 

dynasty 139 

90.  Stele  of  Nefer-oun 140 

91.  Preparation  of  the  victims  and  arrival  of  funeral  gifts 141 

92.  Table  for  offerings 144 

93.  Another  form  of  the  table  for  offerings 144 

94.  Labourers  heaping  up  ears  of  corn 146 

95.  96.  Sepulchral  statuettes 147 

97.  Vignette  from  a  i?////a/ upon  papyrus 149 

98.  Arrival  in  Egypt  of  a  company  of  Asiatic  emigrants 152,  153 

99.  The  tomb  of  Ti ;  women,  representing  the  lands  of  the  deceased,  carrying 

the  funeral  gifts 154 

100.  Lid  of  the  cot^n  of  Entef 158 

loi,  102.  Scarabs 159 

103,  104.  Funerary  amulets 159 

105.  Pillow 160 

106.  Actual  condition  of  a  Mastaba.     The  Tomb  of  Sabou 167 

107.  Three  mastabas  at  Gizeh 168 

108.  Restoration  of  part  of  the  Necropolis  of  Gizeh 169 

109.  The  Mastabat-el-Faraoun 170 

no.  Entrance  to  a  Mastaba  at  Sakkarah 171 

111.  Lintel  of  the  tomb  of  Teta 172 

112.  Plan  of  the  tomb  of  Ti 174 

113.  114.  Mastaba  at  Sakkarah 174 

115.  AVestern  wall  in  the  chamber  of  the  tomb  of  Ptah-Hotep 175 

116.  Plan  of  a  Mastaba  with  four  serdabs 178 

117.  Longitudinal  section  of  the  same  Mastaba 178 

118.  Transverse  section  through  the  chamber 179 

119.  Transverse  section  through  the  serdabs 179 

120.  Figures  in  high  relief,  from  a  Mastaba  at  Gizeh 180 


xiv  List  of  Illustrations. 


FIG.  PAGE 

121.  The  upper  chamber,  well,  and  mummy-chamber i8i 

122.  Double  Mastaba  at  Gizeh 182 


12 


J- 


Sarcophagus  of  Khoo-foo-Ankh 183 

124.  Details  of  the  Sarcophagus  of  Khoo-foo-Ankh 184 

125.  Bas-reHef  from  Sakkarah 185 

126.  Head  of  a  Mummy 188 

127.  Plans  of  the  temples  belonging  to  the  Second  and  Third  Pyramids  ...  193 

128.  Plan  of  the  Pyramid  of  Cheops 198 

129.  The  Great  Pyramid  and  the  small  pyramids  at  its  foot 199 

130.  The  Three  Great  Pyramids ;  from  the  south 201 

131.  The  Pyramid  of  lUahoun,  horizontal  section  in  perspective 205 

132.  Section  of  the  Pyramid  of  Cheops 206 

133.  The  southern  Pyramid  of  Dashour 207 

134.  Section  of  the  Stepped  Pyramid 207 

135.  The  Stepped  Pyramid .  208 

136 — 142,  Successive  states  of  a  pyramid 209 

143.  Section  of  the  Stepped  Pyramid  at  Sakkarah 213 

144.  Construction  of  the  Pyramid  of  Abousir  in  parallel  layers 213 

145.  Partial  section  of  the  Stepped  Pyramid 214 

146.  The  Pyramid  of  IMeidoum 215 

147.  The  Mastabat-el-Faraoun 216 

148.  Funerary  monument  represented  in  the  inscriptions 216 

149.  Plan  and  elevation  of  a  pyramid  at  Meroe 219 

150.  Method  of  closing  a  gallery  by  a  stone  portcullis 220 

151.  Portcullis  closed 220 

152.  Transverse  section,  in  perspective,    through  the   Sarcophagus-chamber 

and  the  discharging  chambers  of  the  Great  Pyramid 221 

153.  Longitudinal  section  through  the  lower  chambers 222 

[54,  Pyramidion 230 

155.  The  casing  of  the  pyramids 233 

156.  Plan  of  the  Pyramids  of  Gizeh  and  of  that  part  of  the  Necropolis  which 

immediately  surrounds  them 237 

157.  The  Sphinx 238 

158.  Pyramid  with  its  inclosure,  Abousir 239 

159.  The  river  transport  of  the  Mummy 243 

160.  Tomb  at  Abydos 244 

161.  Section  of  the  above  tomb 244 

162.  Tomb  at  Abydos 245 

163.  Section  of  the  above  tomb 245 

164.  Stele  of  the  eleventh  dynasty,  Abydos 246 

165.  Stele  of  Pinahsi,  priest  of  Ma;  Abydos 247 

166.  Facade  of  a  tomb  at  Beni-Hassan 250 

167.  Facade  of  a  tomb  at  Beni-Hassan,  showing  some  of  the  adjoining  tombs  251 

168.  Interior  of  a  tomb  at  Beni-Hassan 252 

169.  Plan  of  the  above  tomb 252 

170.  Chess  players,  Beni-Hassan 253 

171.  General  plan  of  Thebes 257 


List  op^  Illustrations.  xv 


FI^-  PAGE 

172.  Rameses  III.  conducting  a  religious  procession,  at  Medinet-Abou    ...  261 

173.  Rameses  III.  hunting 26s 

174.  Rameses  II.  in  battle -.    .    .  271 

X75.  Painting  in  a  royal  tomb  at  Gournah 273 

176.  Amenophis  III.  presenting  an  offering  to  Amen 274 

177.  Flaying  the  funerary  victim 275 

178.  Entrance  to  a  royal  tomb 277 

179.  Plan  of  the  tomb  of  Rameses  II 282 

180.  Horizontal  section  of  the  same  tomb 282 

181.  The  smaller  Sarcophagus-chamber  in  the  tomb  of  Rameses  VI 283 

182.  Entrance  to  the  tomb  of  Rameses  III 284 

183.  Hunting  scene  upon  a  tomb  at  Gournah 286 

184.  The  weighing  of  actions 287 

185.  Anubis,  in  a  funerary  pavilion 288 

186.  Plan  and  section  of  a  royal  tomb 292 

187.  188.  Theban  tombs  from  the  bas-reliefs 294 

189.  Theban  tomb  from  a  bas-relief 295 

190.  A  tomb  of  Apis 296 

191.  The  tomb  of  Petamounoph 297 

192.  The  most  simple  form  of  Theban  tomb 299 

193.  Tomb  as  represented  upon  a  bas-relief 299 

194.  Stele  in  the  Boulak  Museum,  showing  tombs  with  gardens  about  them  .    .  302 

195.  The  sarcophagus  of  a  royal  scribe 303 

196.  Canopic  vase  of  alabaster 305 

197.  View  of  the  grand  gallery  in  the  Apis  Mausoleum 306 

198.  Sepulchral  chamber  of  an  Apis  bull 308 

199.  Section  in  perspective  of  "  Campbell's  tomb  " 312 

200.  Vertical  section  in  perspective  of  the  Sarcophagus-chamber  of  the  above 

tomb 312 

201.  A  Tomb  on  El-Assasif 313 

202.  The  Temple  of  the  Sphinx 324 

203.  Interior  of  the  Temple  of  the  Sphinx 325 

204.  The  Temple  of  the  Sphinx,  the  Sphinx,  and  the  neighbouring  parts  of  the 

Necropolis 331 

205.  Ram,  or  KriospJiinx 336 

206.  Gateway  and  boundary  wall  of  a  temple 339 

207.  Principal  facade  of  the  Temple  of  Luxor 345 

208.  The  Temple   of  Khons  ;  horizontal  and   vertical   section    showing    the 

general  arrangement  of  the  temple 349 

209.  The  Bari^  or  sacred  boat 352 

210.  Portable  tabernacle  of  painted  wood 354 

211.  Granite  tabernacle 355 

212.  General  plan  of  the  Great  Temple  at  Karnak 358 

213.  Longitudinal  section  of  the  Temple  of  Luxor 361 

214.  Plan  of  the  anterior  portion  of  the  Great  Temple  at  Karnak 363 

215.  The  Great  Temple  at  Karnak  ;  inner  portion 367 

216.  Karnak  as  it  is  at  present 369 


xvi  List  of  Illustrations. 

FIG.  PAGE 

217.  Plan  of  the  Temple  of  Luxor 371 

218.  Bird's-eye  view  of  Luxor 373 

219.  Plan  of  the  Ramesseum 377 

220.  The  Ramesseum.     Bird's-eye  view  of  the  general  arrangement     ....  379 

221.  General  plan  of  the  buildings  at  Medinet-Abou 381 

222.  Plan  of  the  Temple  of  Thothmes 382 

223.  Plan  of  the  Great  Temple  at  Medinet-Abou 383 

224.  Plan  of  the  Temple  at  Abydos 387 

225.  Seti,  with  the  attributes  of  Osiris,  between  Amen,  to  whom  he  is  paying 

homage,  and  Chnoum 390 

226.  Plan  of  the  Temple  of  Gournah 392 

227.  Facade  of  the  naos  of  the  Temple  of  Gournah 393 

228.  Logitudinal  section  of  the  Temple  of  Gournah,  from  the  portico  of  the 

naos  to  the  back  wall 393 

229.  Plan  of  the  Temple  of  Elephantine 396 

230.  View  in  perspective  of  the  Temple  of  Elphantine' 397 

231.  Longitudinal  section  of  the  Temple  of  Elephantine 398 

232.  Temple  of  Amenophis  IIL  at  Eilithyia 401 

233.  Temple  of  Amenophis  III.  at  Eilithyia  ;  longitudinal  section 403 

234.  The  speos  at  Addeh • 406 

235.  The  speos  at  Addeh  ;  longitudinal  section 406 

236.  Plan  of  speos  at  Beit-el- Wali 407 

237.  Longitudinal  section  of  the  speos  at  Beit-el-Wali 407 

238.  Plan  of  the  hemispeos  of  Gherf-Hossein 408 

239.  Gherf-Hossein  ;  longitudinal  section 409 

240.  Plan  of  the  hemispeos  of  Derri 409 

241.  Longitudinal  section;  Derri 409 

242.  Facade  of  the  smaller  temple  at  Ipsamboul 411 

243.  Plan  of  the  smaller  temple 413 

244.  Perspective  of  the  principal  Chamber  in  the  smaller  temple 413 

245.  Longitudinal  section  of  the  smaller  temple 413 

246.  Plan  of  the  Great  Temple 413 

247.  Perspective  of  the  principal  Hall  in  the  Great  Temple 414 

248.  Fa9ade  of  the  Great  Temple  at  Ipsamboul 415 

249.  Longitudinal  section  of  the  Great  Temple 417 

250.  Dayr-el-Bahari 419' 

251.  Restoration  in  perspective  of  Dayr-el-Bahari 423 

252.  The  ruins  on  the  Island  of  Philae 431 

253.  The  battle  against  the  Khetas,  Luxor 436 

254.  Rameses  II.  returning  in  triumph  from  Syria 437 

255.  The  goddess  Anouke  suckling  Rameses  II.,  Beit-el-Wali 441 


INTRODUCTION. 


I. 

The  successful  interpretation  of  the  ancient  writings  of  Egypt, 
Chaldaea,  and  Persia,  which  has  distinguished  our  times,  makes 
it  necessary  that  the  history  of  antiquity  should  be  rewritten. 
Documents  that  for  thousands  of  years  lay  hidden  beneath  the 
soil,  and  inscriptions  which,  like  those  of  Egypt  and  Persia,  long 
offered  themselves  to  the  gaze  of  man  merely  to  excite  his 
impotent  curiosity,  have  now  been  deciphered  and  made  to  render 
up  their  secrets  for  the  guidance  of  the  historian.  By  the 
help  of  those  strings  of  hieroglyphs  and  of  cuneiform  characters, 
illustrated  by  paintings  and  sculptured  reliefs,  we  are  enabled  to 
separate  the  truth  from  the  falsehood,  the  chaff  from  the  wheat,  in 
the  narratives  of  the  Greek  writers  who  busied  themselves  with 
those  nations  of  Africa  and  Asia  which  preceded  their  own  in  the 
ways  of  civilization.  Day  by  day,  as  new  monuments  have  been 
discovered  and  more  certain  methods  of  reading  their  inscriptions 
elaborated,  we  have  added  to  the  knowledge  left  us  by  Herodotus 
and  Diodorus  Siculus,  to  our  acquaintance  with  those  empires  on 
the  Euphrates  and  the  Nile  which  were  already  in  old  age  when 
the  Greeks  were  yet  struggling  to  emerge  from  their  primitive 
barbarism. 

Even  in  the  cases  of  Greece  and  Rome,  whose  histories  are 
supplied  in  their  main  lines  by  their  classic  writers,  the  study 
of   hitherto  neglected  writings    discloses    many    new  and    curious 

VOL.  I.  b 


ii  Introduction. 


details.  The  energetic  search  for  ancient  inscriptions,  and  the 
scrupulous  and  ingenious  interpretation  of  their  meaning,  which  we 
have  witnessed  and  are  witnessing,  have  revealed  to  us  many 
interesting  facts  of  which  no  trace  is  to  be  found  in  Thucydides  or 
Xenophon,  in  Livy  or  Tacitus  ;  enabling  us  to  enrich  with  more  than 
one  feature  the  picture  of  private  and  public  life  which  they  have 
handed  down  to  us.  In  the  effort  to  embrace  the  life  of  ancient 
times  as  a  whole,  many  attempts  have  been  made  to  fix  the  exact 
place  in  it  occupied  by  art,  but  those  attempts  have  never  been 
absolutely  successful,  because  the  comprehension  of  works  of  art, 
oi plastic  creations  in  the  widest  significance  of  that  word,  demands 
an  amount  of  special  knowledge  which  the  great  majority  of  his- 
torians are  without ;  art  has  a  method  and  language  of  its  own, 
which  obliges  those  who  wish  to  learn  it  thoroughly  to  cultivate 
their  taste  by  frequenting  the  principal  museums  of  Europe,  by 
visiting  distant  regions  at  the  cost  of  considerable  trouble  and 
expense,  by  perpetual  reference  to  the  great  collections  of  en- 
gravings, photographs,  and  other  reproductions  which  considera- 
tions of  space  and  cost  prevent  the  savant  from  possessing  at 
home.  More  than  one  learned  author  has  never  visited  Italy  or 
Greece,  or  has  found  no  time  to  examine  their  museums,  each  of 
which  contains  but  a  small  portion  of  the  accumulated  remains  of 
antique  art.  Some  connoisseurs  do  not  even  live  in  a  capital, 
but  dwell  far  from  those  public  libraries,  which  often  contain 
valuable  collections,  and  sometimes — when  they  are  not  packed 
away  in  cellars  or  at  the  binder's — allow  them  to  be  studied  by 
the  curious.^  The  study  of  art,  difficult  enough  in  itself,  is  thus 
rendered  still  more  arduous  by  the  obstacles  which  are  thrown  in 
its  way.  The  difficulty  of  obtaining  materials  for  self-improve- 
ment in  this  direction  affords  the  true  explanation  of  the  absence, 
in  modern  histories  of  antiquity,  of  those  laborious  researches 
which  have  led  to  such  gfreat  results  since  Winckelmann  founded 
the  science  of  archaeology  as  we  know  it.  To  take  the  case  of 
Greece,  many  learned  writers  have  in  our  time  attempted  to  retrace 
its  complete  history — England,  Germany,  and  France  have  each 
contributed  works  which,  by  various  merits,  have  conquered  the 
favour  of  Europe.  But  of  all  these  works  the  only  one  which 
betrays  any  deep  study  of  Greek  art,  and  treats  it  with  taste  and 

1  Our  national  library  at  the  British  Museum  is,  perhaps,  the  only  one  which 
does  not  deserve  this  reproach. — Ed. 


Introduction.  in 


competence,  is  that  of  M.  Ernest  Curtius  ;  as  for  Mr.  Grote,  he  has 
neither  a  theoretic  knowledge  of  art,  nor  a  feehncr  for  it.  Here 
and  there,  indeed,  where  he  cannot  avoid  it,  he  alludes  to  the 
question,  but  in  the  fewest  and  driest  phrases  possible.  And  yet 
Greece,  without  its  architects,  its  sculptors,  and  its  painters,  with- 
out in  fact  its  passion  for  beautiful  form,  a  passion  as  warm  and 
prolific  as  its  love  for  poetry,  is  hardly  Greece  at  all. 

Much  disappointment  is  thus  prepared  for  those  who,  without 
the  leisure  to  enter  deeply  into  detail,  wish  to  picture  to  themselves 
the  various  aspects  of  the  ancient  world.  They  are  told  of 
revolutions,  of  wars  and  conquests,  of  the  succession  of  princes  ; 
the  mechanism  of  political  and  civil  institutions  is  explained  to 
them ;  "  literature,"  we  are  told,  "  is  the  expression  of  social 
life,"  and  so  the  history  of  literature  is  written  for  us.  All  this 
is  true  enough,  but  there  is  another  truth  which  seems  to  be 
always  forgotten,  that  the  art  of  a  people  is  quite  as  clear  an 
indication  of  their  sentiments,  tastes,  and  ideas,  as  their  litera- 
ture. But  on  this  subject  most  historians  say  little,  contenting 
themselves  with  the  brief  mention  of  certain  works  and  proper 
names,  and  with  the  summary  statement  of  a  few  general  ideas 
which  do  not  even  possess  the  merit  of  precision.  And  where 
are  we  to  find  the  information  thus  refused  ?  Europe  possesses 
several  histories  of  Greek  and  Roman  literature,  written  with 
great  talent  and  eloquence,  such  as  the  work,  unhappily  left 
unfinished,  of  Ottfried  M tiller  ;  there  are,  too,  excellent  manuals, 
rich  in  valuable  facts,  such  as  those  of  Bernhardy,  Baehr,  and 
Teuffel ;  but  where  is  there,  either  in  England,  in  France,  or 
in  Germany,  a  single  work  which  retraces,  in  sufficient  detail, 
the  whole  history  of  antique  art,  following  it  throughout  its 
progress  and  into  all  its  transformations,  from  its  origin  to  its 
final  decadence,  down  to  the  epoch  when  Christianity  and  the 
barbaric  invasions  put  an  end  to  the  ancient  forms  of  civilization 
and  prepared  for  the  birth  of  the  modern  world,  for  the  evolution 
of  a  new  society  and  of  a  new  art  ? 

To  this  question  our  neighbours  may  reply  that  the  Geschichte 
der  bildenden  Ktinst  of  Carl  Schnaase  ^  does  all  that  we  ask.  But 
that   work   has    one   great    disadvantage    for  those  who    are  not 

^  Geschichte  der  bilde?iden  Kunsf,  2nd  ed.,  corrected  and  augmented,  with  wood 
engravings  in  the  text,  8  vols.  8vo.  1865-1873.  The  first  edition  consisted  of  7  vols., 
and  appeared  between  1843  and  1864. 


iv  Introduction. 


Germans.  Its  great  bulk  will  almost  certainly  prevent  its  ever 
finding  a  translator,  while  it  makes  it  very  tedious  reading 
to  a  foreigner.  It  must,  besides,  be  very  difficult,  not  to  say 
impossible,  for  a  single  writer  to  treat  with  equal  competence 
the  arts  of  Asia,  of  Greece,  and  of  Rome,  of  the  Middle  Ages 
and  of  modern  times.  As  one  might  have  expected,  all  the  parts 
of  such  an  extensive  whole  are  by  no  means  of  equal  value, 
and  the  chapters  which  treat  of  antique  art  are  the  least  satisfac- 
tory. Of  the  eight  volumes  of  which  the  work  consists,  two  are 
devoted  to  ancient  times,  and,  by  general  acknowledgment,  they 
are  not  the  two  best.  They  were  revised,  indeed,  for  the 
second  edition,  by  two  colleagues  whom  Herr  Schnaase  called 
in  to  his  assistance  ;  oriental  art  by  Carl  von  Liitzow,  and  that 
of  Greece  and  Rome  by  Carl  Friedrichs.  But  the  chapters  in 
which  Assyria,  Chaldaea,  Persia,  Phoenicia,  and  Egypt  are  discussed 
are  quite  inadequate.  No  single  question  is  exhaustively  treated. 
Instead  of  well-considered  personal  views,  we  have  vague  guesses 
and  explanations  which  do  nothing  to  solve  the  many  problems 
which  perplex  archaeologists.  The  illustrations  are  not  numerous 
enough  to  be  useful,  and,  in  most  cases,  they  do  not  seem  to  have 
been  taken  from  the  objects  themselves.  Those  which  relate  to 
architecture,  especially,  have  been  borrowed  from  other  well 
known  works,  and  furnish  therefore  no  new  elements  for 
appreciation  or  discussion.  Finally,  the  order  adopted  by  the 
author  is  not  easily  understood.  For  reasons  which  have  decided 
us  to  follow  the  same  course,  and  which  we  will  explain  farther 
on,  he  takes  no  account  of  the  extreme  east,  of  China  and  Japan  ; 
but  then,  why  begin  with  India,  which  had  no  relations  with  the 
peoples  on  the  shores  of  the  Mediterranean  until  a  very  late  date, 
and,  so  far  as  art  was  concerned,  rather  came  under  their  influence 
than  brought  them  under  its  own  ? 

The  fact  is  that  Schnaase  follows  a  geographical  order,  which 
is  very  confusing  in  its  results.  To  give  but  one  example  of  its 
absurdity,  he  speaks  of  the  Phoenicians  before  he  has  said  a 
word  of  Egypt  ;  now,  we  all  know  that  the  art  of  Tyre  and 
Sidon  was  but  a  late  reflection  from  that  of  Egypt  ;  the  work- 
shops of  those  two  famous  ports  were  mere  factories  of  cheap 
Egyptian  art  objects  for  exportation. 

Again,  the  first  part  of  Herr  Schnaase's  work  is  already  seventeen 
years  old,  and  how  many  important  discoveries  have  taken  place 


Introduction. 


since  1865  ?  Those  of  Cesnola  and  Schliemann,  for  instance, 
have  revealed  numberless  points  of  contact  and  transmission 
between  one  phase  of  antique  art  and  another,  which  were  never 
thought  of  twenty  years  ago.  The  book  therefore  is  not  "  down 
to  date."  With  all  the  improvements  which  a  new  edition  might 
introduce,  that  part  of  it  which  deals  with  antiquity  can  never 
be  anything  but  an  abridgment  with  the  faults  inherent  in  that 
kind  of  work.  It  could  never  have  the  amplitude  of  treatment 
or  the  originality  which  made  Winckelmann's  History  of  Art  and 
Ottfried  Mliller's  Manual  of  Artistic  ArchcEology  so  successful  in 
their  day.^ 

Winckelmann's  History  of  Art  among  the  Ancients,  originally 
published  in  1 764,  is  one  of  those  rare  books  which  mark  an  epoch 
in  the  history  of  the  human  intellect.  The  German  writer  was 
the  first  to  formulate  the  idea,  now  familiar  enough  to  cultivated 
intelligences,  that  art  springs  up,  flourishes,  and  decays,  with  the 
society  to  which  it  belongs  ;   in  a  word,  that  it  is  possible  to  write 

^  Germany  had  long  felt  the  want  which  Schnaase  attempted  to  satisfy.  As  early 
as  1841  Franz  Kugler  published  his  Handlnidi  der  Kwistgeschichte,  which  embraces 
the  whole  history  of  art  from  the  earliest  times  down  to  our  own  day.  The  book 
was  successful ;  the  fourth  edition,  revised  and.  corrected  by  Wilhelm  Liibke  (2  vols. 
8vo.  1861,  Stuttgart),  lies  before  us,  but  to  give  an  idea  of  its  inadequacy  as  a  history 
of  ancient  art,  it  is  enough  to  say  that  the  whole  of  the  antique  period,  both  in 
Greece  and  Asia,  occupies  no  more  than  206  pages  of  the  first  volume.  The  few 
illustrations  are  not  very  good  in  quality,  and  their  source  "is  never  indicated  ;  the 
draughtsman  has  taken  little  care  to  reproduce  with  fidelity  the  style  of  the  originals 
or  to  call  attention  to  their  peculiarities ;  finally,  the  arrangements  adopted  betray 
the  defects  of  a  severely  scientific  method.  The  author  commences  with  Celtic 
monuments  (dolmens  and  menhirs),  and  then  passes  to  the  structures  of  Oceania 
and  America  ;  before  commencing  upon  Egypt  he  takes  us  to  Mexico  and  Yucatan. 
Liibke,  whilst  still  occupied  with  the  work  of  Kugler,  wished  to  supply  for  the  use 
of  students  and  artists  a  book  of  a  more  elementary  character  ;  he  therefore  published 
in  i860  an  8vo  volume  which  he  called  Grundriss  der  Ktinstgeschichte ;  the  antique 
here  occupies  208  pages  out  of  720.  His  plan  seems  to  us  to  be  open  to  the  same 
objection  as  that  of  Kugler ;  he  follows  a  geographical  instead  of  an  historical 
arrangement ;  he  begins  with  the  extreme  east ;  he  puts  the  Assyrians  and  the 
Persians  before  Egypt,  and  India  before  Assyria.  His  illustrations  are  sometimes 
better  than  those  of  Kugler,  but  many  of  the  cuts  are  common  to  both  works. 

Under  the  title  Geschichte  der  Flasfik,  Overbeck  and  Liibke  have  each  written  a 
comprehensive  history  of  sculpture.  [The  word  "  comprehensive  "  must  here  be 
understood  in  a  strictly  limited  sense. — Ed.]  The  word  Plastik  in  the  language 
of  German  critics  has  this  special  and  restricted  meaning — it  comprises  sculpture 
only.  The  work  of  Overbeck,  far  superior  to  that  of  Liibke,  deserves  the  success 
which  has  attended  it ;  the  third  edition,  which  contains  the  results  of  the  searches 
at  Olympia  and  at  Pergamus,  is  now  in  course  of  publication. 


vi  Introduction. 


its  history.^  This  great  savant,  whose  memory  Germany  holds  in 
honour  as  the  father  of  classic  archaeology,  was  not  content  with 
stating  a  principle  :  he  followed  it  through  to  its  consequences  ; 
he  beo-an  by  tracing  the  outlines  of  the  science  which  he  founded, 
and  he  never  rested  till  he  had  filled  them  in.  However,  now  that 
a  century  has  passed  away  since  it  appeared,  his  great  work,  which 
even  yet  is  never  opened  without  a  sentiment  of  respect,  marks  a 
date  beyond  which  modern  curiosity  has  long  penetrated.  Winckel- 
mann's  knowledge  of  Egyptian  art  was  confined  to  the  pasticcios 
of  the  Roman  epoch,  and  to  the  figures  which  passed  from  the 
villa  of  Hadrian  to  the  museum  of  Cardinal  Albani.  Chaldsea  and 
Assyria,  Persia  and  Phoenicia,  had  no  existence  for  him  ;  even 
Greece  as  a  whole  was  not  known  to  him.  Her  painted  vases 
were  still  hidden  in  Etruscan  and  Campanian  cemeteries  ;  the  few 
which  had  found  their  way  to  the  light  had  not  yet  succeeded  in 
drawing  the  attention  of  men  who  were  preoccupied  over  more  im- 
posing manifestations  of  the  Greek  genius.  Nearly  all  Winckel- 
mann's  attention  was  given  to  the  works  of  the  sculptors,  upon 
which  most  of  his  comprehensive  judgments  were  founded  ;  and 
yet,  even  in  regard  to  them,  he  was  not  well-informed.  His 
opportunities  of  personal  inspection  were  confined  to  the  figures, 
mostly  of  unknown  origin,  which  filled  the  Italian  galleries. 
The  great  majority  of  these  formed  part  of  the  crowd  of 
copies  which  issued  from  the  workshops  of  Greece,  for  some 
three  centuries  or  more,  to  embellish  the  temples,  the  basilicas, 
and  the  public  baths,  the  villas  and  the  palaces  of  the  masters 
of  the  world.  In  the  very  few  instances  in  which  they  were 
either  originals  or  copies  executed  with  sufficient  care  to  be 
fair  representations  of  the  original,  they  never  dated  from  an 
earlier  epoch  than  that  of  Praxiteles.  Scopas,  and  Lysippus. 
Phidias     and    Alcamenes,    Pseonius     and     Polycletus,    the    great 

1  Winckelmann's  History  of  Ancient  Art  should  be  read  in  connection  with  his 
Remarks  upon  the  History  of  Art,  which  is  a  kind  of  supplement  to  it,  and 
takes  the  place  of  that  new  edition  of  which  the  author's  premature  and  tragic  death 
deprived  the  world.  It  is  an  answer  to  the  objections  which  made  themselves  heard 
on  every  side  ;  the  preface  to  Momtmentiinediti  (Rome,  1867,  2  vols,  in  folio,  with 
208  plates)  should  also  be  read.  The  method  of  Winckelmann  is  there  most  clearly 
explained.  Finally,  the  student  of  the  life  and  labours  of  Winckelmann  may  consult 
with  profit  the  interesting  work  of  Carl  lusti,  Wi?ickeltriann,  sein  Leben,  seine  Werkc, 
und  seme  Zeifgenossen,  which  will  give  him  a  clear  idea  of  the  state  of  archeology 
at  the  time  when  the  German  sava7it  intervened  to  place  it  upon  a  higher  footing. 


Introduction.  vii 


masters  of  the  fifth    century,  were    only  known    to    the    historian 
by  the  descriptions  and  allusions  of  the  ancient  authors. 

In  such  a  case  as  this  the  clearest  and  most  precise  of  verbal 
descriptions  is  of  less  value  than  any  fragment  of  marble  upon 
which  the  hand  of  the  artist  is  still  to  be  traced.  Who  would 
then  have  guessed  that  the  following  generation  would  have 
the  opportunity  of  studying  those  splendid  groups  of  decorative 
sculpture  whose  close  relation  to  the  architecture  of  certain  famous 
temples  has  taught  us  so  much  ?  Who  in  those  days  dreamt 
of  looking  at,  still  less  of  drawing,  the  statues  in  the  pediments 
and  sculptured  friezes  of  the  Parthenon,  of  the  Thesseum,  of  the 
temples  at  yEgina,  at  Phigalia,  or  at  Olympia  ?  Now  if  Winckel- 
mann  was  ignorant  of  these,  the  real  monuments  of  classic 
perfection,  it  follows  that  he  was  hardly  competent  to  recognise 
and  define  true  archaism  or  to  distinguish  the  works  of  sculpture 
which  bore  the  marks  of  the  deliberate,  eclectic,  and  over-polished 
taste  of  the  critical  epochs.  He  made  the  same  mistake  in  speak- 
ing of  architecture.  It  was  always,  or  nearly  always,  by  the  edifices 
of  Rome  and  Italy,  by  their  arrangement  and  decoration,  that  he 
pretended  to  explain  and  judge  the  architecture  of  Greece. 

But  Winckelmann  rendered  a  great  service  to  art  by  founding 
a  method  of  study  which  was  soon  applied  by  Zoega  -^  and  by 
Ennio  Quirino  Visconti,'-^  to  the  description  of  the  works  which 
filled  public  and  private  galleries,  or  were  being  continually  dis- 
covered by  excavation.  These  two  savants  classified  a  vast 
quantity  of   facts  ;    thanks    to    their    incessant    labours,  the    lines 

^  Zoega  busied  himself  greatly  with  Egypt,  and  in  inaugurating  the  study  of 
Coptic  prepared  the  way  for  ChampoUion.  But  the  work  which  gave  him  a 
place  among  the  chief  scholars  of  Winckelmann  is  unfinished ;  the  Bassirilievi 
antichi  di  Roma  (Rome,  2  vols.  4to.  1808)  only  contains  the  monuments  in  the  Villa 
Albani,  engraved  by  Piroli,  with  the  help  of  the  celebrated  Piranesi.  A  volume 
containing  most  of  his  essays  was  given  to  the  world  by  Welcker  in  181 7  {Abhand- 
lungen  herausgegeben  und  mit  Zusdtsen  begleifet,  8vo.  Gottingen),  who  also  published 
his  life  and  a  volume  of  his  correspondence  (Zoega,  Sammluiig  seiner  Briefe  und 
Betirtheihing  seiner  Werke.     2  vols.  8vo.      Stuttgart,  18 19). 

2  //  Mi/seo  Fio- Clementina,  Visconti,  vol.  i.  1782  ;  by  Enn.  Quir.  Visconti,  vols, 
ii.  to  vii.  Rome,  1784  1807  Museutn  Worsleyamwi,  2  vols,  folio.  London, 
1794.  Moniimenti  Gabini  della  Villa  Pinciana,  Visconti,  8vo.  1797.  Description 
des  Antiques  du  Musee  Royal,  begun  by  Visconti  and  continued  by  the  Comte  de 
Clarac.  12 mo.  Paris,  1820.  For  the  collection  of  the  materials  and  the  execution 
of  the  plates  in  the  Jconographie  Grecque  et  Romain,  Visconti  took  advantage  of 
his  opportunities  as  director  of  the  Musce  Napoleon,  into  which  the  art  treasures  of 
all  Europe,  except  England,  were  collected  at  the  beginning  of  this  century. 


viii  Introduction. 


of  the  master's  rough  sketch  were  accented  and  corrected  at 
more  than  one  point ;  the  divisions  which  he  had  introduced 
into  his  picture  were  marked  with  greater  precision  ;  the  groups 
which  he  had  becfun  to  form  were  rendered  more  coherent  and 
compact ;  their  features  became  more  precise,  more  distinct, 
and  more  expressive.  This  progress  was  continuous,  but  after 
the  great  wars  of  the  Revolution  and  the  Empire  its  march  be- 
came much  more  rapid,  and  the  long  peace  which  saw  the 
growth  of  so  rich  a  harvest  of  talent,  was  also  marked  by  a 
great  increase  in  the  energy  with  which  all  kinds  of  historical 
studies  were  prosecuted. 

But  the  widest,  as  well  as  the  most  sudden,  enlargement  of 
the  horizon  was  due  to  a  rapid  succession  of  discoveries,  some 
the  result  of  persevering  searches  and  lucky  excavations,  others 
rendered  possible  by  feats  of  induction  which  almost  amounted 
to  genius.  It  seemed  as  though  a  curtain  were  drawn  up,  and, 
behind  the  rich  and  brilliant  scenery  of  Grseco-Roman  civilization, 
the  real  ancient  world,  the  world  of  the  East,  the  father  of 
religions  and  of  useful  inventions,  of  the  alphabet  and  of  the 
plastic  arts,  were  suddenly  revealed  to  us.  The  great  work 
which  was  compiled  by  the  savants  who  accompanied  Bona- 
parte to  Egypt  first  introduced  the  antiquities  of  that  country 
to  us,  and  not  long  afterwards  Champollion  discovered  the  key  to 
the  hieroglyphics,  and  thus  enabled  us  to  assign  to  the  monuments 
of  the  country  at  least  a  relative  date. 

A  little  later  Layard  and  Botta  freed  Nineveh  from  the  ruins  of 
its  own  buildings,  and  again  let  in  the  light  upon  ancient  Assyria. 
But  yesterday  we  knew  nothing  beyond  the  names  of  its  kings,  and 
yet  it  sprang  again  to  the  day,  its  monuments  in  marvellous 
preservation,  its  history  pictured  by  thousands  of  figures  in 
relief  and  narrated  by  their  accompanying  inscriptions.  These 
did  not  long  keep  their  secrets  to  themselves,  and  their  in- 
terpretation enables  us  to  classify  chronologically  the  works  of 
architecture  and  sculpture  which  have  been  discovered. 

The  information  thus  obtained  was  supplemented  by  careful 
exploration  of  the  ruins  in  Babylonia,  lower  Chaldaea,  and 
Susiana.  These  had  been  less  tenderly  treated  by  time  and 
by  man  than  the  remains  of  Nineveh.  The  imposing  ruins 
of  the  palace  at  Persepolis  and  of  the  tombs  of  the  kings,  had 
been  known  for  nearly  two  centuries,  but  only  by  the  inadequate 


Introduction. 


IX 


descriptions  and  feeble  drawings  of  early  travellers.  Ker- Porter, 
Texier,  and  Flandrin  provided  us  with  more  accurate  and  com- 
prehensive descriptions,  and,  thanks  to  their  careful  copies  of 
the  writings  upon  the  walls  of  those  buildings,  and  upon  the 
inscribed  stones  of  Persia  and  Media,  Eugene  Burnouf  succeeded 
in  reconstructing  the  alphabet  of  Darius  and  Xerxes. 

Thus,  to   the   toils   of  artists   and   learned  men,  who   examined 
the    country   from    the    mountains    of   Armenia    to    the    low    and 
marshy  plains  of  Susiana,  and  from  the  deserts  which  border  the 
Euphrates  to  the  rocks  of  Media  and  Persia,  and  to  the  philologists 
who  deciphered  the  texts  and  classified  the  monumental  fragments 
which  had  travelled  so  far  from   the   scene  of  their  creation,  we 
owe  our  power  to  describe,  upon  a  sound  basis  and  from  authentic 
materials,  the  great  civilisation  which  was  developed  in  Western 
Asia,  in  the  basin  of  the  Persian  Gulf     There  were   still   many 
details  which  escaped  us,  but,  through   the  shadows   which  every 
day  helped  to   dissipate,   the  essential    outlines   and    the    leading 
masses  began   to  be   clearly   distinguished,  and  the  local   distinc- 
tions which,  in    such    a    vast    extent    of   country  and    so    long    a 
succession    of   empires,   were    caused    by  differences    of   race,   of 
time,  and  of  physical  conditions,  began  to  be  appreciated.     But,  in 
spite  of  all  these  differences,  the  choice  of  expressive  means  and 
their  employment,  from  Babylon  to  Nineveh,  and  from  Nineveh  to 
Susa  and  Persepolis,  presented  so  many  points  of  striking  similarity 
as  to  prove  that  the  various  peoples  represented  by  those  famous 
capitals  all  sprang  from  the  same  original  stock.      The  elements  of 
writing  and  of  the  arts  are  in  each  case  identical.      The  alphabets 
were  all  formed  upon  the  same  cuneiform  principle,  notwithstand- 
ing  the  variety  in   the    languages    which    they  served.       In    the 
plastic     arts,     although     the    plans     of    their    buildings    vary    in 
obedience     to     the     requirements     of     different     materials,    their 
sculpture    always    betrays    the    same    way   of    looking    at    living 
forms,    the    same    conventions    and    the    same    motives.       Every 
work  fashioned  by  the  hand  of  man  which   has   been   discovered 
within   the   boundaries  given   above,  displays  community  of  style 
and  unity  of  origin  and  tradition. 

The  result  of  these  searches  and  discoveries  was  to  show 
clearly  that  this  ancient  civilisation  had  sprung  from  two 
original  sources,  the   one   in   the  valley  of  the    Nile,  the   other   in 

VOL.    1.  c 


Introduction. 


Chaldaea.  The  latter  was  the  less  ancient  of  the  two,  and  was 
considerably  nearer  our  own  time  than  the  epoch  which  witnessed 
the  commencement  of  the  long  series  of  Egyptian  dynasties  by 
the  reien  of  Menes.  These  two  civilizations  met  and  inter- 
mingled  through  the  agency  of  the  Phoenicians,  and  an  active 
and  prolific  interchange  of  ideas  and  products  began,  traces  of 
which  are  still  to  be  found  both  in  Egypt  and  Assyria. 

It  still  remained  doubtful,  and  the  doubt  has  but  lately  been 
removed,  how  the  influence  of  these  two  great  centres  of  culti- 
vation was  extended  to  the  still  barbarous  tribes,  the  ancestors  of 
the  Greeks  and  Romans,  who  inhabited  the  northern  and  eastern 
shores  of  the  Mediterranean. 

It  is  only  within  the  last  twenty  years,  since  the  mission  of 
M.  Renan,  that  Phoenicia  has  become  well-known  to  us.  Several 
English  and  French  travellers,  Hamilton,  Fellows,  Texier,  among 
others,  had  already,  in  the  first  half  of  the  century,  described 
the  curious  monuments  of  Lydia,  Phrygia,  Cappadocia,  and  of 
the  still  more  picturesque  Lycia,  whose  spoils  now  enrich  the 
British  Museum  ;  people  vaguely  conjectured  that  through  those 
countries  had  progressed,  stage  by  stage,  from  the  east  to  the 
west,  the  forms  and  inventions  of  a  system  of  civilization  which 
had  been  elaborated  in  the  distant  Chaldaea,  But  it  was  not  till 
1 86 1  that  an  expedition,  inspired  by  the  desire  to  clear  up  this 
very  question,  succeeded  in  demonstrating  the  role  actually  played 
by  the  peoples  inhabiting  the  plateau  of  Asia  Minor.  As  for 
Cyprus,  it  was  but  yesterday  that  the  explorations  of  Lang  and 
Cesnola  revealed  it  to  us,  with  its  art  half  Egyptian  and  half 
Assyrian,  and  its  cuneiform  alphabet  pressed  into  the  service  of 
a  Greek  dialect.  These  discoveries  have  put  us  on  the  alert. 
Not  a  year  passes  without  some  lucky  "  find,"  such  as  that  of  the 
Palaistrina  treasure,  in  the  immediate  neighbourhood  of  Rome, 
or  that  made  by  Salzmann  at  Rhodes.  These  pieces  of  good 
fortune  allow  the  archaeologist  to  supply,  one  by  one,  the  missing 
links  of  the  chain  which  attaches  the  arts  of  Greece  and  Italy 
to  the  earlier  civilizations  of  Egypt  and  Assyria. 

While  the  remains  of  Oriental  antiquity  were  being  thus  recovered 
piece  by  piece,  secrets  no  less  interesting  and  documents  no  less 
curious  were  continually  coming  to  the  surface  to  cast  new  light 
upon  the  history  of  classic  antiquity.      First  came  the  marbles  of 


Introduction.  xi 


the  Parthenon,  transferred  by  Lord  Elgin  to  the  British  Museum 
in  1816.  Both  artists  and  connoisseurs,  after  a  short  pause  of 
hesitation,  agreed  in  asserting  that  the  bas-rehefs  of  the  frieze  and 
the  sculptures  of  the  two  pediments  excelled  anything  which  had 
previously  entered  into  any  European  museum.  Artists  declared 
that  they  experienced  a  sense  of  beauty  never  felt  before  ;  they 
were  face  to  face  for  the  first  time  with  the  ideal  of  the  Greeks, 
as  it  had  been  conceived  and  realised  at  that  happy  period  of 
perfection  which  followed  the  disappearance  of  the  last  traces  of 
archaic  hardness.  That  period  w^s  but  too  short.  It  was  com- 
prised in  a  single  generation,  which  was  followed  by  one  which 
made  the  first  steps  down  the  slope  of  the  decadence.  During  a 
single  lifetime  a  crowd  of  works  were  produced  which,  in  spite  of 
differences  in  material  and  subject,  were  all  stamped  with  the  same 
character  of  easy  and  frank  nobility,  of  sincerity  and  elegant 
severity,  of  simplicity  combined  with  grandeur.  The  death,  or 
even  the  old  age  of  the  great  men  who  had  produced  these  works, 
was  sufficient  to  lower  the  standard.  Emphasis  and  a  striving  for 
effect  took  the  place  of  nobility ;  under  a  pretence  of  sincerity, 
artists  took  to  a  servile  imitation  of  nature,  and  mannerism,  with 
all  its  weaknesses,  beean  to  disfio^ure  their  works.  Art  remained 
at  a  high  level  in  Greece,  however,  longer  than  elsewhere.  The 
word  decadence  can  hardly  be  pronounced  in  connection  with  the 
admirable  works  produced  in  the  fourth  century  before  Christ,  and 
yet  it  cannot  be  denied  that,  so  long  as  we  were  without  original 
examples  from  the  great  epoch  of  Pericles,  we  were  without  that 
most  necessary  material  for  a  history  of  Greek  art,  a  knowledge 
of  the  most  masterly,  the  most  pure,  and  the  most  elevated  of  her 
creations.  The  literary  historian  might  as  well  have  attempted  to 
trace  the  course  of  her  poetry  without  having  read  Sophocles, 
without  having  heard  of  the  Elcctra  or  the  CEdiptis  Rex. 

Attention  being  once  turned  in  this  direction,  discoveries 
followed  each  other  in  rapid  succession.  The  statues  from  the 
pediments  at  yEgina,  so  ably  restored  by  Thorwaldsen,  were 
boupfht  to  form   the  nucleus   of  the  collection  at  Munich.^     The 

o 

1  They  were  discovered  in  181 1  amid  the  ruins  of  one  of  the  temples  at  yEgina, 
by  a  company  of  excavators  presided  over  by  Mr.  Cockerell.  They  were  boiigiTt 
by  Prince  Louis  of  Bavaria  in  181 2,  and  Thorwaldsen  was  occupied  during  several 
years  in  putting  together  and  restoring  them.  They  were  first  exhibited  in  the 
Glyptothek  of  Munich  in  1820. 


xil  Introduction. 


study  of  these  statues  is  very  instructive  in  making  clear  to  us  the 
paths  which  sculptors  had  to  follow  in  their  progress  from  the 
stiffness  and  conventions  of  early  periods  to  the  ease  and  amplitude 
of  classic  perfection.  As  for  the  friezes  from  the  temple  of- 
Apollo  Epicurius,  near  Phigalia,  they  too  are  in  the  British 
Museum.^  Thus  brought  into  immediate  propinquity  with  the 
marbles  from  the  Parthenon,  with  which  they  are  almost  cotem- 
porary,  they  afford  us  some  curious  information.  They  show  us 
what  the  art  of  Phidias  and  Alcamenes  became  when  those 
sculptors  had  to  work  in  what  we  should  call  "  the  provinces  ;  "  how 
much  they  preserved  and  how  much  they  lost  of  their  complete 
excellence  when  employed  upon  buildings  erected  at  less  cost  and 
with  less  care  than  those  of  the  capital.  So  far  as  the  composition 
is  concerned,  the  consummate  facility  and  the  natural  verve  of  the 
master  who  supplied  the  sketches  and  models  is  never  absent,  but 
the  execution,  which  must  have  been  left  to  local  artists,  betrays 
their  inferiority  by  its  inequalities  and  general  weakness.  The 
same  may  be  said  of  the  figures  with  which  Alcamenes  and 
Paeonius  ornamented  the  pediments  and  metopes  of  the  Temple 
of  Zeus  at  Olympia.  Even  before  the  discoveries  at  ^E^gina  and 
Phigalia,  the  results  of  the  French  expedition  to  the  Morea  and 
the  beautiful  fragments  of  sculpture  brought  to  the  Louvre  from 
the  banks  of  the  Alphaeus,  had  given  us  reason  to  suspect  this 
inferiority  of  provincial  art,  and  the  excavations  recently  under- 
taken by  Germany,  after  an  interval  of  about  half  a  century  of 
inaction,  have  finally  removed  all  doubts.  Neither  the  statues  nor 
the  bas-reliefs,  nor  any  other  part  of  the  decoration  of  the  temple 
at  Olympia,  possess  the  nobility  and  purity  which  distinguish  the 
great  buildings  on  the  Athenian  acropolis.  They  show  abundant 
power  and  science,  but  also  perceptible  inequalities,  and  certain 
signs  of  that  exaggerated  objectivity  which  we  now  call  realism. 
Each  fresh  discovery  helps  us  to  comprehend,  not  without  a  cer- 
tain sense  of  surprise,  how  much  freedom  and  variety  Greek  art 
possessed  during  its  best  time.  There  is  none  of  that  dull 
uniformity  which,  with  other  races,  distinguishes  most  of  the 
works  of  a  single  epoch,  none  of  the  tyranny  of  a  single  master  or 
school,  none  of  the  narrowness  of  va^x^  foinnulce. 

1  The  debris  of  the  temple  at  Bassge  was  explored  by  the  same  company  in  the 
year  1812,  and  a  whole  frieze  was  found,  which  was  bought  by  the  British  Museum 
in  1815. 


Introduction.  xiii 


The  memorable  exploration  to  which  we  have  alluded,  and 
many  others  which  it  would  take  too  long  to  enumerate,  have  not 
only  made  known  to  us  the  most  original  and  most  fertile  period  of 
Greek  sculpture,  but  have  given  us  much  information  as  to  that  art 
which,  when  combined  with  the  statues  of  Phidias  and  Alcamenes, 
reared  those  splendid  creations  which  have  been  reconstructed 
with  such  skill  and  care  by  the  artist  and  the  archaeologist ;  we 
mean  Greek  architecture  at  its  best,  the  purest  and  the  most  com- 
plete architecture  which  the  world  has  yet  seen.  Every  year  sees 
the  excellent  example  set  by  Stuart  and  Revett,^  in  the  second 
half  of  the  eighteenth  century,  followed  by  an  increasing  number 
of  imitators.  The  smallest  remains  of  ancient  architecture  are 
measured  and  drawn  with  religious  care  ;  their  arrangements  are 
explained,  their  elements  are  grouped,  their  ensemble  is  restored 
with  a  comprehension  of  their  artistic  conditions  which  steadily 
gains  in  certainty  and  penetration.  Blouet's  interesting  restora- 
tions of  Olympia  and  Phigalia,  published  in  the  account  of  the 
French  expedition  to  the  Morea,^  excited  the  emulation  of  the 
young  architects  at  the  French  Academy  in  Rome,  and  opened  to 
them  a  new  course  of  study.  Until  then  they  had  been  contented 
with  the  monumental  buildings  of  Rome  and  its  neighbourhood,  of 
Latium  and  Campania ;  a  few  of  the  more  adventurous  among 
them  had  penetrated  as  far  as  Paestum  ;  but  it  was  not  till  1845 
that  they  ventured  to  cross  the  sea  and  to  study  the  ruins  of 
Greece  and  Athens  ;  ^  in  later  years  they  have  travelled  as  far 
as  Syria  and  Asia  Minor  in  search  of  objects  for  their  pencils.* 

But  the  occupants  of  the  Villa  Medici  were  not  alone  in  these 
researches.  Doubtless,  the  invaluable  publication  which  contains 
the  results  of  their  labours,  forms  the  most  ample  and  varied 
collection  of  documents  open  to  the  historian  of  architecture 
among   the    ancients.      But    many    other    architects   of    different 

1  The  Antiquities  of  Athens,  Measured  and  Delineated  by  J.  Stuart  and  N.  Revett. 
Folio.     London,  1761. 

-  Expedition  scientifique  de  Moree,  ordonnee  par  le  Gouvernement  Franfais. 
Architecture,  Sculpture,  Inscriptions,  inesurees,  dessinees,  recueillies  et  publices,  par 
A.  Blouet,  A.  Ravoisie,  Alph.  Poirot,  F.  Tre'zel,  et  Fr.  de  Gournay.     Paris,  183 1-7. 

2  The  restoration  of  the  temple  of  Athene  Polias  and  of  the  Parthenon,  by  Ballu 
and  Paccard,  dates  from  1845.  Since  that  time  the  students  of  the  French 
Academy  have  drawn  and  restored  all  the  most  important  monuments  of  Greece. 

*  One  temple  at  Baalbec  was  restored  in  1865  by  M.  Moyau  ;  the  Mausoleum 
of  Halicarnassus  by  M.  Bernier  in  1878,  and  the  temple  of  Athene  at  Priene  by 
M.  I'homas  in  1879. 


xiv  Introduction. 


nationalities    have   given    their    help    to    the    work    of    patiently 
reconstructing  the  past.^ 

Examined  thus  closely,  and  by  the  trained  eyes  of  professional 
artists  provided  with  all  the  necessary  instruments,  the  relics  of 
antiquity  yielded  up  secrets  which  would  never  have  been 
suspected  by  the  casual  observer.  Thus  Mr.  Penrose  discovered 
and  explained  that  those  walls  of  the  Propylsum  and  of  the 
Parthenon,  which  seemed  straight  to  the  eye,  are  in  fact  planned 
on  a  eentle  curve ;  -  he  showed  how  this  subtle  variation  was 
calculated  to  add  to  the  beauty  of  the  buildings,  and  to  augment 
their  effect.  Hittorf  arrived  at  still  more  important  results  through 
the  minute  examination  of  the  Sicilian  ruins.  He  was  the  first 
to  describe  the  important  part  which  painting  played  in  the 
decoration  of  Greek  architecture  ;  he  affirmed  that  in  many  parts 
of  their  buildings  the  stone  or  marble  was  painted  over,  and  that 
the  various  members  of  the  architecture  were  distinguished  by 
differences  of  tint,  which  gave  accent  to  the  mouldings,  and 
force  to  the  figures  in  relief.  These  ideas  were  too  strongly 
opposed  to  modern  habits  of  thought  to  be  received  without 
strong  protestations.  Their  partisans,  too,  did  something  to  retard 
their  acceptance  by  their  absolute  fashion  of  stating  their  con- 
victions, and  by  certain  unhappy  applications  of  their  system  ; 
but  the  polychromatic  principles  of  the  Greeks  are  now  confirmed 
by  too  many  facts  to  be  denied.^ 

Of  the  three  principal  branches  of  ancient  art,  that  of  which 
we  know  least  is  painting,  properly  speaking  ;  the  art  of  Poly- 
gnotus,  of  Zeuxis,  and  of  Apelles.  Of  this  we  have  but  few 
remains,  and  we  are  obliged  to  take  our  ideas  of  its  excellence 
from  the  descriptions  of  ancient  authors.     We  have  indeed    the 

^  In  1872  this  collection  consisted  of  sixty-one  restorations,  comprising  691 
original  drawings  upon  a  very  large  scale,  and  forming  fifty-two  bound  volumes. 
Thanks  to  M.  Jules  Simon,  then  Minister  of  Public  Instruction,  and  M.  Charles 
Blanc,  Director  of  Fine  Arts,  the  publication  of  the  series  in  its  entirety  was  resolved 
upon.  A  commission,  with  M.  Ernest  Vinet  as  secretary,  was  appointed  to  superin- 
tend the  expenditure  of  an  annual  grant  of  20,000  francs  voted  by  the  Chamber. 
But  the  work  progresses  very  slowly.  In  1881  only  five  sections  had  appeared,  the 
most  important  being  the  Restauratmi  des  Temples  de  Fcestum,  by  Labrouste. 

2  F.  C.  Penrose,  An  Investigation  of  the  Principles  of  At/ienian  Architecture. 
Folio,  with  plates.     London,  1851. 

^  J.  J.  Hittorf,  Restitution  du  Temple  d' Empedocle  a  Selinonte;  ou,  l' Architecture 
polychrome  chez  les  Grecs,  4to,  and  plates  in  folio.     Paris,  185 1. 


Introduction.  xv 


wall-paintings  of  those  Campanian  cities  which  were  so  long  burled 
under  the  ashes  of  Vesuvius  ;  paintings  which  were  uncovered 
in  great  numbers  under  the  Napoleonic  domination,  and  have  in 
later  times  been  added  to  every  year,  in  spite  of  the  indolent 
fashion  in  which  the  excavations  have  been  conducted.  Frag- 
mentary mural  paintings  of  the  same  kind  have  also  been 
discovered  in  Rome  and  in  a  few  other  neighbourhoods.  But  after 
all,  great  though  the  interest  may  be  which  attaches  to  these 
works,  it  must  not  be  forgotten  that  they  are  Italian  rather  than 
Greek,  that  they  are  the  decorations  for  the  most  part  of  small 
provincial  cities,  and  that  even  the  best  of  them,  when  compared 
with  the  productions  of  the  fifth  and  fourth  centuries  before  our  era, 
are  examples  of  decadence.  At  the  most  they  enable  us  to  recall, 
with  some  approach  to  probable  truth,  the  taste  and  technical 
methods  of  the  Alexandrian  school.^  Winckelmann  and  his  im- 
mediate successors  saw  the  ashes  cleared  from  the  first  Pompeian 
wall-paintings.  But  they  possessed  no  standards  by  which  they 
could  define  the  styles  of  those  great  schools  of  painting  which 
flourished  in  Greece  between  the  epoch  of  the  Persian  Wars 
and  the  beginning  of  the  Macedonian  supremacy  ;  such  a  defi- 
nition we  may  now  however  attempt  with  at  least  partial  success. 
Since  the  time  of  Winckelmann  hundreds  and  thousands  of  those 
painted  vases  of  burnt  clay,  which  the  public  persist  in  calling 
Etruscan,  have  been  discovered,  classified,  described,  and  ex- 
plained, in  such  a  manner  as  to  leave  unsolved  scarcely  any  of 
the   problems   upon  which  they  could   cast  a  light. 

Gerhard  led  the  way  in  1831  with  his  famous  report  on  the 
Volscian  vases ;  ^  numerous  savants  have  followed  his  example, 
and  nearly  every  day  the  series  which  they  have  established  are 
enriched  by  new  discoveries.  These  vases,  as  we  now  know, 
were  made  in  many  places,  at  Athens,  at  Corinth,  in  the  Greek 
cities  of  Africa  and  of  Magna  Grsecia.  They  were  eagerly 
sought  after  by  some  of  the  races  whom  the  Greeks  considered 
barbarous,  by   the  Grseco-Scythians  of  the  Crimea,  as  well  as   by 

'  See  upon  this  subject  M.  Wolfgang  Helbig's  Uritersiichungen  ucber  die  Cam- 
panische  Wandmalerei.  Leipsic,  1873.  M.  Boissier  has  summed  up  the  leading 
opinions  in  this  matter  in  an  interesting  article  in  the  Rroue  des  Deux  Mondes, 
entitled  Zd'j  Pcintures  d Herculanetcm  et  de  Fo mpei  {OciohQX  i,  1879). 

2  Rapporto  intorno  i  Vasi  Volcenti  {Annali  delV  Insiituto  di  Corrispondenza 
Archeologica,  vol.  iii.  p.  5). 


xvi  Introduction. 


the  Sabellians  and  the  Etruscans  ;  the  latter  imitated  them  now 
and  then  more  or  less  awkwardly,  but  it  is  unanimously  ac- 
knowledged that  they  are  an  essentially  Greek  product,  the  product 
of  an  art  which  sprang  up  with  the  first  awakening  of  the  Greek 
genius,  and  was  extinguished  about  two  hundred  years  before 
Christ,  when  the  nation  ceased  to  be  creative  and  prolific.  From 
analogy  with  all  that  has  passed  elsewhere  we  are  justified  in 
believing  that,  in  each  century,  the  painting  of  these  vases,  which 
would  belong  to  what  we  call  the  industrial  arts,  followed  with 
docility  the  example  set  by  historical  painters,  and  that  it  re- 
produced, so  far  as  its  resources  would  allow,  the  style  and  taste 
of  their  works.  If  we  study  each  series  of  vases  in  the  light  of 
the  judgments  passed  by  the  ancients  upon  the  most  celebrated 
painters  of  Greece,  we  may  find,  by  a  legitimate  induction,  traces 
now  of  the  style  of  Polygnotus,  now  of  that  of  Zeuxis,  and  again 
suggestions  of  the  hands  of  Apelles  or  Protogenes  ;  a  vase  here 
and  there  may  have  even  preserved  more  or  less  faithful  imitations 
of  the  actual  works  of  those  masters.  These  inductions  and 
conjectures  certainly  demand  both  prudence  and  delicacy  of  per- 
ception, but  their  principle  is  incontestable,  and  the  profit  to  be 
obtained  from  them  is  great.  In  the  whole  wreck  of  antiquity 
there  is  no  loss  which  lovers  of  art  find  so  hard  to  bear,  as  the 
complete  annihilation  of  the  works  of  those  great  painters  whom 
the  ancients  put  at  least  upon  the  same  level  as  their  most  famous 
sculptors  ;  and  who  would  not  rejoice  to  be  able,  by  the  remains 
of  contemporary  though  inferior  productions,  to  trace  a  reflection, 
distant  and  feeble  perhaps,  but  yet  faithful  so  far  as  it  goes,  of  a 
whole  art  which  has  been  lost  to  the  world  ? 

The  archaeologists  of  the  eighteenth  century  never  dreamt  of 
such  researches  as  these,  still  less  of  the  results  to  which  they 
might  lead  ;  few  of  them  suspected  what  valuable  aid  might  be 
afforded  to  the  historian  of  art  and  of  antique  civilization,  by  the 
multitude  of  small  objects — vases,  gems,  glass,  mirrors,  bronze 
plaques  and  figures,  terra-cotta  bas-reliefs,  and  statuettes — which 
are  now  so  eagerly  sought  after,  and  which  begin  to  form  valuable 
collections   in   most  of  the  great  museums  of  Europe.  ^      These 

1  One  of  the  first  antiquaries  to  whom  it  occurred  that  the  examination  of  these 
little  objects  might  lead  to  profitable  results  was  the  Comte  de  Caylus,  a  savatit  who 
is  in  some  danger  of  being  forgotten,  and  who  deserves  that  his  claims  to  our 
gratitude  should  be  recalled  to  the  public  mind.     The  work  in  which  he  has  brought 


Introduction.  xvii 


objects,  which  were  in  continual  use,  were  manufactured  in  pro- 
digious quantities  for  thousands  of  years,  and  their  vast  numbers 
gave  them  a  greatly  increased  chance  of  being  preserved.  In 
spite  of  the  rough  usage  of  man,  and  the  slower  progress  of 
destruction  due  to  the  action  of  nature,  a  certain  number  of  them 
were  sure,  from  the  first,  to  find  means  of  escape,  and,  from  so 
many  examples,  a  few  of  each  type  have  therefore  come  down  to 
us.  The  small  size  of  these  objects  also  contributed  to  preserve 
them  from  destruction.  In  times  of  war  and  revolution  the  poor 
and  humble  ones  of  the  earth  easily  avoid  the  catastrophes  which 
overwhelm  those  who  are  richer,  more  powerful,  and  more  con- 
spicuous than  themselves.  So  it  was  with  these  little  memorials 
of  antiquity.  Their  insignificance  was  their  salvation  in  the 
overthrow  of  the  civilisation  to  which  they  belonged.  More 
numerous  and  better  sheltered  than  the  masterpieces  of  fine  art, 
they  survived  when  the  latter  perished.  Thus  it  is  that  so  many 
of  the  lighter  and  more  fragile  products  of  industry  have  sur- 
vived to  our  time,  and  have  made  us  acquainted  with  modes  of 
thought  and  life,  and  with  forms  of  plastic  expression  which  we 
should  never  have  known  without  them.  The  painted  vases,  for 
instance,  have  preserved  for  us  more  than  one  myth  of  which 
no  trace  can  be  found  in  poetry  or  sculpture  ;  and  as  for  terra- 
cottas, to  which  the  Tanagra  statuettes  have  directed  so  much 
attention,  w^e  may  judge  from  the  labours  of  M.  Henzey  of 
the  value  which  they  possess  for  archaeologists,  who,  though 
unable,  like  some  of  our  amateurs,  to  buy  them  with  their  weight 
in  gold,  may  compare  them  one  with  another  and  study  their 
smallest  details.^  Those  statuettes,  which  are  now  classified  in 
museums  in  the  order  of  their  production,  have  shown  us  how 
narrow  and  inadequate  were  the  formulae  by  which  the  early 
historians  of  the  plastic  arts  attempted  to  define  the  genius  of 
the  Greeks.  Even  now,  the  most  accomplished  and  well-informed 
critics  are   not  always  able  to  repress  a  feeling  of  astonishment 

together  the  fruits  of  a  long  life  spent  in  travelling,  in  collecting,  and  in  examining 
the  technical  processes  of  the  ancients,  both  by  himself  and  with  the  help  of  specialists, 
may  be  consulted  with  advantage  i^Reaidl  d'A/itiqiiites  egyptietuies,  etrusques^  grecques, 
et  romains,  6  vols.  4to.  1752-64.     Supplement,  i  vol.  4to,  1767). 

1  Recherches  sur  les  Figures  de  Fetmnes  voilees  dans  t Art  Grec,  4tG.  Paris,  1873. 
Recherches  sur  un  Groupe  de  Praxitele,  d'aprh  les  Figurines  de  ierre  cuite,  8vo. 
Paris,  1875.  ^^^  Figurines  antiques  de  terre  cuite  du  Mus'ce  du  Lowre,  4to. 
1878,  Morel. 

VOL.   I.  d 


xviii  Introduction. 


when  they  examine  a  collection  of  terra-cottas.  Some  of  these 
figures,  no  more  than  a  span  high,  resemble  the  marbles  of  the 
Parthenon  in  dignity  and  grandeur,  others  are  full  of  grace  and 
playfulness  in  their  outlines,  and  show  a  capricious  abandon  which 
disconcerts  for  a  moment  even  those  who  are  least  insensible 
to  their  charm.  At  the  bases  of  such  works  one  is  apt  to  look 
for  the  siofnature  of  some  artist  of  the  Renaissance  or  of  the 
eighteenth  century.  In  reality  they  have  existed  ever  since 
the  fourth  or  third  century  before  our  era,  and  yet  there  is  some- 
thing modern  in  their  appearance.  But  an  indescribable  purity  of 
taste  suffices  to  betray  their  real  origin  to  all  those  who  possess 
knowledge  and  delicate  perceptions.  That  origin  is  still  Greece, 
but  Greece  in  her  lighter  and  more  playful  moments,  when, 
leaving  the  representation  of  gods  and  heroes,  she  condescends 
to  treat  the  familiar  objects  of  domestic  life,  and  does  it  with  an 
ease  of  which  her  great  writers,  notably  Plato  and  Aristophanes, 
had  also  found  the  secret,  when  they  passed  from  epic  tragedy 
to  comedy,  from  the  noblest  eloquence  to  hearty  expressions  of 
enjoyment. 

These  little  statues  interest  the  historian  for  other  reasons  also. 
They  sometimes  give  him,  as  at  Tanagra,  the  most  precise  and 
accurate  information  as  to  dress  and  social  customs  :  sometimes, 
as  at  Tegaea,  they  afford  particulars  of  a  famous  though  obscure 
form  of  worship,  of  a  divinity  and  of  rites  which  are  but  im- 
perfectly described  in  the  writings  of  classic  authors. 

This  extension  of  knowledge  and  the  great  discoveries  upon 
which  it  was  based,  naturally  led  those  who  were  interested  in  the 
study  of  the  remains  of  antique  civilisation,  to  feel  the  necessity  of 
organisation,  of  division  of  labour,  and  of  the  importance  of 
ensuring  a  steady  supply  of  the  best  and  most  trustworthy  in- 
formation. Societies  were  therefore  founded  in  many  different 
centres  with  the  express  object  of  meeting  those  wants.  We  can- 
not, of  course,  enumerate  them  here,  nor  attempt  to  estimate  their 
various  claims  to  our  gratitude,  but  we  may  be  permitted  to  allude 
to  the  good  W'ork  accomplished,  during  fifty  years  of  incessant 
activity,  by  the  Association  which  has  perhaps  done  more  than 
any  other  for  the  progress  of  archaeology,  we  mean  the  Instituio 
di  Corrispondenza  Archeoiogica,  founded  in  Rome  in  1829,  by 
Bunsen,  Gerhard,  and  the  Due  de  Luynes.  Thanks  to  the 
breadth   of  view  which  characterised  its  founders,  this  society  has 


Introduction.  xix 


been,  ever  since  its  inauguration,  an  international  one  in  the  best 
sense  of  the  word  ;  it  brings  together  for  a  common  end  the  most 
eminent  European  savants  and  their  best  pupils ;  it  finds  fellow- 
labourers  and  correspondents  in  every  country.  With  their  aid  it 
soon  established  a  Biillettino,  where,  month  by  month,  all  discoveries 
of  interest  made  at  any  point  of  the  Mediterranean  basin  were 
registered ;  and  volumes,  called  sometimes  Annali,  sometimes 
Memorie,  in  which  really  important  discoveries,  and  the  problems 
to  which  they  give  rise,  were  discussed.  Some  of  these  dissertations 
are  so  elaborate  and  so  full  of  valuable  matter  as  to  have  formed 
epochs  in  the  history  of  science.  They  are  accompanied  by  fine 
plates,  which,  by  their  size,  permit  the  reproduction  of  objects  of 
art  on  a  grander  scale,  and  with  more  fidelity,  than  had  been 
previously  attempted.^ 

While  the  Roman  Instituto  was  thus  devoting  itself  to  research, 
and  assuring  to  its  members  the  advantages  of  a  regular  publicity, 
these  inquiries  were  daily  attracting  a  more  considerable  share  of 
attention  from  the  other  learned  bodies  of  Europe.  The  Acaddmie 
des  Inscriptions  et  de  Belles  Lettres,  the  Academies  of  Berlin, 
Munich,  and  Vienna,  devoted  an  ever-increasing  portion  of  their 
programmes  to  such  studies.  Men  began  everywhere  to  under- 
stand that  the  writings  of  the  classic  authors,  which  had  been  so 
exhaustively  studied  ever  since  the  Renaissance,  were  no  longer 
capable  of  affording  fresh  information.  In  order  to  learn  more  of 
antiquity  than  the  great  scholars  of  the  last  three  centuries,  it  was 
necessary  to  penetrate  into  the  past  by  paths  as  yet  unexplored  ; 
it  was  necessary  to  complement  and  control  the  evidence  of  classic 
authors  by  that  of  public  and  private  inscriptions,  engraved  upon 
bronze,  marble,  or  stone  ;  it  was  above  all  necessary  to  seek  for 
the  expression,  in  their  handiwork,  of  the  wants  and  ideas,  of  the 
personal  sentiments  and  religious  conceptions,  of  the  men  of 
antiquity.     There   are,   in    fact,   nations,   such  as    the   Etruscans, 

^  For  the  history  of  the  histitiito  Archeologico,  the  notice  written  for  the  celebration, 
in  1879,  of  the  fiftieth  anniversary  of  its  foundation,  may  be  consulted.  It  is  from 
the  pen  of  Michaelis,  one  of  the  most  learned  of  modern  German  archaeologists, 
and  bears  the  following  title  :  Storia  deW  Instituto  Archeologico  Gerjiiano,  1829-1879, 
strenna  pubblicata  nelF  occasione  delta  festa  del  21  Aprile,  1879,  dalla  direzione 
centrale  delV  Instituto  Archeologico,  8vo.  Roma,  1879.  It  was  also  published  in 
German.  An  article  by  M.  Ernest  Vinet  in  the  volume  entitled  L  Art  et  V Archeologie 
(pp.  74-91,  8vo.  Didier,  1874),  upon  the  origin  and  labours  of  the  Instituto,  will  also 
be  found  interesting. 


XX  Introduction. 


whose  whole  Hterature  has  perished,  who  are  only  known  to  us  by 
the  relics  of  their  art.  Others,  like  the  Greeks  and  Latins,  have 
indeed  transmitted  to  us  noble  masterpieces  of  literature ;  but 
these  masterpieces  are  few  in  proportion  to  those  which  time 
has  destroyed.  Of  the  thoughts  which  they  expressed  in  their 
immortal  languages,  too  many  have  been  lost  for  ever  with 
the  fragile  strips  of  papyrus  to  which  they  were  confided. 

With  the  ardour  for  knowledge  and  the  heroic  perseverance 
which  are  among  the  virtues  of  our  time,  curiosity  has  refused  to 
resign  itself  to  such  a  loss.  It  has  determined  to  discover  the  un- 
published, to  draw  into  the  light  all  that  has  not  perished  beyond 
recovery,  to  collect  all  that  the  spirit  of  antiquity  has  left  behind  it, 
either  upon  works  hitherto  unnoticed,  or  upon  those  which  have 
been  imperfectly  understood.  The  treasures  of  epigraphy  have 
been  classified  and  shown  in  their  full  value  by  Boeckh,  Borghesi, 
and  others,  and  the  world  is  now  able  to  guess  all  that  history  may 
owe  to  them.  The  study,  however,  of  those  remains  which  bear 
figured  representations  is  still  more  complex  and  formidable.  The 
language  of  forms  is,  in  itself,  less  definite  than  that  of  words, 
and  it  becomes  very  difficult  to  decipher  when  we  have  no  words 
dealing  with  the  same  ideas  to  help  us,  when  we  possess  the  art 
of  a  people  without  a  line  of  their  literature.  Another  difficulty 
springs  from  the  very  abundance  and  variety  of  the  materials  to 
our  hand.  We  feel  oppressed  by  the  ever-growing  accumulation 
of  facts,  and  can  neither  determine  where  to  begin  our  work,  nor 
how  to  leave  it  off :  we  cannot  see  the  forest  for  the  trees ! 


11. 

In  1830,  when  the  Roman  Institute  was  founded,  the  time 
seemed  to  have  come  for  the  formulation  of  all  the  gathered 
facts  and  for  their  arrangement  into  groups,  a  task  which  had 
become  much  more  difficult  than  in  the  time  of  Winckelmann. 
To  conduct  it  to  a  successful  conclusion  a  rare  combination  of 
faculties  was  required  ;  breadth  of  intellect,  aided  by  vast  reading 
and  a  powerful  memory  ;  a  philosophical  spirit,  capable  of  wide 
generalisation,  joined  to  that  passion  for  accurate  detail  which 
distinguishes  the  philologist ;  it  demanded   one  whose  taste  would 


Introduction.  xxi 


survive  the  trying  labour  of  the  cabinet,  a  savant  and  an  artist 
combined  in  one  person.  Books  do  not  teach  everything.  He 
who  wishes  to  speak  of  art  with  intelHgence  must  study  art 
objects  themselves,  must  cultivate  an  intimate  acquaintance  with 
them,  and,  within  himself,  a  love  for  beautiful  forms.  Without  the 
perceptive  powers  which  such  an  educational  process  alone  can 
give,  no  man  can  appreciate  the  subtle  differences  which  dis- 
tinguish styles  and  schools.  He  who  possesses  no  ear,  who  is 
unable  to  perceive  the  intervals  which  separate  one  note  from 
another,  who  knows  that  he  can  neither  recognise  nor  remember 
an  air,  does  not,  unless  he  be  both  presumptuous  and  ignorant, 
dilate  upon  music,  or  attempt  to  write  its  history.  In  the  art  of 
design,  as  in  music,  no  education  can  supply  the  place  of  natural 
aptitudes  ;  but  the  latter  are  not  by  themselves  sufficient  to  form 
a  connoisseur.  Something  more  is  necessary  to  those  who  wish  to 
form  judgments  upon  which  reliance  may  be  placed,  and  to  give 
reasons  for  them  which  will  bear  discussion.  A  special  prepara- 
tion must  be  undergone,  the  rules  and  technical  processes — that 
is  to  say,  the  language  of  art — must  be  learnt.  A  connoisseur 
need  not  be  able  to  compose  an  opera,  or  to  chisel  a  statue, 
but  he  should  be  able  to  read  a  part,  or  to  decide,  for  instance, 
by  the  appearance  of  a  copy  whether  its  original  were  of  bronze 
or  marble. 

At  the  end  of  the  last  century  there  was  born  in  Silesia  a  man 
who,  while  yet  in  his  first  youth,  gave  evidence  of  a  rare  combina- 
tion of  the  gifts  necessary  for  the  successful  accomplishment  of 
the  task  which  we  have  described  ;  we  mean  Carl  Ottfried  Muller, 
who  has  been  called,  without  any  exaggeration,  a  "  scholar  of 
genius."^  A  disciple  of  Niebuhr  and  Boeckh,  he  excelled  all  his 
contemporaries  in  his  efforts  to  embrace  the  whole  of  antiquity  in 
one  view,  to  trace  out  and  realise  for  himself  all  the  varied  aspects 
of  ancient  civilisation.  As  a  philologist,  he  took  the  greatest 
pleasure  in  the  science  which  weighs  words  and  syllables,  w^hich 
collates  manuscripts.  A  poet  in  his  hours  of  leisure,  he  appre- 
ciated both  ancient  and  modern  works  of  literature.  As  a  young 
man  he  studied  with  passion  the  antiques  in  the  Dresden  Museum 
and  the  gallery  of  casts  belonging  to  the  University  of  Gottingen. 

^  Leo  JouBERT,  Essais  de  critique  el  (T histoire  (Vzj'is,  Firmin-Didot,  i  vol.  1863,  p.  4). 
We  shall  never  cease  to  regret  that  politics  have  deprived  literature  of  this  judicious 
and  widely  instructed  critic. 


xxii  Introduction. 


In  the  last  year  of  his  life  he  traversed  Italy  and  Sicily  with 
continual  delight,  and  was  like  one  intoxicated  with  the  beauty  of 
that  Athens  of  which  he  caught  but  a  glimpse,  of  that  Greece 
whose  sun  so  quickly  destroyed  him. 

All  this  knowledge,  all  these  experiences  he  hoped  to  make  use 
of  as  the  lines  and  colours  for  the  great  picture  of  ancient  Greece 
which  he  meditated,  for  the  canvas  upon  which  he  meant  to  portray 
the  Greek  civilization  for  the  benefit  of  the  moderns,  with  all  its 
indivisible  unity  of  social  and  political  life,  of  literary  and  artistic 
production.  In  striking  him  down  in  his  forty-second  year,  death 
put  an  end  to  this  project,  and  the  great  picture,  which  would 
have  been,  perhaps,  one  of  the  capital  works  of  our  century,  was 
never  executed.  But  the  preparatory  sketches  of  the  master 
happily  remain  to  us.  While  he  was  employed  in  collecting 
materials  for  the  work  which  he  meant  to  be  his  highest  title 
to  honour,  he  was  not  shut  up  in  silence  and  meditation,  as  a 
less  prolific  spirit  might  have  been.  His  facility  of  arrange- 
ment and  utterance  was  prodigious  ;  all  that  he  learnt,  all  new 
discoveries  that  he  made  or  thought  he  had  made,  he  hastened 
to  make  public,  either  by  direct  addresses  to  the  auditors  who 
crowded  round  his  chair  at  Gottingen,  or  by  his  pen  to  the  readers 
of  the  numerous  philosophical  periodicals  to  which  he  contributed. 
Like  a  man  who  has  travelled  much  and  who  loves  to  tell  of 
what  he  has  seen,  he  was  ever  ready  to  take  the  public  into  his 
confidence  when  he  embarked  upon  a  new  study.  This  he  gene- 
rally did  by  means  of  papers  full  of  facts  and  ideas,  written 
sometimes  in  German,  sometimes  in  Latin.  In  his  later  years 
he  issued  short  articles  upon  archaeology  and  the  history  of  art, 
in  sufficient  number  to  form  five  substantial  volumes.^  Besides 
this,  he  gave  to  the  world  learned  editions  of  Varro,  of  Festus, 
of  the  EMmenides  of  yEschylus  ;  or  important  monographs  like  his 
GeschicJiten  hellenischer  Stdinrne  ttnd  Stddtc,  including  Orchoinenos 
und  die  Minye^'-  and  Die  Dorier,  the  most  famous  and  most 
actively  discussed  of  his  works  ;  and  finally,  Die  Etrusker,  a  work 
which  was  suggested  to  him  by  one  of  the  publications  of  the 
Berlin  Academy.  There  was  also  Prolegomena  zu  einer  wissen- 
schaftlichen  Mythologie,  which  has  been  fruitful  for  good  even  in 
its  errors,  and  the  GescJiichte  der  griechischen  Literatur,  &c.,  which, 

'  Knnsiarchccologische   Wcrke.     Berlin,  Calvary,  i8mo.  1873. 


Introduction.  xxiii 


incomplete  as  it  is,  has  never  become  obsolete.  Since  the  time 
of  Ottfried  Muller  several  other  critics  have  attempted  to  rival 
his  achievements,  but  they  have  all  lacked  his  breadth  of  view 
and  comprehensiveness  of  exposition,  as  well  as  the  versatility 
wjth  which  he  combined  the  most  accurate  scientific  investigations 
v/ith  a  delicate  appreciation  of  the  beauty  and  originality  of  the 
Greek  authors. 

But  of  all  these  works,  that  which  has  perhaps  rendered  the 
greatest  service  to  the  science  of  archaeology  is  the  Handbuch  der 
Ai^chcEologie  der  Kunst,  which  was  published  in  Breslau  in  1830.^ 
Translated  into  French,  Italian,  and  English,  it  at  once  took  its 
place  as  the  indispensable  guide  for  all  those  who  wished  to  learn 
something  of  antique  art.^  In  all  the  universities  into  which 
archaeology  had  made  good  its  entrance,  this  manual  has  formed 
the  basis  of  the  teaching,  and  also  has  enabled  the  pupils  to 
supplement  for  themselves  the  lessons  which  they  learnt  from 
their  professors.  Even  now  it  has  not  been  superseded,  and  to 
all  appearance  it  will  long  preserve  its  supremacy. 

The  form  of  a  Jiandbnch  or  manual,  which  Ottfried  Muller 
gave  to  his  work,  was  well  and  favourably  known  to  cultivated 
Germans,  but  it  was  not  so  with  the  French.  They  had  nothing 
of  the  kind  but  worthless  epitomes  made  to  facilitate  the  passing 
of  University  examinations.  In  this  matter  the  Germans  are 
better  off  than  any  other  nation  in  Europe.  They  have  manuals 
in  which  every  branch  of  history  and  science  is  treated  by  com- 
petent writers  with  as  much  care  and  skill  as  the  most  ambitious 
publications,  a  few  being  original  works  by  savants  of  the  first 
order.  The  arrangement  of  the  Handbuch  is  very  simple.  It 
opens  with  an  introduction  in  which  the  author  defines  art — more 
especially  the  plastic  arts — divides  it  into  classes,  and  indicates  the 
principal  works  to  be  consulted,  namely,  those  to  which  he  himself 
has  had  continually  to  refer  during  the  progress  of  his  book. 
Then   comes   the  history  of  Greek   and    Roman  art  divided  into 

^  Ha7idlnich  der  Archceologie  der  Kunst,  1  \o\.  8vo. 

2  The  French  translation,  from  the  pen  of  M.  P.  Nicard,  forms  three  volumes  of 
the  collection  of  handbooks  known  under  the  name  of  the  E7icyc}opedie  Roret.  It 
appeared  in  1841,  so  that  the  translator  was  unable  to  make  use  of  the  additions  and 
corrections  with  which  Welcker  enriched  the  edition  of  1848.  But  M.  Nicard's 
edition  has  one  great  advantage  over  the  German  versions  in  the  complete  tables 
with  which  it  is  provided.  The  best  English  translation  is  that  by  J.  Leitch,  the 
second  edition  of  which  appeared  in  1850. — Ed. 


xiv  '  Introduction. 


periods,  and  the  paragraphs  which  are  devoted  to  Etruria  and 
the  East.  To  this  historical  epitome  succeed  the  theoretical 
chapters. 

He  takes  antique  art  as  a  whole,  and  studies  its  constitution, 
the  materials  and  processes  which  it  employs,  the  conditions  under 
which  it  works,  the  characteristics  which  it  gives  to  form,  the 
subjects  of  which  it  treats,  and  the  partition  of  its  remains  over  the 
whole  territory  occupied  by  ancient  civilization.  Greece,  in  her  best 
days,  gave  most  of  its  care  to  the  representation  of  those  beings, 
superior  to  humanity  and  yet  clothed  with  human  forms,  in  which 
her  glowing  imagination  personified  the  forces  and  eternal  laws  of 
nature  and  of  the  moral  world  ;  it  was  in  striving  to  create 
these  types,  and  to  endow  them  with  outward  features  worthy  of 
their  majesty,  that  Grecian  art  produced  its  noblest  and  most  ideal 
works.  It  will,  therefore,  be  seen  that  a  comprehensive  manual 
had  to  include  a  history  of  those  gods  and  heroes  which,  with 
that  of  their  statues,  formed  a  whole  mythology  of  art  ;  and 
this  mythology  occupies  the  larger  portion  of  the  second  part 
of  the  work. 

This  plan  has  been  often  criticised,  but  we  need  here  make  no 
attempt  to  repel  or  even  to  discuss  the  objections  which  have  been 
brought  against  it. 

It  has  doubtless  the  inconvenience  of  leading  to  frequent 
repetition  ;  monuments  which  have  been  necessarily  described  and 
estimated  in  the  historical  division  are  again  mentioned  in  the 
chapters  which  treat  of  theory  ;  but  a  better  plan  has  yet  to  be 
found,  one  which  will  enable  us  to  avoid  such  repetitions  without 
any  important  sacrifice.  The  chief  thing  in  a  work  of  the  kind  is 
to  be  clear  and  complete,  merits  which  the  Handbuch  possesses  in 
the  highest  degree.  Things  are  easily  found  in  it,  and,  by  a 
powerful  effort  of  criticism,  the  author  has  succeeded  in  classifying 
and  condensing  into  a  single  convenient  volume,  all  the  interesting 
discoveries  of  several  generations  of  archaeologists.  Not  that  it 
is  a  mere  compilation,  for  previous  writers  were  far  from  being 
unanimous  as  to  the  dates  and  sisfnificance  of  the  remains  which 
they  had  described,  and  it  was  necessary  to  choose  between  their 
different  hypotheses,  and  sometimes  to  reject  them  all.  In  such 
cases  Miiller  shows  great  judgment,  and  very  often  the  opinion 
to  which  he  finally  commits  himself  had  been  previously  un- 
known.     Without  entering  into  any  long  discussion  he  sustains  it 


Introduction.  xxv 


by  a  few  shortly  stated  reasons,  which  are  generally  conclusive. 
The  plan  of  his  book  prevents  him  from  launching  out,  like 
Winckelmann,  into  enthusiastic  periods  ;  he  makes  no  attempt  at 
those  brilliant  descriptions  which  in  our  day  seem  a  little  over- 
coloured  ;  but  in  the  very  brevity  of  his  judgments  and  his  laconic 
but  significant  phraseology,  we  perceive  a  sincere  and  individual 
emotion,  an  independent  intellect,  a  pure  though  catholic  taste. 
We  need  say  no  more  to  the  objectors  who  attack  the  mere  form 
of  the  book.  Its  one  real  defect  is  that  it  was  written  thirty  or 
forty  years  too  soon.  The  second  edition,  carefully  revised  and 
largely  augmented,  appeared  in  1835  ;  it  was  the  last  issued  during 
the  lifetime  of  Muller.  From  that  moment  down  to  the  day  but 
lately  passed  when  the  excavations  at  Olympia  and  Pergamus 
were  brought  to  an  end,  many  superb  remains  of  Greek,  Etruscan, 
and  Roman  art  have  risen  from  their  temporary  graves  and  ranged 
themselves  in  our  museums.  If,  however,  recent  archaeology  had 
made  no  further  discoveries,  a  few  occasional  corrections  and 
additions,  at  intervals  of  ten  or  fifteen  years,  would  have  sufficed 
to  prevent  the  manual  from  becoming  obsolete.  With  a  little  care 
any  intelligent  editor  could  have  satisfactorily  performed  w^hat  was 
wanted.  For  the  Grseco-Roman  period  especially  Muller  had 
erected  so  complete  a  historical  framework  that  the  new  discoveries 
could  find  their  places  in  it  without  any  difficulty.  Welcker, 
indeed,  published  a  third  edition  in  1848,  corrected  and  completed, 
partly  from  the  manuscript  notes  left  by  the  author  in  his  inter- 
leaved copy,  partly  from  information  extracted  by  the  editor  from 
the  lectures  and  other  writings  of  Muller.  But  why  does  Welcker 
declare,  in  his  advertisement  to  the  reader,  that  but  for  the  respect 
due  to  a  work  which  had  become  classic,  he  would  have  modified 
it  much  more  than  he  had  dared.  And  why,  for  more  than 
thirty  years,  has  his  example  found  no  imitators  ?  Why  have 
we  been  content  to  reprint  word  for  word  the  text  of  that 
third   edition  ? 

A  few  years  ago  one  of  the  most  eminent  of  our  modern 
archaeologists,  Carl  Bernhard  Stark,  was  requested  by  a  firm  of 
publishers  to  undertake  a  new  revision  of  the  Handduck.  Why 
then,  after  having  brought  his  materials  tOQ^ether,  did  he  find  it 
more  useful,  and  even  easier,  to  compose  an  original  work,  a  new 
manual  which  should  fulfil  the  same  requirements  on  a  system  of 
his  own  devising  ? — an  enterprise  which  he  would  have  brought  to 

VOL.  T.  c 


XX  vi  Introduction. 


a  successful  conclusion  had  not  death  interrupted  him  after  the 
publication  of  the  first  part.^ 

The  answer  is  easy.  The  East  was  not  discovered  till  after  the 
death  of  Ottfried  Mullen  By  the  East  we  mean  that  part  of 
Africa  and  Asia  which  is  bordered  by  the  Mediterranean,  or  is  so 
near  to  that  sea  that  constant  communication  was  kept  up  with  its 
shores  ;  we  mean  Egypt,  Syrian  Phoenicia,  and  its  great  colony  on 
the  Libyan  Coast,  Chaldaea  and  Assyria,  Asia  Minor,  and  those 
islands  of  Cyprus  and  Rhodes  which  were  so  long  dependent  upon 
the  empires  on  the  neighbouring  continents.^  It  was  between  1820 
and  1830  that  the  young  savant  conceived  the  ideas  which  he 
developed  in  his  works  ;  it  was  then  that  he  first  took  an  important 
part  in  the  discussion  as  to  the  origin  of  the  Greek  nation,  upon 
which  archaeologists  had  long  been  engaged.  What  part  had 
foreign  example  taken  in  the  birth  and  development  of  the  religion, 
the  arts,  the  poetry,  and  the  philosophy  of  Greece,  of  the  whole 
Hellenic  civilization  "^  How  much  of  it  was  due  to  supfSfestions 
derived  from  those  peoples  who  had  so  long  preceded  the  Greeks 
in  the  ways  of  civil  life  ?  No  historian  has  answered  this  question 
in  a  more  feeble  and  narrow  spirit  than  Ottfried  MUller ;  no  one 
has  been  more  obstinate  than  he  in  insisting  upon  the  originality 
of  the  Greek  genius,  and  in  believing  that  the  Greek  race 
extracted  from  its  own  inner  consciousness  all  that  has  made  Its 
greatness  and  glory. 

When  Miiller  first  attacked  this  question,  Egypt  alone  had 
begun  to  emerge  from  the  obscurity  which  still  enveloped  the 
ancient  civilization  of  the  East.  It  was  not  until  three  years  after 
his  death,  that  Botta  began  to  excavate  the  remains  of  Assyrian 
art  ;  and  nothing  but  the  vaguest  and  most  confused  information 
was  to  be  had  about  the  ruins  In  Chaldsea.  Now,  however,  we 
can  follow  the  course  of  the  Phoenician  ships  along  the  Mediter- 
ranean, from  the  Thracian  Bosphorus  to  the  pillars  of  Hercules. 
From  the  traces  left  by  the  commerce  and  the  industries  of  the 


1  Stark  died  at  Heidelberg  in  October,  1879.  The  title  of  his  work  was  identical 
with  that  of  Miiller:  Handhich  der  Archaologie  det-  Kunst.  The  first  256  pages  of 
the  first  volume  were  published  in  1878  with  the  sub-title  :  Einleitender  und  grund- 
legender  77/d?// (Leipsic,  Engelmann,  8vo).  A  second  instalment  appeared  in  1880, 
by  which  the  introduction  was  completed.  The  entire  work,  which  will  not  be 
continued,  was  to  have  formed  three  volumes.  We  explained  its  plan  and 
made  some  remarks  upon  the  part  already  published  in  the  Reime  Critique  of 
July  14,  1879. 


Introduction.  xxvii 


Syrians  and  Carthaginians,  we  can  estimate  the  duration  of  their 
stay  in  each  of  the  countries  which  they  visited,  and  the  amount  of 
influence  which  they  exercised  over  the  various  peoples  who  were 
tributary  to  them.  Forty  years  ago  this  was  impossible ;  the 
writings  of  ancient  authors  were  our  sole  source  of  knowledge  as 
to  the  style  and  taste  of  Phoenician  art,  and  the  ideas  which  they 
imparted  were  of  necessity  inexact  and  incomplete.  Wherever 
they  passed  the  Phoenicians  left  behind  them  numbers  of  objects 
manufactured  by  them  for  exportation,  and  these  objects  are  now 
eagerly  collected,  and  the  marks  of  the  Sidonian  and  Carthaginian 
makers  examined  and  classified,  and  thus  we  are  enabled  to 
recognize  and  describe  the  industrial  processes  and  the  decorative 
motives,  which  were  conveyed  to  the  Greeks  and  to  the  races  of 
the  Italian  peninsula  by  the  "watery  highway"  of  the  Mediter- 
ranean. Fifty  years  ago  the  land  routes  were  as  little  known  as 
those  by  sea.  The  roads  were  undiscovered  which  traversed  the 
defiles  of  the  Taurus  and  the  high  plateaux  of  Asia  Minor,  to 
bring  to  the  Greeks  of  Ionia  and  y^olia,  those  same  models,  forms, 
and  even  ideas,  and  it  was  still  impossible  to  indicate  their  detours, 
or  to  count  their  stages. 

Leake  had  indeed  described,  as  early  as  182 1,  the  tombs  of  the 
Phrygian  kings,  one  of  whom  bore  that  name  of  Midas  to  which 
the  Greeks  attached  so  strange  a  legend  ;  ^  but  he  had  given  no 
drawings  of  them,  and  the  work  of  Steuart,^  which  did  not  appear 
till  1842,  was  the  first  from  which  any  definite  knowledge  of  their 
appearance  could  be  obtained.  Miiller  knew  nothing  of  the  dis- 
coveries of  Fellows,  of  Texier,  or  of  Hamilton  ;  while  he  was 
dying  in  Greece,  they  were  exploring  a  far  more  difficult  and 
dangerous  region.  A  few  years  afterwards  they  drew  the  attention 
of  European  savants  to  the  remains  which  they  had  discovered, 
dotted  about  over  the  country  which  extends  from  the  shores  of 
the  ^g^ean  to  the  furthest  depths  of  Cappadocia,  remains  which 
recall,  both  by  their  style  and  by  their  symbolic  devices,  the  rock 
sculptures  of  Upper  Assyria.  The  Lycian  remains,  which  give 
evidence    of  a   similar   inspiration    and    are    now    in    the    British 

^  Journal  of  a  Tom-  in  Asia  Minor,  %vith  Comparative  Remarks  ofi  the  Ancient 
and  Modern    Geography   of  that  Country  {\  vol.  in  8vo.     London,  Murray,   182 1, 

PP-  31-33)- 

"  A  Description  of  some  Ancient  Monuments  with  Inscriptions  still  existitig  in  Lydta 

and  P/uygia.     London,  1842,  in  folio. 


xxvlii  Introduction. 


Museum,    were    not    transported    to    Europe    until  after  Mliller's 
death. 

The  clear  intellect  of  Ottfried  Miiller  easily  enabled  him  to 
perceive  the  absurdity  of  attempting  to  explain  the  birth  of  Greek 
art  by  direct  borrowing  from  Egypt.  He  saw  that  the  existing 
remains  in  both  countries  emphatically  negatived  such  a  sup- 
position, but  materials  were  wanting  to  him  for  a  right  judgment 
of  the  intensity  and  duration  of  the  influence  under  which  the 
Greeks  of  the  heroic  age  w^orked  for  many  centuries,  influences 
which  came  to  them  partly  from  the  Phoenicians,  the  privileged 
agents  of  intercourse  between  Egypt  and  the  East,  partly  from 
the  people  of  Asia  Minor,  the  Cappadocians,  Lycians,  Phrygians, 
and  Lydians,  all  pupils  and  followers  of  the  Assyrians,  whose 
dependants  they  were  for  the  time,  and  w^ith  whom  they  com- 
municated by  caravan  routes.  We  may  thus  explain  the  ex- 
travagance of  the  hypothesis  which  Miiller  advocated  in  all  his 
writings  ;  and,  as  the  originality  of  the  Greek  intellect  displayed 
itself  in  the  plastic  arts  much  later  than  in  poetry,  the  partial 
falsity  of  his  views  and  their  incompleteness  is  much  more  obvious 
and  harmful  in  his  handbook  than  in  his  history  of  Greek 
literature. 

In  writing  the  life  of  any  great  man  and  attempting  to  account 
for  his  actions,  it  is  important  to  know  where  he  was  born,  and 
who  were  his  parents  ;  to  learn  the  circumstances  of  his  education, 
and  the  surroundings  of  his  youth.  The  biographer  who  should 
have  no  information  on  these  points,  or  none  but  what  was  false, 
would  be  likely  to  fall  into  serious  mistakes  and  misapprehensions. 
He  would  find  great  difficulty  in  explaining  his  hero's  opinions  and 
the  prejudices  and  sentiments  by  which  he  may  have  been  influenced, 
or  he  would  give  absurd  explanations  of  them.  Peculiarities  of 
character  and  eccentricities  of  idea  would  embarrass  him,  w^hich, 
had  he  but  known  the  hereditary  predisposition,  the  external  cir- 
cumstances during  infancy  and  adolescence,  the  whole  course  of 
youthful  study,  of  the  man  whose  life  he  was  describing,  he  might 
easily  have  understood.  It  is  the  same  with  the  history  of  a 
people  and  of  their  highest  intellectual  manifestations,  such  as 
their  religion,  arts,  and  literature. 

It  was  not  the  fault  of  Ottfried  Miiller,  it  was  that  of  the  time 
in  which  he  lived,  that  he  was  deceived  as  to  the  true  origin  of 
Greek  art.     The  baneful  effects   of   his    mistake   are  evident    in 


Introduction.  xxix 


the  very  first  pages  of  the  historical  section  of  his  work,  in 
the  chapters  which  he  devotes  to  the  archaic  period.  These 
chapters  are  very  unsatisfactory.  Attempt,  under  their  guidance 
alone,  to  study  the  contents  of  one  of  those  museum  saloons 
where  the  remains  of  Oriental  art  are  placed  side  by  side 
with  those  from  Etruria  and  primitive  Greece ;  at  every  step 
you  will  notice  resemblances  of  one  kind  or  another,  similarities 
between  the  general  aspects  of  figures,  between  the  details  of 
forms  and  the  choice  of  motives,  as  well  as  in  the  employ- 
ment of  common  symbols  and  attributes.  These  resemblances 
will  strike  and  even  astonish  you,  and  if  you  are  asked  how 
they  come  to  exist  among  differences  which  become  ever  more  and 
more  marked  in  the  succession  of  the  centuries,  you  will  know  not 
how  to  reply.  In  these  archaic  remains  there  are  many  traits  for 
which  those  who,  like  Ottfried  Miiller,  begin  with  the  history  of 
Greece,  are  unable  to  account.  He  wishes  us  to  believe  that 
Greece  in  the  besfinninQf  was  alone  in  the  world,  that  she  owed  all 
her  glory  to  the  organic  development  of  her  unequalled  genius, 
which,  he  says,  "  displayed  a  more  intimate  combination  than  that 
of  any  other  Aryan  nation  of  the  life  of  sensibility  with  that  of 
intelligence,  of  external  with  internal  life."  He  goes  no  further 
back  than  the  Greece  described  to  us  in  the  heroic  poems  ;  he 
never  has  recourse  to  such  comparisons  as  we  are  now  continually 
making ;  at  most  he  lets  fall  at  lengthy  intervals  a  few  words 
which  seem  to  imply  that  Oriental  civilization  may  have  had  some- 
thing to  do  with  the  awakening  of  Greek  thought  and  the  directing 
of  her  first  endeavours.  He  never  formally  denies  her  indebtedness, 
but  he  fails  to  perceive  its  vast  importance,  or  to  declare  it  with 
that  authoritative  accent  which  never  fails  him  in  the  expression 
of  those  Ideas  which  are  dear  to  him,  of  those  truths  which  he  has 
firmly  grasped. 

This  tendency  is  to  be  seen  even  in  the  plan  of  his  work. 
There  is  nothing  surprising  in  the  fact  that  Mtiller,  in  1830,  or 
even  in  1835,  had  but  a  slight  acquaintance  with  the  art  of  the 
Eastern  Empires ;  but  as  he  thought  it  necessary  not  entirely 
to  ignore  those  peoples  in  a  book  which  pretended  to  treat  of 
antiquity  as  a  whole,  it  would  perhaps  have  been  better  not  to 
have  relegated  them  to  a  few  paragraphs  at  the  end  of  his  his- 
torical section.  He  knew  well  enough  that  the  Egyptians,  the 
Babylonians,  the  Phoenicians,  even  the  Phrygians  and  the  Lydians 


XXX  Introduction. 


were  much  older  than  the  Greeks  ;  why  should  he  have  postponed 
their  history  to  that  of  the  decline  and  fall  of  Grseco-Roman  art  ? 
Would  it  not  have  been  better  to  put  the  little  he  had  to  tell  us  in 
its  proper  place,  at  the  beginning  of  his  book  ? 

This  curious  prejudice  makes  the  study  of  a  whole  series  of  im- 
portant works  more  difficult  and  less  fruitful.  It  prevents  him 
from  grasping  the  true  origin  of  many  decorative  forms  which, 
coming  originally  from  the  East,  were  adopted  by  the  Greeks  and 
carried  to  perfection  by  their  unerring  taste,  were  perpetuated  in 
classic  art,  and  thence  transferred  to  that  of  modern  times  ;  and 
this,  bad  though  it  is,  is  not  the  worst  result  of  Muller's  misappre- 
hension. His  inversion  of  the  true  chronological  order  makes  a 
violent  break  in  the  continuity  of  the  phenomena  and  obscures 
their  mutual  relations.  There  is  no  sequence  in  a  story  so  broken 
up,  falsified,  and  turned  back  upon  itself.  You  will  there  seek  in 
vain  for  that  which  we  mean  to  strive  after  in  this  present  history 
of  antiqiie  art — a  regular  and  uninterrupted  development,  which  in 
spite  of  a  few  more  or  less  brusque  oscillations  and  periods  of 
apparent  sterility,  carried  the  civilization  of  the  East  into  the 
West,  setting  up  as  its  principal  and  successive  centres,  Memphis, 
Thebes,  Babylon,  Nineveh,  Sidon,  Carthage,  Miletus  and  the 
cities  of  Ionia,  Corinth  and  Athens,  Alexandria,  Antioch,  Pergamus, 
and  finally  Rome,  the  disciple  and  heir  of  Greece. 

Ottfried  Muller  saw  clearly  enough  the  long  and  Intimate 
connection  between  Greece  and  Rome,  but  he  did  not  compre- 
hend— and  perhaps  in  the  then  state  of  knowledge  it  was  impossible 
that  he  should  comprehend — that  the  bonds  were  no  less  close 
which  bound  the  Hellenic  civilization  to  the  far  more  ancient 
system  which  was  born  upon  the  banks  of  the  Nile,  and  crept  up 
the  valleys  of  the  Tigris  and  Euphrates,  to  spread  itself  over  the 
plains  of  Iran  on  the  one  hand  and  of  Asia  Minor  on  the  other  ; 
while  the  Phoenicians  carried  it,  with  the  alphabet  which  they  had 
invented  and  the  forms  of  their  own  worship  of  Astarte,  over  the 
whole  basin  of  the  Mediterranean.  His  error  lay  in  his  arbitrary 
isolation  of  Greece,  in  dragging  her  from  the  soil  in  which  her 
roots  were  deeply  imbedded,  from  which  she  had  drawn  her  first 
nourishment  and  the  primary  elements  of  that  varied  and  luxuriant 
vegetation  which,  in  due  time,  became  covered  with  the  fairest 
hues  of  art  and  poetry. 


Introduction.  xxxi 


III. 

Thanks  to  the  numerous  discoveries  of  the  last  fifty  years,  and 
to  the  comparisons  which  they  have  suggested,  thanks  also  to  the 
theories  for  which  they  afford  a  basis,  history  has  been  at  last 
enabled  to  render  justice  to  certain  nations  whose  activity  had 
never  before  been  properly  understood,  to  give  to  them  their 
proper  place  in  the  civilization  of  ancient  times.  But  Greece — the 
Greece  which  Ottfried  M tiller  worshipped,  and  for  which  he  was 
too  ready  to  sacrifice  her  predecessors  and  teachers,  to  whom  she 
herself  was  more  just  in  her  early  legends — has  lost  nothing  by 
the  more  exact  information  which  is  now  at  our  command. 
Served  by  her  situation  on  the  confines  of  Europe  and  Asia  and 
not  far  from  Africa,  by  the  superiority  of  the  genius  of  her  people 
and  the  marvellous  aptitudes  of  her  language,  Greece  was  able  to 
arrange  and  classify  previous  discoveries  and  to  bring  them  to 
perfection,  to  protect  from  destruction  and  oblivion  the  machinery 
of  progress,  the  processes  of  art,  the  newly-born  scientific  methods, 
in  a  word,  all  the  complex  and  fragile  apparatus  of  civilization 
which  was  so  often  threatened  with  final  destruction,  and  which 
has  more  than  once  been  overwhelmed  for  a  time  in  epochs  of 
national  conflict  and  social  decadence. 

This  is  not  the  place  for  insistence  upon  all  that  Greece  has 
accomplished  in  the  domains  of  pure  thought,  philosophy,  and 
science,  nor  even  for  calling  attention  to  her  literature.  We  are 
writing  the  history  of  the  arts  and  not  that  of  letters,  a  history 
which  we  wish  to  conduct  to  the  point  where  M tiller  left  off,  to 
the  commencement  of  those  centuries  which  are  called  the  Middle 
Ages ;  and  Greece  will  occupy  by  far  the  most  important  place  in 
our  work.  We  shall  endeavour  to  bring  the  same  care  and  con- 
science, the  same  striving  after  accuracy,  into  every  division  of  our 
history  ;  but  the  monuments  of  Greece  will  be  examined  and 
described  in  much  greater  detail  than  those  of  Egypt  and  Assyria, 
or  even  those  of  Etruria  and  Latium.  It  was  our  love  for  Greece 
that  drove  us  to  this  undertaking  ;  we  desire  and  hope  to  make 
her  life  better  known,  to  show  a  side  of  it  which  is  not  to  be  found 
in  the  works  of  her  cfreat  writers,  to  o-ive  to  our  readers  new  and 
better  reasons  for  loving  and   admiring  her  than  they  have   had 


xxxii  Introduction. 


before.  A  combination  of  circumstances  that  is  unique  in  the 
history  of  the  world  gave  to  the  contemporaries  of  Pericles  and 
Alexander  the  power  of  approaching  more  nearly  to  perfection,  in 
their  works  of  art,  than  men  of  any  other  race  or  any  other  epoch. 
In  no  other  place  or  time  have  ideas  been  so  clearly  and  com- 
pletely interpreted  by  form  ;  in  no  other  place  or  time  have  the 
intellectual  qualities  been  so  closely  wedded  to  a  strong  love  for 
beauty  and  a  keen  sensibility  to  it.  It  results  from  this  that  the 
works  of  the  Greek  artists,  mutilated  by  time  and  accident  as 
they  are,  serve  as  models  and  teachers  for  our  painters  and  sculp- 
tors, a  role  which  they  will  continue  to  fill  until  the  end  of  time. 
They  form  a  school,  not,  as  some  have  thought,  to  enable  us  to 
dispense  with  nature,  the  indispensable  and  eternal  master,  but  to 
incite  to  such  an  ardent  and  intelligent  study  of  her  beauties,  as 
may  lead  to  the  creation  of  great  works,  works  capable,  like  those 
of  the  Greeks,  of  giving  visible  expression  to  the  highest  thoughts. 

As  the  Greeks  excelled  all  other  nations  in  the  width  and  depth 
of  their  aesthetic  sentiments  ;  as  their  architects,  their  sculptors, 
and  their  painters,  were  superior  both  to  their  pupils  and  their 
masters,  to  the  orientals  on  the  one  hand,  and  the  Etruscans  and 
the  Latins  on  the  other,  we  need  feel  no  surprise  at  their  central 
and  dominating  position  in  the  history  of  antique  art.  Other 
national  styles  and  artistic  manifestations  will  pass  before  the  eye 
of  the  reader  in  their  due  order  and  succession  ;  they  will  all  be 
found  interesting,  because  they  show  to  us  the  continual  struggle 
of  man  against  matter,  and  we  shall  endeavour  to  distinguish  each 
by  its  peculiar  and  essential  characteristics,  and  to  illustrate  it  by 
the  most  strikinof  remains  which  it  has  left  behind.  But  each 
style  and  nationality  will  for  us  have  an  importance  in  proportion 
to  the  closeness  of  its  connection  with  the  art  of  Greece.  In  the 
case  of  those  oriental  races  which  were  the  teachers  of  the  Greeks, 
we  shall  ask  how  much  they  contributed  to  the  foundations  of 
Greek  art  and  to  its  ultimate  perfection  ;  in  the  case  of  the 
ancient  Italians,  we  shall  endeavour  to  estimate  and  describe  the 
ability  shown  by  them  in  apprehending  the  lessons  of  their 
instructors,  and  the  skill  with  which  they  drew  from  their  teachers 
a  method  for  the  expression  of  their  own  peculiar  wants  and 
feelines  and  for  the  satisfaction  of  their  own  aesthetic  desires. 

The  study  of  oriental  art  will  really,  therefore,  be  merely  an 
introduction  to  our  history  as  a  whole,  but  an  introduction  which 


Introduction.  xxxiii 


is  absolutely  required  by  our  plan  of  treatment,  and  which 
will  be  completely  embodied  in  the  work.  The  history  of 
Etruscan  and  Roman  art  will  be  its  natural  and  necessary 
epilogue. 

This  explanation  will  show  how  far,  and  for  what  reasons  we 
mean  to  separate  ourselves  from  our  illustrious  predecessor.  We 
admit,  as  he  did,  we  even  proclaim  with  enthusiasm,  the  pre- 
eminence of  Greece,  the  originality  of  its  genius  and  the  superiority 
of  its  works  of  plastic  art ;  but  we  cannot  follow  him  in  his 
arbitrary  isolation  of  Greece,  which  he  suspends,  so  to  speak,  in 
air.  Our  age  is  the  age  of  history  ;  it  interests  itself  above  all 
others  in  the  sequence  of  social  phenomena  and  their  organic 
development,  an  evolution  which  Hegel  explained  by  the  laws  of 
thought.  It  would  be  more  than  absurd  in  these  days  to  accept 
Greek  art  as  a  thing  self-created  in  its  full  perfection,  without 
attempting  to  discover  and  explain  the  slow  and  careful  stages  by 
which  it  arrived  at  its  apogee  in  the  Athens  of  Pericles.  In  this 
history  of  ours  of  which  we  are  attempting  to  sketch  the  form,  we 
must,  in  order  to  get  at  the  true  origin  of  Greek  art,  penetrate  far 
beyond  its  apparent  origin  ;  to  describe  the  springing  of  Greek 
civilization,  we  must  first  study  the  early  history  of  those  races 
which  surround  the  eastern  basin  of  the  Mediterranean. 

The  Greece  which  we  call  ancient  entered  late  into  history, 
when  civilization  had  already  a  long  past  behind  it,  a  past  of  many 
centuries.  In  this  sense,  the  words  which,  as  we  are  told  by 
Plato,  a  priest  of  Sais  addressed  to  Solon,  were  perfectly  true, 
"You  Greeks,  you  are  but  children!"^  In  comparison  with 
Egypt,  with  Chaldsea,  with  Phoenicia,  Greece  is  almost  modern  : 
the  age  of  Pericles  is  nearer  to  our  day  than  to  that  which  saw  the 
birth  of  Egyptian  civilization. 

Appearing  thus  lately  upon  the  scene,  when  the  genius  of  man 
had,  by  efforts  continued  without  intermission  through  a  long 
procession  of  centuries,  arrived  at  the  power  of  giving  clear  and 
definite  expression  to  his  thoughts,  by  means  either  of  articulate 
sounds  and  the  symbols  which  represent  them  or  by  the  aid  of 
plastic  forms,  the  Greeks  could  only  have  remained  ignorant  of  all 
that  had  been  achieved  before  their  time  if  they  had  sprung  into 
existence  in  some  distant  and  isolated  corner  of  the  world,  or  in 
some    inaccessible    island.       Their    actual    situation    was    a    very 

^   Timceus,  p.  22. 
VOL.    I.  / 


xxxlv  Introduction. 


different  one.  In  the  earliest  epoch  of  which  we  have  any  record 
we  find  them  estabhshed  in  a  peninsula,  which  is  on  one  hand 
upon  the  very  borders  of  Asia,  and  upon  another  seems  to  hold 
out  a  hand  to  Africa  by  the  innumerable  islands  which  surround 
its  shores.  Between  the  shores  of  this  peninsula  and  those  of 
Asia,  these  islands  are  sprinkled  so  thickly  over  the  narrow  seas 
that  nature  seems  to  have  intended  them  for  stepping-stones 
which  should  tempt  the  least  venturesome  to  cross  from  one 
continent  to  the  other. 

The  Greek  race  thus  found  itself,  by  the  accident  of  its 
geographical  situation,  in  contact  with  the  Egyptian,  Assyrian,  and 
Median  empires,  the  masters  of  the  Eastern  Mediterranean  ;  while 
the  insular  or  peninsular  character  of  most  of  the  region  which  it 
inhabited,  together  with  the  numerous  colonies  attached  to  the 
surrounding  coasts  like  vessels  at  anchor,  had  the  effect  of  greatly 
multiplying  the  points  of  contact.  The  Greek  frontier  was  thus: 
one  of  abnormal  extent,  and  was,  moreover,  always  open,  always 
ready  to  receive  foreign  ideas  and  influences.  Her  eyes  were 
ever  turned  outwards ;  the  Greek  nationality  was  not  one  of  those 
which  remain  for  ages  inaccessible  to  foreign  merchandize  and 
modes  of  thought. 

Such  being  the  situation  of  Greece,  it  could  not  but  happen,  that, 
as  soon  as  the  Greek  race  drew  itself  clear  from  primitive 
barbarism,  the  fertile  germs  of  art, — examples,  models,  processes, 
— should  penetrate  into  the  country  from  the  neighbouring  East 
by  all  the  channels  of  communication  which  we  have  m.entioned. 
Seeing;  how  far  civilization  had  advanced,  would  it  not  have  been 
absurd  for  the  Greeks  to  have  turned  a  deaf  ear  and  a  blind  eye  to 
the  experience  of  their  predecessors  ;  to  have  begun  again  at  the 
beginning  ?  Was  it  not  better  to  take  up  the  work  at  the  point 
where  it  had  been  left,  and  to  make  use,  for  future  developments, 
of  those  which  had  already  been  established  ?  Man  progresses  as 
fast  as  he  can  ;  as  soon  as  he  learns  any  new  method  of  satisfying 
his  wants  and  ameliorating  his  life,  he  makes  use  of  it  ;  he  makes 
use  of  it  at  first  in  its  original  form,  but  with  years  and  experience 
he  improves  it  and  brings  it  nearer  to  perfection. 

Thus  then,  the  more  we  study  the  past,  the  more  surely  do  we 
recognize  the  truth  contained  in  those  myths  and  tradifions  which 
betray  the  influence  exercised  upon  Greece  by  the  people  of 
Egypt,   Syria,    and    Asia    Minor.      To    confine    ourselves    to    the 


Introduction.  xxxv 


plastic  arts,  the  historian  of  Greek  art  discovers  survivals,  forms 
and  motives  which  had  been  employed  in  previous  centuries  and 
earlier  civilizations,  in  exact  proportion  to  the  accuracy  of  his 
researches,  and  to  the  number  of  his  elements  for  comparison. 
He  also  finds  that  the  Greeks  borrowed  from  the  same  instructors 
those  industrial  processes  which,  although  not  in  themselves  artistic, 
are  among  the  antecedent  conditions  of  art  ;  namely,  metallurgy, 
ceramics,  smith's  work,  glass-making,  weaving,  embroidery,  stone- 
working  and  carving,  in  a  word  all  those  trades  wdiich  seem  so 
simple  when  their  secrets  are  known,  but  which,  nevertheless,  re- 
present the  accumulated  efforts  of  countless  unknown  inventors. 

It  was  not  only  the  material  outfit  of  civilization  that  the  Greeks 
borrowed  from  their  predecessors  ;  they  obtained,  together  with 
that  alphabet  which  represents  the  principal  sounds  of  the  voice  by 
a  few  special  signs,  another  alphabet  which  has  been  happily  named 
the  alphabet  of  art,  certain  necessary  conventions,  combinations 
of  line,  ornaments,  decorative  forms,  a  crowd  of  plastic  elements 
which  they  had  employed  in  the  expression  of  their  own  ideas  and 
sentiments.  Even  after  Greek  art  had  reached  perfection  and  w^as 
in  the  full  enjoyment  of  her  own  individuality,  we  still  find  traces 
of  these  early  borrowings.  Sometimes  it  is  a  decorative  motive, 
like  the  sphinx,  the  griffin,  the  palm-leaf,  and  many  others,  which, 
invented  on  the  banks  of  the  Nile  or  the  Tigris,  w^ere  transported 
to  Greece  and  there  preserved  to  be  handed  down  to  our  modern 
ornamentist.  The  nearer  we  get  to  the  fountain  head  of  Greek  art, 
the  more  we  are  struck  with  these  resemblances,  which  are  some- 
thing beyond  mere  coincidences.  The  deeper  we  penetrate  into 
what  is  called  archaism,  the  more  numerous  do  those  features 
become  which  are  common  to  oriental,  especially  Assyrian,  art,  and 
that  of  Greece.  We  find  analogous  methods  of  indicating  the 
human  skeleton,  of  accenting  its  articulations,  of  representing  the 
drapery  with  which  the  forms  are  covered.  Greek  taste  had  not 
yet  so  transformed  the  details  of  ornamentation  as  to  prevent  us 
from  reco^nizine  the  motives  which  commerce  had  brouc^ht  for  its 
use  over  the  waves  of  the  yEgean  or  the  mountains  of  Asia  Minor. 
The  marks  of  their  origin  are  continually  visible,  and  yet  a  prac- 
tised eye  can  perceive  that  the  Greeks  were  never  satisfied,  like 
the  Phoenicians,  with  merely  combining  in  various  proportions 
the  materials  furnished  by  the  artisans  of  Egypt  or  Assyria ; 
the    facilities  of   such  a  soulless  and    indiscriminating  eclecticism 


xxxvi  Introduction. 


as  that  could  not   satisfy  the  ambitions   of  a   race    that    already- 
possessed  the  poetry  of  Hesiod  and  Homer. 

The  art  of  Greece  was  profoundly  original  in  the  best  sense  of 
the  word.  It  was  far  superior  to  all  that  went  before  It  ;  It  alone 
deserved  to  become  classic,  that  is,  to  furnish  a  body  of  rules 
and  laws  capable  of  being  transmitted  by  teaching.  In  what 
does  its  superiority  consist  ?  How  does  its  originality  show 
itself,  and  how  can  its  existence  be  explained  ?  These  are  the 
questions  which  we  propose  to  answer  ;  but  In  order  to  arrive 
at  a  just  conclusion  we  must  begin  with  the  study  of  those 
nations  to  whom  the  Greeks  went  to  school,  and  of  whose  art  they 
were  the  heirs  and  continuers.  We  should  be  unable  to  grasp 
the  exclusively  Greek  features  of  Greek  art  did  we  not  begin 
by  defining  the  foreign  elements  which  have  taken  their  part 
In  the  work,  and  that  we  can  only  do  by  going  back  to  the 
civilizations  in  which  they  were  produced  ;  we  must  endeavour 
to  penetrate  into  the  spirit  of  those  civilizations,  to  discover 
whence  they  started  and  how  far  they  progressed  ;  we  must 
first  define  their  ideas  of  the  beautiful,  and  then  show,  by  well- 
chosen  examples,  by  what  means  and  with  how  great  a  measure 
of  success,  they  realised  their  own  conception. 

We  undertake  this  long  detour  in  order  that  we  may  arrive 
in  Greece  instructed  by  all  that  we  have  learnt  on  the  way, 
and  prepared  to  understand  and  to  judge  ;  but  during  the  whole 
voyage  our  eyes  will  be  turned  towards  Greece,  as  those  of  the 
traveller  towards  his  long-desired  goal.  Our  route  will  conduct 
us  from  the  shores  of  the  Nile  to  those  of  the  Euphrates  and 
Tigris,  over  the  plains  of  Medea  and  Persia  and  Asia  Minor  to 
the  shores  of  Phoenicia,  to  Cyprus  and  Rhodes.  But  beyond  the 
obelisks  and  pyramids  of  Egypt,  beyond  the  towers  of  Chaldaea 
and  the  domes  of  Nineveh,  the  lofty  colonnades  of  Persepolis, 
the  fortresses  and  rock-cut  tombs  of  Phrygia  and  Lycia,  beyond 
the  huge  ramparts  of  the  cities  of  Syria,  we  shall  never  cease  to 
perceive  on  the  horizon  the  sacred  rock  of  the  Athenian  acropolis  ; 
we  shall  see  It  before  us,  as  our  history  of  the  past  advances, 
lifting  into  the  azure  sky  the  elegant  severity  of  Its  marble 
porticoes,  the  majesty  of  those  pediments  where  live  and  breathe 
the  gods  of  Homer  and  Phidias. 

When    we   have  crossed   the  threshold  of   the   Propylsea,  and 


Introduction.  xxxvii 


have  visited  the  Parthenon,  the  Erectha^um,  and  the  temple  of 
the  Wingless  Victory  ;  when  we  have  seen  all  Greece  become 
covered  with  monuments  of  architecture  and  sculpture,  which, 
without  rivalling  those  of  Athens  in  purity  of  line  or  finesse  of 
execution,  bear  the  impress  of  the  same  style  and  the  same 
taste ;  when  we  have  seen  Praxiteles  and  Scopas  succeed  to 
Phidias  and  Polycletus,  will  it  not  cost  us  a  struggle  to  quit 
the  scene  of  so  many  wonders  and  conclude  our  voyage  ?  If 
we  leave  the  Athens  of  Cimon,  of  Pericles,  and  Lycurgus,  for 
the  pompous  capitals  of  the  heirs  of  Alexander  ;  if  we  cross 
the  sea  to  visit  Veii  and  Clusium,  to  describe  the  Etruscan 
cemeteries  with  the  fantastic  magnificence  of  their  decoration  ; 
if  at  last  we  find  ourselves  in  imperial  Rome,  among  its  basilicas, 
its  baths,  its  amphitheatres,  and  all  the  sumptuous  evidence  of 
its  luxury,  we  shall  now  and  again  turn  our  eyes  with  regret  to 
what  we  have  left  behind  ;  and,  although  we  shall  endeavour 
to  comprehend  and  to  judge  with  the  liberality  and  largeness  of 
taste  and  sympathy  which  is  the  honour  of  contemporary  criti- 
cism, we  shall  sometimes  sigh  for  that  ideal  of  pure  and  sovereign 
beauty  which  we  adored  in  Greece ;  and  shall  feel,  now  and  again, 
the  nostalgia  of  the  exile. 


IV. 

In  this  sketch  of  our  plan,  we  have  reserved  no  place  for  the 
art  which  is  called  prehistoric,  the  art  of  the  caverns  and  the 
lake  dwellings.  This  omission  may  surprise  some  of  our  readers, 
and  we  therefore  beg  to  submit  for  their  consideration  the  rea- 
sons which,  after  grave  reflection,  have  induced  us  to  refrain  from 
retracing  the  first  steps  of  human  industry,  from  describing 
the  first  manifestations  of  the  plastic  instinct  of  mankind. 

We  are  actuated  by  neither  indifference  nor  disdain.  We  fully 
appreciate  the  importance  of  such  researches,  and  of  the  results 
to  which  they  have  led.  No  sooner  had  it  entered  into  the  mind 
of  man  to  look  for  and  collect  the  humble  remains  upon  which  so 
many  centuries  had  looked  with  indifference,  than  they  were 
found  almost  everywhere,  thickly  dispersed  near  the  surface  of  the 
earth,  heaped  among  the  bones  of  deer  in  the  grottoes  for  which 


xxxviii  Introduction. 


men  and  animals  had  once  contended,  buried  in  peat  marshes  and 
sandy  shores,  sometimes  even  sprinkled  upon  the  surface  of  the 
fields  and  country  roads.  Pieces  of  flint,  bone,  or  horn,  fashioned 
into  instruments  of  the  chase,  into  fishhooks,  and  domestic 
utensils  ;  shells,  perforated  teeth,  amber  balls  which  were  once 
strung  upon  necklaces  and  bracelets  ;  fragments  of  rough  tissue 
and  of  skin  garments  ;  seeds  and  carbonized  fruits  ;  earthen 
vessels  made  by  hand  and  dried  in  the  sun  or  simply  in  the 
open  air.  In  some  of  the  cave  dwellings,  bones  and  pieces 
of  horn  have  been  found  upon  which  the  figures  of  animals  are 
carved  with  a  truth  and  spirit  which  allow  their  species  to  be  at 
once  and  certainly  recognized. 

But  none  of  these  remains  bear  the  slightest  trace  of  a  system 
of  signs  for  the  transmission  of  ideas  or  recollections  ;  there  is 
nothing  which  suggests  writing ;  and,  more  significant  still,  there 
is  a  complete  absence  of  metaL  All  this  is  evidence  that  the 
remains  in  question  belong  to  a  very  remote  antiquity,  to  a  period 
much  nearer  the  primitive  barbarism  than  to  the  civilization  of 
Egypt  and  Chaldsea,  to  say  nothing  of  that  of  Greece  and  Rome. 
The  comparative  method,  which  has  done  so  much  for  natural 
science,  has  also  taken  these  remains  in.  hand,  has  attempted  to 
classify  them  and  to  gain  from  them  some  notion  of  the  life  led  by 
the  early  human  families  which  manufactured  them.  These  arms, 
tools,  and  instruments  which  have  been  recovered  from  the  soil 
of  the  old  and  cultivated  nations  of  Europe,  have  been  carefully 
compared  with  similar  objects  still  in  use  by  the  savage  races 
which  people  the  far  corners  of  the  world.  These  comparisons 
have  enabled  us  to  decide  the  former  use  of  each  of  the  objects 
discovered.  By  collating  the  observations  of  the  various  travellers 
who  have  visited  the  savage  races  in  question,  we  have  been 
enabled  to  form  for  ourselves  a  probably  truthful  picture,  so  far 
as  it  goes,  of  the  life  and  social  habits  of  those  primitive  Euro- 
peans who  made  use  of  similar  tools  and  weapons.  And  that 
is  not  all.  The  general  character  of  those  early  periods  being 
established,  further  examination  brought  to  light  the  local  differ- 
ences which  prevailed  then  as  now.  Thus  the  proneness  to 
plastic  imitation  seems  to  have  been  peculiar  to  a  few  tribes — 
although  traces  of  this  taste  are  found  elsewhere,  it  is  nowhere 
so  marked  as  among  those  primitive  cave-dwellers  of  Perigord 
whom    Christy  and    Edouard    Lartet    have    so    patiently  studied.. 


Introduction.  xxxix 


By  dint  of  careful  classification  and  comparison,  we  have  been 
enabled  to  follow  the  march  of  progress  through  those  countless 
centuries  whose  number  will  never  be  known  to  us,  whose  total  • 
would,  perhaps,  oppress  our  imaginations  if  we  knew  it ;  we  have 
been  enabled  to  discover  the  slow  steps  by  which  mankind  raised 
itself  from  the  earliest,  almost  shapeless,  flint  axe,  found  with  the 
bones  of  the  mammoth  in  the  quaternary  alluvial  deposits,  to 
the  rich  and  varied  equipment  of  *'  lacustrian  civilization,"  as  it 
has  sometimes  been  called.  In  this  unlimited  field,  of  which  one 
side  at  least  must  ever  be  lost  in  unfathomable  obscurity,  the  main 
divisions  have  been  traced  ;  the  stone  age  has  been  defined  and 
divided  into  the  palcsolithic  and  neolithic  epochs  ;  the  age  of 
bronze  followed,  and  then  came  the  iron  age.  With  the  appear- 
ance of  the  former  metal  the  tribes  of  northern  and  central 
Europe  established  a  connection  with  the  civilized  races  which 
surrounded  the  Mediterranean,  and  with  iron  we  are  in  the  full 
classic  period. 

We  can  never  be  too  grateful  for  the   persevering  labours  of 

those  who  have  carried   on  these  researches  in  every  corner  of 

Europe  ;    their   deserts   are    all  the   greater   from    the    fact   that 

they  could   never   count   upon    those    agreeable    surprises    which 

come  now  and  then  to  reward  excavators  on  the  sites  of  ancient 

and    historic   cities.      Their   chances    are  small    of   finding  those 

objects   of   art   which,  by  their  beauty  and  elegance,   repay   any 

amount  of   toil  and  expense.       The   remains    which    they  bring 

to   light   have    little    to    say   to    our   aesthetic    perceptions ;    they 

repeat    a    few    types    with    an    extreme    monotony  ;     but,  on    the 

other  hand,  they  carry  our  thoughts  back  to  a  point  far  nearer  the 

cradle  of  our  race   than  the  myths  of  early  history  or  even   the 

monumental    remains  of  Egypt  and    Chaldaea.     They  cast   some 

slight    illumination   upon   those   distant   ages   of    which   humanity 

has    preserved    no    recollection.       They    people     with    unknown 

multitudes   those    remote    epochs    into   which    scientific    curiosity 

had,  but   yesterday,   no    desire   to  penetrate.     There  can    be    in 

all  this  no  real  question  of  chronology,  but  when  from  the  sands 

of  Abbeville  or  the  caverns  of  Perigord  we  dig  up  the  first  flint 

implements    or   those  fragments   of   bone,   of  ivory,  of   reindeer 

horn,  which  have  preserved  to  us  the  first  attempts  made  by  man 

to   copy   the  outlines    of  living    beings,    it    takes  us    far    beyond 

those    days    of    which    our    only    knowledge    comes    from   vague 


xl  Introduction. 


tradition,  and  still  farther  beyond  those  centuries  which  saw  the 
first  struggling  dawn  of  history. 

We  have,  then,  decided  not  to  embark  upon  these  questions  of 
prehistoric  art,  because,  as  the  title  which  we  have  chosen  declares, 
we  propose  to  write  a  history,  and  the  word  history,  when  the 
human  race  is  in  question,  implies  established  relations  between 
certain  groups  of  facts  and  certain  portions  of  time,  measured  at 
least  with  something  approaching  to  probable  truth. 

We  do  not  yet  possess,  probably  we  shall  never  possess,  any 
means  of  estimating  even  within  five  or  six  thousand  years,  the 
actual  duration  of  the  stone  age.  From  all  analogies  progress  must 
have  been,  in  the  beginning,  exceedingly  slow  ;  like  that  of  a  falling 
body,  the  rapidity  of  industrial  progress  is  continually  accelerating. 
This  acceleration  is  not  of  course  quite  regular  ;  the  phenomena  of 
social  life  are  too  complex,  the  forces  at  work  are  too  numerous  and 
sometimes  too  contrary  to  allow  us  to  express  it  by  the  mathema- 
tical formula  which  may  be  applied  to  movement  in  the  physical 
world  ;  but  on  the  whole  this  law  of  constantly  accelerated  progress 
holds  good,  as  indeed  may  be  historically  proved.  So  long  as  man 
had  to  do  without  metals,  each  generation,  in  all  probability,  added 
but  little  to  the  discoveries  of  that  which  preceded  it ;  most  likely 
after  each  happy  effort  many  generations  succeeded  one  another 
without  any  further  attempt  to  advance.  Ever  since  they  have 
been  under  our  observation,  the  savage  races  of  the  world  have 
been  practically  stationary  except  where  European  commerce  has 
profoundly  modified  the  conditions  of  their  lives.  It  is  probable, 
therefore,  that  more  centuries  rolled  away  between  the  first  chipped 
flints  and  the  well  polished  weapons  which  succeeded  them  than 
between  the  latter  and  the  earliest  use  of  bronze.  But  we  cannot 
prove  that  it  was  so,  nor  satisfy  those  whom  probability  and  a 
specious  hypothesis  will  not  content.  Where  neither  written  evidence 
nor  oral  tradition  exist  there  can  be  little  question  of  historic  order. 
The  remains  of  the  stone  age  are  not  calculated  to  dissipate  the 
silence  which  enshrouds  those  centuries.  In  the  art  of  a  civilized 
people  we  find  their  successive  modes  of  feeling  and  thought 
interpreted  by  expressive  forms  ;  we  may  even  attempt  under  all 
reserve  to  sketch  their  history  with  the  sole  aid  of  their  plastic 
remains.  The  chances  of  error  would  of  course  be  numerous  ;  but 
yet  if  all   other   materials   had,  unhappily,  failed   us,   the  attempt 


Introduction.  xli 


would  have  been  well  worth  making.  The  more  ancient  portions 
of  our  prehistoric  collections  do  not  offer  the  same  opportunities  ; 
they  are  too  simple  and  too  little  varied.  The  primitive  savage 
who  moulded  matter  to  his  will  with  great  and  painful  difficulty, 
could  impress  upon  it  nothing  but  those  gross  instincts  which  are 
common  to  man  and  beast ;  we  can  discover  nothing  from  his  works, 
beyond  the  means  which  he  employed  in  his  struggles  with  his 
enemies,  and  in  his  never-ending  effort  to  procure  food  for  himself. 

The  word  history  cannot  then  be  pronounced  in  connection  with 
these  remote  periods,  nor  can  their  remains  be  looked  upon  in  any 
sense  as  works  of  art.  Art  commences  for  us  with  man's  first 
attempts  to  impress  upon  matter  some  form  which  should  be  the 
expression  of  a  sentiment  or  of  an  idea.  The  want  of  skill  shown 
in  these  attempts  is  beside  the  question  ;  the  mere  desire  on  the 
part  of  the  workman  renders  him  an  artist.  The  most  hideous 
and  diseustine  of  those  idols  in  stone  or  terra-cotta  which  are 
found  in  the  islands  of  the  Greek  Archipelago,  at  Mycenae  and  in 
Boiotia,  idols  which  represent,  as  we  believe,  the  great  goddess 
mother  whose  worship  the  Phoenicians  taught  to  the  Greeks,  are 
works  of  art  ;  but  we  are  unable  to  grive  that  title  to  the  axes 
and  arrowheads,  the  harpoons  and  fish-hooks,  the  knives,  the  pins, 
the  needles,  the  crowds  of  various  utensils  which  we  see  in  the 
glass  cases  of  a  pre-historic  museum  ;  all  this,  interesting  though  it 
be  to  those  who  wish  to  study  the  history  of  labour,  is  nothing  but 
an  industry,  and  a  rudimentary  industry,  which  is  content  with 
supplying  the  simplest  wants.  It  is  not  until  we  reach  the  sculp- 
tures c^f  the  cave-dwellings  that  we  find  the  first  germs  of  artistic 
effort,  and  in  truth,  man  did  not  cut  the  figures  of  animals  upon  the 
handles  of  his  tools  and  upon  those  objects  which  have  been  called, 
perhaps  a  little  recklessly,  batons  of  command,  for  any  utilitarian 
purpose  ;  it  was  to  give  himself  pleasure,  it  was  because  he  found 
true  aesthetic  enjoyment  in  copying  and  interpreting  living  nature. 
Art  was  born,  we  may  acknowledge,  with  those  first  attempts  at 
the  representation  of  life,  and  it  might  fairly  be  expected  that  our 
history  should  commence  with  them,  were  it  not  that  they  offer 
no  sequence,  no  starting  point  for  any  continuous  movement  like 
that  which,  beginning  in  Egypt  and  Chaldaea,  was  prosecuted  in 
Greece  and  led  in  time  to  such  high  developments  ;  even  its 
competent  students  confess  that  the  art  of  the  cave-men  was  an 
isolated  episode  without  fruition  or  consequence.      Specimens  of 

VOL.  I.  ^ 


xlii  Introduction. 


this  art  are  found  at  but  a  few  points  of  the  vast  surface  over 
which  the  vestiges  of  primitive  man  are  spread,  and  neither 
in  the  neohthic  age,  nor  even  in  that  of  bronze — both  far  in 
advance  in  other  ways  of  that  of  the  cave-dwelhngs — -does  it 
ever  seem  to  have  entered  into  the  mind  of  man  to  copy  the  types 
offered  to  him  by  the  organic  world,  still  less  those  of  mankind, 
which,  however,  had  long  before  been  roughly  figured  in  one  or 
two  caves  in  the  Dordogne.^ 

Towards  the  close  of  the  prehistoric  age  the  taste  for  ornament 
becomes  very  marked,  but  that  ornament  is  always  of  the  kind 
which  we  call  geometric.  Hardly  a  single  decorative  motive  is 
taken  from  the  vegetable  world.  Like  the  rude  efforts  of  the  cave- 
men, this  decoration  proves  that  those  by  whom  it  was  imagined 
and  who  frequently  employed  it  with  such  happy  results,  were  not 
contented  with  bare  utility,  but,  so  far  as  they  could,  sought  after 
beauty.  A  secret  instinct  worked  in  them  and  inspired  them  with 
the  desire  to  give  some  appearance  of  elegance  to  the  objects  which 
they  had  in  daily  use.  This  geometrical  style  of  decoration 
prevailed  all  over  central  Europe  until,  in  the  first  place,  the  move- 
ments of  commerce  with  Greece  and  Etruria,  and  secondly  the 
Roman  conquest,  introduced  the  methods  of  classic  art. 

From  what  we  have  said,  it  will  be  seen  that  we  could  not  have 
passed  over  in  silence  this  system  of  ornamentation  ;  but  we  shall 
again  find  it  in  our  path  when  we  come  to  treat  of  that  pre- 
historic Greece  which  preceded  by  perhaps  two  or  three  centuries 
the  Greece  of  Homer.  By  the  help  of  the  discoveries  which 
have  been  lately  made  in  the  Troad,  at  Mycenae,  and  in  other 
ancient  sites,  we  shall  study  the  works  produced  by  the  ancestors 
of  the  Greeks  before  they  went  to  school  to  the  nations  of  the 
East.  But  even  with  the  discoveries  which  carry  us  farthest  back, 
we  only  reach  the  end  of  the  period  in  question,  when  maritime 
commerce  had  already  brought  to  the  islands  and  the  mainland  of 
Greece  objects  of  Egyptian,  Phoenician  or  Chaldaic  manufacture, 
but  before  those  objects  were  sufficiently  numerous  or  the  relations 
with  those  countries  sufficiently  intimate  to  produce  any  great 
effect  upon  the  habits  of  native  workmen.  Among  the  deposits  to 
which  we  have  alluded,  it  is  generally  possible  to  distinguish  those 
works    which  are    of   foreign  origin ;  and    such  works    excluded, 

'  Dictionnaire  arch'eologique  dc  la  Gaule,  vol.  i.  Cavernes,  figure  28.   Al.  Bertrand, 
Archiologie  celtique  et  gauloise  (i  vol.  8vo.     Didier,  1876,  p.  68). 


Introduction.  xliii 


it  is  easy  to  form  a  sufficiently  accurate  general  idea  of  the  art 
practised  by  the  forefathers  of  the  historic  Greeks — by  the  Felas- 
gians,  to  use  a  conventional  term.  So  long  as  it  was  left  to  its 
own  inspiration,  Pelasgic  art  did  not  differ,  in  its  general  character- 
istics, from  that  of  the  various  peoples  spread  over  the  continent  of 
Europe,  and  still  practised  for  centuries  after  the  dawn  of  Greek 
civilization  in  the  great  plains  to  the  north  of  the  Alps  and  the 
Danube,  Its  guiding  spirit  and  its  motives  are  similar.  There  is 
the  same  richness,  or  rather  the  same  poverty,  the  same  combi- 
nations produced  by  a  small  number  of  never-changing  linear 
elements.  One  would  say  that  from  the  shores  of  the  great  ocean 
and  the  Baltic  to  those  of  the  Mediterranean,  all  the  workmen 
laboured  for  the  same  masters.  Struck  by  this  resemblance,  or 
rather  uniformity,  one  of  the  most  eminent  of  German  archaeo- 
logists, Herr  Conze,  has  proposed  that  this  kind  of  ornament  shall 
be  called  Indo-European  ;  he  sees,  in  the  universality  of  the 
system,  a  feature  common  to  all  branches  of  the  Aryan  race,  a 
special  characteristic  which  may  serve  to  distinguish  it  from  the 
Semites. 

Objections  have  been  brought  against  this  doctrine  of  which 
Herr  Conze  himself  has  recognized  the  gravity  ;  by  numerous 
examples  taken  from  the  art  of  nations  which  do  not  belong  to  the 
Aryan  family,  it  has  been  shown  that,  human  nature  being  the 
same  everywhere,  all  those  peoples  whose  development  has  been 
normal,  neither  interrupted  nor  accelerated  by  external  causes, 
have,  at  some  period  of  their  lives,  turned  to  the  style  in  question 
for  the  decoration  of  their  weapons,  of  their  earthenware,  their 
furniture,  their  apparel  and  their  personal  ornaments.  The  less 
richly  endowed  among  them  would  have  stopped  at  that  point 
but  for  the  example  of  their  neighbours,  who  stirred  them  on  to 
new  attempts  and  further  progress  ;  others  advanced  without  im- 
pulse from  other  sources  than  their  own  instincts,  they  reproduced 
vegetable  and  animal  forms,  and  finally  the  human  figure  in  all 
its  beauty  and  nobility.  It  was  the  same  with  letters.  Among 
the  nations  which  have  made  a  name  in  history  how  few  there  are 
that  possess  a  true  literature,  a  poetry  at  once  inspired  and 
critical  !  All  however,  under  one  form  or  another,  have  a  popular 
poetry  which  is  more  or  less  varied  and  expressive. 

The  trace  of  this  earliest  spontaneous  effort,  of  this  first  naive 
product  of  the  imagination,  never  entirely  disappears  in  a  literature 


xliv  Introduction. 


which  is  hfe-Hke  and  sincere  ;  it  is  found  even  in  the  most  perfect 
works  of  its  classic  period.  In  the  same  way  the  most  advanced 
and  refined  forms  of  art  draw  a  part  of  their  motives  and  effects 
from  geometrical  decoration.  This  style  therefore  should  be 
studied  both  for  its  principle  and  for  the  resources  of  which  it 
disposes,  but  as  we  shall  have  to  notice  it  when  we  treat  of  Greece, 
it  seems  to  us  better  to  adjourn  till  then  any  discussion  of  its 
merits.  Both  in  Greece  and  Italy  approximate  dates  can  be  given 
to  the  monuments  which  it  ornaments,  they  can  be  placed  in  their 
proper  historical  position,  which  is  by  no  means  the  case  with  the 
objects  gleaned  throughout  central  Europe. 

There  is  another  consideration  of  still  greater  importance  ;  the 
artistic  remains  of  Greece  form  an  almost  unbroken  series,  from 
the  humble  and  timid  attempts  of  nascent  sculpture  to  the 
brilliant  masterpieces  of  Phidias  and  Polycletus,  and  show  the  steps 
by  which  the  artist  succeeds  in  passing  from  one  style  to  another, 
from  curves  and  interlacing  lines,  from  all  mere  abstract  combi- 
nations, to  the  imitation  of  nature,  to  the  representation  of  bodies 
which  breathe,  feel,  and  speak,  which  move  and  struggle.  Else- 
where force  has  either  been  wanting  for  this  development,  or 
evidence  of  the  transition  has  escaped  our  researches.  Nothing 
can  be  much  more  imperfect  or  more  conventional  than  the  figures 
which  we  find  upon  some  of  the  painted  vases  from  Mycenae  and 
Cyprus,^  upon  which  the  workman's  hand,  accustomed  to  straight 
lines  and  circles,  or  segments  of  circles,  has  succeeded  in  suggesting 
by  those  means  the  figures  of  birds  and  fighting  men.  Nothing 
could  be  farther  from  the  subtlety  and  variety  of  the  contours 
presented  by  living  organisms.  But  in  spite  of  all  this,  art  was 
born  with  the  awakening  of  this  desire  to  reproduce  the  beauty 
and  mobility  of  living  forms.  All  that  had  preceded  it  was  but 
the  vague  murmuring  of  a  wish  which  had  not  yet  become  self 
conscious  ;  but,  at  last  the  intellect  divined  the  use  to  which  it 
might  be  put,  and  guessed  at  the  part  which  might  be  played  by 
the  plastic  instincts  with  which  it  felt  itself  endowed.  All  the  rest 
depended  upon  natural  gifts,  upon  time  and  circumstance  ;  the 
march  along  the  road  of  progress  began,  and  although  its  rapidity 
was  intermittent,   it   was  certain   to  arrive,   if   not  always   at  the 

^  ScHLiEMANN,  MycetKB,  sec  figs.    33   and  213:  Cesxola,   Cyprus,  see  pis.  44 
and  46. 


Introduction.  xl 


V 


production  of  masterpieces  of  divine  beauty,  at  least  at  sufficient 
competence  in  painting  and  modelling  to  transmit  the  types  of  a 
race  and  the  images  of  its  gods  to  posterity. 

The  student  of  plastic  art   finds  in  the  remains   of  prehistoric 
times  rather  a  tendency  to  the  creation  of  art,  than  art  itself;  by 
postponing  our  study  of  this  tendency  until  we  come  to  investigate 
the  origin  of  Greek  and  Italian  art,  we  are  enabled  to  avoid  all  ex- 
cursion beyond  the  limits  implied  by  our  title,  beyond  that  which  is 
generally  called  antiquity.     The  conventional  meaning  of  this  word 
embraces    neither   the    primitive  savages    who    chipped    the    first 
flint,  nor  the  cave-men,  but  it  calls  up  before  our  eyes  the  brilliant 
cities  of  northern  Africa  and  hither- Asia,  of  Greece  and  Italy,  with 
which  our   school-days  have   made  us  familiar  ;   it  reminds  us  of 
those  nations  whose  stories  we  learnt  from  the  sacred  and  profane 
authors   whose  works   we  read  in   our   youth  ;  and   our  thoughts 
revert  to  their  grandiose  monuments  of  architecture  and  sculpture, 
to    their   masterpieces    of  poetry   and  eloquence,   to    those  great 
works  of  literature  in  which  we  took  our  first  lessons  in  the  art  of 
writing  and  speaking.     Behind  all  these  images  and  associations 
the  intelligence  of  an  educated  man  tells  him — and  the  discoveries 
of  science  every  day  make  the  fact  more  certain — that  in  the  ancient 
as  in  the  modern  world,  the  nations  which  figure  upon  the  stage  of 
history    were    not    isolated  ;  they    each    had    neighbours   who  in- 
fluenced them,  or  whom  they  influenced,  by  commerce  or  conquest ; 
each  also  received  something  from  its  predecessors,  and  in  turn 
transmitted  the  results'of  its  labour  to  those  which  came  after  it ; 
in  a  word,  the  work  of  civilization  was  continuous  and  universal. 
The  nations  which,  for  three  or  four  thousand  years,  were  grouped 
round  the  basin  of  the   Mediterranean,  belonged  to  one  historical 
system  ;  to  those  who  take  a  wide  grasp  of  facts  they  are  but  the 
members  and  organs  of  one  great  body,  in  which    the    nervous 
centres,  the   sources  of  life,   of  movement  and  of  thought,  slowly 
gravitated  with  the  efiluxion  of  time  from  the  east  to  the  west, 
from  Memphis  and  Babylon  to  Athens  and  Rome. 
"  As  for  the  populations  which,  long  before  the  opening  of  this 
period  and  during  the  whole  of  its  duration,  lived  on  the  north  of 
the  Danube,  the  Alps,  and  the  Pyrenees,  they  do  not  belong  to  the 
same   system  ;  they  were  attached  to  it  by  the  Roman  conquest, 
but  at  a  very  late  period ;   not  long,  indeed,  before  the  triumph  of 
Christianity,  the  invasion  of   the  barbarians,  and  the  fall  of  the 


Ivi  Introduction. 


xivi 


empire,  led  to  the  dissolution  of  the  antique  system  and  the 
substitution  for  it,  after  centuries  of  confusion  and  violence,  of  the 
wider  and  more  comprehensive  civilization  of  modern  Europe, 
a  civilization  which  was  destined  to  cross  every  sea  and  to  spread 
itself  over  the  whole  surface  of  the  globe.  As  soon  as  the 
victories  of  the  Roman  legions,  and  the  construction  of  the  great 
roads  which  united  Rome  with  her  most  distant  provinces,  had 
brought  them  into  constant  communication  with  the  maritime 
cities  of  the  Mediterranean,  these  barbaric  nations,  who  had 
neither  history,  nor  letters,  nor  expressive  art,  received  them  from 
their  conquerors,  whose  very  language  they  all,  or  nearly  all, 
adopted  ;  and  for  all  this  they  gave  practically  nothing  in  return. 
Elsewhere,  the  old  world  had  almost  finished  its  task.  It  had 
exhausted  every  form  in  which  those  ideas  and  beliefs  could  be 
clothed  which  it  had  kept  unchanged,  or  little  changed,  for  mil- 
lenium  after  millenium.  The  old  world  employed  such  force  and 
vitality  as  remained  to  it  in  giving  birth  to  the  new,  to  that 
religion  which  has  led  to  the  foundation  of  our  modern  social 
and  political  systems.  These  also  were  to  have  their  modes 
of  expression,  rich  and  sonorous  enough,  but  dominated  by 
analysis  ;  they  were  to  have  arts  and  literatures,  which  have  given 
expression  to  far  more  complex  ideas  than  those  of  antiquity. 
The  Celts  and  Teutons,  the  Slavs  and  Scandinavians,  all  those 
tribes  which  the  Romans  called  barbarous,  have,  in  spite  of  the 
apparent  poverty  of  their  share,  made  an  important  contribution  to 
the  civilization  into  which  they  plunged  at  so  late  a  period,  when 
they  did  so  much  to  provide  a  foundation  for  those  modes  of 
thought  and  feeling  which  are  only  to  be  found  in  modern  society. 
These  races  do  not  belong,  then,  to  what  we  call  antiquity. 
They  are  separated  from  it  by  many  things  ;  they  have  no  history, 
they  have  neither  literary  and  scientific  culture  nor  anything  that 
deserves  the  name  of  art.  Hidden  behind  a  thick  curtain  of  moun- 
tains and  forests,  sprinkled  over  vast  regions  where  no  towns 
existed,  they  remained  in  their  isolation  for  thousands  of  years, 
furnishing  to  civilization  nothing  but  a  few  rough  materials  which 
they  themselves  knew  not  how  to  use  ;  they  took  no  part  in  the 
work  which,  throughout  those  ages,  was  being  prosecuted  in  the 
great  basin  of  the  Persian  Gulf  and  the  Mediterranean,  in  that 
accumulation  of  inventions  and  creations  which,  fixed  and  pre- 
served by  writing  and  realized  by  art,  form  the  common  patrimony 


Introduction.  xl 


Vll 


of  the  most  civilized  portion  of  the  human  species.  When,  at  a 
late  hour,  these  nations  entered  upon  the  scene,  it  was  as  disturbers 
and  destroyers, — and  although  they  helped  to  found  modern 
society,  they  produced  none  of  those  elements  left  to  us  by 
antiquity  and  preserved  for  us  by  that  Rome  in  whose  hands  the 
heritage  of  Greece  was  concentrated. 


V. 

We  have  different,  but  equally  valid,  reasons  for  leaving  that 
which  is  called  the  far  East — India,  China,  and  Japan — outside  the 
limit  of  our  studies.  Those  rich  and  populous  countries  have, 
doubtless,  a  civilization  which  stretches  back  nearly  as  far  as  that  of 
Egypt  and  Assyria,  a  civilization  which  has  produced  works  both 
of  fine  and  of  industrial  art  which  in  many  respects  equalled  those 
of  the  nations  with  which  we  are  now  occupied.  In  all  those 
countries  there  are  buildings  which  impress  by  their  mass  and  by 
the  marvellous  delicacy  of  their  ornamentation,  sculptures  of  a 
singular  freedom  and  power,  and  decorative  painting  which  charms 
by  its  skilful  use  of  brilliant  colour  as  well  as  by  the  facility  and 
inventive  fancy  of  its  design.  The  representation  of  the  human 
figure  has  never  reached  the  purity  of  line  or  nobility  of  expression 
of  a  Greek  statue,  but,  on  the  other  hand,  the  science  of  decoration 
has  never  been  carried  farther  than  by  the  wood-carvers,  weavers 
and  embroiderers  of  Hindostan,  and  the  potters  of  China  and 
Japan. 

These  styles  have  their  fanatical  admirers  who  see  nothing  but 
their  brilliant  qualities  ;  they  have  also  their  detractors,  or  at  least 
their  severe  judges,  who  are  chiefly  struck  by  their  shortcomings, 
but  no  one  attempts  to  deny  that  each  of  those  nations  possesses  an 
art  which  is  always  original,  and  sometimes  of  great  and  rare  power. 
Why  then,  it  may  be  asked,  do  we  refuse  to  comprehend  the  more 
ancient  monuments  of  India  and  China,  those  which  by  their  age 
belong  to  the  centuries  with  which  we  are  concerned,  in  this  work  '^ 
Our  motives  may  be  easily  divined. 

W^e  might  allege  our  incompetence  for  such  an  extended  task, 
which  would  be  enough  to  occupy  several  lives.      But  we  have  a 


:lvlii  Introduction. 


still  more  decisive  reason.  Neither  Aryan  India  nor  Turanian 
China  belongs  to  the  antiquity  which  we  have  defined,  and  as  for 
Indo-China  and  Japan  they  are  but  annexes  to  those  two  great 
nations ;  religion,  written  characters,  the  industrial  and  plastic 
arts — all  came  to  them  from  one  or  the  other  of  those  two  ereat 
centres  of  civilization. 

So  far  as  China  is  concerned  no  doubt  or  hesitation  is  possible. 
Down  almost  to  our  own  days  China  and  its  satellites  had  no  deal- 
ings with  the  western  group  of  nations.  It  is  a  human  family  which 
has  lived  in  voluntary  isolation  from  the  rest  of  its  species.  It  is 
separated  from  western  mankind  by  the  largest  of  the  continents, 
by  deserts,  by  the  highest  mountains  in  the  world,  by  seas  once 
impassable,  finally,  by  that  contempt  and  hatred  of  everything 
foreign  which  such  conditions  of  existence  are  calculated  to  engen- 
der. In  the  course  of  her  long  and  laborious  existence  China  has 
invented  many  things.  She  was  the  first  to  discover  several  of 
those  instruments  and  processes  which,  in  the  hands  of  Europeans, 
have,  in  a  few  centuries,  changed  the  face  of  the  world  ;  not  only 
did  she  fail  to  make  ^ood  use  of  her  inventions,  she  guarded  them 
so  closely  that  the  West  had  to  invent  them  anew.  We  may  cite 
printing  as  an  example  ;  nearly  two  hundred  years  before  our  era 
the  Chinese  printed  with  blocks  of  wood.  On  the  other  hand,  every 
useful  discovery  made  in  the  period  and  by  the  group  of  nations  to 
whom  we  mean  to  confine  our  attention,  from  the  time  of  Menes 
and  Ourkham,  the  first  historic  kings  of  Egypt  and  Chaldsea,  to  the 
latest  of  the  Roman  Emperors,  has  been  turned  to  the  profit  of  others 
than  its  authors,  and  forms,  so  to  speak,  part  of  the  public  wealth. 
A  single  alphabet,  that  which  the  Phoenicians  extracted  from  one  of 
the  forms  of  Egyptian  writing,  made  the  tour  of  the  Mediterra- 
nean, and  served  all  the  nations  of  the  ancient  world  in  turn  for 
preserving  their  thoughts  and  the  idiom  of  their  language.  A 
system  of  numerals,  of  weights  and  measures,  was  invented  in 
Babylon  and  travelled  across  western  Asia  to  be  adopted  by  the 
Greeks,  and,  through  the  mediation  of  the  Greek  astronomers  and 
geographers,  has  given  us  the  sexagesimal  division  which  we  still 
employ  for  the  partition  of  a  circumference  into  degrees,  minutes 
and  seconds. 

From  this  point  of  view,  then,  there  is  a  profound  difference 
between  Egypt  or  Chaldaea,  and  China.  The  most  remote  epochs 
in  the  history  of  China  do  not  belong   to   antiquity  as  we  have 


Introduction.  xllx 


defined  the  term.  Without  knowing  It  or  wishing  it,  all  those 
nations  included  In  our  plan  laboured  for  their  neighbours  and 
for  their  successors.  Read  as  a  whole,  their  history  proves  to  us 
that  they  each  played  a  part  in  the  gradual  elaboration  of  civilized 
life  which  was  absolutely  necessary  to  the  total  result.  But  when 
China  is  In  question  our  impression  Is  very  different,  our  intellects 
are  quite  equal  to  imagining  what  the  world  would  have  been  like 
had  that  Empire  been  absolutely  destroyed  centuries  ago,  with  all 
its  art,  literature,  and  material  wealth.  Rightly  or  wrongly,  we 
should  not  expect  such  a  catastrophe  to  have  had  any  great  effect 
upon  civilization  ;  we  should  have  been  the  poorer  by  a  few 
beautiful  plates  and  vases,  and  should  have  had  to  do  without  tea, 
and  that  would  have  been  the  sum  of  our  loss. 

The  case  of  India  is  different.  Less  remote  than  China,  bathed 
by  an  ocean  which  bore  the  fleets  of  Egypt,  Chaldsea,  Persia, 
Greece  and  Rome,  she  was  never  beyond  the  reach  of  the  western 
nations.  The  Assyrians,  the  Persians,  and  the  Greeks  carried 
their  arms  Into  the  basin  of  the  Indus,  some  portions  of  which 
were  annexed  for  a  time  to  those  Empires  which  had  their  centre 
in  the  valley  of  the  Euphrates  and  stretched  westwards  as  far  as  the 
Mediterranean.  There  was  a  continuous  coming  and  going  of  cara- 
vans across  the  plateau  of  Iran  and  the  deserts  which  lie  between  it 
and  the  oases  of  Bactrlana,  Aria,  and  Arachosia,  and  through  the 
passes  which  lead  down  to  what  is  now  called  the  Punjab  ;  between 
the  ports  of  the  Arabian  and  Persian  gulfs  and  those  of  the  lower 
Indus  and  the  Malabar  coast,  a  continual  commercial  movement 
went  on  which,  though  fluctuating  with  time,  was  never  entirely 
interrupted.  From  the  latter  regions  western  Asia  drew  her 
supplies  of  aromatic  spices,  of  metals,  of  precious  woods,  of  jewels, 
and  other  treasures,  all  of  which  came  mainly  by  the  sea  route. 

All  this,  however,  was  but  the  supply  of  the  raw  material  for 
Egyptian,  Assyrian,  and  Phoenician  industries.  There  is  no  evi- 
dence that  up  to  the  very  last  days  of  antique  civilization  the 
Inhabitants  of  Hlndostan  with  all  their  depths  and  originality  of 
thought  ever  exercised  such  Influence  upon  their  neighbours  as 
could  have  made  itself  felt  as  far  as  Greece.  The  grand  lyric  poetry 
of  the  Vedas,  the  epics  and  dramas  of  the  following  epoch,  the 
religious  and  philosophical  speculations,  those  learned  grammatical 
analyses  which  are  now  admired  by  philologists,  all  the  rich  and 

VOL.  I.  .  h 


1  Introduction. 


brilliant  intellectual  development  of  a  race  akin  to  the  Greeks 
and  in  many  ways  no  less  richly  endowed,  remained  shut  up  in 
that  basin  of  the  Ganges  into  which  no  stranger  penetrated 
until  the  time  of  the  Mohammedan  conquest.  Neither  Egyptians, 
Arabs  nor  Phoenicians  reached  the  true  centres  of  Hindoo 
civilization  ;  they  merely  visited  those  sea-board  towns  where 
the  mixed  population  was  more  occupied  with  commerce  than 
with  intellectual  pursuits.  The  conquerors  previous  to  Alexander 
did  no  more  than  reach  the  gates  of  India  and  reconnoitre  its 
approaches,  while  Alexander  himself  failed  to  penetrate  beyond 
the  vestibule. 

Let  us  suppose  that  the  career  of  the  Macedonian  hero  had 
not  been  cut  short  by  the  fatigues  and  terrors  of  his  soldiers. 
So  far  as  we  can  judge  from  what  Megasthenes  tells  us  of 
Palibothra,  the  capital  of  Kala9oka,  the  most  powerful  sovereign  in 
the  valley  of  the  Ganges  in  the  time  of  Seleucus  Nicator,  the  Greeks 
would  not,  even  in  that  favoured  region,  have  found  buildings  which 
they  could  have  studied  with  any  profit,  either  for  their  plan, 
construction,  or  decoration.  Recent  researches  have  proved 
Meeasthenes  to  be  an  intellio-ent  observer  and  an  accurate 
narrator,  and  he  tells  us  that  in  the  richest  parts  of  the  country  the 
Hindoos  of  his  time  had  nothing  better  than  wooden  houses,  or 
huts  of  pise  or  rough  concrete.  The  palace  of  the  sovereign, 
at  Palibothra,  impressed  the  traveller  by  its  situation,  its  great 
extent,  and  the  richness  of  its  apartments.  It  was  built  upon 
an  artificial,  terraced  mound,  in  the  midst  of  a  vast  garden.  It 
was  composed  of  a  series  of  buildings  surrounded  by  porticos, 
which  contained  large  reception  halls  separated  from  one  another 
by  courtyards  in  which  peacocks  and  tame  panthers  wandered 
at  will.  The  columns  of  the  principal  saloons  were  gilt.  The 
general  aspect  was  very  imposing.  The  arrangements  seem 
to  have  had  much  in  common  with  those  of  the  Assyrian  and 
Persian  palaces.  But  there  was  one  capital  distinction  between 
the  two ;  at  Palibothra  the  residence  of  the  sovereign,  like  those 
of  his  subjects,  was  built  of  wood.  With  its  commanding 
position,  and  the  fine  masses  of  verdure  with  which  it  was 
surrounded,  it  must  have  produced  a  happy  and  picturesque 
effect,  but,  after  all,  it  was  little  more  than  a  collection  of 
kiosques.  Architecture,  worthy  of  the  name,  began  with  the 
employment   of  those  solid  and   durable    materials   which   defend 


Introduction.  ]i 


themselves  against  destruction  by  their  weight  and  constructive 
repose. 

The  other  arts    could   not  have    been    much   more   advanced. 
Ignorant    as    they    were   of  the    working   of  stone   for  building, 
these  people  can   hardly    have    been  sculptors,    and   as    to   their 
painting,    we    have    no    information.      There    is,    moreover,    no 
allusion  to  works  of  painting   or   sculpture    in    their   epics    and 
dramas,  there    are    none   of  those   descriptions   of  pictures    and 
statues  which,  in  the  writings  of  the  Greek  poets  and  dramatists, 
show  us  that  the  development  of  the  plastic  arts  followed  closely 
upon    that   of  poetry.     This    difference  between    the    two    races 
may    perhaps   be    explained    by    the    opposition    between    their 
religions  and,  consequently,  their  poetry.      In  giving  to  their  gods 
the  forms  and  features  of  men,  the  oldest  of  the  Greek  singers 
sketched  in  advance  the  figures  to  be  afterwards  created  by  their 
painters  and  sculptors.      Homer  furnished  the  sketch  from  which 
Phidias  took  his  type  of  the  Olympian  Jupiter.     It  was  not  so 
with  the  Vedic  hymns.      In  them  the   persons   of  the  gods  had 
neither  consistence  nor  tangibility.     They  are  distinguished  now 
by   one   set    of   qualities   and    again    by    another  ;    each    of    the 
immortals  who  sat   down    to  the  banquet  on   Olympus,   had  his 
or    her     own    personal    physiognomy,   described    by   poets    and 
interpreted  by  artists,  but  it  was  not  so  with  the  Hindoo  deities. 
The    Hindoo   genius   had   none   of   the   Greek    faculty   for   clear 
and   well-defined  imagery  ;    it  betrays  a  certain   vagueness  and 
want  of  definition  which  is  not  to  be  combined  with  a  complete 
aptitude  for  the  arts  of  design.     It  is  the  business  of  these  arts 
to  render  ideas  by  forms,  and  a  well  marked  limit  is  the  essence 
of  form,   which  is  beautiful   and  expressive   in    proportion  as   its 
contours  are  clearly  and  accurately  drawn. 

Indian  art  then,  for  the  reasons  which  we  have  given,  and  others 
which  are  unknown,  was  only  in  its  cradle  in  the  time  of  Alexander, 
while  the  artists  of  Greece  were  in  full  possession  of  all  their 
powers  ;  they  had  already  produced  inimitable  master-pieces  in 
each  of  the  great  divisions  of  art,  and  yet  their  creative  force  was 
far  from  being  exhausted.  It  was  the  age  of  Lysippus  and 
Apelles  ;  of  those  great  architects  who,  in  the  temples  of  Asia 
Minor,  renewed  the  youth  of  the  Ionic  order  by  their  bold  and 
ingenious  innovations.  Under  such  conditions,  what  would  the 
effect    have    been,  had    these    two    forms    of   civilization    entered 


Hi  Introduction. 


into  close  relations  with  each  other  ?  In  all  probability  the 
result  would  have  been  similar  to  that  which  ensued  when  the 
ancestors  of  the  Greeks  began  to  deal  with  the  more  civilized 
Phoenicians  and  the  people  of  Asia  Minor.  But  in  the  case  of  the 
Hindoos,  as  we  have  said,  the  disciples  had  a  less,  instead  of  a 
greater,  aptitude  for  the  plastic  arts  than  their  teachers,  and, 
moreover,  the  contact  between  the  two  was  never  complete  nor 
was  it  of  long  duration.  The  only  frontier  upon  which  the  inter- 
change of  idea  was  frequent  and  continuous  was  the  north-west, 
which  divided  India  from  that  Bactrian  kingdom  of  which  we 
know  little  more  than  the  mere  names  of  its  princes  and  the  date 
of  its  fall.  But  before  the  end  of  the  second  century  B.C.  this 
outpost  of  Hellenism  had  fallen  before  the  attacks  of  those 
barbarians  whom  we  call  the  Saci.  In  such  an  isolated  position  it 
could  not  long  hope  to  maintain  itself,  especially  after  the  rise  of 
the  Parthian  monarchy  had  separated  it  from  the  empire  of  the 
Seleucidse.  Its  existence  must  always  have  been  precarious,  and 
the  mere  fact  that  it  did  not  succumb  until  the  year  136  B.C.  is 
enough  to  prove  that  several  of  its  sovereigns  must  have  been 
remarkable  men.  Should  their  annals  ever  be  discovered  they 
would  probably  form  one  of  the  strangest  and  most  interesting 
episodes  in  the  history  of  the  Greek  race. 

Through  the  obscurity  in  which  all  the  details  are  enveloped  we 
can  clearly  perceive  that  those  princes  were  men  of  taste.  They 
were,  as  was  natural,  attached  to  the  literature  and  the  arts  which 
reminded  them  of  their  superior  origin  and  of  that  distant 
fatherland  with  which  year  after  year  it  became  more  difficult  to 
communicate.  Athough  they  were  obliged,  in  order  to  defend 
themselves  against  so  many  enemies,  to  employ  those  mercenary 
soldiers,  Athenians,  Thebans,  Spartans  and  Cretans,  which  then 
overran  Asia,  and  to  pay  them  dearly  for  their  services,  they  also 
called  skilful  artists  to  their  court  and  kept  them  there  at  great 
expense ;  the  beautiful  coins  which  have  preserved  their  images 
down  to  our  day  are  evidence  of  this,  the  decoration  of  their  cities, 
of  their  temples,  and  of  their  palaces  must  have  been  in  keeping 
with  these  ;  everywhere  no  doubt  were  Corinthian  and  Ionic 
buildings,  statues  of  the  Greek  gods  and  heroes  mixed  with  those 
portraits  and  historic  groups  which  had  been  multiplied  by  the 
scholars  of  Lysippus,  wall  paintings,  and  perhaps  some  of  those 
easel   pictures  signed   by  famous   masters,  for  which  the   heirs  of 


Introduction.  liii 


Alexander  were  such  keen  competitors.  Artisans,  who  had 
followed  the  Greek  armies  in  their  march  towards  the  East  with  the 
object  of  supplying  the  wants  '  of  any  colonies  which  might  be 
established  in  those  distant  regions,  reproduced  upon  their  vases 
and  in  their  terra-cotta  figures  the  motives  of  the  painting,  the 
sculpture,  and  the  architecture  which  they  left  behind  ;  goldsmiths, 
jewellers  and  armourers  cut,  chased,  and  stamped  them  in  metal. 
And  it  was  not  only  the  Greek  colonists  who  employed  their  skill. 
Like  the  Scythian  tribes  among  whom  the  Greek  cities  of  the 
Euxine  were  planted,  the  nations  to  the  north  of  India  were 
astonished  and  delighted  by  the  elegance  of  their  ornament 
and  the  variety  of  its  forms.  They  imported  from  Bactriana 
these  products  of  an  art  which  was  wanting  to  them,  and  soon 
set  themselves,  with  the  help  perhaps  of  foreign  artists  settled 
among  them,  to  imitate  Grecian  design  in  the  courts  of  the 
Indian  rajahs. 

That  this  was  so  is  proved  by  those  coins  which  bear  on  their 
reverse  such  Hindoo  symbols  as  Siva  with  his  bull,  and  on  their 
obverse  Greek  inscriptions,  and  by  the  remains  of  what  is  now 
called  Graeco-Buddhic  art,  an  art  which  seems  to  have  flourished 
in  the  upper  valley  of  the  Indus  in  the  third  or  second  century 
before  our  era.  These  remains,  formerly  much  neglected,  are  now 
attracting  much  attention.  They  haye  been  carefully  studied  and 
described  by  Cunningham  ^  ;  Dr.  Curtius  has  described  them  and 
published  reproductions  of  the  most  curious  among  them.^  They 
are  found  in  the  north  of  the  Punjab  upon  a  few  ancient  sites 
where  excavations  have  been  made.  Some  of  them  have  been 
transported  to  Europe  in  the  collection  of  Dr.  Leitner,  while  others 
remain  in  the  museums  of  Peshawur,  Lahore,  and  Calcutta.^  In 
those  sacred  buildings  which  have  been  examined  the  plan  of  the 
Greek  temple  has  not  been  adopted,  but  the  isolated  members  of 
Greek  architecture  and  the  most  characteristic  details  of  its 
ornament  are  everywhere  made  use  of.  It  is  the  same  with  the 
sculpture  ;  in  the  selection  of  types,  in  the  arrangement  of  drapery, 
in  the  design,  there  is  the  same  mixture  of  Greek  taste  with  that 
of  India,  of  elements  borrowed  from  foreign,  and  those  drawn  from 


'  Archczological  Survey  of  India,  3  vols.  1871-73. 

2  Archceologische  Zeitung,  1876,  p.  90.     Die  Griechische  Kunst  in  Indien. 

^  The  Louvre  has  lately  acquired  some  curious  examples  of  this  art. 


Hv  Introduction. 


the  national,   beliefs.     The  helmeted  Athene  and    Helios   in  his 
quadriga  figure  by  the  side  of  Buddha. 

Traces  of  the  same  influence  are  to  be  found  in  a  less  marked 
degree  in  other  parts  of  India.  Near  the  mouth  of  the  Indus  and 
upon  the  Malabar  coast,  the  native  sculptors  and  architects  were 
able  to  obtain  more  than  one  useful  suggestion,  more  than  one 
precious  hint  as  to  their  technique,  from  the  works  of  art  brought 
in  the  ships  of  maritime  traders.  It  is  even  possible  that  Greek 
workmen  may  thus  have  been  introduced  into  seaport  towns,  and 
there  employed  upon  the  decoration  of  palaces  and  temples. 
However  this  may  be  it  is  incontestable  that  all  the  important 
sacred  edifices  of  that  region,  whether  stone-built  or  carved  in  the 
living  rock,  date  from  a  period  more  recent  than  that  of  Alexander, 
and  that  most  of  them  show  details  which  imply  acquaintance  with 
Greek  architectural  forms  and  their  imitation.  We  are  thus  on 
all  hands  forced  to  this  conclusion  :  that,  in  the  domain  of  the  plastic 
arts,  Greece  owed  nothing  to  India,  with  which  she  made  ac- 
quaintance very  late  and  at  a  period  when  she  had  no  need  to  take 
lessons  from  others.  That,  moreover,  India  had  little  or  nothing 
to  give  ;  that  her  arts  were  not  developed  till  after  her  early 
relations  with  Greece,  and  it  would  even  seem  that  her  first 
stimulus  was  derived  from  the  models  which  Greece  put  within 
her  reach. 

From  all  this  it  will  be  seen  that  we  need  not  go  as  far  as  China, 
or  even  as  the  Punjab,  in  order  to  explain  the  origin  of  Greek  art. 
During  the  period  with  which  we  are  concerned,  China  might  as 
well  have  been  in  the  planet  Saturn  for  all  she  had  to  do  with  the 
ancient  world,  and  we  need  refer  to  her  no  more,  except  now  and 
then  perhaps  for  purposes  of  illustration.  We  cannot  treat  India 
quite  in  the  same  fashion,  because  there  were,  as  we  have  said, 
certain  points  of  contact  and  reciprocal  influences  at  work  between 
her  and  the  group  of  nations  we  are  about  to  treat.  But  as  Greece 
borrowed  nothing  from  India,  at  least  in  the  matter  of  art,  the 
little  which  we  shall  have  to  say  of  the  products  of  the  Hindoos 
will  not  be  connected  with  our  discussion  of  the  origin  of  Greek 
art.  A  curious  though  hardly  an  important  episode  in  history,  is 
seen  in  the  reaction  by  which  the  Greek  genius,  when  arrived  at 
maturity,  threw  itself  at  the  command  of  Alexander  upon  that  East 
from  which  it  had  received  its  first  lessons. 


Introduction.  Iv 


None  of  those  philosophical  discussions  to  which  Ottfried 
Miiller  and  Stark  thought  it  necessary  to  give  so  large  a  place 
will  be  found  in  our  introduction  ;  both  of  those  authors  devoted  a 
long  chapter  to  the  definition  of  art  and  its  principal  manifestations. 
Stark  went  so  far  as  to  discuss,  with  much  patience  and  ingenuity, 
the  definitions  of  art  and  of  its  essential  forms  which  had  been 
given  by  previous  writers.  We  shall  attempt  nothing  of  the 
kind  ;  we  have  not  undertaken  a  work  of  criticism  or  aesthetic 
demonstration.  We  wish  to  build  up  the  history  of  ancient 
civilization  through  the  study,  description,  and  comparison  of  its 
plastic  remains. 

Neither  do  we  feel  sure  that,  in  such  a  question  as  this,  definitions 
do  not  lead  to  confusion  rather  than  to  clearness.     When  short, 
they  are  vague  and  obscure,  and  only  acquire  precision  through 
distinctions    and    developments    which    have    to    be    discussed   at 
length  ;  and  again  they  generally  lead,  on  one  hand  or  the  other, 
either    to    objections    or    reservations.      Omnis   definitio   in  jure 
periadosa,  says  an  old  maxim,  which  is  certainly  true  in  matters 
of  art.     Why  should  we  attempt,  unless  we  are  obliged,  to  define 
terms   which    awake   sufficiently  clear   and    distinct    ideas    in    all 
cultivated  minds  .^     No  satisfactory  definition  has  ever  been  given 
of  the  word  architecture,  and  yet,  when  we  use  it,  every  one  knows 
what   we   mean.     Architecture,  sculpture,  painting,  each  of  these 
sounds   has   a  precise   meaning  for  those   to  whom   our  work   is 
addressed,  and  we  may  say  the  same  of  certain  other  expressions, 
such  as  iiidustrial  arts,  decoration,  style,  Jiistorical  painting,  genre 
painting,   landscape  painting,   which  will   often   be  found   in    our 
pages.     We    must    refer   those    who    want    definitions    of    these 
phrases    to    the   Gra77imaire  des  Arts  du  Dessin  of   M.   Charles 
Blanc  and  kindred  works.      It  will  suffice  for  us  that  these  words 
should  be  taken  in  the  ordinary  meaning  which  they  bear  in  the 
conversation    of    cultivated    men.      If    our   ideas    of  art   and   its 
different  branches  diverge  here  and  there  from  those  which  are 
commonly  received,  those  divergencies  will  become  evident,  and 
will  be  discussed  and  justified  to  the  best  of  our  ability  as  the 
work  proceeds.      But  on  all  occasions  we  shall  do  our  best  to  avoid 
the   abstract    and    pedantic    terminology    which    makes    Ottfried 
Muller's  first  chapter  so  difficult  to  read. 

We  have  now  declared  the  aim  of  our  work  and  the  route  which 


Ivi  Introduction. 


we  propose  to  follow.  In  order  to  increase  our  chances  of  success, 
I  have  sought  and  obtained  the  collaboration  of  M.  Charles 
Chipiez,  whose  special  knowledge  is  well  calculated  to  neutralise 
my  own  deficiencies.  To  his  Histoire  critique  des  Origines  et  de 
la  Formation  des  Ordres  grecques,  was  awarded,  in  1877,  one  of  the 
highest  prizes  of  the  Academic  des  Inscriptions,  and  in  the  Salons 
of  1878  and  1879  he  confirmed  his  double  reputation  as  a  skilful 
draughtsman  and  a  learned  theorist ;  his  Essais  de  Restoration 
dun  Temple  grecque  hypHlire,  et  des  tonrs  a  dtages  de  la  C/ialdde, 
was  much  noticed  and  discussed  by  connoisseurs.  It  would  not 
be  fitting,  however,  to  praise  it  here.  I  must  confine  myself  to 
saying  how  fortunate  I  am  in  having  obtained  a  help  which  I  have 
found  more  helpful,  more  single-minded,  more  complete,  than  I 
had  dared  to  hope  for.  In  all  that  has  to  do  with  architecture, 
I  have  not  written  a  line  until  after  consulting  M.  Chipiez  upon  all 
technical  points.  He  has  also  taken  an  active  part  in  the  revision  of 
the  text  of  certain  chapters.  As  for  the  plates  and  illustrations  in 
the  text,  we  have  together  chosen  the  objects  to  be  represented, 
and  M.  Chipiez,  as  a  professional  man  and  able  draughtsman,  has 
personally  superintended  the  execution  of  the  drawings.  It 
remains  for  me  to  explain  the  role  which  we  have  assigned  to  our 
illustrations. 


VI. 

In  the  single  edition  of  his  great  work  which  appeared  during 
his  own  lifetime,  Winckelmann  inserted  but  a  small  number  of 
illustrations,  and  those  for  ornament  rather  than  for  instruction.  One 
of  his  translators,  M.  Huber,  tells  us  that  their  execution  gave 
great  dissatisfaction  to  the  author.^  In  our  days,  on  the  other  hand, 
those  who  undertake  a  work  of  this  kind  make  use  of  the  great 
progress  which  has  taken  place  in  engraving  and  typography,  to 
insert  numerous  figures  in  their  text,  to  which  they  offer  an 
indispensable  and  animated  commentary.  Without  their  help 
many  descriptions  and  observations  might  remain  unnecessarily 
obscure  and  doubtful.  When  forms  are  to  be  defined  and  com- 
pared, mere  words,  in  whatever  language  spoken  or  written,  can 
never  suffice. 

^  Histoire  de  V Art ;  Huber's  preface  to  his  translation,  p.  xxxii. 


Introduction.  Ivii 


/ 


With  well  chosen  phrases  we  may  awake  the  recollections 
of  others,  and  give  renewed  life  to  any  impression  which  they 
may  have  received  from  some  striking  natural  phenomenon  or 
some  fine  work  of  art.  Their  imaginations  will  call  up  for  a 
moment  some  landscape,  picture,  or  statue  which  has  formerly 
charmed  them.  But  if  we  wish  to  explain  the  complicated  plan  of 
some  great  building,  its  design  and  its  proportions,  the  slightest 
sketch  will  be  of  more  use  than  the  longest  and  most  minute 
descriptions.  So  it  will,  if  we  wish  to  make  clear  the  character- 
istics which  distinguish  one  style  from  another,  the  Assyrian  from 
the  Egyptian,  the  archaic  Greek  style  from  that  of  the  Phidian 
epoch  or  of  the  decadence,  an  Ionic  column  from  the  Erechtheum 
from  one  of  the  same  order  treated  by  a  Roman  architect. 
Between  the  contour  of  a  figure  from  a  Memphite  bas-relief  and 
that  of  one  from  Nineveh,  what  difference  is  there  ?  A  tenth  of 
an  inch  more  or  less,  a  slight  difference  in  the  sweep  of  a  line  in 
order  to  mark  more  strongly  the  junction  of  the  thigh  and  the 
knee.  If  we  placed  three  nude  torsos  side  by  side,  one  of  the 
sixth  century,  another  of  the  fifth  century,  and  the  third  of  the  time 
of  Hadrian,  a  practised  eye  would  at  once  assign  its  true  date  to 
each,  in  accordance  with  the  manner  in  which  the  skeleton  was 
indicated  under  the  liesh,  and  the  muscles  drawn  over  it  and 
attached  to  it.  Supposing  that  the  same  model  had  served  all 
three  artists,  it  would  show  in  the  one  case  a  lively  sentiment  of 
form  combined  with  some  dryness  and  rigidity ;  in  another  a 
freer,  larger,  and  more  subtle  treatment,  and  in  the  third  a  want 
of  vigour  and  firmness  :  but  it  would  be  difficult  to  give  by  words 
a  clear  idea  of  what  caused  the  difference.  Between  the  contour 
which  satisfies  us  and  that  which  does  not  there  is  hardly  the 
difference  of  a  hair ;  by  leaning  a  little  harder  with  the  chisel  the 
aspect  of  the  one  surface  might  have  been  made  identical  with  the 
other.  By  its  double  astragali,  by  the  fine  chiselling  of  its 
gorgerin,  by  the  elegant  curve  which  unites  the  two  volutes,  and 
by  the  general  delicacy  of  its  ornament,  a  capital  from  the 
Erechtheum  is  distinguished  above  a  Roman  Ionic  capital  ;  it  is 
at  once  finer  in  design  and  richer  in  ornamentation  :  by  the  side 
of  it  a  capital  from  the  theatre  of  Marcellus  or  the  Coliseum 
would  look  mean  and  poor. 

The   whole   history  of  art   consists   of  the   succession   of  subtle 
changes  like  these,  and  it  would  be  impossible  to  convey  them  to 

VOL.  I.  i 


Iviii  Introduction, 


the  reader  by  the  utmost  precision  of  technical  language  or  the 
most  brilliant  and  life-like  descriptions.  The  best  thing  that 
can  be  done  is  to  make  one's  remarks  in  the  presence  of  the 
statues,  pictures  and  buildings  concerned.  But  it  is  rarely  that 
we  find  ourselves  in  such  favourable  conditions  for  teaching  and 
explaining  our  ideas.  But,  in  default  of  the  objects  themselves, 
we  may  at  least  give  the  most  faithful  images  of  them  which  can 
be  obtained,  and  that  we  shall  attempt  to  do  throughout  the  course 
of  this  history. 

We  shall,  then,  give  a  large  number  of  figures,  in  which  absolute 
accuracy  and  justice  of  proportion  will  be  aimed  at  rather  than 
picturesque  effects.  It  is  not  very  long  since,  in  collections  of 
drawings  from  antique  remains,  they  were  all  presented  under  one 
aspect,  so  far  as  the  subtleties  of  style  were  concerned.  The  hand 
of  the  engraver  spread  a  technical  uniformity  over  them  all  in 
which  differences  of  school  and  date  disappeared,  just  as  the 
delicate  carvings  and  coloured  ornament  of  the  middle  ages  and 
the  renaissance,  which  gave  to  each  building  an  individuality  of  its 
own,  were  reduced  to  dull  monotony  by  the  undiscriminating 
brush  of  the  whitewasher.  It  seemed  to  the  artist  natural  enough 
to  clothe  the  monuments  of  the  past  in  the  style  of  his  own  day — 
and  it  required  much  less  care  than  would  have  been  needed  for 
the  successful  expression  of  all  the  diversities  of  style  in  his  models. 
We  have  now,  however,  grown  more  exacting.  We  demand  from 
the  draughtsman  who  pretends  to  interpret  a  work  of  art  the  same 
devotion  and  the  same  self-sacrifice  as  from  the  writer  who  is 
charged  with  the  translation  of  a  work  of  literature  from  one 
language  into  another — we  require  him  to  forget  himself,  so  that 
we  may  say  of  him,  as  the  Latin  poet  says  of  his  Proteus  : 

"  Omnia  transformat  sese  in  miracula  rerum." 

We  require  him  to  change  his  style  with  every  change  of 
subject,  to  copy  the  gesture,  the  accent,  and  even  the  faults  of 
his  model  ;  to  be  Chinese  in  China,  Greek  in  Greece,  and  Tuscan 
when  he  takes  us  to  Siena  or  Florence.  But  we  have  indicated 
an  ideal  which  is  not  often  reached.  Every  one  of  us  has  his 
preferences  and  natural  affinities,  every  artist  his  own  methods 
and  personal  modes  of  thought.  One  will  be  conspicuous  for  his 
interpretation  of  the  nobility  and  purity  of  the  antique,  another  for 
his  treatment  of  oriental  art  or  of  the  elecrance  of  our  eiirhteenth 


Introduction.  llx 

century.  But  the  mere  enunciation  of  the  principle  is  of  value, 
for  a  great  effect  follows  the  praise  which  those  who  treat  their 
model  with  scrupulous  and  intelligent  respect  are  sure  to  obtain, 
and  the  blame  to  which  they  who  are  less  conscientious  expose 
themselves. 

Fidelity  in  interpretation  is,  in  fact,  the  honesty  of  the  draughts- 
man ;  it  may  become,  if  carried  to  a  great  height,  his  honour,  and 
even  his  glory.  So  far  as  we  are  concerned,  we  demand  it  from 
all  those  who  are  associated  with  us  in  this  task  ;  and,  so  far  as 
existing  methods  will  allow,  we  shall  see  that  we  obtain  it.  Unless 
our  illustrations  had  that  merit  they  would  obscure  the  text  instead 
of  making  it  more  comprehensible.  Our  readers  would  search  in 
vain  for  the  features  and  characteristics  to  which  we  might  call 
their  attention,  and  many  of  our  remarks  and  theories  would 
become  difficult  to  understand.  We  should  be  in  the  same 
position  as  an  incompetent  barrister  who  has  made  a  bad  choice  oi 
witnesses  ;  witnesses  who,  when  in  the  box,  prove  either  to  know 
nothing  or  to  know  only  facts  which  tell  against  the  party  who  has 
called  them. 

Our  aim  in  choosing  our  illustrations  will  be  to  place  before 
our  readers  good  reproductions  of  most  of  the  objects  which  are 
discussed  in  our  text.  We  shall,  of  course,  be  unable  to  figure 
everything  that  is  of  interest,  but  we  can  at  least  ensure  that  those 
figures  which  we  give  shall  each  be  interesting  in  some  particular 
or  another.  So  far  as  possible,  we  shall  select  for  illustration 
such  objects  as  have  not  previously  been  reproduced,  or  have  been 
ill  reproduced,  or  have  been  figured  in  works  which  are  difficult  of 
access.  W^e  shall  sometimes,  of  course,,  find  it  necessary  to 
reproduce  some  famous  statue  or  some  building  which  is  familiar 
to  most  people  ;  but  even  then  we  shall  endeavour  to  give 
renewed  interest  to  their  beauties  by  displaying  them  under  some 
fresh  aspect  and  by  increased  care  in  the  delineation  of  their  forms. 
Views  in  perspective,  of  which  we  shall  make  frequent  use,  give 
the  general  aspect  of  buildings  with  much  greater  truth  and 
completeness  than  a  mere  plan,  or  a  picturesque  sketch  of  ruinous 
remains,  or  even  than  an  elevation. 

Most  of  the  more  important  perspectives  and  restorations  due  to 
the  learned  pencil  of  M.  Chipiez  will  be  given  in  plates  separate 
from  the  text,  as  well  as  the  most  curious  or  significant  of  the 
works  in  sculpture  or  painting  to   which    we   shall  have   to  refer. 


Ix  Introduction. 

Some  of  these  plates  will  be  coloured.  But  the  majority  of  our 
illustrations  will  consist  of  engravings  upon  zinc  and  wood,  which 
will  not,  we  hope,  fall  short  of  their  more  elaborate  companions  in 
honesty  and  fidelity. 

From  the  earliest  Egyptian  dynasties  and  from  fabled  Chaldsea 
to  imperial  Rome,  from  the  Pyramids  and  the  Tower  of  Babel  to 
the  Coliseum,  from  the  Statue  of  Chephren  and  the  bas-reliefs  of 
Shalmaneser  III.  to  the  busts  of  the  Caesars,  from  the  painted 
decorations  of  the  tomb  of  Ti,  and  the  enamelled  bricks  of 
Nineveh  to  the  wall-paintings  of  Pompeii,  we  shall  review  in  due 
succession  all  the  forms  which  the  great  nations  of  antiquity  made 
use  of  to  express  their  beliefs,  to  give  shape  to  their  ideas,  to 
satisfy  their  instincts  for  luxury  and  their  taste  for  beauty,  to  lodge 
their  eods  and  their  kines,  and  to  transmit  their  own  likenesses 
to  posterity. 

We  propose  to  trace  and  explain  the  origin  of,  and  to  describe, 
without  aesthetic  dissertations  or  excessive  use  of  technical  terms, 
those  processes  which  imply  the  practice  of  art  ;  the  creation 
and  descent  of  forms  ;  the  continual  changes,  sometimes  slight  and 
sometimes  great,  which  they  underwent  in  passing  from  one 
people  to  another,  until,  among  the  Greeks,  they  arrived  at  the 
most  happy  and  complete  perfection  which  the  world  has  seen. 
We  hope,  too,  by  the  judicious  choice  and  careful  execution  of  our 
figures,  to  give  a  fair  idea  of  this  course  of  development  even  to 
those  artists  who  have  neither  time  nor  patience  to  follow  our 
criticisms  and  descriptions. 

I  conceived  the  plan  of  this  history,  of  which  the  first  instalment 
is  now  submitted  to  the  public,  at  the  time  when  M.  Wallon,  who 
is  secretary  to  the  Acad(^iuie  des  Inscriptions  et  Belles-Lettres, 
entrusted  me  with  the  inauguration,  at  the  Sorbonne,  of  the 
teaching  of  classic  archaeology.  But  before  it  could  be  realized 
two  conditions  had  to  be  fulfilled.  I  had  to  find  an  associate  in 
the  work,  a  companion  who  would  help  me  in  the  necessary  labour 
and  study,  and  I  found  him  among  my  auditors  in  those  first 
lectures  at  the  Sorbonne.  I  had  also  to  find  a  publisher  who 
would  understand  the  wants  of  the  public  and  of  the  critics  in  such 
a  matter.  In  this,  too,  I  have  succeeded,  and  I  am  free  to 
undertake  a  work  which  is,  I  hope,  destined  to  carry  far  beyond 
the    narrow   limits   of  a   Parisian   lecture   room,    the   methods  and 


Introduction, 


XI 


principal  results  of  a  science,  which,  having  made  good  its  claims  to 
the  gratitude  of  mankind,  is  progressing  with  a  step  which  becomes 
daily  more  assured.  The  task  is  an  arduous  one,  and  the  continual 
discoveries  which  are  reported  from  nearly  every  quarter  of  the 
ancient  world,  make  it  heavier  every  day.  As  for  my  colleague 
and  myself,  we  have  resigned  ourselves  in  advance  to  seeing 
omissions  and  defects  pointed  out  even  by  the  most  benevolent 
critics,  but  we  are  convinced  that  in  spite  of  such  imperfections  as 
it  may  contain,  our  work  will  do  good  service,  and  will  cause  one 
of  the  aspects  of  ancient  civilization  to  be  better  understood.  This 
conviction  will  sustain  us  through  the  labours  which,  perhaps  with 
some  temerity,  we  have  taken  upon  us.  How  far  shall  we  be 
allowed  to  conduct  our  history  ?  That  we  cannot  tell,  but  we  may 
venture  tcT  promise  that  it  shall  be  the  chief  occupation  and  the 
dearest  study  of  all  that  remains  to  us  of  life  and  strength. 

Georges  Perrot. 


TO    THE    READER. 


We  have  been  in  some  doubt  as  to  whether  we  should  append  a  special  biblio- 
graphy to  each  section  of  this  work,  but  after  mature  reflection  we  have  decided 
against  it.  We  shall,  of  course,  consider  the  art  of  each  of  the  races  of  antiquity  in 
less  detail  than  if  we  had  undertaken  a  monograph  upon  Egyptian,  upon  Assyrian, 
or  upon  Phoenician  art ;  but  yet  it  is  our  ambition  to  neglect  no  source  of  information 
which  is  likely  to  be  really  valuable.  From  many  of  the  books  and  papers  which 
we  shall  have  to  consult  we  may  reproduce  nothing  but  their  titles,  but  we  hope  that 
no.  important  work  will  escape  us  altogether,  and  in  every  case  we  shall  give 
references  which  may  be  easily  verified.  Under  these  circumstances  a  formal  list  of 
works  would  be  a  mere  repetition  of  our  notes  and  would  only  have  the  effect  of 
giving  a  useless  bulk  to  our  v'olumes. 

Whene\er  oar  drawings  have  not  been  taken  directly  from  the  originals  we  have 
been  careful  to  indicate  the  source  from  which  we  obtained  them^  and  we  have 
made  a  point  of  borrowing  only  from  authors  of  undoubted  authority.  Those 
illustrations  which  bear  neither  an  artist's  name  nor  the  title  of  a  book  have  been 
engraved  from  photographs.  As  for  the  perspectives  and  restorations  supplied  by 
M.  Chipiez,  they  are  in  every  case  founded  upon  the  study  and  comparison  of  all 
accessible  documents ;  but  it  would  take  too  long  to  indicate  in  each  of  these 
drawings  how  much  has  been  borrowed  from  special  publications  and  how  much 
has  been  founded  upon  photographic  evidence.  M.  Chipiez  has  sometimes 
employed  the  ordinary  perspective,  sometimes  that  which  is  called  axonometric 
perspective.     The  difference  will  be  at  once  perceived. 

Egyptologists  may,  perhaps,  find  mistakes  in  the  hieroglyphs  which  occur  in  our 
illustrations.  These  hieroglyphs  have  been  as  a  rule  exactly  transcribed,  but  we 
do  not  pretend  to  offer  a  collection  of  texts  ;  we  have  only  reproduced  these 
characters  on  account  of  their  decorative  value,  and  because  without  them  we  could 
not  have  the  general  appearance  of  this  or  that  monument.  It  will  thus  be  seen 
that  our  object  is  not  affected  by  a  mistake  or  two  in  such  matters. 

We  may  here  express  our  gratitude  to  all  those  who  have  interested  themselves 
in  our  enterprise  and  who  have  helped  us  to  make  our  work  complete.  Our  dear 
and  lamented  Marielte  had  promised  us  his  most  earnest  help.  During  the  winter 
that  we  passed  in  Egypt,  while  he  still  enjoyed  some  remains  of  strength  and  voice, 
we  obtained  from  his  conversation  and  his  letters  some  precious  pieces  of  informa- 
tion.    We  have  cited  the  works  of  M.  Maspero  on  almost  every  page,  and  yet  we 


]xiv  To  THE   Reader. 


have  learnt  more  from  his  conversation  than  from  his  writings.  Before  his 
departure  for  Egypt — whither  he  went  to  succeed  Marietta — M.  Maspero  was 
our  perpetual  counsellor  and  referee  ;  whenever  we  were  embarrassed  we  appealed 
to  his  well  ordered,  accurate,  and  unbiased  knowledge.  We  are  also  deeply 
indebted  to  M.  Pierret,  the  learned  conservateur  of  the  Louvre  ;  not  only  has  he 
done  everything  to  facilitate  the  work  of  our  draughtsmen  in  the  great  museum,  he 
has  also  helped  us  frequently  with  his  advice  and  his  accumulated  knowledge. 
M.  Arthur  Rhone  has  lent  us  a  plan  of  the  temple_  of  the  Sphinx,  and  M.  Ernest 
Desjardins  a  view  of  the  interior  of  that  building. 

The  artists  who  have  visited  Egypt  have  helped  us  as  cordially  as  the  learned  men 
who  have  deciphered  its  inscriptions.  M.  Gerome  opened  his  portfolios  and  allowed 
us  to  take  three  of  those  drawings,  which  express  with  such  truthful  precision  the 
character  of  Egyptian  landscape  from  them.  M.  Hector  Leroux  was  as  generous 
as  M.  Gerome,  and  if  we  have  taken  but  one  illustration  from  his  sketch-books  it  is 
because  the  arrangements  for  this  volume  were  complete  before  we  had  the  chance 
of  looking  through  them.  M.  Brune  has  allowed  us  to  reproduce  his  plans  of 
Karnak  and  Medinet-Abou. 

We  have  had  occasion,  in  the  work  itself,  to  express  our  acknowledgments  to 
MM.  J.  Bourgoin,  G.  Benedite,  and  Saint-Elme  Gautier,  who  have  drawn  for  us 
the  principal  monuments  of  the  Boulak  and  Louvre  Museums.  For  the  architecture 
we  must  name  M.  A.  Gue'rin,  a  pupil  of  M.  Chipiez,  who  prepared  the  drawings 
under  the  direction  of  his  master,  and  M.  Tomaszkievicz,  whose  light  and  skilful 
point  has  so  well  engraved  them.  If  the  process  of  engraving  upon  zinc  has  given 
results  which,  as  we  hope,  will  satisfy  our  readers,  much  of  the  honour  belongs  to  the 
untiring  care  of  M.  Comte,  whose  process  has  been  employed ;  all  these  plates  have 
been  reviewed  and  retouched  by  him  with  minute  care.  The  steel  engravings  are 
by  MM.  Ramus,  Hibon,  Guillaumot  pere  and  Sulpis.  Li  order  that  the 
polychromatic  decoration  of  the  Egyptians  should  be  rendered  with  truth  and 
precision  in  its  refined  tones  and  complicated  line,  we  begged  M.  Sulpis  to  make 
use  of  a  process  which  had  almost  fallen  into  disuse  from  its  difficulty  and  want 
of  rapidity;  we  mean  that  which  is  called  aquatint.  Our  plates  IL,  XIIL, 
and  XIV.  will  perhaps  convince  our  readers  that  its  results  are  superior  to  those 
of  chromo-lithography,  which  is  now  so  widely  employed. 


A    HISTORY    OF 
ART    IN    ANCIENT    EGYPT 


CHAPTER  I. 

THE    GENERAL    CHARACTER    OF    EGYPTIAN    CIVILIZATION. 

§   I.  Egypt's  Place  in  the  History  of  the   World. 

Egypt  is  the  eldest  dau£rhter  of  civilization.  In  imdertakine  to 
group  the  great  nations  of  antiquity  and  to  present  them  in  their 
proper  order,  in  attempting  to  assign  to  each  its  due  share  in  the 
continuous  and  unremitting  labour  of  progress  until  the  birth  of 
Christianity,  we  have  no  alternative  but  to  commence  with  the 
country  of  the  Pharaohs. 

In  studying  the  past  of  mankind,  we  have  the  choice  of  several 
points  of  view.  We  may  attempt  to  determine  the  meaning  and 
value  of  the  religious  conceptions  which  succeeded  one  another 
during  that  period,  or  we  may  give  our  attention  rather  to  the 
literature,  the  arts  and  the  sciences,  to  those  inventions  which  in 
time  have  done  so  much  to  emancipate  mankind  from  natural 
trammels  and  to  make  him  master  of  his  destiny.  One  writer 
will  confine  himself  to  a  description  of  manners,  and  social  and 
political  institutions  ;  another  to  the  enumeration  and  explanation 
of  the  various  changes  brought  on  by  internal  revolutions,  by  wars 
and  conquests ;  to  what  Bossuet  calls  "  la  suite  des  empires!' 
Finally,  he  who  has  the  highest  ambition  of  all  will  attempt  to 
unite  all  these  various  features  into  a  single  picture,  so  as  to 
show,  as  a  whole,  the  creative  activity  of  a  race  and  the  onward 

^y'VOL.   L  B 


A    History  of  Art  in  Ancient  Egypt. 


movements  of  its  eenius  in  the  continual  search  for  "  the  best." 
But  in  any  case  the  commencement  must  be  made  with  Egypt. 
It  is  Egypt  that  has  preserved  the  earhest  attempts  of  man 
towards  outward  expression  ;  it  is  in  Egypt  that  those 
monuments  exist  which  contain  the  first  permanent  manifestation 
of  thought  by  written  characters  or  plastic  ^  forms  ;  and  it  is  in 
Egypt  that  the  historian  of  antique  art  will  find  the  earliest 
materials  for  study. 

But,  in  the  first  place,  we  must  give  some  account  of  the  curious 
conditions  under  which  the  people  lived  who  constructed  and 
ornamented  so  many  imposing  monuments.  We  must  begin, 
then,  by  describing  the  circumstances  and  the  race  characteristics 
under  which  this  early  civilization  was  developed. 


§  2.    The   Valley  of  the  Nile  and  its  Inhabitants. 

The  first  traveller  in  Egypt  of  which  we  have  any  record  is 
Herodotus  ;  he  sums  up,  in  an  often  quoted  phrase,  the  impression 
which  that  land  of  wonders  made  upon  him  :  "  Egypt,"  he  says, 
"is  a  present  from  the  Nile."  ^  The  truth  could  not  be  better 
expressed.  "  Had  the  equatorial  rains  not  been  compelled  to  win 
for  themselves  a  passage  to  the  Mediterranean,  a  passage  upon 
which  they  deposited  the  mud  which  they  had  accumulated  on 
their  long  journey,  Egypt  would  not  have  existed.  Egypt  began 
by  being  the  bed  of  a  torrent ;  the  soil  was  raised  by  slow 
degrees  ....  man  appeared  there  when,  by  the  slow  accumu- 
lation of  fertile  earth,  the  country  at  last  became  equal  to  his 
support  .  .   .  .  ^ 

Other  rivers  do  no  more  than  afford  humidity  for  their 
immediate  borders,  or,  in  very  low-lying  districts,  for  a  certain 
narrow  stretch  of  country  on  each  hand.  When  they  overflow 
their  banks  it  is  in  a  violent  and  irregular  fashion,  involving  wide- 
spread ruin  and  destruction.     Great  floods  are    feared  as    public 

^  The  word  "  plastic "  is  used  throughout  this  work  in  its  widest  significance, 
and  is  not  confined  to  works  "in  the  round." — Ed. 

^  Herodotus,  ii.  7. 

2  Mariette,  Itineraire  de  la  haute  Egypfe,  p.  10  (edition  of  1872,  r  vol. 
Alexandria,  Mourt;s). 


The  Valley  of  the  Nile  and  its   Inhabitants. 


3 


misfortunes.,  It  is  very  different  with  the  Nile.  Every  year,  at 
a  date  which  can  be  almost  exactly  foretold,  it  begins  to  rise 
slowly  and  to  spread  gently  over  the  land.  It  rises  by  degrees  until 
its  surface  is  eisfht  or  nine  metres  above  low-water  mark  ;  ^  it  then 
begins  to  fall  with  the  same  tranquillity,  but  not  until  it  has 
deposited,  upon  the  lands  over  which  it  has  flowed,  a  thick  layer 
of  fertile  mud  which  can  be  turned  over  easily  with  the  lightest 
plough,  and  in  which  every  seed  will  germinate,  every  plant 
spring  up  with  extraordinary  vigour  and  rapidity. 


-.:r/l:^*^^*=^£fi5is^->-'^;-'^"'r^~^^  ^Stfjffs.f^-^-"--'- ^--^^fiiirfri?'* 


i'fF 


^e 


Fig.  I. — During  the  Inundation  of  the  Nile. 


Thus  nature  has  greatly  facilitated  the  labour  of  the  Egyptian 
agriculturist  ;  the  river  takes  upon  itself  the  irrigation  of  the 
country  for  the  whole  width  of  its  valley,  and  the  preparation  of 
the  soil  for  the  autumnal  seed-time  ;  it  restores  the  virtue  annually 
taken  out  of  the  ground  by  the  crops.  Each  year  it  brings  with 
it  more  fertility  than  can  be  exhausted  in  the  twelve  months,  so 

1  The  river  should  rise  to  this  height  upon  the  Nilometer  at  Cairo  if  there  is  to  be 
a  "  good  Nile."  In  upper  Egypt  the  banks  of  the  river  are  much  higher  than  in 
middle  Egypt.  In  order  to  flow  over  those  banks  it  must  rise  to  a  height  of  some 
eleven  or  twelve  metres,  and  unless  it  rises  more  than  thirteen  metres  it  will  not 
have  a  proper  effect. 


A  History  of  Art  in  Ancient  Egypt. 


that  there  is  a  constantly  accumulating  capital,  on  both  banks  of 
the  river,  of  the  richest  ves^etable  earth. 


Fig.  2. — Hoeing  ;  Beni-Hash^an.     (Champollion,  pi.  381  bis?) 

Thus    the    first    tribes    established    themselves    in    the   country 
under  singularly  favourable  conditions  ;  thanks  to  the  timely  help 


Fig.    3. — Ploughing;    from    the  Necropolis    of  Memphis.      [Description    de  FEgypte,    ant.   V., 

pi.   17.) 

of  the  river  they  found  themselves  assured  of  an  easy  existence.^ 
We  know  how  often  the  lives  of  those  tribes   who  live  by  fishing 

^  This  work  of  Champollion's,  to  which  we  are  greatly  indebted,  is  entitled :  Monu- 
ments de  lEgypte  et  de  la  Niibie,  4  vols,  folio.  It  contains  511  plates,  partly  coloured, 
and  was  published  between  the  years  1833  and  1845.  The  drawings  for  the  plates 
were  made  by  members  of  the  great  scientific  expedition  of  which  Champollion  was 
the  head.  INIany  of  those  drawings  were  from  the  pencil  of  Nestor  L'Hote,  one  of 
those  who  have  most  sympathetically  rendered  the  Egyptian  monuments. 

2  This  advantage  was  thoroughly  appreciated  by  the  ancients.  Diodorus  Siculus, 
speaking  of  the  Egyptians,  says  that  "At  the  beginning  of  all  things,  the  first  men 
were  born  in  Egypt,  in  consequence  of  the  happy  climate  of  the  country  and  the 
physical  properties  of  the  Nile,  whose  waters,  by  their  natural  fertility  and  their 
power  of  producing  various  kinds  of  aliment,  were  well  fitted  to  nourish  the  first 


The  Valley  of  the  Nile  and  its  Inhabitants. 


and  the  chase  are  oppressed  by  care  ;  there  are  some  days  when 
game  is  not  to  be  found,  and  they  die  of  hunger.  Those  who  Hve 
a  pastoral  Hfe  are  also  exposed  to  cruel  hardships  from  the 
destruction  of  their  flocks  and  herds  by  those  epidemics  against 
which  even  modern  science  sometimes  struQfOfles  in  vain.  As 
for  agricultural  populations,  they  are  everywhere,  except  in 
Egypt,  at  the  mercy  of  the  weather ;  seasons  which  are  either  too 
dry  or  too  wet  may  reduce  them  to  famine,  for  in  those  distant 
times  local  famines  were  far  more  fatal  than  in  these  days,  when 
facility  of  transport  and  elaborate  commercial  connections  ensure 
that  where  the  demand  is,  thither  the  supply  will  be  taken.  In 
Egypt  the  success  of  the  crops  varied  with  the  height  of  the 
Nile,  but  they  never  failed  altogether.      In  bad  years  the  peasant 


«^>^ 

&^i 


Fig.  4. — Harvest  scene  ;  from  a  tomb  at  Gizeh.     (Champollion,  pi.  417.) 


may  have  had  the  baton  of  the  tax-collector  to  fear,  but  he 
always  had  a  few  onions  or  a  few  ears  of  maize  to  preserve  him 
from  starvation.^ 

beings  who  received  the  breath  of  life.  .  .  .  It  is  evident  that  from  the  foundation 
of  the  world  Egypt  was,  of  all  countries,  the  most  favourable  to  the  generation  of 
men  and  women,  by  the  excellent  constitution  of  its  soil  "  (i.  10). 

^  In  all  ages  the  rod  has,  in  Egypt,  played  an  important  part  in  the  collection  of 
the  taxes.  In  this  connection  M.  Lieblein  has  quoted  a  passage  from  the  well- 
known  letter  from  the  chief  guardian  of  the  archives  of  Ameneman  to  the  scribe 
Pentaour,  in  which  he  says  :  "  The  scribe  of  the  port  arrives  at  the  station  ;  he 
collects  the  tax  ;  there  are  agents  with  rattans,  and  negroes  with  branches  of  palm  ; 
they  say  '  Give  us  some  corn  ! '  and  they  are  not  to  be  repulsed.  The  peasant  is 
bound  and  sent  to  the  canal ;  he  is  driven  on  with  violence,  his  wife  is  bound  in  his 
presence,  his  children  are  stripped ;  as  for  his  neighbours,  they  are  far  off  and  are 
busy  over  their  own  harvest."  {Les  Rccits  de  Recolte  dans  rancienne  Egypie,  annmc 
Elements  chronologiqiie^  in  Recneil  de  Tnivaiix  j-elatifs  a  la  Fhi/olo^ie  el  a  V  Archcologic 
egypiiennes  ct  assyrieunes,  t.  i.  p.  149). 


A   History  of  Art  in   Ancient  Egypt. 


The  first  condition  of  civilization  is  a  certain  measure  of  security 
for  life.  Now,  thanks  to  the  beneficent  action  of  the  king-  of 
rivers,  that  condition  was  created  sooner  in  Egypt  than  elsewhere. 
In  the  valley  of  the  Nile  man  found  himself  able,  for  the  first 
time,  to  calculate  upon  the  forces  of  nature  and  to  turn  them 
to  his  certain  profit.  It  is  easy  then  to  understand  that  Egypt 
saw  the  birth  of  the  most  ancient  of  those  civilizations  whose 
plastic  arts  we  propose  to  study. 

Another  favourable  condition  is  to  be  found  in  the  isolation  of 
the  country.  The  tribes  who  settled  there  in  centuries  so  remote 
that  they  are  beyond  tradition  and  even  calculation,  could  live 
in  peace,  hidden  as  it  were  in  a  narrow  valley  and  protected 
on   all   sides,   partly  by  deserts,  partly   by  an  impassable  sea.      It 


Fig.  5. — The  Bastinado :;  Beni-Hassan.      (ChampoIUon,  pi.  390.) 


would  perhaps  be  well  to  give  some  idea  of  the  natural  features 
of  their  country  before  commencing  our  study  of  their  art.  The 
terms.  Lower-,  Middle-,  and  Upper- Egypt,  the  Delta,  and 
Ethiopia  will  continually  recur  in  these  pages,  as  also  will  the 
names  of  Tanis  and  Sais,  Memphis  and  Heliopolis,  Abydos 
and  Thebes,  and  of  many  other  cities  ;  it  is  important  therefore 
that  our  readers  should  know  exactly  what  is  meant  by  each  of 
these  time-honoured  designations  ;  it  is  necessary  that  they  should 
at  least  be  able  to  find  upon  the  map  those  cities  which  by  their 
respective  periods  of  supremacy  represent  the  successive  epochs 
of  Egyptian  history. 

"  Egypt  is  that  country  which,  stretching  from  north  to  south, 
occupies  the  north-east  angle  of  Africa,  or  Libya  as  the  ancients 


The  Valley  of  the  Nile  and  its   Inhabitants.  7 

called  it.  It  is  joined  to  Asia  by  the  isthmus  of  Suez.  It  is 
bounded  on  the  east  by  that  isthmus  and  the  Red  Sea  ;  on  the 
south  by  Nubia,  the  Ethiopia  of  the  Greeks,  which  is  traversed 
by  the  Nile  before  its  entrance  into  Egypt  at  the  cataracts  of 
Syene  ;  on  the  west  by  the  desert  sprinkled  here  and  there  with 
a  few  oases,  and  on  the  north  by  the  Mediterranean.  The  desert 
stretches  as  far  north  on  the  west  of  the  country  as  the  Red  Sea 
does  on  the  east. 

*'  It  penetrates  moreover  far  into  the  interior  of   Egypt  itself. 
Strictly  speaking  Egypt  consists  simply  of  that  part  of  this  corner 
of  Africa   over    which    the  waters    of   the    Nile  flow  during  the 
inundation,   to  which  may  be  added  those  districts  to   which  the 
water  is  carried  by  irrigation.     All  outside  this  zone  is  uninhabited, 
and  produces  neither  corn  nor  vegetables  nor  trees  nor  even  grass. 
No  water  is  to  be  found  there  beyond  a  few  wells,  all  more  or  less 
exposed  to  exhaustion  in  an  ever-parching  atmosphere.      In  Upper 
Egypt  rain  is  an   extremely  rare   phenomenon.     Sand  and   rock 
cover   the  whole  country,  except  the   actual  valley   of  the    Nile. 
Up  to  the  point  where  the  river  divides   into   several  arms,  that 
is  to   say  for  more  than  three-quarters   of  the  whole    length  of 
Egypt,  this  valley  never  exceeds   an  average  width  of  more  than 
four  or  five  leagues.      In  a  few  districts  it  is  even  narrower  than 
this.      For   almost    its    whole    length    It    is    shut   In  between  two 
mountain  chains,  that  on   the  east  called   the   Arab,  that  on  the 
west  the  Libyan  chain.     These  mountains,  especially  towards  the 
south,  sometimes  close  in  and  form  defiles.     On  the  other  hand,  in 
Middle  Egypt  the   Libyan  chain  falls   back   and  becomes  lower, 
allowing  the  passage  of  the  canal  which  carries  the  fertilizing  waters 
into  the  Fayoum,  the  province  in  which  the  remains  of  the  famous 
reservoir    which    the    Greek    writers    called    Lake    Moeris    exist. 
Egypt,  which  was  little  more  than  a  glen  higher  up,  here  widens 
out  to  a  more  imposing  size.      A  little  below  Cairo,  the  present 
capital    of     Egypt,    situated    not    far   from    the    site    of    ancient 
Memphis,    the    Nile    divides    into    two    branches,  one    of  which, 
the    Rosetta  branch,    turns   to  the   north-west,  the  other,  that  of 
Damietta,  to  the  north  and  north-east  ....   The  ancients  knew 
five  others  which,  since  their  time,  have  either  been  obliterated  or 
at    least    have    become   non-navigable  ....   All  these  branches 
took  their  names  from  towns  situated  near  their  mouths.      A  large 
number  of  less  Important  watercourses  threaded  their  way  through 


8  A   History  of  Art  in  Ancient  Egypt. 

Lower  Egypt ;  but  as  the  earth  there  is  marshy,  their  channels 
have  shifted  greatly  from  age  to  age  and  still  go  on  changing. 
The  Nile  forms  several  lagunes  near  the  sea,  shut  in  by  long 
tongues  of  earth  and  sand,  and  communicating  with  the  Mediter- 
ranean by  openings  here  and  there.  The  space  comprised 
between  the  two  most  distant  branches  of  the  river  is  called  the 
Delta,  on  account  of  its  triangular  form,  which  is  similar  to  that  of 
a  capital  Greek  delta  (A)."  ^ 

At  one  time  the  waves  of  the  Mediterranean  washed  the 
foot  of  the  sandy  plateau  which  is  now  crowned  by  the  Great 
Pyramid  ;  the  Nile  flowed  into  the  sea  at  that  time  slightly  to 
the  north  of  the  site  upon  which  Memphis  was  afterwards  built. 
With  the  slow  passage  of  time  the  particles  of  earth  which  it 
brought  down  from  the  mountains  of  Abyssinia  were  deposited 
as  mud  banks  upon  the  coast,  and  gradually  filling  up  the  gulf, 
created  instead  wide  marshy  plains  intersected  by  lakes.  Here 
and  there  ancient  sand  ridges  indicate  the  successive  water- 
courses.  The  never-ceasing  industry  of  its  floods  had  already, 
at  the  earliest  historic  period,  carried  the  mouths  of  the  Nile 
far  beyond  the  normal  line  of  the  neighbouring  coasts.  The 
Egyptian  priests — whose  words  have  been  preserved  for  us  by 
Herodotus — had  a  true  idea  as  to  how  this  vast  plain  had  been 
created,  a  plain  which  now  comprises  twenty-three  thousand  square 
kilometres  and  is  continually  being  added  to ;  but  they  were 
strangely  deceived  when  they  thought  and  declared  that  Menes 
or  Mena,  the  first  of  all  kings,  found  almost  all  Egypt  under  the 
waters.  The  sea,  they  said,  penetrated  in  those  days  beyond 
the  site  of  Memphis,  and  the  remainder  of  the  country,  the 
district  of  Thebes  excepted,  was  an  unhealthy  morass.^  The 
Delta  had,  in  fact,  existed  long  before  the  appearance  of  Menes, 
and  perhaps  it  may  have  shown  pretty  much  its  present  form 
when  the  Egyptian  race  first  appeared  in  the  valley  of  the  Nile.^ 
As  to  the  origin  of  that  race,  we  need  not  enter  at  length 
into  a  question  so  purely  ethnographical.  It  is  now  generally 
allowed  that  they  were  connected  with  the  white  races  of  Europe 

1  RoBiou,  Histoire  ancienne  des  Peiiples  de  /'  Orient,  ch.  v. 

-  Herodotus,  ii.  4. 

^  Maspero,  Histoire  ancienjie  des  Peuples  de  r Orient,  pp.  6  and  7.  In  such  general 
explanations  as  are  unavoidable  we  shall  content  ourselves  with  paraphrasing 
M.  Maspero. 


The  Valley  of  the  Nile  and  its   Inhabitants. 


and  Western  Asia ;  the  anatomical  examination  of  the  bodies 
recovered  from  the  most  ancient  tombs,  and  the  study  of  their 
statues,  bas-rehefs,  and  pictures,  all  point  to  this  conclusion.  If 
we  take  away  individual  peculiarities  these  monuments  furnish 
us  with  the  following  common  type  of  the  race  even  in  the 
most  remote  epochs  : — 

"  The  average  Egyptian  was  tall,  thin,  active.  He  had  large 
and  powerful  shoulders, ^  a  muscular  chest,  sinewy  arms  termi- 
nating in  long  and  nervous  hands,  narrow  hips,  and  thin  muscular 
legs.  His  knees  and  calves  were  nervous  and  muscular,  as  is 
generally  the  case  with  a  pedestrian  race  ;  his  feet  were  long, 
thin,  and  flattened,  by  his  habit  of  going  barefoot.  The  head, 
often  too  large  and  powerful  for  the  body,  was  mild,  and  even 
sad  in  its  expression.  His  forehead  was  square  and  perhaps  a 
little  low,  his  nose  short  and  round  ;  his  eyes  were  large  and  well 
opened,  his  cheeks  full  and  round,  his  lips  thick  but  not  turned 
out  like  a  negro's ;  his  rather  large  mouth  bore  an  habitually 
soft  and  sorrowful  expression.  These  features  are  to  be  found 
in  most  of  the  statues  of  the  ancient  and  middle  empires,  and 
in  all  the  later  epochs.  Even  to  the  present  day  the  peasants, 
or  fellaJis,  have  almost  everywhere  preserved  the  physiognomy 
of  their  ancestors,  although  the  upper  classes  have  lost  it  by 
repeated  intermarriage  with  strangers."  ^ 

When  Mariette  discovered  in  the  necropolis  at  Memphis  the 
famous  wooden  statue  of  a  man  standing  and  holding  in  his 
hand  the  baton  of  authority,  the  peasants  of  Sakkarah  recog- 
nised at  once  the  feature  and  attitude  of  one  of  themselves, 
of  the  rustic  dignitary  who  managed  the  corvdes  and  apportioned 
the  taxation.  An  astonished  fellah  cried  out :  "  The  Sheikh-el- 
Beled!"  His  companions  took  up  the  cry,  and  the  statue  has 
been  known  by  that  name  ever  since.^ 

Increased  knowledge  of  the  Egyptian  language  has  enabled  us 
to  carry  our  researches  much  farther  than  Champollion  and  his 
successors.       By  many   of  its  roots,   by  its    system   of  pronouns, 

^  Their  exceptional  breadth  of  shoulder  has  been  confirmed  by  an  exammation  of 
the  skeletons  in  the  mummies.  See  on  this  subject  a  curious  note  in  Bonomi's  So/ne 
ObseiTations  on  the  Skeleton  of  an  Egyptian  Alum  my  ( Transactions  of  the  Society  of 
Biblical  Airhceolog^<,\o\.  iv.  pp.  251 — 253).         ^  Maspero,  Histoire  ancienne,  p.  16. 

^  Notice  des  principaux  Afonuments  exposes  dans  les  Galcries  provisoires  dii  Musee 
£A7itiquites  egyptiennes  de  S.  A.  le  Vice-Roi,  a  Boulaq  (1876),  No.  492.  The  actual 
statue  holds  the  baton  in  its  left  hand. 

VOL.    I.  C 


lO 


A   History  of  Art  in  Ancient  Egypt. 


by  its  nouns  of  number,  and  by  some  of  the  arrangements  of 
its  conjugations,  it  seems  to  have  been  attached  to  the  Semitic 
family   of    languages.      Some    of    the    idioms    of    these    Semitic 

tongues  are  found  in  Egyptian 
in  a  rudimentary  state.  From 
this  it  has  been  concluded 
that  Egyptian  and  its  cognate 
languages,  after  having  belonged 
to  that  group,  separated  from  it 
at  a  very  early  period,  while  their 
grammatical  system  was  still  in 
course  of  formation.  Thus,  dis- 
united and  subjected  to  diverse 
influences,  the  two  families  made 
a  different  use  (gf  the  elements 
which  they  possessed  in  common. 
There  would  thus  seem  to 
have  been  a  community  of  root 
between  the  Egyptians  on  the 
one  part  and  the  Arabs,  Hebrews, 
and  Phoenicians  on  the  other, 
but  the  separation  took  place  at 
such  an  early  period,  that  the 
tribes  who  came  to  establish 
themselves  in  the  valley  of  the 
Nile  had  both  the  time  and  the 
opportunity  to  acquire  a  very 
particular  and  original  physi- 
ognomy of  their  own.  The 
Egyptians  are  therefore  said  to 
belong  to  the  proto-Semitic 
races. 
This  opinion  has  been  sustained  with  more  or  less  plausi- 
bility by    MM.    Lepsius,    Benfey,  and    Bunsen,   and  accepted  by 


Fig.  6. — Statue  from  the  Ancient  Empire, 
in  calcareous  stone.  (Boulak.^)  Drawn 
by  G.   Benedite. 


^  Notice  des  pr'nicipaux  Monuments  exposes  dans  les  Galeiies  provisvires  dii  Mnsee 
d'Antiquitcs  cgyptiennes  de  S.  A.  le  Vice-Roi,  a  Botdaq  (1S76),  p.  582.  With  the 
exception  of  a  few  woodcuts  from  photographs  the  contents  of  the  museums  at 
Cairo  and  Boulak  have  been  reproduced  from  drawings  by  M.  J.  Bourgoin.  The 
Boulak  Museum  will  be  referred  to  by  the  simple  word  Boulak.  The  reproductions 
of  objects  in  the  Louvre  are  all  from  the  pencil  of  M.  Saint-Elme  Gautier. 


Yi^.^j.—Th.^  Sheikh-cl-Belcd.     (Boulak.)     Drawn  by  J.  Bourgoin. 


The  Valley  of  the  Nile  and  its   Inhabitants.  13 

M.  Maspero.^  But  other  critics  of  equal  authority  are  more 
impressed  by  the  differences  than  by  the  resemblances,  which, 
however,  they  neither  deny  nor  explain.  M.  Renan  prefers  to 
rank  the  Copts,  the  Tuaregs,  and  the  Berbers  in  a  family  which 
he  would  call  Chamitic,  and  to  which  he  would  refer  most  of 
the  idioms  of  Northern  Africa.^  A  comparison  of  the  languages 
is,  then,  insufficient  to  decide  the  question  of  origin. 

The  people  whose  physical  characteristics  we  have  described 
and  whose  idiom  we  have  defined,  came  from  Asia,  to  all  appear- 
ance, by  the  Isthmus  of  Suez.  Perhaps  they  found  established 
on  the  banks  of  the  Nile  another  race,  probably  black,  and 
indigenous  to  the  African  continent.^  If  this  were  so  the  new 
comers  forced  the  earlier  occupants  of  the  country  southwards 
without  mixing  with  them,  and  set  themselves  resolutely  to  the 
work  of  improvement.  Egypt  must  then  have  presented  a  very 
different  sight  from  its  richness  and  fertility  of  to-day.  The 
river  when  left  to  itself,  was  perpetually  changing  its  bed,  and 
even  in  its  highest  floods  it  failed  to  reach  certain  parts  of  the 
valley,  which  remained  unproductive  ;  in  other  districts  it 
remained  so  long  that  it  changed  the  soil  into  swamp.  The 
Delta,  half  of  it  drowned  in  the  waters  of  the  Nile,  the  other 
half  under  those  of  the  Mediterranean,  was  simply  a  huge 
morass  dotted  here  and  there  with  sandy  islands  and  waving 
with  papyrus,  reeds,  and  lotus,  across  which  the  river  worked 
its  sluggish  and  uncertain  way ;  upon  both  banks  the  desert 
swallowed  up  all  the  soil  left  untouched  by  the  yearly  inun- 
dations. From  the  crowding  vegetation  of  a  tropical  marsh 
to  the  most  absolute  aridity  was  but  a  step.  Little  by  little 
the  new  comers  learnt  to  control  the  course  of  the  floods,  to 
bank  them  in  and  to  carry  them  to  the  farthest  corners  of  the 
valley,  and  Egypt  gradually  arose  out  of  the  waters  and  became 
in  the  hand  of  man  one  of  the  best  adapted  countries  in  the 
world  for  the  development  of  a  great  civilization,'^ 

How  many  generations  did  it  require  to  create  the  country 
and    the    nation  ?      We   cannot   tell.      But  we  may  affirm  that  a 

^  Maspero,  Histoire  ancienne,  p.  17. 
2  Histoire  des  Longiies  sanitiqiies,  Book  i.  ch.  ii.  §  4. 

^  See  Lepsius,  Ucber  die  Atwahme  ei?ies  sogenatmtcn  prehistorischcn  Stcinaltcrs  in 
ALgypten  (in  the  ZeitscJuHft  fi'ir  ALgypiische  Sprache,  1870,  p.  113,  at  seq.). 
^  Maspero,  p.  18. 


H 


A  History  of  Art  in  Ancient  Egypt. 


commencement  was  made  by  the  simultaneous  establishment  at 
several  different  points  of  small  independent  states,  each  of  which 
had  its  own  laws  and  its  own  form  of  worship.  These  districts 
remained  almost  unchanged  in  number  and  in  their  respective 
boundaries  almost  up  to  the  end  of  the  ancient  world.  Their 
union   under  one  sceptre  formed   the  kingdom  of  the   Pharaohs, 


jBI»V^4;^,>^C^C/J^"^  -^i  H^A^" 


I   >^>5 


Fig.  8. — Hunting  in  the  Marthes  ;  from  a  tas-relief  in  tlie  tomb  of  Ti. 


the  country  of  Khemi,  but  their  primitive  divisions  did  not  there- 
fore disappear  ;  the  small  independent  states  became  provinces 
and  were  the  foundation  of  those  local  administrative  districts 
which  the  Greeks  called  nomes. 

Besides    this    division    into    districts,    the    Egyptians    had    one 
other,   and    only    one — the    division    into    Lower    Egypt,  or  the 


Tpie  Valley  of  the  Nile  and  its  Inhabitants. 


15 


North  Country  {Toiiiera,  or  To-meJi),  and  into  Upper  Egypt, 
or  the  South  Country  {To-res).  Lower  Egypt  consisted  of  the 
Delta;  Upper  Egypt  stretched  from  the  southernmost  point  of 
the  Delta  to  the  first  cataract.  This  division  has  the  advantage 
of  corresponding  exactly  to  the  configuration  of  the  country  ; 
moreover,  it  preserves  the  memory  of  a  period  before  the  time 
of  Menes,  during  which  Egypt  was  divided  into  two  separate 
kingdoms — that  of  the  North  and  that  of  the  South,  a  division 
which  in  later  times  had  often  a  decisive  influence  upon  the 
course  of  events.  This  state  of  things  was  of  sufficiently  long 
duration  to  leave  an  ineffaceable  trace  upon  the  official  language  of 
Egypt,  and  upon  that  which  we  may  call  its  blazonry,  or  heraldic 
imagery.     The  sovereigns  who  united   the  whole  territory  under 


.■-■■:r->i.T?^';;fc 


Fig.  9. — Shadouf ;  machine  for  irrigatihg  the  land  above  the  level  of  the  canals. 


one  sceptre  are  always  called,  in  the  royal  protocols,  the  lords 
of  Upper  and  Lower  Egypt  ;  they  carry  on  their  heads  two 
crowns,  each  appropriate  to  one  of  the  two  great  divisions  of 
their  united  kingdom.  That  of  Upper  Egypt  is  known  to 
egyptologists  as  the  white  crown,  because  of  the  colour  which 
it  bears  upon  painted  monuments  ;  that  of  the  North  is  called 
the  red  crown,  for  a  similar  reason.  Combined  with  one  another 
they  form  the  complete  regal  head-dress  ordinarily  called  the 
pschent.  In  the  hieroglyphics  Northern  Egypt  is  indicated  by 
the  papyrus  ;  Southern   by  the  lotus. 

During  the  Ptolemaic  epoch  a  new  administrative  division  into 
Upper,  Middle,  and   Lower  Egypt  was  established.      The  Middle 


i6 


A  History  of  Art  in  Ancient  Egypt. 


Egypt  of  the  Greek  geographers  began  at  the  southern  point 
of  the  Deha,  and  extended  to  a  Httle  south  of  HermopoHs. 
Although  this  latter  division  was  not  estabhshed  until  after  the 
centuries  which  saw  the  birth  of  those  monuments  with  which 
we  shall  have  to  deal,  we  shall  make  frequent  use  of  it,  as  it 
will  facilitate  and  render  more  definite  our  topographical  expla- 
nations.     For  the  contemporaries  of  the  Pharaohs  both  Memphis 


Fig.  io. — The  White  Crown.  Fig.  ii. — The  Red  Crown.  Fig.  12. — The  Pschent. 


and  Thebes  belong  to  Upper  Egypt,  and  if  we  adopted  their 
method  of  speech  we  should  be  under  the  continual  necessity 
of  stopping  the  narration  to  define  geographical  positions  ;  but 
with  the  tri-partite  division  we  may  speak  of  Beni-Hassan  as  in 
Middle,  and  Abydos  as  in  Upper  Egypt,  and  thus  give  a 
sufficient  idea  of  their  relative  positions. 


§   3.    The   Great  Divisions  of  Egyptian  History. 

In  enumerating  and  analysing  the  remains  of  Egyptian  art,  we 
shall  classify  them  chronologically  as  well  as  locally.  The 
monuments  of  the  plastic  arts  will  be  arranged  into  groups  de- 
termined by  the  periods  of  their  occurrence,  as  well  as  by  their 
geographical  distribution.  We  must  refer  our  readers  to  the  works 
of  M.  Maspero  and  others  for  the  lists  of  kings  and  dynasties,  and 
for  the  chief  events  of  each  reicrn,  but  it  will  be  convenient  for  us 
to  give  here  a  summary  of  the  principal  epochs  in  Egyptian  history. 
Each  of  those   epochs   corresponds  to   an  artistic    period    with  a 


The  Great  Divisions  of  Egyptian   History.  17 

special  character  and  individuality  of  its  own.  The  following 
paragraphs  taken  from  the  history  of  M.  Maspero  give  all  the 
necessary  information  in  a  brief  form. 

"  In  the  last  years  of  the  prehistoric  period,  the  sacerdotal  class 
had  obtained  a  supremacy  over  the  other  classes  of  the  nation. 
A  man  called  Menes  (Menha  or  Mena  in  the  Egyptian  texts) 
destroyed  this  supremacy  and  founded  the  Egyptian  monarchy. 

"  This  monarchy  existed  for  at  least  four  thousand  years,  under 
thirty  consecutive  dynasties,  from  the  reign  of  Menes  to  that  of 
Nectanebo  (340  years  before  our  era).  This  interval  of  time,  the 
longest  of  which  political  history  takes  note,  is  usually  divided  into 
three  parts  :  the  Ancient  Empire,  from  the  first  to  the  eleventh 
dynasty  ;  the  Aliddle  Empire,  from  the  eleventh  dynasty  to  the  in- 
vasion of  the  Hyksos  or  Shepherds;  the  New  Empire  from  the 
shepherd  kings  to  the  Persian  conquest.  This  division  is  incon- 
venient in  one  respect ;  it  takes  too  little  account  of  the  sequence 
of  historical  events. 

"  There  were  indeed,  three  great  revolutions  in  the  historical 
development  of  Egypt.  At  the  beginning  of  its  long  succession 
of  human  dynasties  (the  Egyptians,  like  other  peoples,  placed  a 
number  of  dynasties  of  divine  rulers  before  their  first  human  king) 
the  political  centre  of  the  country  was  at  Memphis  ;  Memphis  was 
the  capital  and  the  burying-place  of  the  kings  ;  Memphis  imposed 
sovereigns  upon  the  rest  of  the  country  and  was  the  chief  market 
for  Egyptian  commerce  and  industry.  With  the  commencement  of 
the  sixth  dynasty,  the  centre  of  gravity  began  to  shift  southwards. 
During  the  ninth  and  tenth  dynasties  it  rested  at  Heracleopolis, 
in  Middle  Egypt,  and  in  the  time  of  the  eleventh  dynasty,  it  fixed 
itself  at  Thebes.  From  that  period  onwards  Thebes  was  the 
capital  of  the  country  and  furnished  the  sovereign.  From  the 
eleventh  to  the  twenty-first  all  the  Egyptian  dynasties  were  Theban 
with  the  single  exception  of  the  fourteenth  Xoite  dynasty.  At  the 
time  of  the  shepherd  invasion,  the  Thebaid  became  the  citadel  of 
Egyptian  nationality,  and  its  princes,  after  centuries  of  war  against 
the  intruders,  finally  succeeded  in  freeing  the  whole  valley  of  the 
Nile  for  the  benefit  of  the  eighteenth  dynasty,  which  opened  the 
era  of  great  foreign  wars. 

"  Under  the  nineteenth  dynasty  an  inverse  movement  to  that 
of  the  first  period  carried  the  political  centre  of  the  country  back 
towards  the  north.     With  the  twenty-first  Tanite  dynasty,  Thebes 

VOL.    I.  D 


1 8  A   History  of  Art  in  Ancient  Egypt. 

5 ^ . 

ceased  to  be  the  capital,  and  the  cities  of  the  Delta,  Tanis,  Bubastis, 
Mendes,  Sebennytos,  and  above  all  Sais,  rose  into  equal  or  superior 
importance.  From  that  time  the  political  life  of  the  country  con- 
centrated itself  in  the  maritime  districts.  The  nomes  of  the 
Thebaid,  ruined  by  the  Ethiopian  and  Assyrian  invasions,  lost  their 
influence  ;  and  Thebes  itself  fell  into  ruin  and  became  nothing 
more  than  a  rendezvous  for  curious  travellers. 

"  I  propose,  therefore,  to  divide  Egyptian  history  into  three 
periods,  each  corresponding  to  the  political  supremacy  of  one  town 
or  province  over  the  whole  of  Egypt : — 

"  First  Period,  Memphite  (the  first  ten  dynasties).  The 
supremacy  of  Memphis  and  of  the  sovereigns  furnished  by 
her. 

"  Second  Period,  Theban  (from  the  eleventh  to  the  twentieth 
dynasties  inclusive).  Supremacy  of  Thebes  and  the  Theban  kings. 
This  period  is  divided  into  two  sub-periods  by  the  Shepherd 
dynasties. 

''a.  The  old  Theban  empire,  from  the  eleventh  to  the  sixteenth 
dynasties. 

'' b.  The  new  Theban  empire,  from  the  sixteenth  to  the  twentieth 
dynasties. 

"  Third  Period,  Sait  (from  the  twenty-first  to  the  thirtieth 
dynasties,  inclusive).  Supremacy  of  Sais  and  the  other  cities  of  the 
Delta.     This  period  is  divided  into  two  by  the  Persian  invasion  : — 

"  First  Sait  period,  from  the  twenty-first  to  the  twenty-sixth 
dynasties, 

"  Second  Sait  period,  from  the  twenty-seventh  to  the  thirtieth 
dynasties."  ^ 

Mariette  places  the  accession  of  Mena  or  Menes  at  about  the 
fiftieth  century  before  our  era,  while  Bunsen  and  other  Egyptologists 
bring  forward  his  date  to  3,600  or  3,500  b.c.  as  they  believe  some 
of  the  dynasties  of  Manetho  to  have  been  contemporary  with  each 
other.  Neither  Mariette  nor  Maspero  deny  that  Egypt,  in  the 
course  of  its  long  existence,  was  often  partitioned  between  princes 
who  reigned  in  Upper  and  Lower  Egypt  respectively;  but,  guided 

^  Histoire  ancienne  des  Peuples  de  V  Orient,'^.  ^2^.  We  believe  that  the  division 
proposed  by  M.  Maspero  is,  in  fact,  the  best.  It  is  the  most  suggestive  of  the  truth 
as  to  the  successive  displacements  of  the  political  centre  and  the  movement  of 
history.  We  sliall,  however,  have  no  hesitation  in  making  use  of  the  terms  Ancient, 
Middle,  and  New  Empire,  as  occasion  arises. 


The  Great  Divisions  of  Egyptian   History.  19 

by  circumstances  which  need  not  be  described  here,  they  inchne  to 
beheve  that  Manetho  confined  himself  to  enumerating  those 
dynasties  which  were  looked  upon  as  the  legitimate  ones.  The 
work  of  elimination  which  has  been  attempted  by  certain  modern 
savants,  must  have  been  undertaken,  to  a  certain  extent,  in  Egypt 
itself  ;  and  some  of  the  collateral  dynasties  must  have  been  effaced 
and  passed  over  in  silence,  because  the  monuments  still  remaining 
preserve  the  names  of  reigning  families  which  are  ignored  by 
history. 

Whatever  may  be  thought  of  this  initial  date,  Egypt  remains,  as 
has  been  so  well  said  by  M.  Renan,  "  a  lighthouse  in  the  profound 
darkness  of  remote  antiquity."  Its  period  of  greatest  power  was 
long  anterior  to  the  earliest  traditions  of  the  Greek  race  ;  the  reign 
of  Thothmes  III.,  who,  according  to  a  contemporary  expression, 
"drew  his  frontiers  where  he  pleased,"  is  placed  by  common  con- 
sent in  the  seventeenth  century,  b.c.  The  Egyptian  empire  then 
comprised  Abyssinia,  the  Soudan,  Nubia,  Syria,  Mesopotamia,  part 
of  Arabia,  Khurdistan,  and  Armenia.  Founded  by  the  kings  of 
the  eighteenth  dynasty,  this  greatness  was  maintained  by  those  of 
the  nineteenth.  To  this  dynasty  belonged  Rameses  II.,  the 
Sesostris  of  the  Greeks,  who  flourished  in  the  fifteenth  century.  It 
was  the  superiority  of  its  civilization,  even  more  than  the  valour  of 
its  princes  and  soldiers,  which  made  Egypt  supreme  over  Western 
Asia. 

This  supremacy  declined  during  the  twenty-first  and  twenty- 
second  dynasties,  but,  at  the  same  time,  Egyptian  chronology 
becomes  more  certain  as  opportunities  of  comparison  with  the  facts 
of  Hebrew  history  increase.  The  date  of  980,  within  a  year  or 
two,  may  be  given  with  confidence  as  that  of  the  accession  of 
Sheshonk  I.,  the  contemporary  of  Solomon  and  Rehoboam.  From 
that  date  onwards,  the  constant  struggles  betw^een  Egypt  and  its 
neighbours,  especially  with  Assyria,  multiply  our  opportunities  for 
synchronic  comparison.  In  the  seventh  century  the  country  was 
opened  to  the  Greeks,  the  real  creators  of  history,  who  brought 
with  them  their  inquiring  spirit  and  their  love  for  exactitude. 
After  the  accession  of  Psemethek  I.,  the  founder  of  the  twenty- 
sixth  dynasty,  in  656,  our  historical  materials  are  abundant.  For 
that  we  must  thank  the  Greek  travellers  who  penetrated  every- 
where, taking  notes  which  they  afterwards  amplified  into  narratives. 
It  is  a  singular  thing,  that  even  as  late  as  the  Ptolemies,  when  the 


20  A  History  of  Art  in  Ancient  Egypt. 

_ 

power  of  the  Macedonian  monarchy  was  fully  developed,  the 
Egyptians  never  seem  to  have  felt  the  want  of  what  we  call  an 
era,  of  some  definite  point  from  which  they  could  measure  the 
course  of  time  and  the  progress  of  the  centuries.  "  They  were 
satisfied  with  calculating  by  the  years  of  the  reigning  sovereign, 
and  even  those  calculations  had  no  certain  point  of  departure. 
Sometimes  they  counted  from  the  commencement  of  the  year  which 
had  witnessed  the  death  of  his  predecessor,  sometimes  from  the 
day  of  his  own  coronation.  The  most  careful  calculations  will 
therefore  fail  to  enable  modern  science  to  restore  to  the  Egyptians 
that  which,  in  fact,  they  never  possessed."  ^ 

Even  thus  summarily  stated,  these  historical  indications  are 
enough  to  show  how  little  foundation  there  is  for  the  opinion  which 
was  held  by  the  ancient  Greeks,  and  too  long  accepted  by  modern 
historians.  It  was,  they  said,  from  Ethiopia  that  Egyptian 
civilization  had  come.  A  colony  of  Ethiopian  priests  from  the 
island  of  Meroe  in  Upper  Nubia,  had  introduced  their  religion, 
their  written  characters,  their  art  and  their  civil  institutions  into  the 
country.  The  exact  opposite  of  this  is  the  truth.  "It  was  the 
Egyptians  who  advanced  up  the  banks  of  the  Nile  to  found  cities, 
fortresses,  and  temples  in  Ethiopia  ;  it  was  the  Egyptians  who 
carried  their  civilization  into  the  midst  of  savage  negro  tribes. 
The  error  was  caused  by  the  fact  that  at  one  epoch  in  the  history 
of  Egypt  the  Ethiopians  played  an  important  part. 

"If  it  were  true  that  Egypt  owed  its  political  existence  to 
Ethiopia,  we  should  be  able  to  find  in  the  latter  country  monuments 
of  a  more  remote  antiquity,  and  as  we  descended  the  Nile,  we 
should  find  the  remains  comparatively  modern  ;  but,  strangely 
enough,  the  study  of  all  these  monuments  incontestably  proves 
that  the  sequence  of  towns,  holy  places,  and  tombs,  constructed  by 
the  Egyptians  on  the  banks  of  their  river,  follow  each  other  in 
such  chronological  order  that  the  oldest  remains,  the  Pyramids, 
are  found  in  the  north,  in  Lower  Egypt,  near  the  southern  point 
of  the  Delta,  The  nearer  our  steps  take  us  to  the  cataracts  of 
Ethiopia,  the  less  ancient  do  the  monuments  become.  They  show 
ever  increasing  signs  of  the  decadence  of  art,  of  taste,  and  of  the 
love  for  beauty.  Finally,  the  art  of  Ethiopia,  such  as  its  still  existing 
monuments  reveal  it  to  us,  is  entirely  wanting  in  originality.  A 
glance  is  sufficient  to  tell  us  that  it  represents  the  degeneracy  only 
^  Mariette,  Apcf(n  de  l' Histoire  dEgypfe,  p.  66. 


The  Constitution  of  Egyptian  Society.  21 

of  the  Egyptian  style,  that  the  spirit  of  Egyptian  forms  has  been 
weakly  grasped,  and  that  their  execution  is  generally  mediocre."  ^ 

We  may  condense  all  these  views  into  a  simple  and  easily 
remembered  formula ;  we  may  say  that  as  we  mount  towards  the 
springs  of  the  Nile,  we  descend  the  current  of  time.  Thebes  is 
younger  than  Memphis,  and  Meroe  than  Thebes.  The  river 
which  Egypt  worshipped,  and  by  which  the  walls  of  its  cities  were 
bathed,  flowed  from  the  centre  of  Africa,  from  the  south  to  the 
north  ;  but  the  stream  of  civilization  flowed  in  the  other  direction, 
until  it  was  lost  in  the  country  of  the  negro,  in  the  mysterious 
depths  of  Ethiopia.  The  springs  of  this  latter  stream  must  be 
sought  in  that  district  where  the  waters  of  the  Nile,  as  if  tired  by 
their  long  journey,  divide  into  several  arms  before  falling  into  the 
sea;  In  that  district  near  the  modern  capital,  over  which  stretch 
the  long  shadows   of  the  Pyramids. 


§  4.    The  Constitution  of  Egyptian  Society— I nfluence  of  that 
Constitution  upon  Monuments  of  Art. 

During  the  long  sequence  of  centuries  which  we  have  divided 
into  three  great  periods,  the  national  centre  of  gravity  was  more 
than  once  displaced.  The  capital  was  at  one  time  in  Middle 
Egypt,  at  another  in  Upper,  and  at  a  third  period  in  Lower  Egypt, 
in  accordance  with  its  political  necessities.  At  one  period  the 
nation  had  nothing  to  fear  from  external  enemies,  at  others  It  had 
to  turn  a  bold  front  to  Asia  or  Ethiopia.  At  various  times  Egypt 
had  to  submit  to  her  foreign  foes  ;  to  the  shepherd  invaders,  to  the 
kings  of  Assyria  and  Persia,  to  the  princes  of  Ethiopia,  and  finally 
to  Alexander,  to  whom  she  lost  her  Independence  never  again  to 
recover  it.  And  yet  it  appears  that  the  character  and  social 
condition  of  the  race  never  underwent  any  great  change.  At 
the  time  of  the  pyramid-builders,  Egypt  was  the  most  absolute 
monarchy  that  ever  existed,  and  so  she  remained  till  her  final 
conquest. 

^  Brugsch-Bey,  Histoirc  de  V Egypt,  pp.  6  and  7.  Maspero's  Histoire  ancienne, 
p.  382,  may  also  be  consulted  upon  the  character  of  the  Ethiopian  kingdom  and  the 
monuments  of  Napata.  A  good  idea  of  this  process  of  degradation  may  be  gained 
by  merely  glancing  through  the  plates  to  part  v.  of  Lepsius's  Doikmickr ;  plate  6, 
for  example,  shows  what  the  caryatid  became  at  Napata. 


2  2  A  History  of  Art  in  Ancient  Egypt. 

"  Successor  and  descendant  of  the  deities  who  once  reigned 
over  the  valley  of  the  Nile,  the  king  was  the  living  manifestation 
and  incarnation  of  God  :  child  of  the  sun  [Se  Rd),  as  he  took  care 
to  proclaim  whenever  he  wrote  his  name,  the  blood  of  the  gods 
flowed  in  his  veins  and  assured  to  him  the  sovereign  power."  ^ 
He  was  the  priest  above  all  others.  Such  a  form  of  w^orship  as 
that  of  Egypt,  required  no  doubt  a  large  sacerdotal  class,  each 
member  of  which  had  his  own  special  function  in  the  complicated 
and  gorgeous  ceremonies  in  which  he  took  part  ;  but  the  king 
alone,  at  least  in  the  principal  temples,  had  the  right  to  enter  the 
sanctuary  and  to  open  the  door  of  the  kind  of  chapel  in  which  the 
symbolical  representation  of  the  divinity  was  kept ;  he  alone  saw 
the  god  face  to  face,  and  spoke  to  him  in  the  name  of  his  people.^ 
The  pre-eminent  dignity  of  this  priestly  office  did  not,  however, 
prevent  the  king  from  taking  his  proper  share  in  war  or  political 
affairs  generally.  The  army  of  scribes  and  various  functionaries, 
wdiose  titles  may  still  be  read  upon  the  most  ancient  monuments  of 
the  country,  depended  upon  him  for  their  orders  from  one  end  of 
the  country  to  the  other,  and  in  w^ar,  it  was  he  wdio  led  the  serried 
battalions  of  the  Egyptian  army.  The  king  was  thus  a  supreme 
pontif,  the  immediate  chief  of  all  civil  and  military  officers  ;  and, 
as  the  people  believed  that  his  career  was  directed  by  the  gods 
with  whom  he  held  converse,  he  became  to  them  a  visible  deity 
and,  in  the  words  of  an  inscription,  "'  the  representative  of  Ra 
among  the  living."  His  divinity,  begun  on  earth,  was  completed 
and  rendered  perpetual  in  another  life.  All  the  dead  Pharaohs 
became  gods,  so  that  the  Egyptian  pantheon  obtained  a  new  deity 
at  the  death  of  each  sovereign.  The  deceased  Pharaohs  thus 
constituted  a  series  of  pfods  to  whom  the  rei^nino-  sovereign  would 
of  course  address  himself  when  he  had  anything  to  ask  ;  hence  the 

1  Maspero,  Ilistoire  ancie/ine,  p.  58.  This  affiliation  of  the  king  to  the  god  was 
more  than  a  figure  of  speech.  In  an  inscription  which  is  reproduced  both  at 
Ipsamboul  and  at  Medinet-Abou,  Ptah  is  made  to  speak  in  the  following  terms  of 
Rameses  II.  and  Rameses  III.  respectively:  "I  am  thy  father,  as  a  god  I  have 
begotten  thee  j  all  thy  members  are  divine  ;  when  I  approached  thy  royal  mother  I 
took  upon  me  the  form  of  the  sacred  ram  of  IMendes"  (line  3rd).  This  curious 
text  has  lately  been  interpreted  by  E.  'i>i:xv\\\e  {Society  of  Biblical  Arc/iaoIogy,\o\.  vii. 
pp.  1 19-138).  The  monarchy  of  the  Incas  was  founded  upon  an  almost  identical 
belief 

2  See  the  account  of  the  visit  to  Heliopolis  of  the  conquering  Ethiopian,  Piankhi- 
Mer-Amen ;  we  shall  quote  the  text  of  tliis  famous  inscription  in  our  chapter  upon 
the  Egyptian  temple. 


o 


CJ 


'3 


,J0 


mi   ^ 


C/1 


The  Constitution  of  Egyptian  Society, 


25 


monuments  upon  which  we  find  living  Pharaohs  offering  worship 
to  their  predecessors.^ 

The  prestige  which  such  a  theory  of  royalty  w^as  calculated  to 
give  to  the  Egyptian  kings  may  easily  be  imagined.  They 
obtained  more  than  respect ;  they  were  the  objects  of  adoration,  of 
idolatry.  Brought  up  from  infancy  in  this  religious  veneration,  to 
which  their  hereditary  qualities  also  inclined  them,  generation 
succeeded  generation  among  the  Egyptians,  without  any  attempt 
to  rebel  against  the  royal  authority  or  even  to  dispute  it.  Ancient 
Egypt,  like  its  modern  descendant,  was  now  and  then  the  scene  of 
military  revolts.  These  were  generally  provoked  by  the  presence 
of  foreign  mercenaries,  sometimes  by  their  want  of  discipline  and 
licence,  sometimes  by  the  jealousy  which  they  inspired  in  the 
native  soldiery  ;  but  never, 
from  the  time  of  Menes 
to  that  of  Tewfik-Pacha, 
has  the  civil  population, 
whether  of  the  town  or 
of  the  fields,  shown  any 
desire  to  obtain  the 
slightest  guarantee  for 
what  we  should  call  their 
rights  and  liberties.  Dur- 
ing all  those  thousands  of 
years  not  the  faintest  trace 
is  to  be  discovered  of  that 

spirit  from  which  sprung  the  republican  constitutions  of  Greece 
and  ancient  Italy,  a  spirit  which,  in  yet  later  times,  has  led  to  the 
parliamentary  governments  of  Christian  Europe.  The  Egyptian 
labourer  or  artisan  never  dreamt  of  calling  in  question  the  orders 
of  any  one  who  might  be  master  for  the  time.  Absolute  obedience 
to  the  will  of  a  single  man — such  was  the  constant  and  instinctive 
national  habit,  and  by  it  every  movement  of  the  social  machine, 
under  foreign  and  native  kings  alike,  was  regulated. 

From  the  construction  of  the  pyramids  of  Cheops  and  Chephren, 
and  the  cutting  of  a  new  canal  between  the  two  seas  under  Nekau, 
to    the    Mahmoudieh    canal    of    Mehemet-Ali    and    that    abortive 

1  Fr.  Lenormant,  Manuel  d Histoire  ancienne,  t.  i,  pp.  485-486.  The  most 
celebrated  of  these  is  the  famous  Chamber  of  Ancestors  from  Karnak,  which  is  now 
preserved  in  the  Bibliothcqiie  Nationale  at  Paris. 

VOL.    I.  E 


Fjg.  14. 


-Rameses  II.  in  adoration  before  Seti. 
From  Abydos  (Mariette). 


26 


A   History  of  Art  in  Ancient  Egypt. 


enterprise,  the  barrage  of  the  Nile,  the  only  method  thought  of  for 
obtaining  the  necessary  labour  was  compulsion,^  An  order  is 
received  by  the  governor,  who  has  it  proclaimed  from  one  village 
to  another  throughout  his  province  ;  next  day  the  w^hole  male 
population  is  driven,  like  a  troop  of  sheep,  to  the  workshops. 
Each  man  carries  a  bag  or  basket  which  holds  his  provisions  for  a 
fortnight  or  a  month,  as  the  case  may  be  ;  a  few  dry  cakes,  onions, 
garlic,  and  Egyptian  beans,  as  the  Greeks  called  the  species  of 
almond  which  is  contained  in  the  fruit  of  the  lotus.  Old  men  and 
children,  all  had  to  obey  the  summons.  The  more  vigorous  and 
skilful  among  them  dressed  and  put  in  place  the  blocks  of  granite 
or  limestone  ;  the  weakest  were  useful  for  the  transport  of  the 
rubbish  to  a  distance,  for  carrying  clay  and  water  from  the  Nile  to 


Fig.  15  — Homage  to  Amenophis  III.     (From  Prisse.-) 


the  brickmakers,  for  arranging  the  bricks  in  the  sun  so  that  they 
might  be  dried  and  hardened. 

Under  the  stimulus  of  the  rod,  this  multitude  worked  well  and 
obediently  under  the  directions  of  the  architect's  foreman  and  of 
skilled  artisans  who  were  permanently  employed  upon  the  work  ; 
they  did  all  that  could  be  done  by  men  without  special  education. 
At  the  end  of  a  certain  period  they  were  relieved  by  fresh  levies 

1  The  beaters  for  the  great  hunts  which  took  place  in  the  Delta  and  the  Fayoum 
were  procured  in  the  same  fashion.  These  hunts  were  among  the  favourite  pleasures 
of  the  kings  and  the  great  lords.  See  Maspero,  Le  Papyrus  Mallet,  p.  58  (in 
Recueil  de  Travaux,  etc.  t.  i). 

^  The  work  to  which  we  here  refer  is  the  Histoire  de  V Art  Egyptien  d'apres  les 
Monuments,  2  vols,  folio.  Arthus  Bertrand,  1878.  As  the  plates  are  not  numbered, 
we  can  only  refer  to  them  generally. 


The  Constitution  of  Egyptian  Society, 


27 


from  another  province,  and  all  who  had  not  succumbed  to  the  hard 
and  continuous  work,  returned  to  their  own  places.  Those  who 
died  were  buried  in  hasty  graves  dug  in  the  sands  of  the  desert  by 
the  natives  of  their  own  village. 

The  massive  grandeur  of  some  of  the  Egyptian  monuments  is 
only  to  be  explained  by  this  levy  en  masse  of  every  available  pair 
of  hands.  The  kings  of  the  ancient  empire,  at  least,  were 
unable  to  dispose  of  those  prisoners  of  war  captured  in  myriads, 
in  whole  races,  by  the  Assyrian  kings,  and  apparently  employed 
by  them  in  the  construction  of  Nineveh.  Now,  it  is  impossible 
that  such  works  as  the  Pyramids  could  have  been  begun  and 
finished  in  the  course  of  a  single  reign  by  free  and  remunerated 
labour,  even  if  it  had  the  help  of  numerous  slaves.  Certain 
arrangements  in  their  design  and  the  marvellously  exact  execution 
of  the  more  important  details  of  the  masonry,  prove  that  architects 


Fig.  16. — Construction  of  a  Temjile  at  Thebes,     (From  Prisse.) 


of  great  ability  and  skilful  workmen  were,  indeed,  employed  upon 
those  gigantic  works  ;  but  the  great  bulk  of  the  task  must  have 
required  the  collective  effort  of  a  whole  population  ;  of  a  popu- 
lation devoting  themselves  night  and  day  to  complete  the  work 
when  once  begun,  like  ants  over  their  subterranean  city  or  bees 
over  their  comb. 

Even  supposing  that  history  had  been  silent  upon  this  subject, 
the  architect  could  easily  divine,  from  these  monuments  them- 
selves, how  they  had  been  constructed.  Cast  your  eyes  upon 
the  ruins  of  the  Athenian  Acropolis  ;  their  dimensions  will  seem 
to  you  small  in  the  extreme  if  you  compare  them  with  the 
buildings  of  Egypt  and  Assyria  ;  on  the  other  hand  'their 
workmanship  is  equally  careful  throughout  ;  it  is  as  exact  and 
perfect  in  the  concealed  parts  of  the   structure  as   in  those  which 


28 


A   History  of  Art  in  Ancient  Egypt. 


were  to  be  visible,  in  the  structural  details  as  in  the  ornamental 
painting  and  sculpture.  By  these  signs  you  may  recognize  at 
once,  that,  from  its  foundation  to  Its  completion,  the  whole  work 
was  in  the  hands  of  artisans  whom  long  practice  had  made  perfect 
in  their  trade,  and  that  each    single  individual  among  them  had 


1]G.  17. — Columns  in  the  Hypostyle  Hall,  Karnak 


made  it  a  point  of  honour  to  acquit  himself  worthily  of  the  task 
entrusted  to  him.  In  the  gangs  of  docile  labourers  who  succeeded 
each  other  in  the  workshops  of  Memphis  or  Thebes,  there  was,  of 
course,  a  certain   sprinkling  of  men  who  had  become  qualified  by 


The  Constitution  of  Egyptian  Society. 


29 


experience  for  the  special  work  upon  which  they  were  employed  ; 
but  the  great  majority  were  men  suddenly  taken  from  very 
different  occupations,  from  the  oar,  the  plough,  the  management 
of  cattle  ;  who  therefore  could  have  nothing  but  their  unskilled 
labour  to  bestow.  To  such  men  as  these  a  great  part  of  the  work 
had  perforce  to  be  confided,  in  order  that  it  might  be  complete  at 
the  required  time.  In  spite  of  the  strictest  supervision,  the  almost 
religious  care  in  the  placing  and  fixing  of  masonry,  v/hich  might 
be  fairly  expected  from  the  practised  members  of  a  trade  guild, 
could  not  be  ensured.  Hence  the  singular  inequalities  and 
inconsistences    which    have    been    noticed    in    most    of  the   great 


Figs.  18,  19. — Scribes  registering  tlie  yield  of  the  harvest.     From  a  tomb  at  Sakkarah.     (Boulak, 

95  inches  high.     Drawn  by  Bourgoin. ) 


Egyptian  buildings  ;  sometimes  it  is  the  foundations  which  are 
in  fault,  and,  by  their  sinking,  have  compromised  the  safety  of  the 
whole  building  ;  ^  sometimes  It  is  the  built  up  columns  of  masonry, 
which,  when   deprived  by   time  of  their  coating  of  stucco,  appear 

1  "The  foundations  of  the  great  temple  at  Abydjs,  commenced  by  Seti  I.  and 
finished  by  Rameses  IF.,  consist  of  but  a  single  course  of  generally  ill-balanced 
masonry.  Hence  the  settling  which  has  taken  place,  and  the  deep  fissure  which 
divides  the  building  in  the  direction  of  its  major  axis." — Mariette,  Voyage  dans  la 
Haute-Egypie,  p.  59.  The  same  writer  speaks  of  Karnak  in  a  similar  strain  :  "The 
Pharaonic  temples  are  built,  as  a  rule,  with  extreme  carelessness.  The  western 
pylon,  for  instance,  fell  because  it  was  hollow,  which  made  the  inclination  of  the 
walls  a  source  of  weakness  instead  of  strength." — liina-aire,  p.  179. 


o 


o 


A   History  of  Art  in  Ancient  Egypt. 


very  poor  and  mean.  The  infinite  foresight  and  self-respect,  the 
passionate  love  for  perfection  for  its  own  sake,  which  is  charac- 
teristic of  Greek   work   at   its   best  time,  is  not  here  to  be  found. 


ric.  20. — Culosbi  of  Amenophis  111.  (^talues  of  Meninon)  at  Thebes. 


But  this  defect  was  inseparable   from  the  system  under  which  the 
Egyptian  buildings  were  erected. 

The  absolute  and  dreaded  master  whose  gesture,  whose  single 
w^ord,  was  sufficient  to  depopulate  a  province  and  to  fill  quarries 


The  Constitution  of  Egyptian  Society, 


and  workshops  with  thousands  of  men,  the  sovereign  who,  in 
spite  of  his  mortahty,  was  looked  up  to  by  his  people  as  one  so 
near  akin  to  the  gods  as  to  be  hardly  distinguishable  from  them, 
the  high  priest  and  father  of  his  people,  the  king  before  whom 
all  heads  were  bent  to  the  earth  ;  filled  with  his  own  glory  and 
majesty  the  buildings  w^hich  he  caused  to  spring,  as  if  by  magic, 
from  the  earth.  His  effigy  was  everywhere.  Seated  in  the  form 
of  colossal  statues  in  front  of  the  temples,  in  bas-reliefs  upon 
pylons,  upon  the  walls  of  porticos  and  pillared  halls,  he  was 
represented  sometimes  offering  homage  to  the  gods,  sometimes 
leading  his  troops  to  battle  or  bringing  them  home  victorious. 
The  supreme  efforts  of  architect  and  sculptor  were  directed  to 
constructing  for  their  prince  a  tomb  which  should  excel  all  others 
in  magnificence  and  durability,  or  to  immortalizing  him  by  a  statue 
which  should  raise  its  head  as  much  above  its  rivals  as  the  royal 


-i»  aa^\  \a/sA 


/fVf  _  /        . — -V'J  i  HTTSfssi  ft         :^ 


'Pr*> 


Fig.  2r. — Scribe  registering  merchandize.     Sakkarah.     (9^  inches  high.     Drawn  by  Bourgoin.) 


power  surpassed  the  power  and  dignity  of  ordinary  men.  The  art 
of  Egypt  was,  in  this  sense,  a  monarchical  art  ;  and  in  so  being 
it  was  the  direct  expression  of  the  sentiments  and  ideas  of  the 
society  which  had  to  create  it  from  its  foundations. 

After  the  king  came  the  priests,  the  soldiers,  and  the  scribes  or 
royal  functionaries,  each  receiving  authority  directly  from  the  king 
and  superintending  the  execution  of  his  orders.  These  three 
groups  formed  what  we  may  call  the  upper  class  of  Egyptian 
society.  The  soil  w^s  entirely  in  their  hands.  They  possessed 
among  them  the  whole  valley  of  the  Nile,  with  the  exception  of 
the  royal  domain.  The  agriculturists  were  mere  serfs  attached 
to  the  soil.  They  cultivated,  for  a  payment  in  kind,  the  lands 
belonging  to  the  privileged  classes.  They  changed  masters 
with  the  lands  upon  which  they  lived,  which  they  were  not 
allowed  to  quit  without  the  permission  of   the  local   authorities. 


32 


A   History  of  Art  in  Ancient  Egypt. 


Their  position  did  not  greatly  differ  from  that  of  the  modern 
fellahs,  who  cultivate  the  Egyptian  soil  for  the  benefit  of  the 
effendis,  beys,  and  pachas  or  for  that  of  the  sovereign,  who  is  still 
the  greatest  landowner  in  the  country. 

The  shepherds,  the  fishermen  and  boatmen  of  the  Nile,  the 
artisans  and  shopkeepers  of  the  cities  were  in  a  similar  condition. 
They  lived  upon  their  gains  in  the  same  way  as  the  peasant  upon 
the  share  of  the  harvest  which  custom  reserved  for  his  use.  As  a 
natural  consequence  of  their  life  in  a  city  and  of  the  character  of 
their  occupations,  small  traders  and  artisans  enjoyed  more  liberty 
and  independence,  more  power  of   coming   and   going   than    the 


Fig.   22. — Boatmen.     Tomb  of  Ra-ka-pou,   5th  dyna^y.     (Boulak,    16  inches  high.     Drawn  by 

Bourgoin.) 


agriculturists,  although  legal  rights  were  the  same  in  both  cases. 
The  burden  of  forced  labour  must  have  pressed  less  heavily  upon 
the  latter  class,  and  they  must  have  had  better  opportunities  of 
escaping  from  it  altogether. 

In  consequence  of  a  mistaken  interpretation  of  historic  evidence, 
it  was  long  believed  that  the  Egyptians  had  castes,  like  the 
Hindoos.  This  notion  has  been  dispelled  by  more  careful  study 
of  their  remains.  The  vigorous  separation  of  classes  according  to 
their  social  functions,  the  enforced  heredity  of  professions,  and 
the  prohibition  of  intermarriage  between  the  different  groups, 
never  obtained  a  footing  in  Egypt.     We  often  find,  in  Egyptian 


The  Constitution  of  Egyttian  Society. 


33 


writings,  two  members  of  a  single  family  attached  one  to  the  civil 
service   and  the  other  to  the  army,  or  the  daughter  of  a  general 


Fig.    23. — Cattle  Drovers. 


P>om  the  tomb  of  Ra-ka-pou,   Sakkarah,   5th  dynasty.      (BouLik. 
Drawn  by  Bourgoin.) 


marrying    the    son    of  a  priest.      Nay,  it  often  happens  that  the 
offices  of  soldier  and  priest,  of  priest  and  civil  servant,  or  of  civil 


Fig.  24. — Bakers.     From  a  tomb.     (Boulak,  9I  inches  high.     Bourgoin.) 


servant  and  soldier,  are  united  in  the  person  of  a  single  individual. 
In  families  which  did  not  belong  to  these  aristocratic  classes  there 

.  VOL.   I.  F 


A  History  of  Art  in  Ancient  Egypt. 


was,  in  all  probability,  more  heredity  of  occupation  ;  in  the  ordi- 
nary course  the  paternal  employment  fixed  that  of  the  children, 
but  yet  there  was  nothing  approaching  to  an  absolute  rule.  The 
various  trades  were  formed  into  corporations  or  guilds,  rather  than 
castes  in  the  strict  sense  of  the  word.  From  this  it  resulted  that 
great  natural  talents,  fortunate  circumstances,  or  the  favour  of  the 
sovereign  could  raise  a  man  of  the  lowest  class  up  to  the  highest 
dignities  of  the  state.  In  the  latter  days  of  the  monarchy  we 
have  an  example  of  this  in  the  case  of  Amasis,  who,  born  among 
the  dregs  of  the  population,  finally  raised  himself  to  the  throne.^ 


Fig.  25. — Women  at  a  loom.     From  a  tomb  at  Beni-Hassan.     (Champollion,  381  bis  ) 

Such  events  were  of  frequent  occurrence  in  all  those  oriental 
monarchies  where  the  will  of  the  sovereign  was  the  supreme  and 
undisputed    law.      Even    in    our    own    days,   similar    things    have 

1  Herodotus,  ii.  172.  For  an  earlier  epoch,  see  the  history  of  a  certain  Ahmes, 
son  of  Abouna,  as  it  is  narrated  upon  his  sepulchral  inscription,  which  dates  from 
the  reign  of  Amosis,  the  founder  of  the  eighteenth  dynasty  (De  Rouge,  Mhnoire 
sur  r Inscription  d Ahmes,  Chef  des  N'autoniers,  ^Vo.  1851,  and  Brugsch,  Histoire 
dEgypte,  t.  i.  p.  80).  Starting  as  a  private  soldier  for  the  war  against  the  Shepherds, 
undertaken  for  the  re-conquest  of  Avaris,  he  was  noticed  by  the  king  for  his 
frequent  acts  of  gallantry,  and  promoted  until  he  finally  became  something  in  the 
nature  of  high  admiral. 


The  Constitution  of  Egyptian  Society. 


35 


taken  place  in  Turkey  and  Persia  to  the  surprise  of  none  but 
Europeans.  When  the  master  of  all  is  placed  so  high  above  his 
fellow  men  that  his  subjects  seem  mere  human  dust  about  his  feet, 
his  caprice  is  quite  sufficient  to  raise  the  most  insignificant  of  its 
atoms  to  a  level  with  the  most  illustrious. 

The  priests  of  the  highest  rank,  the  generals  and  officers  of  the 
army  and  the  great  civil  functionaries,  while  they  made  no  effort 
to  rival  the  splendour  of  the  royal  creations,  consecrated  steles, 
images  of  the  deity,  and  chapels,  at  their  own  expense.  It  was 
upon  their  tombs,  however,  that  most  of  their  care  was  lavished. 
These  tombs  furnish  numberless  themes  of  ereat  interest  to  the 
historian.  The  tombs  of  the  Memphite  kings  have  not  preserved 
for  us  anything  that  can  fairly  be  called  sculpture.     All  that  we 


■^ilMli— g' 


Fig.  26. — Netting  birds.     From  a  tomb.     (Boulak.     Drawn  by  Bourgoin.) 


know  of  the  style  and  methods  of  that  art  in  those  early  times  we 
owe  to  the  burial-places  which  the  members  of  the  governing 
classes  were  in  the  habit  of  preparing  during  their  lifetime  in 
the  necropolis  of  Memphis.  We  may  say  the  same  of  the  early 
centuries  of  the  Middle  Empire.  The  Egypt  of  the  great  kings 
belonging  to  the  twelfth  dynasty  has  been  preserved  for  us  upon 
the  tombs  of  Ameni  and  Num-Hotep,  the  governors  of  the  novies 
in  which  they  were  buried.  It  is  to  the  burial  chambers  at  Gizeh, 
at  Sakkarah,  at  Meidoum,  and  at  Beni-Hassan  that  we  must  otq  for 
complete  types  of  sepulchral  architecture  at  those  epochs  ;  to  the 
statues  in  the  recesses  of  their  massive  walls  and  to  the  bas-reliefs 
in  their  narrow  chambers,  we  must  turn  for  those  features  of 
early   Egyptian    civilization    which   remained    for   many    centuries 


36 


A   History  of  Art  in  Ancient  Egypt. 


without  material  change ;  by  these  monuments  we  are  enabled  to 
build  up  piece  by  piece  a  trustworthy  representation  of  the  Egyptian 
people  both  in  their  labours  and  in  their  pleasures.  Finally  it  is 
from  these  tombs  of  private  individuals  that  the  best  works  of 
Egyptian  artists  have    been  obtained,   the  works  in  which    they 


Fig.  27. — Shepherds  in  the  fields.     From  a  tomb  at  Sakkarah.     (Boulak.     8J  inches  hiijh. 

Drawn  by  Bom"goin.) 


approached   most  nearly  to  the  ideal  which  they  pursued  for  so 
many  centuries. 

Thanks  to  these  monuments  erected  at  the  expense  of  the  great 
lords  and  rich  burghers  of  Egypt,  thanks  also  to  the  climate  and  to 


^v  ^=- 


Fig.  28. — Winnowing  corn.     From  a  tomb  at  Sakkarah.     (Boulak.     Drawn  by  Bom'goin. 


the  desert  sand  which  has  preserved  them  without  material  injury, 
the  art  of  Egypt  appears  to  us  more  comprehensive  and  varied  than 
that  of  any  other  nation  of  which  we  shall  have  to  treat  ;  than 
that  of  Assyria  for  instance,  which  represents  little  but  scenes  of 
battle  and  conquest.     A  faithful  mirror  of  Egyptian  society,  it  has 


The  Constitution  of  Egyptian  Society, 


Z7 


preserved  for  us  an  exhaustive  record  of  the  never-ceasing  activity 
which  created  and  preserved  the  wealth  of  the  country  ;  it  has  not 
even  neglected  the  games  and  various  pleasures  in  which  the 
laborious  Egyptian  sought  for  his  well  earned  repose.  The  king 
indeed,  preserved  his  first  place  by  the  importance  of  the  religious 
buildings   which    he  raised,  by  the  size   of  his   tomb,   and   by   the 


Height   I2|  incher 


Height  63  inches. 


Fig.  29. — Herdsmen.     From  a  tomb  at  Sakkarah,  s;h  dynasty.     (Boulak.) 


number  and  dimensions  of  the  reproductions  of  his  features  ;  re- 
productions which  show  him  in  the  various  aspects  demanded  by 
the  complex  nature  of  the  civilization  over  which  he  presided. 
But  in  the  large  number  of  isolated  figures,  groups,  and  scenes 
which  have  come  down  to  us,  we  have  illustrations  of  all  classes 
that  helped  in  the  work  of  national  development,  from  the  plough- 
man  with  his  ox,  to  the  scribe  crouching,  cross-legged,  upon  his 


38  A  History  of  Art  in  Ancient  Egypt. 

mat,  from  the  shepherd  with  his  flock  or  the  hunter  pushing  his 
shallop  through  the  brakes  of  papyrus,  to  the  directors  of  the  great 
public  works  and  the  princes  of  the  blood  who  governed  conquered 
provinces  or  guarded  the  frontiers  of  the  country  at  the  head  of 
ever  faithful  armies. 

The  art  of  Egypt  resembled  that  of  Greece  In  being  a  complete 
and  catholic  art,  seeing  everything  and  taking  an  interest  in  every- 
thing. It  was  sensitive  to  military  glory,  and  at  the  same  time  it 
did  not  scorn  to  portray  the  peaceful  life  of  the  fields.  It  set  itself 
with  all  sincerity  to  interpret  the  monarchical  sentiment  in  its  most 
enthusiastic  and  exaggerated  form,  but  while  it  placed  kings  and 
princes  above  and  almost  apart  from  humanity,  it  did  not  forget 
the  "  humble  and  meek,"  on  the  contrary,  it  frankly  depicted  them 
in  their  professional  attitudes,  with  all  those  ineffaceable  character- 
istics, both  of  face  and  figure  which  the  practice  of  some  special 
trade  so  certainly  imparts.  Looked  at  from  this  point  of  view 
Egyptian  art  was  popular,  it  might  even  be  called  democratic, 
but  that  such  a  phrase  would  sound  curious  when  used  in  con- 
nection with  the  most  absolute  monarchy  which  the  world  has 
ever  seen. 

This  absolute  power,  however,  does  not  seem,  speaking  generally, 
to  have  been  put  in  force  in  a  hard  or  oppressive  manner  either  by 
the  king  himself  or  by  his  agents.  M.  Maspero  and  others  who, 
like  him,  live  in  intimate  communion  with  the  ancient  Egyptians, 
declare  that  they  were  by  no  means  unhappy.  They  tell  us  that 
the  confidences  whispered  to  them  in  the  pictured  tomb-houses  of 
Sakkarah  and  Memphis  complain  of  no  misery,  from  the  time  of 
Mena  to  that  of  Psemethek,  except  during  a  few  violent  reigns 
and  a  few  moments  of  national  crisis.  The  country  suffered  only 
on  those  comparatively  rare  occasions  when  the  sceptre  passed  into 
the  hands  of  an  incapable  master  or  into  those  of  some  insatiable 
warrior  who  thought  only  of  satisfying  his  own  ambition,  and 
sacrificed  to  the  day  the  resources  of  the  future.  Egypt,  with  her 
river,  her  teeming  soil  and  her  splendid  climate,  found  life  easy  as 
long  as  she  enjoyed  an  easy  and  capable  administration.  She  then 
gave  to  her  princes  almost  without  an  effort  all  they  could  desire  or 
demand. 

It  was  one  of  the  fundamental  principles  of  Egyptian  morality 
that  those  who  were  powerful  should  treat  the  poor  and  feeble  with 
kindness  and  consideration.      Their  sepulchral  inscriptions  tell  us 


The  Constitution  of  Egyptian  Society. 


39 


that  their  kings  and  princes  of  the  blood,  their  feudal  lords  and 
functionaries  of  every  grade,  made  it  a  point  of  honour  to  observe 
this  rule.  They  were  not  content  with  strict  justice,  they  practised 
a  bountiful  charity  which  reminds  us  of  that  which  is  the  chief 
beauty  of  the  Christian's  morality.  The  "  Book  of  the  Dead" — 
that  passport  for  Egyptians  into  the  other  world  which  is  found 
upon  every  mummy — gives  us  the  most  simple,  and  at  the  same 
time  the  most  complete  description  of  this  virtue.  "  I  have  given 
bread  to  the  hungry,  I  have  given  water  to  the  thirsty,  I  have 
clothed  the  naked  ...  I  have  not  calumniated  the  slave  in  the 
ears  of  his  master."  The  lengthy  panegyrics  of  which  some 
epitaphs  consist,  are,  in  reality  no  more  than  amplifications  of  this 


iU^ 


-c 


Fig.  30. — Fi-orn  the  tomb  of  Menofre,  at  Sakkarah.     (Champollion,  pi.  408.) 


theme.  "As  for  me,  I  have  been  the  staff  of  the  old  man,  the 
nurse  of  the  infant,  the  help  of  the  distressed,  a  warm  shelter  for 
all  who  were  cold  in  the  Thebaid,  the  bread  and  sustenance  of  the 
down-trodden,  of  whom  there  is  no  lack  in  Middle  Egypt,  and  their 
protector  against  the  barbarians."^  The  prince  Entef  relates  that 
he  has  "  arrested  the  arm  of  the  violent,  used  brute  force  to  those 
who  used  brute  force,  showed  hauteur  to  the  haughty,  and  lowered 
the  shoulders  of  those  who  raised  them  up,"  that  he  himself  on  the 
other  hand,  "  was  a  man  in  a  thousand,  wise,  learned,  and  of  a  sound 
and  truthful  judgment,  knowing  the  fool  from  the  wise  man,  paying 

^  Louvre,  c.  i.     Cf.  Maspero,  ««  G&uverneur  de  Thebes  au  temps  de  la  douziemc 
dynastie. 


40  A   History  of  Art  in  Ancient  Egypt. 

attention  to  the  skilful  and  turning  his  back  upon  the  ignorant, 
.  .  .  the  father  of  the  miserable  and  the  mother  of  the  motherless, 
the  terror  of  the  cruel,  the  protector  of  the  disinherited,  the  defender 
of  those  whose  goods  were  coveted  by  men  stronger  than  them- 
selves, the  husband  of  the  widow,  the  asylum  of  the  orphan."  ^ 

Amoni,  hereditary  prince  of  the  nome  of  Meh,  talks  in  the  same 
fashion.  "  I  have  caused  sorrow  to  no  youth  under  age,  I  have 
despoiled  no  widow,  nor  have  I  repelled  any  labourer,  I  have 
imprisoned  no  shepherd,  I  have  never  taken  for  the  labour  gangs 
the  serfs  of  him  who  had  but  five,  there  have  been  no  paupers,  nor 
has  any  man  or  woman  starved  in  my  time  ;  for,  although  there 
have  been  years  of  scarcity,  I  have  caused  all  the  tillable  land  in 
Meh  to  be  tilled,  from  the  northern  frontier  to  that  of  the  south, 
and  have  made  such  arrangements  and  such  provision  for  the 
people  that  there  has  been  no  famine  among  them  ;  I  have  given 
to  the  widow  and  to  the  married  woman  alike,  and  I  have  never 
made  any  distinction  between  the  great  and  the  small  in  my 
gifts."  = 

Doubtless  these  laudatory  self-descriptions  may  be  exaggerated 
in  some  respects  ;  hyperbole  has  ever  been  a  favourite  figure  with 
the  composers  of  epitaphs,  and  those  of  Egypt  formed  no 
exception  to  the  rule.  As  M.  Maspero  remarks  in  connection 
with  this  question,  "  The  man  as  he  is,  often  differs  very  greatly 
from  the  man  as  he  thinks  he  is."  But  we  may  safely  say  that  the 
Egyptian  realized  some  portion  of  the  ideal  which  he  set  before 
himself.  If  only  to  obtain  admiration  and  esteem,  he  would 
practice,  to  a  certain  extent,  the  virtues  of  w^hich  he  boasted. 
Many  signs  combine  to  tell  us  that  the  Egyptians  of  all  classes 
possessed  a  large  fund  of  tenderness  and  good-will.  The  master 
was  often  both  clement  and  charitable  ;  the  peasant,  the  servant, 
and  the  slave,  were  patient  and  cheerful,  and  that  in  spite  of 
the  fatiofue  of  labours  which  could  never  enrich  them.  In  a 
country  so  favoured  by  nature,  men  had  so  few  wants  that  they 
had   no  experience   of  all   that  is   implied   by  that   doleful    word 

^  Quoted  by  Maspero,  Conference  sur  I ' Histoii-e  des  Ames  dans  V Egypte  ancienne, 
dapres  ks  Monuments  dii  Musee  du  Louvre  {Association  scientifique  de  Frajice,  Bulletin 
hebdofnadaire,  No.  594;   23  Alars^  1879)- 

2  Translated  by  Maspero  {la  Grande  Inscription  de  Beni-Hassan,  in  the  Recueil 
de  Travaux  relatifs  a  la  Pliilologie  it  a  l Arch'eologie  cgyptienne  et  assyrienne  (t.  i. 
PP-  173-174)- 


The  Constitution  of  Egyptian  Society.  41 

poverty,  with  us.  The  pure  skies  and  brilHant  sunshine,  the  deep 
draughts  of  Nile  water,  and  the  moments  of  repose  under  the 
shadows  of  the  sycamores,  the  freshness  of  the  evening  bath,  the 
starry  night  with  its  reinvigorating  breezes,  were  all  enjoyments 
which  the  poorest  could  share. 

We  need  feel  no  surprise  therefore  at  the  vivacity  with  which 
one  of  the  most  learned  of  the  historians  of  Egypt,  Brugsch-Bey, 
protests  against  the  common  misconception  of  the  Egyptians  "  as 
a  race  grave,  serious,  morose,  exclusive,  religious,  thinking  much 
of  the  next  world,   and  little  of  this  ;  living,  in   a  word,  like  the 
Trappists  of  former  days.     Are  we  to  believe,"  he  cries,  "  that  this 
majestic  river  and  the  fertile  soil  through  which  it  flows,  this  azure 
sky  with  its  unclouded  sun,  produced  a  nation  of  living  mummies, 
a  race  of  solemn  philosophers  who  looked  upon  life  in  this  world 
as  a  burden  to  be  shuffled  off  as  quickly  as  possible  ?     Travel  over 
Egypt  ;  examine  the  scenes  painted  and  sculptured  upon  the  walls 
of  sepulchral  chambers  ;  read  the  inscriptions  carved  upon  stone 
or  traced  in   ink   upon   the  rolls   of   papyrus,    and   you   will    find 
yourself  compelled  to  modify  the  false  notions  you  have  imbibed  as 
to  the  Egyptian  philosophers.      Nothing  could  be  more  cheerful, 
more  amusing  or  more  frank,  than  the  social  life  of  this  pleasure- 
loving  people.     Far  from  wishing  to  die,  they  prayed  to  the  gods 
for  a  long  life  and  a  happy  old  age  ;  they  prayed  that,  "  if  possible, 
they  might  live   to    the    perfect   age  of  one   hundred  and   ten." 
They  were  addicted  to  all  kinds  of  pleasures.     They  drank,  they 
sang,  they  danced,  they  were  fond  of  excursions  into  the  country, 
where  the  sports  of  hunting  and  fishing  were  specially  reserved  for 
the  upper  class.     As  a  natural  effect  of  this  desire  for  enjoyment, 
gay  conversation  and  pleasantry  which  was  sometimes  rather  free, 
jokes  and  what  we  should  call  chaff,  were  much  in  vogue  :  even 
their  tombs  were  not  sacred  from  their  desire  for  a  jest."  ^ 

The  worst  government,  the  sternest  oppression,  could  never  ex- 
tinguish this  natural  gaiety  ;  it  was  too  intimately  connected  with 
the  climate  and  the  natural  conditions  of  the  country,  conditions 
which  had  never  changed  since  the  days  of  Menes.  Never  were 
the  Egyptians  more  roughly  treated  than  under  Mehemet  Ali  and 
the  late  viceroy  ;  their  condition  was  compared,  with  justice,  to 
that  of  the  negroes  in  Carolina  and  Virginia,  who,  before  the 
American  civil  war,  laboured  under  the  whips  of  their  drivers,  and 

1   Brucsch-Bky,  Histoire  (f  Egypte,\t\i.  14.  15. 
VOL.    I.  G 


A- 


A  History  of  Art  in  Ancient  Egypt. 


enjoyed  no  more  of  the  fruits  of  their  own  labour  than  what  was 
barely  sufficient  to  keep  life  in  their  bodies.  Torn  from  their 
homes  and  kept  by  force  in  the  public  works,  the  fellahs  died  in 
thousands  ;  those  who  remained  in  the  fields  had  to  pay  the  taxes 
one  or  two  years  in  advance  ;  they  were  never  out  of  debt, 
nominally,  to  the  public  treasury,  and  the  rattan  of  the  collector 
extorted  from  them  such  savings  as  they  might  make  during  years 
of  plenty,  up  to  the  last  coin.  But  still  laughter  did  not  cease  in 
Egypt !  Look,  for  instance,  at  the  children  in  the  streets  of  Cairo 
who  let  out  mounts  to  sight-seeing  Europeans.  Let  the  tourist 
trot  or  gallop  as  he  will,  when  he  stops  he  finds  his  donkey-boy  by 
his  side,  full  of  spirits  and  good  humour ;  and  yet  perhaps  while 


,V'-^ki'^-^^"=^'^^.^_,.^^  .,^/'i^~~^ 


Fig.  31. — Water  Tournament,  from  a  tomb  at  Khoum-el-Ahmar.    (From  Prisse.) 


running  behind  his  "  fare  "  he  has  been  making  his  midday  meal 
upon  a  few  grains  of  maize  tied  up  in  a  corner  of  his  shirt. 

In  1862  I  returned  from  Asia  Minor  in  company  with  M. 
Edmond  Guillaume,  the  architect,  and  M.  Jules  Delbet,  the  doctor, 
of  our  expedition  to  Ancyra.  We  took  the  longest  way  home,  by 
Syria  and  Egypt.  At  Cairo,  Mariette,  after  having  shown  us  the 
museum  at  Boulak,  wished  to  introduce  us  to  his  own  "  Serapeum." 
He  took  us  for  a  night  to  his  house  in  the  desert,  and  showed  us 
the  galleries  of  the  tomb  of  Apis  by  torchlight.  We  passed  the 
next  afternoon  in  inspecting  those  excavations  in  the  necropolis  of 
Sakkarah  which  have  led  to  the  recovery  of  so  many  wonders  of 
Egyptian  art.  The  works  were  carried  on  by  the  labour  of  four 
hundred  children  and  youths,  summoned  by  the  corvde  for  fifteen 


The  Constitution  of  Egyptian  Society.  43 

days  at  a  time  from  some  district,  I  forget  which,  of  Middle  Egypt. 
At  sunset  these  young  labourers  quitted  their  work  and  seated 
themselves  in  groups,  according  to  their  native  villages,  upon  the 
still  warm  sand.  Each  drew  from  a  little  sack,  containing  his 
provision  for  two  or  three  weeks,  a  dry  cake  ;  those  whose  parents 
were  comfortably  off  had  also,  perhaps,  a  leek  or  a  raw  onion. 
But  even  for  such  gourmands  as  those,  the  repast  was  not  a  long 
one.  Supper  over,  they  chattered  for  a  time,  and  then  went 
to  rest ;  the  bigger  and  stronger  among  them  took  possession  of 
some  abandoned  caves,  the  others  stretched  themselves  upon  the 
bare  earth ;    but,  before  going  to  sleep  they  sang ;    they  formed 


Fig.  32. — Mariette's  House. 

themselves   into   two    choirs  who    alternated    and    answered    one 
another,  and  this  they  kept  up  to  an  advanced  hour  of  the  night. 

I  shall  never  forget  the  charm  of  that  night  in  the  desert,  nor 
the  weird  aspect  of  the  moonlight  upon  the  sea  of  sand.  Were  it 
not  that  no  star  was  reflected  upon  its  surface,  and  that  no  ray 
scintillated  as  it  does  even  on  the  calmest  sea,  we  might  have 
thought  ourselves  in  mid  ocean.  Sleep  came  to  me  reluctantly. 
While  I  listened  to  the  alternate  rise  and  fall  of  the  chorus 
outside,  I  reflected  upon  how  little  those  children  required ;  upon 
the  slender  wants  of  their  fathers  and  mothers,  who,  like  them, 
sink  into  their  nightly  sleep  with  a  song  upon  their  lips.  I 
compared  this  easy  happiness   with  the  restless  and  complicated 


44  A  History  of  Art  in  Ancient  Egypt. 

existence  which  we  should  find,  at  the  end  of  a  few  days,  in  the 
ambitious  cities  of  the  West,  and  I  regretted  that  our  year  of 
travel,  our  twelve  months  of  unrestrained  life  in  the  desert  or  the 
forest,  had  come  to  an  end. 


^  5. — The  Egyptian   Religion  and  its  Influence  upon  the 

Plastic  Arts. 

We  have  still  to  notice  the  profoundly  religious  character  of 
Egyptian  art.  "  The  first  thing  that  excites  our  surprise,  when 
we  examine  the  reproductions  of  Egyptian  monuments  which  have 
been  published  in  our  day,  is  the  extraordinary  number  of  scenes 
of  sacrifice  and  worship  which  have  come  down  to  us.  In  the 
collection  of  plates  which  we  owe  to  contemporary  archaeologists, 
we  can  hardly  find  one  which  does  not  contain  the  figure  of  some 
deity,  receiving  with  impassive  countenance  the  prayers  or 
offerings  of  a  prostrate  king  or  priest.  One  would  say  that  a 
country  with  so  many  sacred  pictures  and  sculptures,  must  have 
been  inhabited  by  gods,  and  by  just  enough  men  for  the  service 
of  their  temples.^  The  Egyptians  were  a  devout  people.  Either 
by  natural  tendency  or  by  force  of  education,  they  saw  God 
pervading  the  whole  of  their  universe  ;  they  lived  in  Him  and  for 
Him.  Their  imaginations  were  full  of  His  greatness,  their  words 
of  His  praise,  and  their  literature  was  in  great  part  inspired  by 
gratitude  for  the  benefits  which  He  showered  upon  them.  Most 
of  their  manuscripts  which  have  come  down  to  us  treat  of 
religious  matters,  and  even  in  those  which  are  ostensibly 
concerned  only  with  profane  subjects,  mythological  names  and 
allusions  occur  on  every  page,  almost  at  every  line."  ^ 

An  examination  into  the  primitive  religious  beliefs  of  the 
Egyptians    is    full    of  difficulty.      In    discovering    new   papyri,  in 

'  The  saying  of  one  of  the  characters  of  Petronius  might  be  appHed  to  Egypt : 
"  This  country  is  so  thickly  peopled  with  divinities  that  it  is  easier  to  find  a  god  than 
a  man."  The  place  held  by  religious  observances  in  the  daily  life  of  Egypt  is  clearly 
indicated  by  Herodotus  (ii.  37):  "The  Egyptians,"  he  says,  "are  very  religious; 
they  surpass  all  other  nations  in  the  adoration  with  which  they  regard  their  deities." 

2  Maspero,  Hisioire  ancienne,  pp.  26,  27. 


The  Egyptian  Religion  and  the  Plastic  Arts. 


45 


determining  the  signification  of  signs  which   have  been  puzzHng 
egyptologists,   the  inquirer  will  undoubtedly  do   good  work,   and 


s : =■ 


"N 


Fig.  33. — Anienhotep  or  Amenophis  III.   presented  by  Phre  to  Amen-Ra  ;  Thebes. 

(Champollion,  pi.  344.) 


will  establish  facts  which  are  sure  not  to  lack  interest  and  even 
importance  ;  but  even  when  documents  abound  and  when   every 


46  A  History  of  Art  in  Ancient  Egypt. 

separate  word  they  contain  is  understood,  even  then  it  is  very- 
difficult  to  penetrate  to  the  root  of  their  meaning.  A  glimpse  will 
be  caught  of  it,  I  admit,  by  one  of  those  efforts  of  inductive 
divination  which  distinguish  modern  research  ;  but  even  then  it  will 
remain  to  explain  the  primitive  and  only  half-understood  notions 
of  five  or  six  thousand  years  ago  in  the  philosophical  vocabularies 
of  to-day.  It  is  here  that  the  most  difficult  and  irksome  part  of 
the  task  begins.  We  who  represent  the  old  age,  or,  perhaps,  the 
prime,  of  humanity,  think  of  these  matters  and  speak  of  them  as 
abstractions,  while  the  Egyptians,  who  were  children  compared 
to  us,  thought  of  them  under  concrete  forms.  Their  very  ideals 
were  material,  more  or  less  vague  and  refined  perhaps,  but  still 
material.  Their  only  conception  of  a  deity  was  of  a  figure  larger, 
more  vigorous  and  more  beautiful  than  mortals  ;  the  powers  and 
attributes  with  which  it  was  endowed  were  all  physical.  If  we 
attempt  to  express  their  conceptions  in  abstract  terms,  we  falsify 
their  meaning.  We  cannot  avoid  altering  it  to  a  certain  extent, 
for  exact  equivalents  are  not  to  be  found,  and,  in  spite  of  all 
precautions,  we  give  to  the  confused  and  childish  ideas  of  ancient 
religion,  a  precision  which  is  entirely  modern. 

If,  under  these  reserves,  we  study  the  Egyptian  theology  in  its 
most  learned  and  refined  form — namely,  that  which  it  attained 
during  the  eighteenth  and  nineteenth  dynasties — we  shall  dimly 
perceive  that  it  implies  a  belief  in  the  unity  of  the  First  Cause  of 
all  life.  But  this  belief  is  obscured  behind  the  numerous  gods 
who  are,  in  fact,  emanations  from  its  substance  and  manifestations 
of  its  indefatigable  activity.  It  is  in  the  person  of  these  gods  that 
the  divine  essence  takes  form.  Each  of  them  has  his  own  name, 
his  own  figure,  and  his  own  special  share  in  the  management  of. 
the  universe  ;  each  of  them  presides  over  the  production  of  some 
particular  order  of  phenomena  and  insures  their  regularity. 
These  gods  are  related  to  each  other  as  fathers,  mothers,  and 
sons.  They  thus  form  a  vast  hierarchy  of  beings,  superior  to 
man,  and  each  enjoying  a  dignity  corresponding  to  his  rank  in 
the  series.  There  is,  so  to  speak,  most  of  divinity  in  those  who 
are  nearest  to  the  "one  God  in  heaven  or  earth  who  was  not 
begotten."  These  deities  are  divided  into  groups  of  three,  each 
group  constituting  a  family,  like  those  of  earth,  consisting  of 
father,  mother,  and  son.  Thus  from  triad  to  triad,  the  concealed 
god  develops  his  sovereign  powers  to  all  eternity,  or,  to  use  an 


The  Egyptian  Religion  and  the  Plastic  Arts.  47 

expression  dear  to  the  religious  schools  of  ancient  Egypt,  "  he 
creates  his  own  members,  which  are  themselves  gods."^ 

How  should  the  science  of  comparative  religion  class  this 
form  of  faith  ?  Should  it  be  called  polytheism  or  pantheism  ? 
The  answer  is,  perhaps,  not  of  great  importance,  and  this  is  hardly 
the  place  for  its  discussion.  It  is  certain  that,  practically,  the 
Egyptians  were  polytheists.  The  Egyptian  priests,  indeed,  had, 
by  dint  of  long  reflection,  arrived  at  the  comprehension,  or  at 
least  at  the  contemplation,  of  that  First  Cause  which  had  started 
the  river  of  life — that  inexhaustible  stream  of  which  the  Nile  with 
its  fertilising  waves  was  the  concrete  image — in  its  long  journey 
across  time  and  space.  But  the  devotion  of  the  people  themselves 
never  succeeded  in  mounting  above  the  minor  divinities,  above 
those  intermediaries  in  whom  the  divine  principle  and  attributes 
became  personified  and  put  on  the  tangibility  of  body  necessary 
to  make  them  intelligible  to  childish  understandings.  So,  too, 
was  it  with  artists,  and  for  still  more  powerful  reasons ;  as 
by  forms  only  could  they  express  the  ideas  which  they  had 
conceived.  Even  in  those  religions  which  are  most  clearly  and 
openly  monotheistic  and  spiritual,  such  as  Christianity,  art  has 
done  something  of  the  same  kind.  Aided  in  secret  by  one  of  the 
most  powerful  instincts  of  the  human  soul,  it  has  succeeded,  in 
spite  of  all  resistance  and  protestation,  in  giving  plastic  expression 
to  those  parts  of  our  belief  which  seem  least  fitted  for  such 
treatment ;  and  it  has  caused  those  methods  of  expression  to  be 
so  accepted  by  us  that  we  see  nothing  unnatural  in  the  represent- 
ation under  the  features  of  an  old  man,  of  the  first  Person  of  the 
Trinity, — of  that  Jehovah  who,  in  the  Old  Testament,  proscribed  all 
graven  images  with  such  impartial  rigour  ;  who,  in  the  Evangel, 
described  Himself  as  "  the  Truth  and  the  Life." 

In  Egypt,  both  sculptors  and  painters  could  multiply  their 
images  to  infinity  without  coming  into  collision  with  dogma, 
without  provoking  the  regrets  or  censures  of  its  most  severe 
interpreters.  Doctrine  did  not  condemn  these  personifications, 
even  when  it  had  been  refined  and  elaborated  by  the  speculative 

^  This  formula  frequently  occurs  in  the  texts.  To  cite  but  one  occasion,  we  find 
upon  a  Theban  invocation  to  Amen,  translated  by  P.'  Pierret  {Recueil  de  Tmvaux 
relatifs  a  la  Philologie  et  a  r Archcologie  egyptienne  et  assyrieniie,  t.  i.  p.  70),  at  the 
third  line  of  the  inscription :  "  Sculptor,  thou  modelest  thine  own  members ;  thou 
begettest  them,  not  having  thyself  been  begotten." 


48  A  History  of  Art  in  Ancient  Egypt. 


theologians  of  Thebes  and  Hehopohs.  In  the  interior  of  the 
temples,  there  was  a  small  class  of  mystics  who  took  pleasure  in 
contemplating  "  the  '  One '  who  exists  by  his  own  essential  power, 
the  only  being  who  substantially  exists."  Even  then  men  tried, 
as  they  have  often  done  since,  to  define  the  undefinable,  to  grasp 
the  incomprehensible,  to  perceive  the  supreme  "  I  am  "  through 
the  shifting  and  transparent  veil  of  natural  phenomena.  But 
those  refined  metaphysics  never  touched  and  influenced  the  crowd, 
and  never  will.  The  deity,  in  order  to  be  perceived  by  them 
and  to  touch  their  feelings,  must  have  his  unity  broken  ;  he  must, 
if  the  expression  be  admissible,  be  cut  up  into  morsels  for  them. 

By  a  process  of  abstraction  which  is  as  old  as  religion  itself,  the 
human  intelligence  is  led  to  consider  separately  each  of  the  qualities 
of  existence,  each  of  the  forces  which  it  perceives  to  be  at  work 
either  within  man  himself  or  in  the  exterior  world.  At  first  it 
thinks  those  forces  and  qualities  are  distributed  impartially  to  all 
creation.  It  confounds  existence  with  life.  Hence  the  reign  of 
fetishism,  when  man  believes,  as  young  children  do,  that  thought, 
passion,  and  volition  like  his  own,  are  to  be  found  in  everything  he 
meets.  His  own  imaofe  seems  to  him  reflected  as  in  a  mirror  with 
a  thousand  converging  facets,  and  he  is  unable  to  distinguish  the 
real  condition  of  things  outside  it. 

Certain  celestial  and  terrestrial  bodies  make  a  particularly  strong 
impression  upon  his  mind  by  their  size,  their  beauty,  by  their  evil 
or  beneficial  effects  upon  himself.  They  fill  him  with  more  than 
the  average  gratitude,  admiration,  or  terror.  Driven  by  the  illusion 
which  possesses  him,  he  places  the  origin  of  those  qualities  which 
seem  to  him  the  highest  and  most  important,  in  the  bodies  which 
have  made  so  deep  an  impression  upon  his  senses ;  to  them  he 
attributes  the  friendly  or  hostile  influences  which  alternately  excite 
his  desire  and  his  fear.  According  to  circumstances  a  fetish  might 
be  a  mountain,  a  rock  or  a  river,  a  plant  or  an  animal.  It  might 
be  those  heavenly  bodies  which  exercised  much  more  influence  over 
the  life  of  primitive  man  than  they  do  over  us  ;  it  might  be  the 
moon  and  stars,  which  tempered  the  darkness  of  the  night  and 
diminished  its  terrors ;  it  miorht  be  the  cloud,  from  whose  bosom 
came  rain  and  thunder  ;  above  all,  it  might  be  the  sun  which  re- 
turned every  morning  to  light  and  warm  the  world.  Difierences 
of  climate  and  race  had  their  modifying  effect,  but  everywhere  one 
common   characteristic   is  to  be  found.      It   was   always  to  some 


The  Egyptian   Religion  and  the  Plastic  Arts. 


49 


material  and  visible  object  that  the  human  intellect  referred  those 
forces  and  qualities  which  it  drew  from  its  own  consciousness  ; 
forces  which,  when  thus  united  with  something  tangible,  constituted 
the  first  types  of  those  divine  beings  whom  mankind  have  so  long 
adored,  to  whom  they  have  turned  for  ages  in  their  hope  and  fear. 

As  the  years  passed  away,  man  advanced  beyond  his  primitive 
conceptions.  He  did  not  entirely  renounce  them — we  may  indeed  see 
reminiscences  of  them  all  around  us — but  he  super  imposed  others 
upon  them  which  were  m.ore  complex.  His  powers  of  observa- 
tion, still  imperfect  though  they  were,  began  to  insinuate  into  his 
mind  a  disbelief  in  the  activity  of  inanimate  matter,  and  those 
objects  which  were  nearest  to  him,  which  he  could  touch  with  his 
hand,  were  the  first  victims  of  his  disenchantment.  Thus  began 
a  long  course  of  intellectual  development,  the  result  of  which  we 
know,  although  the  various  stages  of  its  progress  are  difficult  to 
follow  at  this  distance  of  time.  It  appears  certain,  however,  that 
star  worship  formed  the  transition  ho-iwic^n  fetishism  diwd polytheism. 
Men  no  longer  attributed  vital  forces  and  pre-eminent  qualities 
generally  to  bodies  with  which  they  themselves  were  in  immediate 
contact,  to  stones  and  trees  ;  but  they  found  no  difficulty  in  con- 
tinuing to  assign  them  to  those  great  luminaries  whose  distance 
and  beauty  placed  them,  so  to  speak  outside  the  material  world. 
As  they  gradually  deprived  inanimate  matter  of  the  properties 
with  which  they  had  once  gifted  it,  they  sought  for  new  objects  to 
which  they  might  attach  those  properties.  These  they  found  in 
the  stars  which  shone  in  the  firmament  century  after  century, 
and  knew  neither  old  age  nor  death  ;  and  especially  in  the  most 
brilliant,  the  most  beneficent,  and  the  most  necessary  of  them  all, 
in  that  sun  whose  coming  they  awaited  every  morning  with  an 
impatience  which  must  once  have  been  mixed  with  a  certain 
amount  of  anxiety. 

The  attributes  which  awakened  intelligence  had  taken  away 
from  the  inanimate  objects  of  the  world  could  not  be  left  floating 
in  space.  They  became  gradually  and  imperceptibly  grouped  in 
men's  minds  around  the  great  luminary  of  day,  and  a  bond  of  union 
was  found  for  the  difi"erent  members  of  the  group  by  endowing 
the  sun  with  a  personality  modelled  upon  that  of  man.  This 
operation  was  favoured  by  the  constitution  of  contemporary 
language,  by  its  idioms  made  up  entirely  of  those  images  and 
metaphors   which,  by  their  frank  audacity,  surprise  and  charm  us 

VOL,    I.  li 


;o  A   History  of  Art  in  Ancient  Egypt. 


in  the  works  of  the  early  poets.  It  commenced  with  the  first 
awakening  of  thought,  when  man  endowed  all  Yisible  nature  with 
the  bounding  life  which  he  felt  in  his  own  veins.  No  effort  of 
intelligence  was  required  for  its  commencement  or  for  its  pro- 
secution. The  sun  became  a  young  hero  advancing,  full  of  pride 
and  vigour,  upon  the  path  prepared  for  him  by  Aurora  ;  a  hero  who 
pursued  his  daily  path  in  spite  of  all  obstacle  or  hindrance,  who, 
when  evening  came,  went  to  his  rest  amid  all  the  glories  of  an 
eastern  sunset,  and  amid  the  confidence  of  all  that  after  his  hours  of 
sleep  he  would  take  up  his  eternal  task  with  renewed  vigour.  He 
was  an  invincible  warrior.  He  was  sometimes  an  angry  master, 
whose  glance  killed  and  devoured.  He  was  above  all  the  untiring 
benefactor  of  mankind,  the  nurse  and  father  of  all  life.  Whether 
as  Indra  or  as  Amen-Ra,  it  was  the  same  cry  that  went  up  to  him 
from  Egypt  and  Hindostan  ;  the  prayers  which  we  find  in  the 
Vedas  and  in  the  papyri,  breathe  the  same  sentiments  and  were 
addressed  to  the  same  god.^ 

This  solar  god  and  the  divinities  who  resemble  him,  form  the 
transition  from  the  simple  fetish  to  complete  deities,  to  those  gods 
who  played  such  an  important  part  in  the  Egyptian  religion,  and 
attained  to  their  highest  and  most  complete  development  in  the 
Hellenic  mythology.  In  some  respects,  the  luminous  globe  of  the 
sun  with  its  compulsory  course,  belonged  to  the  same  category  as 
the  material  objects  which  received  the  first  worship  of  humanity. 
But  its  brilliance,  its  tranquil  and  majestic  movement,  and  the 
distance  which  conceals  its  real  substance  from  the  eye  of  man, 
allowed  his  imagination  to  endow  it  with  the  purest  and  noblest 
characteristics  which  the  finest  examples  of  humanity  could  show ; 
while  the  phenomena  which  depend  upon  its  action  are  so  numerous 
that  there  was  no  hesitation  in  assigning  to  it  qualities  and  energies 
of  the  most  various  kinds. 

This  type  when  once  established  was  used  for  the  creation  of 
other  deities,  which  were  all,  so  to  speak,  cast  in  the  same  mould. 
As  the  intellect  became  more  capable  of  abstraction  and  analysis, 
the  personality  and  moral  individuality  of  these  gods  gradually 
threw  off  its  astral  or  physical  characteristics,  although  it  never 
lost  all  trace  of  their  existence.  It  resulted  that,  both  in  Egypt 
and  in   Greece,   there   were  deities   who   were   mere   entities,  the 

^  See   the   fine  hymns   quoted  and  translated    by  ^I.    Maspero   in   his    Histoire 
anciennc^  \)\).  30-37. 


The  Egyptian   Religion   and  the  Plastic  Arts. 


51 


simple  embodiment  of  some 
power,  some  quality,  or 
some  virtue.  It  requires 
all  the  subtle  finesse  of 
modern  criticism  to  seek 
out  and  distinoruish  the 
obscure  roots  which  attach 
these  divinities  to  the  natu- 
ralistic beliefs  of  earlier 
ages.  Sometimes  absolute 
certainty  is  not  to  be  at- 
tained, but  we  may  safely 
say  that  a  race  is  poly- 
theistic when  we  find  these 
abstract  deities  amone  their 
gods,  such  deities  as  the 
Ptah,  Amen,  and  Osiris  of 
the  Egyptians,  and  the 
Apollo  and  Athene  of  the 
Greeks.^ 

We  may,  then,  define 
polytheism  as  the  partition 
of  the  highest  attributes  of 
life  between  a  limited 
number  of  agents.  The 
imagination  of  man  could 
not  give  these  agents  life 
without  at  the  same  time  en- 
dowing them  with  essential 
natural  characteristics  and 
with  the  human  form,  but, 
nevertheless,  it  wished  to 
regard  them  as  stronger, 
more  beautiful  and  less 
ephemeral  than  man.      The 

^  Several  of  the  bronzes  which  we  reproduce  may  belong  to  the  Ptolemaic  epoch  ; 
but  they  are  repetitions  of  types  and  attributes  which  had  been  fixed  for  many 
centuries  by  tradition.  It  is  in  this  Capacity  chiefly  that  we  reproduce  them,  as 
examples  of  those  forms  which  seemed  to  the  Egyptian  imagination  to  offer  the  most 
satisfactory  emblems  of  their  gods. 


Fig.  34. — Amen  or  Ammon,  from  a  bronze  in  the 
Louvre.     Heiijht  22 "04  inches. 


52 


A  History  of  Art  in  Ancient  Egypt. 


system  had  said  its  last  word  and  was  complete,  when  it 
had  succeeded  in  embodying  in  some  divine  personality  each 
of  those  forces  whose  combined  energy  produces  movement  in 
the  world-  or  guarantees  its  duration. 

When  religious  evolution  follows 
its  normal  course,  the  work  of 
reflection  goes  on,  and  in  course 
of  time  makes  new  discoveries.  It 
refers,  by  efforts  of  conjecture,  all 
phenomena  to  a  certain  number  of 
causes,  which  it  calls  gods.  It  next 
perceives  that  these  causes,  or  gods, 
are  of  unequal  importance,  and  so 
it  constitutes  them  into  a  hierarchy. 
Still  later  it  begins  to  comprehend 
that  many  of  these  causes  are  but 
different  names  for  one  thine,  that 
they  form  but  one  force,  the  appli- 
cation of  a  single  law.  Thus  by 
reduction  and  simplification,  by 
logic  and  analysis,  is  it  carried  on 
to  recognize  and  proclaim  the  unity 
of  all  cause.  And  thus  monotheism 
succeeds  to  polytheism. 

In  Egypt,  religious  speculation 
arrived  on  the  threshold  of  this 
doctrine.  Its  depths  were  dimly 
perceived,  and  it  was  even  taught 
by  the  select  class  of  priests  who 
were  the  philosophers  of  those 
days  ;  but  the  monotheistic  con- 
ception never  penetrated  into  the 
minds  of  the  great  mass  of  the 
people.^  Moreover,  by  the  very 
method  in  which  Egyptian  myth- 
ology described  it,  it  was  easily  adapted  to  the  national  poly- 
theism, or  even   to   fetish   worship.     The  theory   of  emanations 


iSn.< 


Fig.   35, — Ptali,   from  a  bronze  in  the 
Louvre.     Actual  size. 


1  In  his  work  entitled  Des  deux  Yeux  du  Disqiie  solaire,  M.  Grkbaut  seems  to  have 
very  clearly  indicated  how  far  we  are  justified  in  saying  that  Egyptian  religious 
speculation  at  times  approached  monotheism  {Recueil  de   Travaux,  etc.,  t.  i.  p.  120). 


The  Egyptian   Religion  and  the  Plastic  Arts. 


53 


reconciled  everything.  The  different  gods  were  but  the  different 
qualities  of  the  eternal  substance,  the  various  manifestations  of 
one  creative  force.  These  qualities 
and  energies  were  revealed  by 
being  imported  into  the  world  of 
form.  They  took  finite  shape  and 
were  made  comprehensible  to  the 
intellect  of  man  by  their  mysterious 
birth  and  generation.  It  was 
necessary,  if  the  existence  of  the 
gods  were  to  be  brought  home  to 
mankind,  that  each  of  them  should 
have  a  form  and  a  domicile.  I  masfi- 
nation  therefore  did  well  in  com- 
mencing to  distinguish  and  define 
the  gods  ;  artists  were  piously 
occupied  when  they  pursued  the 
same  course.  They  gave  precision 
of  contour  to  the  forms  roughly 
sketched,  and  by  the  established 
definition  which  they  gave  to  each 
divine  figure,  we  might  almost  say 
that  they  created  the  gods. 

Their  task  was,  in  one  sense, 
more  difficult  than  that  of  the 
Greek  artists.  When  newly  born 
Greek  art  first  began  to  make  re- 
presentations of  Greek  deities,  the 
work  of  intellectual  analysis  and 
abstraction  had  already  come  to  a 
state  of  maturity  which  it  never 
reached  in  Egypt.  The  divinities 
were  fewer  in  number  and  conse- 
quently more  fixed  and  decided  in 
their  individual  characteristics.  The 
Egyptian  polytheism  was  always 
more  mixed,  more  strongly  tinged 
with  fetishism  than  that  of  Greece. 

Even  in  those  centuries  in  which  the  ideas  of  the  Egyptian  people 
were  most  elevated  and  refined,  the  three  successive  stages,  which 


Fig.    36. — Osiris,  from  a  bronze  in  the 
Louvre.      Height  22*8  inches 


54 


A   History  of  Art  in  Ancient  Egypt. 


are  always  found  in  the  development  of  religious  life,  co-existed  in 
the  mind  of  the  nation.  A  few  more  or  less  isolated  thinkers  were 
already  seeking  to  formulate  monotheism.  The  elite  of  the  nation 
— the  king,  the  priest,  and  the  military  class — were  adoring  Amen 

and  Ptah,  Khons  or  Khonsu, 
Mouth,  Osiris  and  Horus, 
Sekhet,  Isis,  Nephtisand  many 
other  divinities  ;  all  more  or 
less  abstract  in  their  nature, 
and  presiding  over  special  phe- 
nomena. As  for  the  lower 
orders  of  the  people,  they  knew 
the  names  of  these  deities  and 
associated  themselves  with  the 
great  public  honours  which 
were  paid  to  them  ;  but  their 
homage  and  their  faith  were 
more  heartily  rendered  to  such 
.concrete  and  visible  crods  as 
the  sacred  animals,  as  the  bulls 
Apis  and  Mnevis,  the  goat  of 
Mendes,  the  ibis,  the  hawk,  &c. 
None  of  the  peculiarities  of 
Egyptian  civilization  struck 
Greek  travellers  with  more 
amazement  than  this  semi- 
wofship  of  animals.^ 

Later     theology     has     suc- 
ceeded   in  orivino-  more  or  less 
subtle    and    specious   explana- 
tions of  these  forms  of  worship. 
Each    of    these    animals    has 
been   assigned,    as    symbol    or 
attribute,  to  one  of  the  greater 
deities.      As    for   ourselves   we 
have    no     doubt     that     these 
objects  of  popular  devotion  were  no  more  than  ancient  fetishes. 
In  the  long   prehistoric   centuries,  while   the    Egyptian   race   was 
occupied  in   making  good   its   possession  of  the    Nile   valley  and 

'  Herodotus,  ii.  75-86. 


Fig.  37. — The  goddess  Bast.     (From  a  bronze  in 
the  Louvre.     Actual  size.) 


The  Egyptian   Religion  and  the  Plastic  Arts.  55 

brino-ine  it  into  cultivation,  imacrination  deified  these  animals, 
some  for  the  services  which  they  rendered,  others  for  the 
terror  which  they  inspired  ;  and  it  was  the  same  with  certain 
vegfetables. 

We  find  traces  of  this  phenomenon,  which  at  first  seems  so 
inexplicable,  among  the  other  races  of  antiquity,  but  it  is  nowhere 
else  so  marked  as  it  is  in  Egypt.  When  Egypt,  after  being  for 
three  centuries  subject  to  the  influence  and  supremacy  of  the 
Greek  genius,  had  lost  all  but  the  shadow  of  its  former  indepen- 
dence and  national  life  ;  when  all  the  energy  and  intellectual 
activity  which  remained  to  it  was  concentrated  at  the  Greco-Syrian 
rather  than  Egyptian  Alexandria,  the  ancient  religion  of  the  race 
lost  all  its  highest  branches.^  The  aspirations  towards  mono- 
theism took  a  form  tliat  was  either  philosophical  and  Platonic  or 
Christian  ;  and  as  for  the  cultivated  spirits  who  wished  to  continue 

'  We  do  not  mean  to  say  that  the  higher  qualities  of  the  Egyptian  religion  were 
then  altogether  lost.  In  Roman  Egypt  the  fetish  superstitions  were  no  doubt 
predominant,  but  still  it  had  not  lost  all  that  theological  erudition  which  it  had 
accumulated  by  its  own  intellectual  energy.  In  an  inscription  cut  in  the  time  of 
Philip  the  Arab,  we  find  an  antique  hymn  transcribed  in  hieroglyphs  upon  the  wall 
of  a  temple.  We  find  abstract  and  speculative  ideas  in  all  those  Egyptian  books 
which  have  come  down  to  us,  in  a  form  which  betrays  the  last  two  centuries  of  the 
Empire.  Alexandria  had  its  Egyptian  Serapeum  by  the  side  of  its  Greek  one. 
Monuments  are  to  be  found  there  which  are  Egyptian  in  every  particular.  Gnostic- 
ism was  particularly  successful  in  Egypt,  which  was  predestined  to  accept  it  by  the 
whole  of  its  past.  Certain  doctrines  of  Plotinus  are  thus  best  explained.  More  than 
one  purely  Egyptian  notion  may  be  found  interpreted  in  the  works  of  Alexandrian 
philosophers  and  in  the  phraseology  of  Greek  philosophy.  The  principal  sanctuaries 
did  not  allow  their  rites  and  ceremonies  to  fall  into  disuse.  Although  Thebes  was 
nothing  but  a  heap  of  ruins,  a  dead  city  visited  for  its  relics  of  the  past,  the 
worship  of  Vulcan,  that  is  of  Ptah,  at  Memphis,  was  carried  on  up  to  the  establish- 
ment of  Christianity.  That  of  Isis,  at  PhilcC,  lasted  until  the  time  of  Justinian. 
Diocletian  negotiated  a  treaty  with  the  Blemmyes,  those  people  of  Nubia  who  were 
at  one  time  such  redoubtable  soldiers,  which  guaranteed  to  them  the  free  use  of  that 
temple.  It  was  not  converted  into  a  church  until  after  the  destruction  of  the 
Blemmyes  by  Silco  and  the  Christian  kings  of  Ethiopia. 

The  old  religion  and  theology  of  the  Egyptians  did  not  expire  in  a  single  day.  It 
was  no  more  killed  by  the  Roman  conquest  than  it  was  by  that  of  the  Ptolemies. 
But  although  its  rites  did  not  cease,  and  some  of  its  elaborate  doctrines  still  continued 
to  be  transmitted,  its  vitality  had  come  to  an  end.  It  exercised  some  remains  of 
influence  only  on  condition  of  being  melted  down  and  re-modelled  in  the  crucible 
of  Greek  philosophy.  A  little  cofen'e  of  thinkers  set  themselves  to  complete  this 
transfusion,  but  the  great  mass  of  the  people  returned  to  simple  practices  which 
had  been  sanctified  by  thousands  of  years,  and  formed  nearly  the  whole  of  their 
religion. 


56  A   History  of  Art  in  Ancient  Egypt. 

the  personification  of  the  eternal  forces  of  the  world  and  of  the 
laws  which  govern  them,  these  laws  and  forces  presented 
themselves  to  their  minds  in  the  forms  which  had  been  figured 
and  described  by  the  sculptors,  painters,  and  writers  of  Greece. 
They  accepted,  without  hesitation  or  dispute,  the  numbers  and 
physical  characteristics  of  the  divine  types  of  Greece.  From 
end  to  end  of  the  habitable  earth,  as  the  Greeks  boasted,  the 
gods  of  the  Hellenic  pantheon  absorbed  and  assimilated  all 
those  of  other  nationalities  ;  within  the  boundaries  of  the  Roman 
Empire,  at  least,  its  polytheism  became  a  kind  of  universal 
religion  for  civilized  humanity,  and  was  adopted  by  nations  of  the 
most  diverse  oriofin  and  lanofuaofe.  The  lower  classes  alone,  who 
read  neither  Homer  nor  Hesiod  and  were  unable  to  admire  the 
statues  of  the  Greek  sculptors,  were  kept  free  from  the  powerful 
and  softening  influence  of  poetry  and  art.  They  guarded  with 
obstinacy  the  ancient  foundations  of  their  early  faith,  and  in  the 
void  left  by  the  disappearance  of  the  national  gods,  their  primitive 
beliefs  seem  to  have  put  on  a  new  life  and  to  have  enjoyed  a 
restored  prestige.  Thus  we  may  see,  in  forest  clearings,  the 
ancient  but  still  vigorous  stumps  of  great  trees  which  have  been 
felled  send  out  fresh  shoots  to  renew  their  youth. 

This  persistence,  this  apparent  recrudescence  of  fetishism  made 
itself  felt  in  Egypt  alone.  It  amazed  and  scandalized  both  pagans 
and  Christians  during  the  early  centuries  of  Christianity.  They 
mocked  at  a  people  who  ''hardly  dared  to  bite  a  leek  or  an 
onion  ;  who  adored  divinities  which  grew  in  their  own  gardens,"  ^ 
and  a  god  which  was  nothing  but  a  "  beast  wallowing  on  a  purple 
carpet."  ^  Guided  by  a  more  critical  knowledge  of  the  past,  we  are 
now  better  able  to  understand  the  origin  of  these  beliefs  and  the 
secret  of  their  lonor  duration.  We  are  enabled  to  account  for  them 
by  that  inexperience  which  falsifies  all  the  judgments  of  infancy, 
in  the  race  as  well  as  in  the  individual ;  we  see  that  they  are  the 
exaororeration  of  a  natural  sentiment,  which  becomes  honourable 
and  worthy  of  our  sympathy  when  it  is  addressed  to  the  useful 
and  laborious  helpers  of  man,  to  domestic  animals,  for  instance, 
such  as  the  cow  and  the  draueht  ox. 


t3 


'  Porrum  et  cffipe  nefas  violare  et  frangere  morsu. 
O  Sanctas  gentes,  quibus  haec  nascuntur  in  hortis 
Numina  ! — Juvenal,  xv.  9-11. 
'  Clemexs  Alexandrinus,  quoted  by  Maspero,  Histoire  anaeiuie,  p.  46. 


The  Egyptian   Religion  and  the  Plastic  Arts.  57 

It  would  be  interesting  to  know  why  these  beHefs  were  so 
curiously  tenacious  of  life  in  Egypt ;  perhaps  the  reason  is  to  be 
found  in  the  prodigious  antiquity  of  Egyptian  civilization.  That 
civilization  was  the  oldest  which  the  world  has  seen,  the  least 
remote  from  the  day  of  man's  first  appearance  upon  the  earth.  It 
may  therefore  be  supposed  to  have  received  more  deeply,  and 
maintained  more  obstinately,  those  impressions  which  characterize 
the  infancy  of  men  as  well  as  of  mankind.  Add  to  this,  that  other 
races  in  their  efforts  to  emerge  from  barbarism,  were  aided  and 
incited  by  the  example  of  races  which  had  preceded  them  on  the 
same  road.  The  inhabitants  of  the  Nile  Valley,  on  the  other 
hand,  were  alone  in  the  world  for  many  centuries  ;  they  had  to 
depend  entirely  upon  their  own  internal  forces  for  the  accomplish- 
ment of  their  emancipation  ;  it  is,  therefore,  hardly  surprising  that 
they  should  have  remained  longer  than  their  successors  in  that 
fetish  w^orship  which  we  have  asserted  to  be  the  first  stage  of 
religious  development.^ 

This  stage  must  never  be  forgotten,  if  we  wish  to  understand 
the  part  which  art  played  in   the   figuring  of  the   Egyptian  gods. 

^  This  was  perceived  by  the  President  de  Brasses,  a  savant  with  few  advantages  but 
a  bold  and  inquiring  spirit,  to  whom  the  language  is  indebted  for  the  use  of  the  term 
fetishism  as  a  name  for  a  definite  state  of  religious  conception.     We  can  still   read 
with  interest  the  book  which  he  published  anonymously  in  1760,  under  the  title:  Du 
Culte  des  Dieux  fetiches  ;  ou,  Parallele  de  V Ancienne  Religion  de  F Egypte  avecla  Religion 
actiielle  de  Nigritie  (i2mo).      The  study  of  the  fetish  elements  of  the   Egyptian 
religion  has  been  resumed  lately  with  competent  knowledge  and  talent  by  a  German 
egyptologist,  Herr  Pietschmann,  in  an  essay  which  appeared  in  1878  in  the  Zdt- 
schriftfilr  Ethnologie,  which  is  published  in  Berlin  under  the  direction  of  M.  Virchow. 
It    is    called    Der   /Egyptische    Fetischditnst    iind    Gotterglaube — Prolegoinena   zur 
^gyptischen  Mythologie  (28  pp.   8vo).     A  great  many  judicious   observations  and 
curious  facts  are  to  be  found  in  it ;  the  realistic  and  materialistic  character  of  the 
Egyptian  conceptions  are  very  well  grasped ;  it  is  perhaps  to  be  regretted  that  the 
author  has  not  endeavoured  to  make  the  creeds  to  which  he  gives  this  name  of 
fetichisme  somewhat  clearer,  and  to  show  by  what  workings  of  the   mind  they  were 
adopted  and  abandoned.     With  regard  to  the  Egyptian  religion,  we  shall  find  treated, 
in  the  excellent  Manuel  de  F  Histoire  des  Religions,  by  Tiele,  which  M.  jNIaurice  Vernes 
has  just  translated  from  the  Dutch  (i  vol.  i2mo,  Ernest  Leroux,  1880),  views  much  the 
same  as  those  which  we  have  just  described.     The  author  denominates  the  religious 
state  which  we  call  fetishism  animism,  but  he  points  out  the  fact  that  this  class  of 
conceptions  had  a  perennial  influence  over  the  Egyptian  mind.     "  The  Egyptian 
religion,"  he  says,  "  like  the  Chinese,  was  nothing  to   begin  with   but  an  organised 
animism.'''     He  finds  traces  of  this  animis?n  in  the  worship  of  the  dead,  the  deifica- 
tion of  the  kings,   and  the   adoration  of    animals.       From  his   point   of  view  the 
custom   of  placing  a  symbol  of  the  divinity  rather  than  an  image  in   the  temple, 
must  be  traced  to  fetishism  (pp.  44  and  45  of  the  French  version). 

VOL.    I.  I 


58 


A  History  of  Art  in  Ancient  Egypt. 


In  most  of  the  types  which  it  created  it  mixed  up  the  physical 
characteristics  of  man  and  beast.  Sometimes  the  head  of  an 
animal  surmounts  the  body  of  a  man  or  woman  ;  sometimes, 
though  more  rarely,  the  opposite  arrangement  obtains.  The 
Sphinx,  and  the  bird  with  a  human  head  which  symbolizes  death, 
are  instances  of  the  latter  combination.  The  usual  explanation  of 
these  forms  is  as  follows.  When  men  began  to  embody  for  the 
eye  of  others  the  ideas  which  they  had  formed  of  the  divine 
powers,  they  adopted  as  the   foundation   for  their  personifications 

the  noblest  living  form  they 
knew,  that  of  man.  In  the 
next  place  they  required  some 
easy  method  for  distinguishing 
their  imaginary  beings  one  from 
another.  They  had  to  give  to 
each  deity  some  feature  which 
should  be  peculiar  to  him  or 
her  self,  and  should  allow  of 
his  beino-  at  once  identified  and 
called  by  his  own  name.  The 
required  result  was  obtained 
in  a  very  simple  manner,  by 
adding  to  the  constant  quantity 
the  human  figure,  a  varying 
element  in  the  heads  of  different 
animals.  These  the  fauna  of 
Egypt  itself  afforded.  In  the 
case  of  each  divinity,  the  par- 
ticular animal  was  selected 
which  had  been  consecrated 
to    it,    which    was    its    symbol 


Fig.  38. — Painted  bas-relief.     Boulak 
by  Bourgoin.) 


(Di-a\vn 


or  at  least  its  attribute,  and  the  head  or  body,  as  the  case 
might  be,  was  detached  in  order  to  form  part  of  a  complex  and 
imaginary  being.  The  special  characteristics  of  the  animal  made 
use  of  were  so  frankly  insisted  upon  that  no  confusion  could 
arise  between  one  deity  and  another.  Even  a  child  could  not 
fail  to  see  the  difference  between  Sekhet,  with  the  head  of  a  cat 
or  a  lioness,  and  Hathor,  with  that  of  a  cow. 

We  do  not  refuse  to  accept  this  explanation,  but  yet  we  may 
express  our  surprise  that  the  Egyptians,  who  were  able,  even  in 


The  Egyptian   Religion  and  the  Plastic  Arts. 


59 


the  days  of  the  ancient  empire,  to  endow  the  statues  of  their  kings 
with  so  much  purity  and  nobihty  of  form,  were  not  disgusted  by 
the  strangeness  of  such  combinations,  by  their  extreme  grotesque- 
ness,  and  by  the  disagreeable  results  which  they  sometimes 
produced.     A  certain  beauty  may  be  found  in  such  creations  as 


Fig.   39. — Sekhet.     Louvre.     (Granite.     Height  0*50  metre?.) 


the  Sphinx,  and  a  few  others,  in  which  the  human  face  is  allied  to 
the  wings  of  a  bird,  and  the  trunk  and  posterior  members 
of  the  most  graceful  and  powerful  of  quadrupeds.  But  could  any 
notion  be  more  unhappy  than  that  of  crowning  the  bust  of  a  man 


6o 


A   History  of  Art  in  Ancient   Egypt. 


or   woman   with  the  ugly  and  ponderous   head   of  a  crocodile,  or 

with  the  slender  neck  and  flat  head  of  a  snake  ? 

Every  polytheistic  nation  attacked 
this  problem  in  turn,  and  each  solved 
it  in  its  own  manner.  The  Hindoos 
multiplied  the  human  figure  by  itself, 
and  painted  or  carved  their  gods  with 
three  heads  and  many  pairs  of  arms 
and  legs,  of  which  proceeding  traces 
are  to  be  found  among  the  Western 
Asiatics,  the  Greeks,  and  even  the 
Latins.  The  Greeks  represented  all 
their  gods  in  human  form,  and  yet  by 
the  delicacy  of  their  contours  and  the 
general  coherence  of  their  character- 
ization, they  were  enabled  to  avoid  all 
confusion  between  them.  With  them, 
too,  costume  and  attributes  helped  to 
mark  the  difference.  But  even  where 
these  are  absent,  our  minds  are  never 
left  in  doubt.  Even  a  fragment  of  a 
torso  can  be  at  once  recognized  at 
sight  as  part  of  a  statue  of  Zeus,  of 
Apollo,  or  of  Bacchus,  and  a  head  of 
Demeter  or  Hera  would  never  be 
confounded  with  one  of  Artemis  or 
Pallas. 

It  may  be  said  that  the  artists  of 


Egypt 


were     lacking-     in     the    skill 


necessary  for    all    this,    or    that    they 
generalized    their    forms    to    such    a 
degree  as  to  leave  no  scope  for  such 
subtle  differences.      But,   in  fact,   we 
find  in   their  oldest  statues  a  facility 
of  execution  which  suggests  that,  had 
they  chosen,  they  could  have  expressed 
anything  which  can  be  expressed  by 
the   chisel.      That  they  did    not    do   so,    we   know.      They  con- 
tented   themselves    with    plastic    interpretations    so    rough    and 
awkward  that,    perhaps,    we  should  rather  seek  their  explanation 


Fig.   40. — Isis-Hathor.      Louvre. 
(Bronze.     Actual  size.) 


_   -  f       >^ 


o 


I 

.S 
'5 


> 

o 


4< 

'E 

rt 

<u 
o 


X 


6 


The  Egyptian  Religion  and  the  Plastic  Arts. 


In  some  hereditary  predisposition,  some  habit  of  thought  and 
action  contracted  In  the  Infancy  of  the  race  and  fortified  by  lono- 
transmission. 

We  have  already  spoken  of  that  which  we  beheve  to  be  the 
cause  of  the  pecuhar  forms  under  which  the  Egyptians  figured 
their  deities,  namely,  the  fetish  worship,  which  was  the  earliest,  and 
for  many  centuries  the  only,  form  of  religion  which  they  possessed. 
That  worship  had  struck  Its  roots  so  deeply  Into  the  souls  of  the 
people,  that  it  could  not  be  torn  up  even  when  a  large  part  of  the 
nation  had  gradually  educated  itself  to 
the  comprehension  of  the  highest  re- 
ligious conceptions.  Its  practices  never 
fell  Into  total  neglect,  and  Its  Influence 
was  so  far  maintained  that  during  the 
decadence  of  the  nation  It  aeain  became 
the  ruling  faith,  so  that  foreign  observers 
were  led  to  believe  that  the  Egyptian 
religion  began  and  ended  In  the  adora- 
tion of  plants  and  sacred  animals.  The 
eyes  and  the  imagination  being  thus 
educated  by  immemorial  custom,  it  Is 
not  surprising  that  even  the  most  culti- 
vated section  of  the  people  should  have 
seen  nothing  offensive  In  the  representa- 
tion of  their  gods  sometimes  under  the 
complete  form  of  an  animal  (Horus  Is 
often  symbolized  under  the  likeness  of  a 
hawk),  sometimes  as  composite  monsters 
with  human  bodies  and  animal  heads. 

Take,    for    a    moment,    the    bird    to 
which     we    have    just    alluded.       The 

hawk,  like  the  vulture,  plays  an  important  part  in  Egyptian 
art.  The  vulture  symbolizes  Maut,  the  spouse  of  Amen.  It 
furnishes  the  sign  by  which  her  name  Is  written,  and  sometimes,  as 
the  symbol  of  maternity,  its  head  appears  over  the  brow  of  the 
goddess,  Its  wings  forming  her  head-dress.  The  goddess  Nekheb, 
who  symbolizes  the  region  of  the  South,  Is  also  represented  by  a 
vulture.^  So  it  is  with  the  ibis.  It  supplies  the  character  by 
which  the  name  Thoth  is  written,  and  that  [jod  is  fio^ured  with  the 
■^  PiERRET,  Dictlonnnirc  d" Archeologie  Eg)'ptienne 


"^  ^^  \  I'lii 


Fig.  42. — Touaris.     Boulak. 
(Drawn  by  G.  Benedite. ) 


64  A  History  of  Art  in  Ancient  Egypt, 


head  of  an  Ibis.  The  part  played  by  these  birds  in  the  repre- 
sentation of  the  gods,  both  in  the  plastic  arts  and  in  writing,  is  to 
be  explained  by  the  sentiments  of  gratitude  and  religious  venera- 
tion of  which  they  were  the  objects,  sentiments  which  were  the 
natural  outcome  of  the  practical  services  which  they  rendered  to 
mankind. 

When  the  early  fathers  of  the  nation  first  established  themselves 
upon  the  banks  of  the  Nile,  they  found  invaluable  allies  in  those 
energetic  birds  of  prey,  and  the  alliance  has  been  continued  to 
their  latest  descendants.  After  the  annual  inundation  the  damp 
earth  was  overrun  by  toads  and  frogs,  by  snakes  and  lizards  and 
all  kinds  of  creeping  things.  Fishes,  left  by  the  retreating  flood 
in  pools  which  were  soon  dried  up  by  the  blazing   sun,  perished, 

and,  decomposing,  rendered  the  air  noisome  and 
malarious.  In  addition  to  this  there  were  the 
corpses  of  v/ild  and  domestic  animals,  and  the 
offal  of  every  kind  which  accumulated  round  the 
dwellings  of  the  peasantry  and  rapidly  became 
putrid  under  the  sun  of  Egypt.  If  left  to 
decompose  they  would  soon  have  bred  a  pesti- 
lence, and  in  those  days  human  effort  was  not 
to  be  reckoned  upon  in  the  work  of  sanitation. 
To  birds  of  prey,  then,  was  assigned  the  in- 
dispensable work  of  elimination  and  transforma- 
tion, an  office  which  they  yet  fill  satisfactorily 
(from  wiikinsoii)"  in  the  towns  and  villages  of  Africa.  Thanks  to 
their  appetite  and  to  the  powerful  wings  which 
carried  them  in  a  twinkling  to  wherever  their  presence  was  re- 
quired, the  multiplication  of  the  inferior  animals  was  kept  within 
due  limits,  and  decomposing  matter  was  recalled  into  the  service 
of  organic  life.  Had  these  unpaid  scavengers  but  struck  work  for 
a  day,  the  plague,  as  Michelet  puts  it,  would  soon  have  become  the 
only  inhabitant  of  the  country.^ 

1  See  in  LOiscau  the  chapter-headed  1! ApJiration.  With  his  genius  for  history 
and  poetry  Michelet  has  well  understood  the  sentiment  which  gave  birth  to  these 
primitive  forms  of  worship,  forms  which  have  too  long  provoked  unjust  contempt. 
The  whole  of  this  beautiful  chapter  should  be  read  ;  we  shall  only  quote  a  few  lines  : 
"  In  America  the  law  protects  these  public  benefactors.  Egyptian  law  does  still 
more  for  them — it  respects  them  and  loves  them.  Although  they  no  longer  enjoy 
their  ancient  worship,  they  receive  the  friendly  hospitality  of  man  as  in  the  time 
of  Pharaoh.      If  you  ask  an  Egyptian  fellah  why  he  allows  himself  to  be  besieged 


The   Egyptian   ReliCxIon  and  the  Plastic  Arts. 


65 


The  worship  of  the  hawk,  the  vulture,  and  the  ibis,  had,  then, 
preceded  by  many  centuries  that  of  the  gods  who  correspond  to 
the  personages  of  the  Hellenic  pantheon.  Rooted  by  long  custom 
in   the  minds  of  the   people,   it  did  not  excite  the  ire  of  the  wise 


Fig.  44. — IIoru>  ;   from  a  brouzc  iu  the  Posuo  collection.      (Height  38  inches.) 


men  of  Heliopolis  or  Thebes.  The  doctrine  of  emanation  and 
of  successive  incarnations  of  the  deity,  permitted  their  theology  to 

and  deafened  by  birds,  why  he  patiently  suffers  the  insolence  of  the  crow  perched 
upon  the  horn  of  the  buffalo,  on  the  hump  of  a  camel,  or  fighting  upon  the  date-trees 
and  shaking  down  the  fruit,  he  will  say  nothing.  Birds  are  allowed  to  do  anything. 
Older  than  the  Pyramids,  they  are  the  ancients  of  the  country.  Man's  existence 
depends  upon  them,  upon  the  persevering  labour  of  the  ibis,  the  stork,  the  crow 
and  the  vulture." 

VOL.   I.  K 


66 


A   History  op^  Art  in  Ancient  Egypt. 


explain  and  to  accept  anything,  even  those  things  which  at  a  later 
epoch  seemed  nothing  more  than  the  grossest  creations  of  popular 
superstition.  These  objects  of  veneration  were  therefore  enabled 
to  maintain  their  places  by  the  side  of  the  superior  gods,  to  repre- 
sent  them    in    written    characters    and   in    plastic    creations,    and, 

in  the  latter  case,  to  be  blended 
with  the  forms  of  man  himself  To 
us,  accustomed  as  we  are  to  the 
types  created  by  Greek  anthro- 
pomorphism, these  figures  are  sur- 
prising ;  but  to  the  Egyptians  they 
seemed  perfectly  natural,  for  they 
offered  the  characteristic  features  of 
the  animals  which  they  had  loved, 
respected,  and  adored  ever  since 
the  birth  of  their  civilization. 

It  is  difficult  for  us  to  see  things 
with  the  same  eyes  as  the  con- 
temporaries of  Cheops  or  even  of 
Rameses  ;  to  enter  into  their  ideas 
and  sentiments  so  as  to  feel  with 
them  and  to  think  with  their  brains. 
Let  us  attempt  to  do  so  for  a 
moment  ;  let  us  make  one  of  those 
intellectual  efforts  which  are  de- 
manded from  the  historian,  and 
we  shall  then  understand  how  it 
was,  that  the  Egyptians  were  not 
offended  by  a  combination  of  two 
classes  of  forms  which,  to  us,  seem 
so  differently  constituted  and  so 
unequal  in  dignity.  The  deity 
took  the  form  of  an  animal  and 
revealed  himself  in  it,  just  as  he  took 
that  of  a  man,  or  of  a  statue  which 
he  was  supposed  to  animate,  and  to  which  he  was  attached.  In 
one  of  his  most  curious  and  most  penetrating  essays,  M,  Maspero 
explains  that  the  sacred  animal  was — like  the  king,  the  son  of 
Amen  ;  like  the  statue  fashioned  by  the  hands  of  a  sculptor — the 
manifestation  of  the  deity,  the  strength  and  support  of  his  life,  his 


Fig.  45.- 


•Thoth.     Louvre.      ICnamtlled 
clay.     Actual  size. 


The  Egyptian  Religion  and  the  Plastic  x'\rts.         67 


dotibh\  to  use  an  expression  dear  to  the  Egyptians.  At  Memphis, 
Apis  repeated  and  constantly  renewed  the  Hfe  of  Ptah  ;  he  was, 
in  a  word,  his  living  statue.^ 

Egyptian  art  was,  then,  the  faithful  and  skilful  interpretation  of 
the  ideas  of  the  people.  What  the  Egyptians  wished  to  say,  that 
they  did  say  with  great  clearness  and  a  rare  happiness  of  plastic 
expression.  To  accuse  them,  as  they  have  been  sometimes 
accused,  of  a  want  of  taste,  would  be  to  form  a  very  narrow 
conception  of  art,  to  sin  against  both  the  method  and  the  spirit  of 
modern  criticism.  This  latter  seeks  for  originality  and  admires  it, 
and  all  art  which  is  at  once  powerful  and  sincere  arouses  its 
interest.     We    do   not,   however, 

wish   to   deny  that  their  concep-  ^^^^^^^^^.^^^^.^ 

tion  of  divinity  is  less  favourable  ^^^^iliPl-ltStK 

to  the  plastic  arts  than  the  anthro- 
pomorphism of  the  Greeks.  No 
more  simple  method  of  dis- 
tinsfuishinof  one  ofod  from  another 

could  well  be  imagined  than  that  W$§MlilkM'^^'^^ 

of    orivinor    to    each,    as    his    ex-  ^MraW^.f^^^ 

elusive    property,     the^   head    of  ||p5||™f^T;T^^^^^ 

some     well-known     animal;     the  rr;(!^./j5i 


fe£  ?t^  ■iVv  ^i^  ■  j-'^-r  ii  ^Tv'iy^ 

WMimMA 


■^' -''■"--;:ai 


employment  or    such  an    unmis-  W§yif':'i5i\}.^^:4\^^^ 

takable    sign    rendered    the   task  fi|p|&|^^iiwC^^ 

of  the  artist  too  easy,  in  giving  {:p||%W:;::B|gS^ 

nmi  assurance  that  his   meanme  ^fe?^vk^^:^^x:'^V:  ■:..;—— —^—— I" ■ 

would  be  understood  at  a  glance  ^'^-   ^~-^^-^-'^=-—  -    -^■-^^-^^ 

without     any    particular     effort     on  Fig.  46.— Sacrifice  to  Apis,  from  Marictte. 

his   part. 

The  value  of  an  artistic  result  is  in  proportion  to  the  difficulty  of 
its  achievement.  The  Greek  sculptor  had  nothing  beyond  the 
bodily  form  and  the  features  of  man  with  which  to  give  a  distinct 
individuality  to  each  god  and  goddess  of  his  mythology ;  he 
was  therefore  oblio-ed  to  make  use  of  the  most  delicate  and 
subtle  distinctions  of  feature  and  contour.  This  necessity  was  a 
great  incentive  to  perfection  ;  it  drove  him  to  study  the  human 
form  with  a  continuous  energy  which,  unhappily  for  himself,  was 
not  required  of  the  Egyptian  sculptor  or  painter. 

'  Maspero,  Notes  su7-  diffcrents  Points  de  Graininairc  et  d' Histoirc  daiislc  Recucil  de 
Travaiix  rdatifs  a  la  Philologie  et  a  l Archcologie  cgypticnne  etassyiic/me,  vol.  i.  p.  157. 


68  A   History  of  Art  in  Ancient   Egypt. 


Art  and  religion  have  ever  been  so  closely  allied  that  it  was 
necessary  that  we  should  give  some  account  of  the  original 
characteristics  of  the  Egyptian  beliefs,  but  we  shall  make  no 
attempt  to  describe,  or  even  to  enumerate,  the  chief  divinities  of  the 
Egyptian  pantheon ;  such  an  attempt  would  be  foreign  to  the 
purposes  which  we  have  in  view.  We  have,  however,  already 
mentioned  most  of  the  chief  deities  of  Egypt,  and  we  shall  have 
occasion  to  draw  the  attention  of  our  readers  to  others,  in  speak- 
ing of  the  tombs  and  temples,  the  statues  and  bas-reliefs,  of  the 
country.  Now,  each  of  these  gods  began  by  being  no  more  than 
the  local  divinity  of  some  particular  nome  or  city.  As  a  city  grew 
in  importance,  so  did  its  peculiar  god,  and  sometimes  it  came  about 
that  both  a  dynasty  of  kings  and  a  divinity  were  imposed  upon 
Egypt  by  the  power  of  what  we  may  call  their  native  city.  In  the 
course  of  time  a  number  of  successive  deities  thus  held  the 
supreme  place,  each  of  whom  preserved,  even  after  his  fall, 
some  of  the  dignity  which  he  had  acquired  during  his  period 
of  supremac)'. 

The   two   first  dynasties,   the   authors   of   Egyptian   unity,   had 
their  capital  in  the  nome  of  Abydos,  the   nome  which   contained 
the  tomb  of  Osiris  ;  and  it  was  in  their  reign  that,  from  one  end 
to  the  other  of  the   Nile  valley,  spread  the  worship  of  that  god  ; 
of  that  Osiris  who,  with  Isis,  seemed  to  Herodotus  to  be  the  only 
deity    whom  all  the  Egyptians  combined  to  adore. ^      Under  the 
following   dynasties,   whose  capital  was   Memphis,   Ptah  rose   into 
the  first  place  ;    but,  as  if  by  a  kind  of  compromise,   his   dignity 
is  combined  with  that  of   the   great    god    of    Abydos    under    the 
names   of    Ptah-Osiris    and    Ptah-Sokar-Osiris.      Toum,   the   chief 
deity  of    Heliopolis,  never  rose  above   the   second  rank   because 
Heliopolis  Itself  was  neither  a  royal  city  nor  even  the  birthplace 
of  any  powerful  dynasty.      During  all  this  period  we  hear  nothing 
of  Amen,   the  local  deity  of   Thebes  ;    his  name   is  hardly   to  be 
found  upon  any  monument  earlier  than  the  eleventh  dynasty,  but, 
with  the  rise  of  the  Theban  empire  he  began  to  be  a  conspicuous 
figure  in   Egypt.      During    the   domination   of  the    Hyksos,   their 
national  deity,  Soutekh  or  Set,  overshadowed  the  ancient  divinities 
of   the   soil  ;    but   the    final    victory   of   Thebes   under  Ahmes    I. 
installed  Amen    as  the  national  god,   and   we    shall   see  hereafter 
what  magnificent  temples  were  raised  in  his  honour  by  the  kings 

^  Herodotus,  ii.  42. 


The  Egyptian  Religion  and  the  Plastic  Arts.         69 


of  the  brilliant  Theban  dynasties.  His  successor  would  no  doubt 
have  been  Aten,  the  solar  disc,  had  Tell-el-Amarna,  the  new 
capital  of  Amenophis  IV.,  and  the  worship  which  was  there 
inaugurated,  enjoyed  a  less  ephemeral  existence  ;  but  Thebes 
and  Amen  soon  regained  their  supremacy.  Again,  when  the 
Egyptian  centre  of  gravity  was  transported  to  the  Delta,  the  local 
deities  of  the  district,  and  especially  Neith,  conquered  the  first  place 
in  the  religious  sentiments  of  the  people.  Under  the  Persians 
they  returned  to  Amen,  as  to  the  protector  who  could  give  back 
to  the  nation  its  former  independence  and  power.  Under  the 
Ptolemies,  Horus  and  Hathor  were  in  the  ascendant,  and  later 
still,  under  the  Roman  emperors,  the  worship  of  the  I  sis  of  Philae 
became  popular  and  was  prolonged  in  that  island  sanctuary  until 
the  sixth  century  of  our  era. 

The  movement  of  religious  thought  in  Egypt  was  very  different 
from  what  we  shall  find  in  Greece.  We  find  no  god,  like  that  of 
the  Hellenes,  whose  pre-eminence  dates  back  to  the  remote  origin 
of  the  Aryan  race,  a  pre-eminence  which  was  never  menaced  or 
questioned  ;  1  we  find  no  Zeus,  no  Jupiter,  whose  godhead  was 
conceived  from  century  to  century  in  an  ever  larger  and  more 
purified  spirit,  until  at  last  it  was  defined  in  the  famous  hymai  of 
Cleanthe  as  that  "  which  governed  all  things  according  to  law." 
We  have  pointed  out  how  greatly  the  Greek  artists  profited  by 
their  efforts  to  endow  the  piety  of  their  countrymen  with  an  image 
of  this  great  and  good  being,  which  should  be  worthy  of  the 
popular  faith  In  him  as  the  father  of  gods  and  men.  The 
Egyptian  artist  could  find  no  such  inspiration  in  a  long  succession 
of  gods,  no  one  of  whom  succeeded  in  concentrating  supreme  power 
in  his  hands.  No  such  ideal  existed  for  them  as  that  which  the 
popular  conscience  and  the  genius  of  the  national  poets  created  in 
the  lord  of  Olympus.  Neither  Thebes  nor  Sals  could  give  birth 
to  a  Phidias  ;  to  an  artist  who  should  feel  himself  spurred  on  by 
the  work  of  all  previous  generations  to  produce  a  masterpiece  In 
which  the  highest  religious  conception,  to  which  the  intelligence  of 
the  race  had  mounted  by  slow  degrees,  should  be  realized  in 
visible  form. 

^  James  Darmesteter,  Le  Dieii  supreme  dans  la  Mythologte  indo-mropcetme  (in 
the  Revue  de  F Histoire  des  Religions,  i8So). 


yo  A  History  of  Art  in  Ancient  Egypt. 


§  6.    T/ia^  Egyptian  Art  did  not  escape  the  Law  of  Change,  and 
that  its  History  may  therefore  be  i^^ritten. 

It  may  be  well,  before  embarking  upon  the  study  of  Egyptian 
architecture,  sculpture,  and  painting,  to  dispel  a  prejudice  which  in 
spite  of  recent  discoveries  still  exists  in  some  minds  ;  we  mean,  the 
belief  in  the  immobility  of  Egyptian  art.  This  mistake  is  a 
very  ancient  one.  The  Greeks  were  the  first  to  make  it,  and  they 
transmitted  their  error  to  us.  In  regard  to  this  we  must  cite  the 
famous  passage  of  Plato  ^  : — "  Long  ago  they  appear  to  have 
recognized  the  very  principle  of  which  we  are  now  speaking — 
that  their  vounof  citizens  must  be  habituated  to  forms  and 
strains  of  virtue.  These  they  fixed,  and  exhibited  patterns  of 
them  in  their  temples  ;  and  no  painter  or  artist  is  allowed  to 
innovate  upon  them,  or  to  leave  the  traditional  forms  and  invent 
new  ones.  To  this  day  no  alteration  is  allowed,  either  in  these 
arts  or  in  music,  at  all.  And  you  will  find  that  their  works 
of  art  are  painted  or  moulded  in  the  same  forms  that  they 
had  ten  thousand  years  ago — (this  is  literally  true  and  no  exaggera- 
tion)— their  ancient  paintings  and  sculptures  are  not  a  whit 
better  or  worse  than  the  work  of  to-day,  but  are  made  with 
just  the  same  skill," 

This  strange  assertion  was  long  accepted  without  question  even  in 
modern  times.  We  need  not  qto  back  to  the  archseoloorists  of  the 
last  century,  whose  credulity  is  to  be  accounted  for  by  their  lack  of 
materials  for  the  formation  of  a  better  judgment.  In  1S28  in  his 
first  lecture  at  the  Bibliotheque  Royale,  Raoul-Rochette  turned  his 
attention  to  Egypt.  He  had  before  his  eyes,  in  the  Parisian 
museums  and  in  the  Description  de  I  Egyptc,  works  which  dated 
from  the  finest  periods  of  the  Theban  dynasties,  although  the  still 
more  ancient  monuments  which  now  form  the  glory  of  the  Boulak 
INIuseum  were  not  yet  discovered  ;  he  might  have  perceived  and 
pointed  out  the  difference  between  the  statues  of  Ousourtesen, 
Thothmes,  and  Rameses  on  the  one  hand,  and  those  of  the  Sait 
epoch  ;  still  more  should  he  have  remarked  upon  the  characteristics 

'    Ti]v  avTYiv    8e   Tiy\'rjV   aTreipyacr/xe'ra,   etc.      Laws,  656.       D.    E.      [^^^e    have    quoted 

from  Professor  Jowett's  English  version,  p.  226,  vol.  v. — Ed.] 


Ceiange  Observable  in   Egyptian  Art.  71 


which  distinguish  the  monuments  of  independent  Egypt  from 
those  which  were  erected  under  the  Ptolemies  and  the  Roman 
emperors.  What  he  did  say,  however,  and  say  with  consummate 
confidence  was  :  "  From  the  first  of  the  Pharaohs  to  the  last  of  the 
Ptolemies,  the  art  of  Egypt  never  varied."  ^ 

Such  crude  notions  as  this  can  no  longer  be  upheld,  M. 
Marriette  protests  in  the  following  almost  indignant  terms  against 
certain  utterances  of  M.  Renan  which  seemed  to  him  to  imply  the 
same  doctrine.  "  M.  Renan  loves  ^  to  represent  ancient  Egypt  as 
a  sort  of  China,  walled  in  and  fortified  against  the  exterior  world, 
immovable,  old  even  in  its  infancy,  and  arrived  by  a  single  spring 
at  a  degree  of  civilization  which  it  never  surpassed.  He 
looks  upon  the  country  as  a  great  plain,  green  indeed  and  fertile, 
but  without  accidents  of  contour  to  break  the  monotony  of  the 
landscape.  And  yet  Egypt  had  periods  of  grandeur  and  decadence 
more  marked  than  those  of  other  countries.  Her  civilization  went 
through  all  the  different  phases  ;  it  went  through  many  complete 
transformations,  it  had  its  sudden  moments  of  brilliancy  and  its 
epochs  of  eclipse.  Its  art  was  not  so  stationary  as  to  prevent  us 
from  writing  its  history.  The  influence  of  Egypt  was  felt  from 
Mesopotamia  to  the  equator.  Thothmes,  in  a  word,  was  no 
Chinaman.  Egypt  perished  because  in  attacking  foreign  nations 
she  provoked  a  reaction  which  was  fatal  to  her."  ^ 

Now  that  we  are  enabled  to  contrast  the  statues  of  the  Ptolemaic 
period  with  those  of  the  pyramid  builders,  we  find  nothing 
surprising  in  Mariette's  language  ;  but  even  before  these  means  of 
study  were  open  to  us,  criticism  should  have  cast  more  than  doubt 
upon  the  assertions  of  Plato  ;  it  should  have  appealed  from  a 
theorv  which  was  at  variance  with  all  historical  analogies  to  the 
monuments  themselves  to  tell  the  truth,  to  those  monuments 
which  were  best  known  and  understood.  Was  it  likely,  was  it 
possible,  that  such  a  people  as  that  which  created  these  monu- 
ments, should  remain  for  more  than  forty  centuries  unaffected  by 
the  law  of  continual,  even  if  almost  insensible,  change  ? 

'  Cours  (T Archcologie,  8vo.  1829,  pp.  10,  11.  This  critic's  ideas  upon  Egyptian 
art  were  both  superficial  and  false.  "  Egyptian  art,"  he  says,  "  never  attempted  any 
realistic  imitation."  We  even  find  sentences  utterly  devoid  of  meaning,  such  as,  for 
instance,  "  The  fundamental  principle  of  Egyptian  art  was  the  absence  of 
art."  (p.  12.) 

"  See  the  Revue  des  Deux  Monies  of  April  i,  1865. 

^    Voyage  dans  la  Haute  Egypte,  vol.  i. 


72  A   History  of  Art  in  Anxiext  Egypt. 


What  right  have  we  thus  to  place  Egypt  and  China  apart  from 
the  rest  of  humanit)'  ?  There  are,  it  is  true,  some  peoples  who  are 
more  attached  than  others  to  traditional  customs  and  ancient 
institutions  ;  they  are  more  conservative,  to  use  the  modern  phrase. 
But,  although  their  evolution  is  a  slower  process,  it  is  there  ;  our 
eyes  cannot  perceive  any  movement  in  the  small  hand  of  a  watch, 
but  yet  it  does  move  exactly  in  the  same  fashion  as  that  which 
marks  the  seconds.  Upon  the  banks  of  the  Pelho  as  upon  those 
of  the  Nile,  upon  the  whole  surface  of  our  planet,  man  is  not  ;  he 
becomes,  to  borrow  one  of  the  favourite  expressions  of  German 
philosophy.  History  can  admit  no  exception  to  this  law  either 
for  China  or  Egypt.  In  the  cases  of  both  those  countries  there  is 
a  certain  illusion,  which  is  to  be  explained  by  our  ignorance.  We 
are  not  well  enough  acquainted  with  them  to  grasp  the  different 
periods  of  their  political  and  social,  their  artistic  and  literary 
development.  For  one  who  is  too  far  off  or  very  short-sighted 
the  details  of  the  most  varied  landscape  become  obliterated  or 
confused  ;  waste  land  and  smiling  helds  are  blended  together  ; 
hollows  and  hillocks  lose  the  vigour  of  their  contours. 

China,  as  we  have  said,  does  not  enter  into  our  purview  ;  and 
as  for  Egypt,  the  deeper  we  penetrate  into  her  history  the  more 
are  we  convinced  that  her  long  career  was  troubled  by  moments 
of  crisis  similar  to  those  which  have  come  to  other  human 
societies.  The  narratives  of  the  Greek  historians  orive  us  reason 
to  suspect  that  it  was  so,  and  the  monuments  which  have  been 
discovered  insist  upon  the  same  truth,  and  compel  us  to  accept 
it.  For  certain  epochs  these  are  very  abundant,  beautiful,  and 
varied.  Afterwards  they  become  rare  and  clumsy,  or  altogether 
wanting  ;  and  again  they  reappear  in  great  numbers  and  in  their 
full  nobility,  but  with  a  different  general  character.  These 
contrasts  and  temporary  eclipses  occur  again  and  again.  How, 
then,  can  we  doubt  that  here,  as  elsewhere,  there  were  alter- 
nations of  grandeur  and  poverty,  of  periods  of  conquest  and 
expansion  and  epochs  of  civil  war  or  of  defeat  by  foreign  in- 
vaders ?  May  we  not  believe  that  through  the  clouds  which 
obscure  the  causes  of  such  changes  we  may  catch  glimpses  of 
those  periods  of  decadence  and  renascence  which,  following  one 
upon  the  other,  exhausted  in  the  end  the  genius  of  the  race  ? 

Let  us  take  a  single  example — the  most  striking  of  all.  "  After 
the    sixth    dynasty    all    documents    cease  ;     they   are    absolutel}' 


Change  Observable  in   Egyptian  Art. 


o 


wanting  until  the  eleventh,  the  first  of  the  Middle  Empire. 
This  is  one  of  those  sudden  interruptions  in  the  history  of 
Egypt  which  may  be  compared  to  the  temporary  disappearance 
of  those  curious  rivers  which  run  partly  underground."^ 


1^,! 


Slip 


'mm 


Fig.  47. — Statue  from  the  Ancient  Empire,  in  limestone.     Boulak.     Drawn  by  Bourgoin. 


When  historians,  living  as  long  after  our  nineteenth  century  as 
we  do  after  the  epochs  of  Memphite  and  Theban  supremacy  in 
Egypt,  come  to  treat  the  history  of  the  past,  they  will  perhaps 

'  M.  Melchoir  de  Vogue,  C/iez  /es  Pharaons  {Rrcue  des  Deux  Mondes  of 
Jan.  15,  1S77). 

VOL.    I.  I^ 


74 


A   History  of  Art  in  Ancient  Egypt. 


look  upon  the  ages  which  rolled  away  between  the  fall  of  Graeco- 
Roman  civilization  and  the  revival  of  learning  in  the  fifteenth  and 
sixteenth  centuries  as  no  longer  than  that  which  divided  the 
ancient  from  the  middle  empire  of  Egypt,  or  the  latter  from  the 
dynasties  of  Thebes.  In  the  distant  future  men  will  know,  in  a 
vague  fashion,  that  between  the  fall  of  Rome  and  the  discovery 
of  printing,  or  that  of  America,  there  were  great  movements 
among  the  nations,  and  an  apparent  recoil  of  civilization ;  but 
memory  and  imagination  will  leap  without  effort  over  the  gap, 
over  that  period  which  we  call  the  Middle  Ages.      The  Roman 


-c-  D ^lr,i'^ft^lll>*^^  -i|i5t5,  „,„, -^^ 


Fig.  48. — Woman  kneading  dough.     Statuette  from  the  Ancient  Empire,  in  limestone. 

Drawn  by  Bourgoin. 


empire  will  seem  to  touch  our  modern  civilization,  and  many  of 
the  differences  which  strike  us  so  strongly  will  be  imperceptible. 
They  will  perceive  that  we  had  a  new  religion  and  new  inven- 
tions, but  they  will  take  more  account  of  the  resemblances  than 
of  the  differences.  Our  languages,  manners,  laws,  and  forms  of 
government  will  seem  to  them  continuations  of  those  of  Greece 
and  Rome.  In  that  which  we  call  antiquity,  and  in  Christian 
Europe,  they  will  find  similar  literary  habits  and  standards  of 
criticism,  the  same  judicial  nomenclature,  the  same  terms  for 
monarchy,  empire,   and   republic,  the    same    titles    for   kings  and 


Change  Observable  in  Egyptian  Art. 


/o 


Caesars.  These  different  civ'ilizations  are  like  star  clusters.  To 
us  who  are  among  them  they  seem  distinct  enough,  but  to  gene- 
rations which  are  divided  from  them  by  a  vast  space  of  time  they 
will  seem  to  form  but  one  nebulous  body. 

Egypt,  then,  had  her  great  convulsions  like  the  rest  of  the 
world.  She  met  with  disasters,  and  underwent  periods  of  con- 
fusion like  those  which  overtook  tlie  nations  of  the  West  between 
the  reigns  of  Trajan  and 
Charlemao^ne.  Wars  and 
invasions,  the  action  and 
reaction  of  civilization,  had 
upon  her  the  same  influence 
as  upon  them,  and,  in  trans- 
forming her  sentiments  and 
ideas,  caused  their  plastic 
expression  to  pass  through 
a  series  of  changes  in  taste 
and  style.  The  Theban 
tomb  of  the  time  of  Rameses 
is  very  different  from  that  of 
Memphis  and  the  ancient 
empire  ;  the  new  empire  con- 
structed no  buildings  like 
the  greater  pyramids,  but 
its  temples  were  larger  and 
more  magnificent  than  any 
of  their  predecessors.  It 
was  the  same  with  sculpture. 
A  cultivated  eye  has  no 
need  to  run  to  inscriptions 
to  enable  it  to  distinofuish 
between  works  of  the  ancient 
and  of  the  middle  empire  ; 
nor   will  it   confound   works 

created  in  either  of  those  periods  with  those  of  the  Sait 
epoch.  The  differences  are  almost  as  well  marked  as  those 
which  enable  archaeologists  to  distinguish  a  torso  of  the  time  of 
Phidias  from  one  of  the  school  of  Praxiteles  or  Lysippus. 
These  differences  it  will  be  our  duty  to  describe  hereafter, 
but    our    readers    may    perhaps    discover     them    for    themselves 


Fig.  49. — The  Scribe  Chaphre.     Fifth  dynasty. 
Boulak.     Limestone. 


76 


A  History  of  Art  in  Ancient  Egypt. 


if    they     examine    the    illustrations    to    this    chapter,    which    are 
arrano-ed    in    chronoloi^ical    order. 

Variety  is  universal  in 
Egypt,  local  variety  as  well 
as  that  of  different  periods. 
Lanofuaore  had  its  dialects  as 
well  as  art.  The  pronuncia- 
tion of  Upper  and  that  of 
Lower  Egypt  was  quite 
dissimilar,  except  in  the  case 
of  a  few  letters.  In  the 
same  way  different  cities 
had  distinct  schools  of  sculp- 
ture and  painting,  which 
were  distinguished  from  one 
another  by  their  traditional 
methods  of  conception  and 
execution.  Neither  under 
Ousourtesen  nor  under 
Rameses,  had  art  the  same 
character  in  the  cities  of  the 
Delta,  in  Memphis,  and  in 
Thebes,  Among  the  works 
in  sculpture  executed  for 
Rameses  II,,  those  of 
Abydos  were  more  elegant 
and  refined  than  those  of 
Thebes. 

How,    then,     are    we    to 
explain  the  error  committed 
by  Plato,  and  by  him  trans- 
mitted   to    posterity  ?     The 
explanation     is     easy.      The 
Greeks    visited     Egypt    too 
late  in  its  history  to  form  a 
true   judgment.      In    Plato's 
time  the  Egyptians  were  still 
trying,  by  violent  but  spasmodic  efforts,  to  reconquer  the  indepen- 
dence which  had  been  destroyed  by  the  successor  of  Cyrus.      But 
the  moment  was  at  hand  when   even  these  intermittent  struggles 


Fig,   50. — Tlie  Lady  Nai.     Wouden  statue  from 
the   19th  or  20th  dynasty.     Louvre. 


Change  Observable  in  Egyptian  Art. 


were  to  be  abandoned,  and  they  were  to  finally  succumb  to 
sovereiens  of  foreign  blood.  Their  still  brilliant  civilization  miorht 
deceive  a  passing  stranger,  but  the  decadence  had  commenced 
— a  decadence  slow  indeed,  but  none  the  more  remediable. 

Some  years  after  the  visit  of  Plato,  the  two  Nectanebos, 
more  especially  the  second,  devoted  themselves  with  energetic 
ardour  to  the  restoration  of  the  ancient  buildings  of  the  country 
and  to  the  construction  of  new  ones,  such  as  the  temple  at 
Philce.  Buildinofs  sio^ned  with  their  name  are  to  be  found  all 
over  Egypt  ;  but  these  simultaneous  undertakings  seem  to 
betray  a  sense  of  vanishing  power,  an  uncertainty  of  the 
morrow,  a  feverish  activity  seeking  to  deceive  itself  and  to 
hide  its  own  weakness.  Nothing  could  be  more  precarious  than 
the  political  conditions  under  which  this  activity  was  displayed. 
The  independence  of  the  country  was  maintained  by  the  dearly 
bought  services  of  Spartan  and  Athenian  mercenaries.  Twice 
already  had  Persia  crushed  Egyptian  revolts,  and  she  was, 
perhaps,  but  watching  her  opportunity  to  cast  the  hordes  of  Asia 
upon  the  unhappy  country  for  a  third  time.  Ill  obeyed  as  he  was, 
the  "  Great  King  "  could  always  find  troops  to  take  part  in  the 
spoiling  of  a  country  whose  riches  had  proved  so  inexhaustible. 
And  if,  by  any  remote  chance,  the  Persians  should  fail  in  their 
enterprise,  another  and  a  graver  danger  would  menace  the 
Egyptian  monarchy  from  the  rapid  growth  of  the  Greek  power 
in  the  Mediterranean.  Since  the  period  of  the  Persian  wars, 
the  language,  the  literature,  the  arts,  the  mythology  of  Greece, 
had  spread  with  great  rapidity  ;  and  the  moment  might  be  foreseen 
when  a  supremacy  founded  upon  intellectual  worth  would  be 
confirmed  by  military  triumph  and  the  creation  of  a  vast  Hellenic 
empire.  The  conquest  of  Egypt  was  begun  by  the  Ionian 
soldiers  and  merchants  who  were  introduced  into  the  Nile  valley 
by  Psemethek  ;  it  was  bloodlessly  completed  by  the  arms  of 
Alexander.  For  three  centuries  the  Egyptians  had  been  accus- 
tomed to  see  the  Greeks  freely  coming  and  going  among  them 
as  merchants,  as  mercenary  officers,  as  travellers  eager  for 
instruction.  The  latter  posed  as  disciples  before  the  priests  of 
Memphis  and  Heliopolis,  and  freely  expressed  a  warmth  of  admira- 
tion which  could  not  fail  to  flatter  the  national  vanity.  The  Greeks 
would  be  better  masters  than  their  rivals  from  Persia.  From  them 
the  Egyptians  would,  at  least,  obtain  good  administration  and  com- 


yS  A   History  of  Art  in  Ancient  Egypt. 


plete  freedom  in  the  exercise  of  their  rehgion  in  return  for  their 
taxes.  The  Greeks  were  clear-sighted  enough  to  understand  their 
own  interests  ;  they  were  too  philosophical  and  large  minded  for 
any  fanatical  persecution  of,  or  even  hindrance  to,  the  national 
religion  ;  they  were  too  much  of  connoisseurs  to  fail  in  respect 
to  a  form  of  civilization  whose  prodigious  antiquity  they  divined, 
and  before  which  the  most  eminent  among  them  were  ever  inclined 
to  bow,  like  youths  before  an  old  man,  or  a  parvenu  before  the 
descendant  of  a  long  line  of  kings. 

Thus  Egypt  gradually  fell  into  the  hands  of  strangers  after  the 
commencement  of  the  fourth  century  before  Christ.  Ethiopians, 
Assyrians  and  Persians  had  by  turns  overrun  the  country.  Great 
numbers  of  the  Phoenicians  had  established  themselves  in  it,  and, 
after  the  fall  of  Jerusalem  and  Samaria,  many  Jews  followed 
their  example.  Finally,  the  Greeks  came  in  by  thousands  through 
the  breaches  which  their  predecessors  had  made,  penetrating  into 
all  parts,  and  making  everywhere  felt  the  superiority  of  a  people 
who  had,  by  appropriating  the  useful  results  cbtained  in  a  long 
succession  of  centuries  by  more  ancient  races,  become  wealthier, 
stronger,  and  better  instructed  than  any  of  their  forerunners. 

Thus  Egypt  lost  her  power  of  national  rejuvenation,  her  power 
of  rising  again  after  calamity.  She  existed  on  through  the  centu- 
ries by  mere  force  of  habit,  but  she  lived  no  more.  Her  population 
was  so  homogeneous,  and  her  institutions  were  so  solid,  that  the 
social  conditions  of  the  country  could  not  be  changed  in  a  day 
or  even  in  a  century.  The  teachings  of  her  religion  had  been 
established  by  so  long  a  course  of  development,  and  the  hands  of 
her  artists  were  so  well  practised,  that  the  monumental  types  which 
had  been  created  in  more  fertile  periods  of  her  history  were 
reproduced  until  a  late  date,  in  a  machine- like  and  instinctive 
fashion.  Imagination  was  dead,  and  the  best  that  could  be  hoped 
for  was  the  faithful  repetition  of  those  forms  which  the  genius  of 
the  race  had  conceived  in  its  last  moments  of  original  thought. 

Under  the  Salt  princes,  under  the  Psemetheks  and  Nekau, 
under  Apries  and  Amasis,  Egypt  was  delivered  from  her  enemies 
and  again  became  mistress  of  Syria  and  of  the  Island  of  Cyprus.  She 
thus  recovered  confidence  in  herself  and  in  her  future,  and  a  period 
ensued  which  had  an  art  of  its  own  with  distinctive  features 
which  we  shall  endeavour  to  trace.  In  the  intervals  of  pre- 
carious   repose  which   characterized   the    Persian   domination,    the 


Change  Observable  in   Egyptian  Art. 


79 


Egyptians  had  leisure  neither  to  invent  nor  to  improve.  They 
copied,  as  well  as  they  could,  the  monuments  of  the  twenty-sixth 
dynasty.  Art  became  a  mere  collection  of  technical  precepts,  kept 
together  and  transmitted  in  the  intercourse  of  the  studio,  by 
instruction   and    practice ;    it  became   a    mere    matter    of  routine 


Fig.  51. — Ouah-ab-ra,  26th  dynasty.     Louvre.     Grey  granite,  height  37  inches. 

implying,  perhaps,  great  technical  skill,  but  displaying  no  sincere 
and  personal  feeling  Nature  was  no  longer  studied  or  cared  for. 
Artists  knew  that  the  human  figure  should  be  divided  into  so  many 
parts.  They  knew  that  in  the  representation  of  this  or  that  god 
a  certain  attitude  or  attribute  was  necessary;  and  they  carved  the 


o  A   History  of  Art  in  Ancient  Egypt. 


statues  required  of  them  after  the  traditional  ■  recipes.  Thus 
Egyptian  art  became  conventional,  and  so  it  remained  to  the  end. 
So  it  was  in  the  time  of  Diodorus.  The  sculptors  whom  that 
historian  saw  at  work  in  Memphis  and  Thebes,  during  the  reign  of 
Augustus,  carved  a  statue  as  a  modern  mechanic  would  make  the 
different  parts  of  a  machine  ;  they  worked  wuth  a  rapidity  and  an 
easy  decision  more  characteristic  of  the  precise  workman  than 
of  the  artist.^  Thought  was  no  longer  necessary  to  them.  The 
due  proportions  and  measurements  had  been  ascertained  and  fixed 
many  centuries  before  their  time. 

But  research  must  still  precede  discovery.  We  admit  that  a 
day  arrived  when  convention  was  supreme  in  Egyptian  art,  but 
it  could  not  have  begun  with  convention  any  more  than  the  arts 
of  other  nations.  We  must  here  define  the  terms  which  we  shall 
have  occasion  to  employ.  Every  work  of  art  is  an  interpretation 
of  nature.  Let  us  take  the  example  of  the  human  figure.  In  the 
works  of  a  single  period  and  of  a  single  people,  it  is  always  full 
of  striking  similarity  ;  and  yet  two  original  artists  never  look  at  it 
with  the  same  eyes.  One  will  look  at  it  in  certain  aspects  and 
will  bring  out  certain  qualities,  which  another,  although  his  con- 
temporary and  fellow-countryman,  will  leave  in  the  obscurity  of 
shadows.  One  will  devote  himself  to  the  beauty  of  form,  another 
to  the  accidents  of  colour  or  the  expression  of  passion  and  thought. 
The  original  remains  the  same,  although  its  interpretations  are 
so  varioys.  And  these  varieties  become  still  more  marked  when 
we  compare  the  arts  of  difterent  races  or  of  different  periods — the 
art  of  Egypt  with  that  of  Assyria  or  Greece,  antique  art 
with  that  of  modern   times. 

On  the  other  hand,  the  great  resemblance  which  the  arts  of 
a  single  time  and  country  bear  to  each  other,  is  accounted  for  by 
the  fact  that  their  creators  look  upon  the  external  facts  of  life 
through  a  glass,  if  we  may  put  it  so,  tinted  with  the  colours  of  the 
national  genius.  They  bring  to  their  study  of  an  eternal  model 
the  same  transient  prejudices,  the  same  preoccupations,  the  same 
desires.  And  yet  among  those  highly  gifted  races  where  art 
holds  or  has  held  a  lofty  place,  groups  of  artists  are  formed, 
either  successively  or  simultaneously,  which  we  call  schools. 
Each  of  these  groups  professes  to  make  a  fresh  reference  to 
nature,    to    interpret    her    works    more    faithfully    than    its    pre- 

'  Diodorus,  i.  98,  7,  8. 


Change  Observable  in  Egyptian  Art. 


8i 


decessors,  and  to  draw  from  them  typical  forms  which  shall  be 
more  expressive  of  the  real  desires  and  sentiments  of  the  public 
for  which  it  caters.  Between  the  works  of  these  different  schools, 
there  are,  however,  many  similarities,  which  are  to  be  explained  by 
the  indentity  of  race  and  belief  There  are  also  diversities  which 
are  caused  either  by  different  conditions  or  by  the  influence  of 
some  master  spirit.  Wherever  these  schools  spring  up,  art  lives, 
moves,  and  progresses.      But  sooner  or  later  comes  a  time  when 


Fig.  52. — Sculptor  at  work  upon  an  arm,  Thebes.     (Champollion,  pi.  iSo.) 


this  ardour  comes  to  an  end,  and  exhaustion  takes  its  place. 
The  civilization  to  which  it  belongs  becomes  old  and  languid,  and 
its  creative  power  ceases  like  the  imperceptible  sinking  of  a  flood. 
Now,  it  often  happens  that  just  before  this  period  of  lassitude, 
in  the  last  days  of  reproductive  strength  and  healthy  maturity, 
a  rich  and  brilliant  school  springs  up,  which  interprets  the 
characteristic  sentiments  of  the  civilization  to  which  it  belongs, 
with  the  greatest  vigour  and  by  admirably  selected  means.  If 
such  an   interpretation    be  found  satisfactory  at    all    points,  why 


VOL.  I. 


M 


82  A  History  of  Art  in  Ancient  Egypt. 

should  a  better  be  souis^ht  for  at  the  risk  of  choosinof  a  worse  ? 
This  question  is  but  a  confession  of  impotence  on  the  part  of 
those  who  ask  it.  From  that  moment  convention  will  be  supreme, 
and  convention  in  the  sense  of  an  artificial  set  of  rules  which 
will  release  the  artist  from  his  obligation  of  continual  reference 
to  nature. 

Such  a  revolution  is  not  the  work  of  a  day.  Art  requires  time 
thus  to  inclose  itself  in  mere  mechanical  dexterity.  As  a  nation 
grows  old,  its  art,  like  its  literature,  continually  becomes  more  and 
more  conventional.  Every  great  period  or  school  leaves  to  the 
generations  that  come  after  it  types  which  have  made  a  vivid 
impression  upon  taste  and  imagination.  As  time  goes  on  these 
types  become  more  numerous  and  more  brilliant,  and  their  prestige 
increases  until  it  becomes  little  less  than  tyranny.  Society  can 
only  escape  from  its  thrall  at  the  expense  of  some  great  religious 
or  philosophical  revolution,  or  by  the  infusion  of  new  blood 
from  without.  And  these  changes  western  civilization  had  to 
undergo  in  the  early  centuries  of  our  era,  in  the  establishment 
of  Christianity,  the  invasion  of  the  barbarians,  and  the  fall  of 
the  Roman  Empire. 

Thanks  to  the  peculiar  circumstances  of  the  country,  Egyptian 
society  was  enabled  to  maintain  the  originality  of  its  genius  and 
the  vitality  of  its  institutions  with  unusual  success.  After  each 
period  of  internal  commotion  or  foreign  invasion,  the  Egyptians 
set  themselves  to  renew  the  chain  of  their  national  traditions. 
In  spite  of  the  foreign  elements  which  had  been  received  among 
them,  the  great  mass  of  the  people  remained  the  same  down 
to  the  latest  days  of  antiquity.  Heterogeneous  constituents  were 
absorbed  by  the  nation  without  leaving  any  apparent  trace.  The 
ideas  which  the  people  had  formed  for  themselves  of  the  ultimate 
destiny  of  humanity  were  developed,  indeed,  and  in  successive 
agfes  varied  sli^htlv  in  greneral  colour,  but  in  none  of  their 
variations  did  they  give  rise  to  a  new  religion,  as  Brahmanism 
gave  birth  to  Buddhism. 

As  often  as  a  new  dynasty  of  kings  succeeded  in  driving  out 
the  foreign  conqueror  and  in  re-establishing  the  unity  of  the 
kingdom,  so  often  was  there  a  complete  restoration.  The  aim 
which  they  had  in  view  was  ever  to  restore,  in  all  its  parts,  a 
regime  which  was  founded  upon  national  pride.  Enjoying  a 
civilization  which  for  agfes  had  been  alone  in  the  world,  it  was  in 


Change  Observable  in   Egyptian  Art. 


8 


its  full  and  glorious  past  that  Egyptian  society  found  the  ideal  to 
which  it  clung  in  spite  of  all  obstacles  and  misfortunes.  Its  gaze 
was  turned  backwards  towards  those  early  sovereigns  who  seemed 
transfigured  by  distance,  but  whose  presence  in  the  memory  kept 
alive  the  perpetual  worship  which  had  been  vowed  to  them. 

Every  restoration  is  inspired  by  a  more  or  less  blind  and 
superstitious  reverence  for  the  past.  This  has  often  been  asserted 
in    connection    with    politics    and    religion,    and    the    assertion    is 


Fig.  53. — Sculptoi-  carving  a  statue,  Thebes.     (Champollion,  pi.  180.) 


equally  true  in  respect  to  art.  Each  of  those  dynasties  to 
which  Egypt  owed  its  political  restoration,  set  themselves  to 
repair  the  temples  which  had  been  destroyed,  and  to  replace 
upon  their  pedestals  the  statues  of  gods  or  ancestors  which  had 
been  overthrown.  When  new  temples  and  new  statues  were 
to  be  erected,  the  first  idea  of  the  artists  employed  was  to  study 
the  ancient  monuments  and  to  try  to  equal  them.  As  long  as 
Egypt  preserved  her  vitality,  the  wants  of  the  present  and 
external    influences    no    doubt    had    their    effect    in    introducing: 


84  A   History  of  Art  in  Ancient  Egypt. 

certain  changes,  both  in  the  arrangement  of  her  buildings,  and  in 
the  modelHng,  movement,  and  expression  of  the  statues  which 
adorned  them.  Ancient  types  were  not  servilely  copied,  but  the 
temptation  to  borrow  from  them  a  point  of  departure,  at  least, 
for  new  attempts  at  progression,  was  too  strong  to  be  resisted. 
It  was  necessary  that  all  buildings  and  statues  should  be  in 
harmony  with  the  remains  which  subsisted  from  previous  ages, 
and  from  this  it  resulted  that  each  new  creative  effort  began  by 
imitating  what  had  gone  before.  The  '  school '  in  process  of 
foundation  accepted  on  trust  the  architectural  disposition  left  by 
its  predecessor,  as  well  as  its  methods  of  looking  at  nature.  And 
this  is  equivalent  to  saying  that,  from  its  first  moment,  it  must 
have  been  conventional  in  a  certain  degree. 

This    conventionality    must    have    increased    at    every    fresh 
renascence,  because  each  new  development  had  its  own  processes 
to  transmit  to  posterity  as  well  as  those  of  its  ancestors.     After 
each  recoil  or  pause  in  the  progress  of  art,  the  weight  of  the  past 
must  have  seemed  heavier  to  those  who  attempted  to  revive  the 
onward  movement.     On  the  one  hand,   the  more  ancient  of  the 
traditional    elements    had    acquired,   by  their  constant   and   often 
repeated  transmission,  a  prestige  and  authority  which  jDlaced  them 
above   discussion  ;  on    the   other,   the   legacy  of  admitted   princi- 
ples and  processes  was  continually  increasing,  until  it  became  a 
source  of  embarrassment  to  the  artist,  and  of  destruction  to  his 
liberty.     When  at  last  the  decadence  of  the  race  had  advanced  so 
far  that  all  initiative  power  and  independence  of  thought  had  dis- 
appeared, the  time  arrived  when  convention  was  everything,  like 
one  of  those  elaborate  rituals  which  regulate  every  word,  and  even 
gesture,  of  the  officiating  priest.     When  Plato  visited  Egypt,  the 
schools  of  sculpture  were  nothing  more  than  institutions  for  teach- 
ing pupils,  who  were  remarkable  for  docility  and  for  dexterity  of 
hand,  to  transmit  to   their  successors  an   assemblage  of  precepts 
and   receipts  which   provided   for  every   contingency  and  left  no 
room  for  the  exercise  of  fancy  or  discretion. 

At  that  very  time  Greek  art  was  progressing  with  a  power  and 
rapidity  which  has  never  been  rivalled.  To  the  school  of  Phidias, 
a  school  established  in  that  Athens  which  yet  possessed  so  many 
works  of  the  archaic  period,  had  succeeded  those  of  Praxiteles 
and  Scopas.  The  Greeks  found  means  to  improve,  or  at  least  to 
innovate,    upon   perfection  itself.      Plato  did   not,  and  could  not, 


Change  Observable  in  Egyptian  Art. 


85 


perceive,  in  his  hasty  journey  through  the  Egyptian  cities,  that 
they  too  had  seen  their  periods  of  change,  their  different  schools 
and  developments  of  style,  less  marked,  perhaps,  than  those  of 
Greece,  and  certainly  less  rapid,  but  yet  quite  perceptible  to  the 
practised  observer.  We  are  now  in  a  better  position  to  estimate 
these  differences.  Monuments  have  been  brought  before  our  eyes 
such  as  Plato  never  saw  ;  namely,  the  statues  of  the  ancient 
empire  which   were   hidden  for  so  many  ages  in  the  thickness  of 


Fig.  54. — Artist  painting  a  statue,  Thebes.     (Champollion,  pi.  iSo.) 


walls  or  in  the  depths  of  sepulchral  pits.  Even  now  these  statues 
have  not  reached  the  age  of  ten  thousand  years  so  persistently 
attributed  by  the  Greek  philosopher  to  the  early  works  which  he 
did  see,  works  which  seemed  to  him  exactly  the  same  as  those 
which  were  being  made  in  his  presence.  But  although  the  statues 
of  the  early  empire  were  then  no  more  than  some  thirty  centuries 
old,  Plato  could  not  have  helped  seeing,  if  he  had  seen  them  at  all, 
that  they  were  quite  distinct  from  the  works  which  the  sculptors  of 


86  A  History  of  Art  in  Ancient  Egypt. 

Nectanebo  had  in  progress,  always  supposing  that  he  looked  at 
them  with  reasonable  attention.  The  art  of  the  pyramid  builders, 
an  art  which  possesses  in  a  very  high  degree  certain  qualities  for 
which  the  Egyptians  have  been  too  commonly  refused  credit,  is 
known  to  us  chiefly  through  the  excavations  of  Mariette  and  the 
contents  of  the  Boulak  museum.  But  even  before  Cheops, 
Chefren,  and  their  subjects  had  risen  from  their  tombs,  the 
historian  might  have  divined  by  analogy,  and  described  by  no  very 
bold  conjecture,  the  essential  charateristics  of  Egyptian  art  during 
its  first  centuries.  Whether  we  speak  of  an  individual,  of  a 
school,  or  of  a  people,  every  artistic  career  which  follows  its 
natural  course  and  is  not  rudely  broken  through,  ends  sooner  or 
later  in  conventionality,  in  that  which  is  technically  called 
mannerisjn.  But  mannerism  is  never  the  beginning  of  art.  Art 
always  begins  by  humble  and  sincere  attempts  to  render  what  it 
sees.  Its  awkwardness  is  at  first  extreme  and  its  power  of 
imitation  very  imperfect.  But  it  is  not  discouraged  ;  it  tries 
different  processes  ;  it  takes  account  now  of  one,  now  of  another 
aspect  of  life  ;  it  consults  nature  incessantly  and  humbly,  taking 
note  of  her  answers  and  modifying  its  work  in  obedience  to  their 
teaching.  This  teaching  is  not  always  rightly  understood,  but  it 
is  ever  received  with  docility  and  good  faith. 

Every  work  which  bears  the  marks  of  frank  and  loyal  effort  is 
interesting  ;  but  the  moment  in  an  artistic  career  which  gives  birth 
to  real  chefs  cCoeiivre  is  towards  the  end  of  that  period,  when  the 
eye  has  become  sure,  and  the  hand  sufficiently  well  practised,  for 
the  faithful  interpretation  of  any  model  whose  beauty  or  original 
expression  may  have  caught  the  fancy.  Success  is  then  achieved, 
always  provided  that  the  model  is  never  lost  sight  of  or  studied 
with  anything  short  of  passionate  devotion.  But  the  time  comes 
when  this  devotion  is  relaxed.  The  artist  thinks  that  such  con- 
stant reference  to  nature  is  no  longer  required  when  he  has  made 
his  final  choice  between  the  different  methods  which  his  art 
employs.  In  devoting  himself  to  the  reproduction  of  certain 
features  for  which  he  has  a  marked  preference,  he  has  himself 
produced  types  which  he  thenceforward  takes  pleasure  in  repeating, 
as  if  they  were  in  themselves  an  epitome  of  nature's  infinite 
diversity. 

In  the  case  of  Egypt,  even  those  discoveries  which  carry  us 
back  farthest  do  not  enable  us  to  grasp,  as  we  can  in  the  case  of 


Fig.  55. — Ibis  nursing  Horus.     Ptolemaic  bronze  ;  in  the  Louvre.     Height,  19  inches. 


The  Monuments  of  the  Memphite  Period.  89 

Greece,  the  first  attempts  at  plastic  expression,  the  first  rude  efforts 
of  the  modeller  or  painter ;  but  they  carry  us  to  the  end  of  that 
period  which,  in  the  case  of  other  countries,  we  call  archaic  ;  and 
above  all  they  transport  us  into  the  centre  of  the  epoch  which  was 
to  Egypt  what  the  fifth  century  was  to  Greece,  namely,  the  age  of 
perfection.  The  Egyptian  people  had  already  lived  so  long  and 
worked  so  hard  that  they  could  not  free  their  work  from  certain 
common  and  irrepressible  characteristics.  In  the  plastic  arts  and 
in  poetry  they  had  their  own  style,  and  that  style  was  both  indi- 
vidual and  original  in  an  extraordinary  degree.  This  style  was 
already  formed,  but  it  was  not  yet  robbed  of  its  vitality  by  indolent 
content  or  petrified  by  mannerism  ;  it  had  neither  renounced  its 
freedom  nor  said  its  last  word. 


•  e 


§   7.   Of  the  place  held  in  this  work  by  the  inoiitinie7its  of  the 
Memphite  period,  and  of  the  limits  of  our  inquiry. 

It  will  be  found  that  a  very  large  space  in  the  present  work, 
some  may  say  too  large  a  space,  is  devoted  to  the  pre-conventional 
art  of  the  ancient  empire.  We  had  reasons  for  taking  such  a 
course,  and  reasons  that  may  be  easily  divined. 

This  early  art  is  much  less  known  than  that  of  the  later  epochs. 
While  the  great  museums  of  Europe  are  filled  with  statues  and 
reliefs  from  Thebes,  or,  at  least,  contemporary  with  the  Theban 
and  Sait  dynasties,  monuments  from  the  Memphite  period  are 
still  rare  out  of  Egypt.  Thanks  to  Mariette  and  Lepsius,  Paris 
and  Berlin  are  not  without  remarkable  examples  of  the  art  in 
question,  but  it  is  in  Egypt  itself,  at  the  Boulak  museum,  that  any 
detailed  study  must  be  made.  It  is  there  that  the  masterpieces  of 
an  art  whose  very  existence  was  unsuspected  by  Champollion,  are 
to  be  found  ;  the  Chephren,  the  two  statues  from  Meidoum,  the 
bas-reliefs  from  the  tomb  of  Ti,  and  many  others  of  similar  style 
and  value.  These  figures  have  been  drawn  for  our  readers  by  two 
skilful  artists,  MM.  Bourgoin  and  Benedite.  They  have  rendered 
with  fidelity  and  sincerity  more  than  one  object  which  had  never 
before  been  reproduced,  either  by  photography  or  otherwise.  A 
few  specimens  of  these  treasures,  selected  by  him  who  had  been 
the  means  of  brin^ino-  them  to  lieht  and  whom  we  now  mourn, 
were   seen  at  the    Universal    Exhibitions   of  1867  and    1878,  but 

VOL.  I.  N 


90 


A  History  of  Art  in  Ancient  Egypt. 


'^1^ 


they  soon  returned  to  Cairo,  and  western  archaeologists  had  but 
sh'ght  opportunity  to  become  acquainted  with  their  characteristics. 
The  art  of  the  early  dynasties  has  thus  been  practically  ignored 
by  those  who  have  never  visited  Egypt.  The  lifelike  and  en- 
thusiastic descriptions  of  M.  Eugene  Melchior  de  Vogue  and 
others  have  done  something  to  arouse  the  attention  of  connoiseurs  ; 
but  in  such  a  matter  the  slightest  sketch,  provided  it  be  correct  so 
far  as  it  goes,  is  of  more  value,  as  a  definition  of  style,  than  the 
most  picturesque  or  eloquent  writing. 

These  reflections  would  by  themselves  justify  our  efforts  to  in- 
corporate in  our  pages,  reproductions  of 
all  the  more  important  objects  with  which 
the  necropolis  at  Memphis  has  enriched 
the  museum  at  Boulak ;  but  we  were 
impelled  by  other  motives  also.  The 
extant  monuments  of  the  ancient  empire 
are  less  numerous  than  those  of  the 
Theban  and  Salt  dynasties ;  they  are  of 
comparatively  modest  dimensions,  and, 
with  rare  exceptions  they  all  belong  to 
one  category,  that  of  works  relating  to 
death  and  burial.  They  also  have  a 
special  interest  of  their  own.  They  enable 
us  to  protest,  and  to  give  tangible  justifi- 
cation for  our  protestations,  against  a 
prejudice  which  dates  back  to  a  remote 
antiquity  ;  even  if  all  evidence  had  perished 
I  I  the   critic   would   have   no   great  difficulty 

Fig.  56.— Chephren.   Sketched     in    Casting    doubt    upou    assertious    which 
byBourgoin.   See  also  Fig.  460.     ^^^^  j^  themselves  extremely  improbable, 

but  his  task  is  rendered  much  easier  when 
he  is  able  to  point  to  existing  monuments  in  support  of  his 
contention,  and  his  pleasure  is  great  in  seeing  the  certainty 
of  his  critical  methods  borne  out,  and  Egyptian  art  replacing 
itself,  as  if  of  its  own  motion,  under  the  normal  conditions  of 
historic   development. 

This  volume,  then,  will  treat  of  the  remains  of  early  Egyptian 
art  at  a  length  which  would  seem  at  first  sight  out  of  due  pro- 
portion to  their  number,  but  later  ages  will  also  be  represented  by 


a  series  of  monuments,  which  will  bring  us  down  to   the   Persian 


The  Limits  of  our  Inquiry. 


91 


conquest.      This  limit  will  hardly  be  over-passed  in  our  choice  of 
examples  for  study,  and  that  for  two  reasons. 

The  first  is,  that  at  the  latter  period  the  evolution  of  Egyptian 
art  was  complete,  it  had  created  all  that  it  could  and  had  become 
a  slave  to  its  own  past.  Disposing  under  the  Ptolemies  of 
all  the  resources  of  a  great  empire,  it  indeed  introduced  certain 
architectural  changes  v/hich  do  not  seem  to  have  been  borrowed 
frOm  previous  buildings,  but  those  changes  were  of  no  very  great 
importance  and  were  mostly  in  matters  of  detail.      In   sculpture 


Fig.  57.^ — Ti,  with  his  wife  and  son. 


and  painting  we  can  easily  see  that  it  abandoned  itself  to  mere 
copying,  to  the  repetition  of  a  lesson  learnt  by  rote.  Whatever 
had  to  be  done,  was  done  in  accordance  with  fixed  tradition,  and 
one  monument  only  differed  from  another  in  the  amount  of  care 
and  manual  dexterity  bestowed  upon  it. 

Our  second  reason  is  this,  that  Egypt  was  opened  to  the  Greeks 
in  the  time  of  the  Sait  princes.  From  the  year  650  B.C.  onwards, 
there  was  constant  communication  between  Ionia  and  the  cities  of 


92  A   History  of  Art  in  Ancient  Egypt. 


the  Delta.  If  at  anytime  Greek  art  borrowed  directly  from  that 
of  Egypt,  it  was  during  the  second  half  of  the  seventh  century 
and  the  first  half  of  the  sixth.  By  the  end  of  the  sixth  century, 
it  had  become  so  oricjinal  and  so  skilful  in  the  manao-ement  of  its 
selected  methods  of  expression  that  it  could  not  have  been  very 
receptive  to  foreign  influences.  After  the  Persian  wars  such  in- 
fluences would  be  still  more  powerless.  In  the  Ptolemaic  era  the 
state  of  things  was  reversed  ;  Greece  imposed  her  language,  her 
literature,  her  religious  conceptions  and  their  visible  symbols  upon 
the  whole  eastern  world.  Even  then  the  art  of  Egypt  could  de- 
fend, and  even  perpetuate  itself,  by  the  power  of  custom  and  of  a 
tradition  which  had  been  handed  down  through  so  many  centuries, 
but  the  day  was  past  when  it  could  provoke  imitation. 

As  for  the  indirect  borrowings  of  forms  and  motives  which 
Greece  received  from  Egypt  through  the  Phoenicians,  their  trans- 
mission had  come  to  an  end  before  the  Persian  conquest,  even 
before  the  time  of  Psemethek.  Egypt  was  represented,  either 
immediately  or  through  the  imitative  powers  of  the  Syrian  manu- 
facturers, in  the  first  textiles,  jewels,  and  vases  of  clay  or  metal, 
carried  by  the  Sidonian  merchants  to  the  savage  ancestors  of  the 
Greeks.  In  this  roundabout  manner  she  had  probably  more  in- 
fluence over  Greece  than  in  their  periods  of  more  direct  commu- 
nication. The  rays  kindled  upon  her  hearth,  the  earliest  of  civili- 
zation, fell  upon  the  Hellenic  isles  as  refracted  rays,  after  passing 
through  the  varied  media  of  Chaldaea,  Assyria  or  Phcenicia. 

Thus  if  we  wish  thoroughly  to  understand  Greece,  we  must  start 
from  Memphis  and  go  through  Babylon  and  Nineveh,  Tyre  and 
Sidon.  But  Greece  will  be  the  aim  of  our  voyage,  and  Egypt  will 
interest  us  less  on  her  own  account  than  on  account  of  that  unique 
and  unrivalled  people  who  inherited  her  inventions  and  discoveries, 
and  made  them  the  foundation  for  a  productiveness  in  which  are 
summarized  all  the  useful  labours  of  antiquity.  Egyptian  art  will 
be  followed  by  us  down  to  the  moment  in  which  it  lost  its  creative 
force  and  with  it  its  prestige.  We  shall  rarely  have  occasion  to 
speak  of  the  Ptolemaic  remains  of  Egyptian  art.  Now  and  then 
we  shall  go  to  theni  for  examples  when  any  particular  detail  which 
we  desire  to  mention  has  not  been  preserved  for  us  by  earlier 
monuments,  but  even  then  we  shall  require  to  have  good  reasons 
for  believing  that  such  detail  did  in  fact  originate  in  the  creative 
periods  of  the  national  history. 


The  Limits  of  our   Inquiry. 


93 


The  Egypt  of  the  Pharaohs  has  not  even  yet  been  entirely 
explored.  Are  we  to  believe  that  the  splendid  edifices  reared  in 
the  cities  of  the  Delta,  and  especially  at  Sais,  by  the  twenty-sixth 
dynasty,  have  perished  to  the  last  stone  ?  We  are  loth  to  think 
that  it  is  so,  but  no  remains  have  yet  been  discovered.  Some 
day,  perhaps,  well  directed  excavations  may  bring  to  light  the 
temples  which  Herodotus  so  greatly  admired  ;  and  who  knows 
but  that  we  may  find  in  them  more  than  one  of  those  motives 
and  arrangements  which  at  present  are  only  known  to  exist  in  the 
buildings  of  the  Ptolemies  and  of  the  Roman  emperors  ? 


CHAPTER  II. 

PRINCIPLES      AND       GENERAL       CHARACTERLSTICS       OF       EGYPTIAN 

ARCHITECTURE. 

§  I. — Method  to  be  Employed  by  iis  in  our  Stiidy  of  this  Architecture. 

In  the  enterprise  which  we  have  undertaken  the  study  of  oriental 
art  is  but  an  introduction  to  that  of  Greece.  Without  an  attentive 
examination  of  its  remains  we  should  be  unable  to  distinguish  the 
original  elements  in  the  work  of  the  Greek  genius  from  those 
which  it  borrowed  from  other  nations.  We  must  pass  in  review 
the  whole  artistic  production  of  several  great  nations  who  occupied 
a  vast  surface  of  the  globe,  and  whose  fertility  was  prolonged 
through  a  long  course  of  centuries,  but  we  shall  not  attem[)t  to 
describe  singly  the  great  buildings  of  Egypt  and  Assyria,  of  Persia 
and  Phoenicia,  as  such  an  attempt  would  perhaps  cause  us  to  lose 
sight  of  the  main  object  of  our  work. 

Our  task  is  no  easy  one.  While  limiting  our  study  in  the 
fashion  which  has  been  described,  we  must  not  fail  to  extend 
our  purview  to  every  fact  which  may  help  to  justify  the  comparison 
which  we  propose  to  institute  between  the  arts  of  Greece  and 
those  of  the  nations  by  whose  teachings  she  profited.  There  is 
but  one  road  to  success  in  this  double  task.  We  must  devote  the 
greatest  possible  care  to  our  study  of  the  details  in  question,  and 
then  give  the  general  results  of  that  study  ;  we  must  make  our- 
selves thoroughly  acquainted  with  all  the  phenomena,  but  must 
confine  our  exposition  to  the  general  laws  which  governed  them, 
such  as  our  minute  inquiries  have  presented  them  to  us.  No 
circumstantial  description  need,  therefore,  be  looked  for  in  these 
pages  even  in  the  case  of  the  most  important  and  famous  buildings 
of  Egypt.  No  monograph  upon  any  tomb  or  temple  will  be 
found,  but  we   shall  ourselves    have   examined   many  tombs  and 


Our  Method  in  the  Study  of  this  Architecture.      95 

temples  ;  we  shall,  to  speak  figuratively,  have  taken  them  to  pieces, 
and  by  means  of  the  knowledge  acquired  we  shall  endeavour  to 
make  our  readers  acquainted  with  the  notions  of  the  Egyptians 
upon  sepulchral  and  religious  architecture,  and  with  the  changes 
which  those  conceptions  underwent  in  the  course  of  centuries. 

Thus,  for  example,  we  have  explored  the  pages  of  Lepsius  ^  and 
Prisse  d' Avenues '  for  information  relating  to  the  sepulchres  of  the 
first  six  dynasties,  and  further  researches  have  been  made  on  the 
spot  expressly  for  the  present  work,  but  we  shall  not  give  any 
descriptions  or  illustrations  of  those  works  individually  ;  we  shall 
merely  use  them  for  an  ideal  restoration  of  the  characteristic  tomb- 
house  of  the  ancient  empire.  We  may,  perhaps,  for  this  purpose, 
make  a  more  particular  reference  to  one  or  two  sepulchres  which 
are  in  unusually  good  preservation,  but  only  for  the  sake  of  giving 
firm  definition  to  the  type  and  to  its  main  variations. 

By  this  analytical  method  of  treatment  we  shall  be  enabled  to 
give  an  account,  which  shall  be  at  once  accurate  and  not  too  long, 
of  the  constructive  processes  employed  by  the  Egyptians,  of  the 
general  aspect  of  their  buildings,  and  of  the  modifications  enforced 
by  the  decorative  forms  of  which  they  made  use.  We  shall  be 
enabled  to  see  how  far  those  forms  were  decided  by  natural 
conditions,  by  ancient  tradition,  or  by  special  wants.  We  shall 
thus  include  in  a  single  chapter  all  that  relates  to  principal  or 
accessory  openings,  to  doors  and  their  construction,  to  those  loftily 
placed  windows  which  were  calculated  to  give  so  little  light.  In 
another  chapter  we  shall  discuss  the  column  and  its  capital ;  we 
shall  describe  the  variations  produced  by  time  and  materials  upon 
its  proportions  and  its  entasis.  Each  assertion  will  be  justified  by 
reference   to  characteristic    examples.       In    this    matter  our   only 

1  DenkvicEler  aus  ^^^ypten  und  ^thiopie)i  (from  drawings  of  the  expedition  sent 
into  Egypt  in  1842,  which  remained  there  till  1845),  12  vols,  folio.     Berlin,  no  date. 

2  Histoire  de  I'Art  egyptien  d'aprh  les  Monuments  depuis  les  Temps  les  plus  recuces 
jusqua  la  Domination  romaine,  2  vols.  Paris,  Arthus  Bertrand,  1878.  The  text 
(i  vol.  4to.),  published  after  the  death  of  Prisse,  has  this  great  inconvenience,  that 
i-t  is  not  always  easy  to  distinguish  what  belongs  to  the  editor,  M.  Ivlarchandon  de 
la  Faye,  from  the  contributions  of  Prisse,  who  was  one  of  the  most  practical  and  ex- 
perienced of  egyptologists.  The  papers,  sketches,  and  drawings  left  by  Prisse  became 
the  property,  in  1880,  of  the  Bibliotheque  Nationale ;  when  they  are  classified  and 
published  we  shall  probably  find  among  them  several  interesting  documents ;  we 
have  only  been  able  hurriedly  to  look  through  them,  when  the  illustrations  to 
this  work  were  already  prepared.  It  is  desirable  that  a  complete  inventory  of  these 
collections  should  be  made  as  soon  as  possible. 


96  A  History  op^  Art  in  Ancient  Egypt. 

difficulty  will  be  an  embarras  de  richesse,  a  difficulty  of  choice 
among  the  vast  number  of  remains  still  existing  of  ancient  Egypt 
from  the  time  of  Menes  to  that  of  the  Persian  conquest. 

In  order  to  avoid  repetition  and  to  put  before  the  reader  ideas 
which  he  w^ill  have  no  difficulty  in  assimilating,  we  shall  push  our 
work  of  analysis  and  generalization  farther  still.  Before  we 
embark  upon  the  study  of  any  special  class  of  buildings  we  shall 
endeavour  to  define  the  general  and  unchanging  characteristics 
of  Egyptian  architecture  as  a  whole ;  characteristics  which  were 
fixed  by  the  idiosyncracy  of  the  race,  by  its  beliefs  and  social 
customs,  by  the  nature  of  the  climate,  and  of  the  materials  of 
which  the  architect  could  dispose.  We  shall  do  the  same  for 
Assyria  and  Chaldaea,  for  Persia  and  Phcenicia,  for  each,  indeed, 
of  the  nationalities  which  are  to  be  considered  in  our  history. 

These  theoretical  chapters  will  be  illustrated  in  the  same  fashion 
as  the  others,  except  that  the  illustrations  will  partake  of  the 
generalized  and  abstract  character  of  the  text  which  they  accom- 
pany. In  most  cases  they  will  be  simple  diagrams  composed  for 
the  express  purpose  of  illustrating  the  definitions  or  descriptions 
to  which  they  belong.  They  will  each  refer  to  some  essential 
element  in  the  national  architecture,  to  some  element  which  is 
not  peculiar  to  any  one  edifice  more  than  another,  but  is  to  be 
found  in  all  those  which  have  similar  aims  and  are  constructed  of 
the  same  materials.  Such  elements  are  above  and  outside  such 
accidental  variations  as  may  be  found  in  details  of  plan  or 
ornament  ;  they  form  part  of  the  substantial  inner  constitution  of 
the  arts  of  Egypt  and  Chaldaea,  and  make  their  originality 
indisputable. 

§  2. —  General  Principles  of  Form. 

The  external  forms  of  Egyptian  edifices  are  pyramidoid ;  in 
other  words,  the  outward  surfaces  of  their  walls  affect  the  form  of 
a  trapezium.  Thus  if  we  prolong  these  surfaces  vertically  we  find 
that  they  unite  at  last  in  a  point,  in  the  case  of  square  buildings 
(Fig.  58),  and  in  a  ridge  in  those  which  are  oblong  in  plan 
(Fig.  59).-^     A  square  building   will  sometimes  end  in  a  ridge,  or 

1  Lois  generales  de  V Itidinaison  des  Coloimes  datis  la  Constniciion  des  Temples  grecs 
de  rA?itiquite,  dedicated  to  his  Majesty,  Otho  I.,  by  Charles  Villeroi,  engineer. 
Athens,  1842,  8vo. 


General  Principles  of  Form. 


97 


arete,  when  the  principal  facade  and  the  corresponding  one  in  its 
rear  are  vertical,  the  other  two  being  incHned. 

Horizontal  lines  piedominate  over  inclined  or  vertical  lines,  and 
buildings,  therefore,  tend  to  develop  in  length  and  depth  rather 
than  in  height.  To  this  general  ride,  however,  the  pylons  afford 
exceptions. 


Fig.  58. — Square  building. 

The  terminations  of  their  edifices  w^ere  also  horizontal.  There 
was  no  necessity  for  sloping  roofs,  as,  away  from  the  immediate 
proximity  of  the  sea,  it  hardly  ever  rains  in  Egypt.  Moreover, 
the  natural  conformation  of  the  country  had  its  influence  upon 
the  creations  of  its  inhabitants.  The  unforeseen  and  sudden 
variations,    the   contrasts    of   hill   and   plain,   which    we   find   in    a 


--iriL 


— =^-.---5il^j 


Fig.  59  — Rectangular  and  oblong  building. 


mountainous  country  like  Greece,  are  here  unknown.  Lower 
Egypt  is  a  verdant  plain,  intersected  by  canals,  and  stretching 
from  the  sea  to  the  desert;  in  Upper  and  Middle  Egypt  the  lazy 
river  is  accompained  throughout  its  journey  from  south  to  north 
by  two  long  chains  of  hills,  the  Arab  chain  and  the  Libyan,  whose 
summits  form  an  almost  unbroken  line.      Between   these  aspects 

VOL    I.  o 


98 


A   History  of  Art  in  Ancient  Egypt. 


of  nature  and  the  works   of  man   which   they  enframe,  there  is 
a  striking"  general  sympathy.^ 

The  pecLiHar  character  of  Egyptian  architecture  is  owing  to 
its  laternal  extension,  and  to  those  wide-spreading  bases  and 
foundations  which  suggest  the  inchnation  of  the  superincumbent 
walls.  In  looking  at  one  of  these  buildings,  we  feel  that  it  is 
capable  of  infinite  extension  horizontally,  and  that  but  one  of  its 
dimensions,  that  of  height,  is  limited  by  its  essential  forms. 
These   characteristics   give  a  look  of  sturdy  power  to    Egyptian 


Fic,.  60. — The  Libyan  chain,  above  the  necropolis  of  Thebes. 


architecture   which    is   peculiar  to   itself,   and   suggests   an   idea  of 
unbounded  durability.^ 

1  Egyptian  landscape  is  well  characterised  in  these  lines  of  M.  Ch.  Blanc,  taken 
from  the  Voyage  de  la  Haute- Egypte  (p.  116):  "Pour  le  moment,  notre  plaiser  se 
borne  a  regarder  un  paysage  simple,  monotone,  mais  grand  par  sa  simplicite  meme  et 
par  sa  monotonie.  Ces  lignes  planes  qui  s'allongent  et  se  prolongent  sans  fin,  et  qui 
s'interrompent  un  instant  pour  reprendre  encore  leur  niveau  et  se  continuer  encore, 
impriment  a  la  nature  un  caractere  de  tranquillite  qui  assoupit  I'imagination  et  qui 
apaise  le  coeur.  Par  une  singularite  peut-etre  unique  au  monde,  les  varie'te's  qui 
viennent  rompre  de  distance  en  distance  la  vaste  uniformite'  de  la  terre  egyptienne 
se  reproduisent  toujours  les  memes."  [We  have  refrained  from  translating  this  piece 
of  word  painting,  lest  its  suggestive  rhythm  should  vanish  in  the  process. — Ed. J 

2  Similar  notions  are  expressed  by  M.  Ch.  Blanc  in  his  Grainmaire  des  Arts  dii 
Dessin  (book  i.  ch.  viii.).  "  The  wide-spreading  base  is  the  distinguishing  charac- 
teristic of  the  Egyptian  monuments.  Wall,  pier,  and  column,  all  the  constructive 
members  of  Egyptian  architecture,  are  short  and  thick  set.  To  add  to  this  appear- 
ance of  solidity  the  relative  size  of  the  base  is  increased  by  that  tendency  towards 
the  pyramid  which  is  to  be  found  in  every  Egyptian  building.  The  pyramids  of 
Memphis,    one  of  them  the  greatest  building  upon  earth,  stand  upon  enormous 


General  PRiNcirLEs  of  Form. 


99 


An  appearance  of  incomparable  gravity,  of  solemnity,  is  also 
stamped  upon  it  by  the  small  number  of  openings  for  the 
admission  of  light  of  which  it  makes  use,  and  also  by  their 
arrangement.  Compared  to  our  modern  architecture,  in  which 
windows  play  such  an  important  part,  that  of  Egypt  is  prison- 
like in  its  gloom  ;  but,  in  consequence  of  its  rare  openings  and 
their  small  size,  it  presents  more  imposing  walls  than  any 
other  style. 

One  of  the  essential  arrangements  of  Egyptian  architecture 
is  shared  by  many  other  countries,  that  of  the  p07'tico,  by  which 
we  mean  an  alternation  of  voids  and  solids  in  certain  well  defined 


iryg[Ini:!jlTUjjj(.j]g^^ 


Fig.  6i. — General  appearance  of  an  Egyptian  Temple.* 


proportions,  either  for  ornamenting  the  exterior  and  providing 
a  covered  way,  or  for  dividing  the  halls  of  the  interior  and 
supporting   their   roofs. 

The    relation    between    voids    and     solids     in     any    style    of 
architecture   is  one  of  the  most  vital  characteristics. 

bases.  Their  height  is  far  less  than  their  largest  horizontal  diameter.  The  pyramid 
of  Cheops,  for  instance,  is  233  metres  along  one  side  of  its  base,  and  only  146  in 
height,  i.e.,  its  base  is  to  its  height  as  8  to  5.  All  Egyptian  monuments,  even  the 
most  lofty,  are  more  remarkable  for  the  ground  they  cover  than  for  their  height 
[except  the  monoliths  ! — Ed.],  and  this  extension  of  their  bases  gives  them  an 
appearance  of  absolutely  eternal  durability." 

^  This  illustration  has  been  compiled  in  order  to  give  a  general  idea  of  the  more 
persistent  characteristics  of  the  Egyptian  temple. 


TOO 


A   History  of  Art  int  Ancient  Egypt. 


In  the  case  of  Egypt  this  relation  gives  rise  to  the  following- 
remarks  : 

I.  Supports  of  the  same  kind  and  of  the  same  diameter  may 
have  very  different  heights  in  one  and  the  same  building 
(Fig.   62). 


r3m 


I         M 


Fig.  62.— Temple  of  Khons,  at  Thebes.     (Description  de l' Agypte,  t.  iii.,  pi.  55.) 


t 

"^^ 

ii#Wfl#ffl* 

Ji- 


Fig.  63. — Temple  of  Khons,  Thebes.     {Descr. 
derEg)'pte,  t.  iii.,  pi.  55.) 


Fig.  64. — Temple  of  Khons,  Thebes.     \Drscr. 
deV&^ypte,  t.  iii.,  pi.  55.) 


In  a  single  edifice  supports  of  different  kinds  but  of  the  same 
diameter,  have  no  fixed  proportions,  one  to  the  other.  A  column 
with  a  lotus  capital  may  be  higher  than  one  with  a  bell-shaped 
termination,  and  vice  versa  (Figs.  63  and  64),  while,  again  in  a 
single  building,  we  may  find  these  two  differently  shaped  columns 


General  Principles  of  Form. 


lOT 


equal    to    each    other    both  in    average    diameter    and    in    height 
(Fig.  65). 


T-V 


Fig.  65.— From  the  second  court  of  Medinet-Abou,  Thebes.     {Description,  t.  ii.,  pi.  6.) 

2.  The  spaces,  or  voids,  between  columns  of  one  size  and 
similar  design,  may  vary  considerably  (Fig.  66),  and  the  entabla- 
tures which  they  support  may  differ  greatly  in  height  (Fig.  66). 


Fig.  66. — Ramesseum,  Thebes.     {Description  deVEgypte,  t.  ii.,  pi.  28.) 

The  proportional  combinations  of  these  elements  are  such  that 
they  cannot  be  methodically  classified,  and  in  this  the  architecture 
of  Egypt  is  distinguished  from  that  which  we  call  classic.    In  Greek 


I02 


A   History  of  Art  in  Ancient  Egypt. 


art  there  is  a  modulus  which. determines  the  quantitative  relation 
of  forms  to  each  other,  and  fixes  a  mutual  and  invariable  inter- 
dependence. This  modulus  is  found  in  the  diameter  of  the 
column,  and  the  standard  of  proportion  which  is  based  upon  it 
is  called  a  canon.  In  Egypt,  as  in  other  countries,  there  must 
have  been  a  certain  connection  between  the  diameter  of  a  column 
and  its  height,  but  there  was  no  approach  to  that  rigid  and 
immutable  law  which  had  its  effect  upon  every  detail  of  a  Greek 
temple.  The  modulus,  in  Egyptian  art,  was  used  with  such 
freedom,  and  gave  rise  to  such  varied  proportions,  that  we  may  say 
that  no  canon  existed.     The  elementary  forms   of  an    Egyptian 


,ip'4p"*"''* 


Fio.  67. — The  Egyptian  Gorge  or  Cornice. 


edifice  had  so  little  dependence  upon  the  modulus  that  we  need 
not  take  it  into  consideration,  and,  in  this  sense,  the  art  of 
Egypt  was  not  mathematical,  like  that  of  Greece. 

Finally,  all  Egyptian  buildings  are  crowned  by  the  same 
entablature,  an  architrave  and  the  moulding  which  is  called  the 
Ecryptian  gorge  (Fig.  67). ^  An  architectural  member,  the  plain 
quadrangular  architrave,  is  invariably  inserted  between  this 
termination  and  the  upper  extremity  of  the  voids  and  points 
of  support. 

^  We  know  but  one  or  two  exceptions  to  this  rule.     It  will  suffice  to  quote  the 
Royal  Pavilion  of  Medinet  Abou,  which  is  crowned  by  a  row  of  battlements. 


X 

2; 


o 


33 


General  Principles  of  Construction. — Materials.     103 


§  3.   General  Principles  of  Consh^uction. — Materials. 

In  studying  a  natural  architecture  and  in  attempting  to  assign 
reasons  for  its  particular  characteristics,  many  circumstances  have 
to  be  taken  into  consideration.  The  innate  genius  of  the  race, 
the  physical  and  moral  conditions  of  its  development,  the  perfection 
of  its  civilization,  the  spirit  of  its  religion,  and  the  ardour  of  its 
faith  ;  none  of  these  must  be  forgotten,  but  some  of  them  act  in 
such  a  complex  fashion  that  they  are  extremely  difficult  to  follow. 
In  its  aspirations  towards  the  infinite  and  the  eternal,  the 
Egyptian  religion  raised  from  the  surface  of  the  earth  many 
buildings  which  varied  as  greatly  in  form  and  aspect  as  they  did 
in  date  and  situation.  The  climatic  conditions  of  the  world  have 
changed  but  little  since  the  beginning  of  the  historic  period,  and 
every  nation  has  to  take  them  into  the  first  consideration  in 
deciding  upon  its  own  architectural  forms  and  principles.  We 
have  here  a  problem  whose  data  do  not  vary,  and  yet  its  solutions 
have  not  always  been  the  same  even  in  a  single  country.  Without 
ever  being  absolutely  incorrect,  they  attached  themselves  now  to 
one  principle,  now  to  another,  and  so  gave  much  variety  to  the 
appearance  of  successive  buildings  under  one  sky  and  destined  for 
similar  uses. 

As  for  the  materials  employed,  we  cannot  go  so  far  as  to  say 
that  their  different  properties  absolutely  determined  the  charac- 
teristics of  Egyptian  building  in  advance.  Stone,  the  chief  of  all 
materials,  can  lend  itself  to  forms  of  great  variety  in  principle  ; 
and  so,  too,  can  brick  and  wood.  But  although  no  material  can 
narrowly  confine  a  skilful  architect,  there  are,  nevertheless,  certain 
systems  and  constructions  which  are  only  possible  with  those  which 
possess  certain  properties. 

To  give  but  a  single  example,  neither  the  hypo-style  halls  of 
Egypt  and  Persepolis,  nor  the  Greek  temples,  with  their  archi- 
traves resting  upon  widely  spaced  columns,  with  the  coffered  roofs 
of  their  porticos,  and  their  decorative  and  expressive  sculpture,  could 
have  been  carried  out  in  brick.  In  stone,  or  rather  in  marble, 
alone,  could  the  typical  temple,  such  as  the  Parthenon,  have  been 
realised  ;  without  such  a  material  the  Greeks  could  never  have 
created  that  incomparable  ensemble   whose  different  parts  are  so 


I04 


A   History  of  Art  in  Ancient  Egypt. 


intimately  allied  one  with  another,  in  which  the  richest  decoration 
is  in  complete  unity  with  the  constructive  forms  which  it  accen- 
tuates and  embellishes.  Brick  could  never  have  led  to  the 
invention  or  employment  of  these  forms.  Those  who  try  to 
imitate  them  in  any  such  material  have  to  make  up  for  its 
deficiencies  by  various  ingenious  devices.  The  joints  between  the 
bricks  have  to  be  hidden  under  stucco,  the  mouldings  and  carved 


sjM-JJX  JMfMU}i^d^S^,MM^^i&J&J!^iis^&Je^ 


50 


Fu;.  68. — Capital  and  entablature  of  the  temple  of  the  Deus  Rediculus  at  Rome. 


ornaments  of  stone  have  to  be  replaced,  as  in  the  temple  of  the 
Deus  Rediculus,  by  moulded  terra  cotta  (Fig.  68).  The  result 
is  sometimes  pleasing  enough,  especially  by  the  surprise  which  it 
causes.      Santa    Maria  delle   Grazie,  at   Milan,  is  a  masterpiece  of 

1  From  the  work  of  the  Abbe  Uggeri,  entitled  :  Ze  Detail  des  Matcriaux  dont se 
servaient  Irs  Anciais  pour  la  Construction  de  leurs  B'ltiments  (Rome,  oblong  folio, 
i8oo,  pi.  v.). 


Gp:neral  Principles  of  Construction. — Materials.      105 

its  kind,  thanks  to  the  skill  and  tact  displayed  by  Bramante  in  the 
management  of  the  burnt  clay  which  was  the  only  material 
afforded  him  by  the  plains  of  Lombardy ;  but  where  Bramante 
succeeded,  less  skilful  artists  have  failed.  They  have  demanded 
effects  from  brick  which  it  was  unable  to  give,  with  a  profound 
discord  between  form  and  matter  as  a  result. 

Of  all  the  causes  which  modify  the  forms  of  architecture  and 
determine  its  character,  the  most  important  is  the  nature,  the 
genius,  if  we  may  say  so,  of  the  materials  used.  So,  before  we 
can  arrive  at  a  correct  judgment  of  the  rules  and  principles  of  any 
style,  we  must  begin  by  appreciating  and  describing  the  materials 
of  which  it  disposes.  We  never  forget  this  in  the  case  of 
sculpture,  still  less  should  we  do  so  in  the  case  of  architecture, 
where  the  material  is  still  more  despotic. 

The  materials  made  use  of  by  the  Egyptians  were  granite,^ 
sandstone,-  and  limestone.'^  A  softer  stone,  namely  alabaster,  was 
often  employed  for  lining."* 

Sandstone  and  limestone,  especially  the  latter,  are  used  nearly 
everywhere ;  granite  is  of  less  frequent  occurrence  and  suggests  an 
important  observation. 

Granite  is  not  a  sedimentary,  stratified  rock  like  limestone  ;  it 
is  a  material  compacted  in  great  masses,  to  a  depth  or,  to  speak 
more  accurately,  in  a  volume  which  is  practically  unlimited  ;  the 
dimensions  of  the  stones  which  may  be  cut  from  these  masses  are 
therefore  infinite  to  all  intents  and  purposes.^ 

The  Egyptians  also  made  use  of  both  burnt  and  unburnt  brick. 

^  The  only  granite  quarries  that  were  worked  in   antiquity  were   those  of  Syene 
now  Assouan,  in  Upper  Egypt,  upon  the  riglit  bank  of  the  Nile. 

-  Sandstone  was  chiefly  obtained  from  two  localities,  Djebel-Ahmar,  near  Cairo, 
and  Djebel-Silsili  in  Upper  Egypt. 

2  The  Arab  Chain  is  almost  entirely  calcareous.  Near  the  sites  of  all  the  ancient 
cities  it  shows  numerous  excavations  bearing  witness  to  the  activity  of  the  ancient 
builders.  The  most  celebrated  of  these  quarries  is  that  at  Mokattam,  near  Cairo. 
The  stone  of  which  the  body  of  tha  pyramids  is  composed  was  drawn  from  it. 

*  The  alabaster  cjuarries  of  to-day  are  all  in  the  Arab  Chain,  between  the  southern 
slopes  of  the  mountain  Mahsarah,  near  Cairo,  and  the  springs  of  the  Wady-Siout, 
opposite  the  town  of  that  name. 

'•'  The  obelisk  of  Queen  Hatasu.  at  Karnak,  is  105  ft.  8  in.  high  :  the  statue  of 
Rameses  II.  at  Thebes,  on  the  left  bank  of  the  river,  is  a  monolith  55  ft.  5  in. 
high,  and  weighing  about  1,200  tons.  [The  obelisk  which  still  remains  at  Syene, 
never  having  been  completely  detached  from  the  rock  in  which  it  was  quarried,  is 
nearly  96  ft.  high  and  ii  fi.  i^^r  in.  diameter  at  its  base. — Ed.] 

VOL.    I.  p 


io6  A  History  of  Art  in  Ancient  Egypt. 

The  employment  of  these  different  materials  gave  birth  to 
what  we  may  call  '*'  dressed  construction,"  that  is,  construction 
the  elements  of  which  are  squared  upon  each  face  and  put  into 
close  juxtaposition  one  with  another. 

Concrete  or  pise,  compressed,  as  in  the  pylons,  between  moulds 
or  caissons  of  woodwork,  was  also  made  use  of  by  the  Egyptians. 
This  material  gave  rise  to  what  we  may  call  com.pact  construction. 

Again,  although  trees,  except  the  palm,  were  rare  enough  in  the 
valley  of  the  Nile,  the  Egyptians  built  also  in  wood,  by  which  a 
third  kind  of  construction,  called  construction  by  assemblage,  in 
which  the  elementary  units  were  held  together  by  being  introduced 
one  into  another,  was  obtained. 

In  a  few  buildings  of  the  latter  class  metal  seems  to  have  been 
employed,  sometimes  in  the  construction,  sometimes  for  lining,  and 
sometimes  for  outward  decoration. 


§  4. — Dressed    Constritction. 

The  constructive  elements  which  enter  into  the  composition  of 
this  first  class  of  buildings  are  stone  and  brick. 

In  the  first  place,  these  elements  are  horizontal  or  vertical. 

The  horizontal  elements  constitute  the  planes,  as  they  cover  the 
voids  by  horizontal  superposition. 

They  consist  of  courses  and  architraves. 

The  courses  form  the  walls.  They  are  arranged  in  horizontal 
bands,  with  vertical  and  sometimes  sloping  joints.  The  separate 
stones  are  often  bound  together  upon  their  horizontal  surfaces  by 
dovetails  or  tenons  of  wood.  The  blocks  made  use  of  in  this  form 
of  construction  are  usually  of  large  dimensions,  but  the  Egyptians 
also  made  use  of  small  stones  or  rubble,  lined  on  the  exterior  by 
large  flat  ones  which  concealed  the  meanness  of  the  material 
behind  them.^     (Fig-  /O-) 

Various  peculiarities  of  construction  wdiich  are  comparatively 
seldom  met  with  will  be  noticed  when  we  come  to  describe  the 
monuments  in  which  they  are  to  be  found. 

Architraves  were  stone  beams  used  to  bridge  over  the  voids  and 


&' 


^  We  find  this  construction  in  the  so-called  Temple  of  the  Spliinx,  near  the  Great 
Pvramid. 


Dressed  Construction. 


id: 


to  support  the  covering  of  the  building,  which  latter  was  composed 
of  long  and  heavy  slabs. 

The  vertical  elements  support  the  architraves  and  combine  them 
one  with  another.  These  vertical  supports  vary  greatly  in  size. 
Those  of  small  or  medium  dimensions  are  monoliths  ;  others  are 


Fig.   69. — The  Eg3'ptian    "bond."' 


composed    of    many  courses  of  stone   one  upon  another,  courses 
which  in  this  case  take  the  name  of  dr2uns. 

Upon  exterior  surfaces,  supports  take  various  forms  of  develop- 
ment which  may  all  be  referred  to  the  type  which  we  have  defined, 
namely,  the  portico.  In  the  interiors  the  form  of  support  is  a 
logical  consequence    of   the    material    employed.      Whenever    the 


ro8 


A   History  of  Art  in  Ancient  Egypt. 


stones  which  form  the  roof  are  too  small  to  bridge  over  the  whole 
of  the  space  comprised  within  two  walls,  they  must  be  made  to  rest 
upon  intermediate  supports  ;  and  this  necessity  springs  up  in  every 

building  of  any  importance. 
This  very  elementary  com- 
bination fulfils  all  the  re- 
quirements of  circulation. 
The  number  of  supports 
depends  upon  the  number  of 
rows  of  the  flat  stones  which 
form  the  roof.  They  are 
sometimes  multiplied  to  such 
an  extent  that  they  remind 
us  of  that  planting  arrange- 
ment in  our  gardens  which 
we  call  a  quincunx. 
We  cannot,  however,  affirm  that  the  number  of  supports  is 
invariably  decided  by  the  length  of  the  architraves,  or  of  the  roof- 
ing stones.      Some  very  long   monoliths   are   supported  at  regular 


Fk;.    70. — Dcuble-faccd  wall. 


Figs.  71,  72. — Elements  of  the  portico. 


intervals,  lest  they  should  break  with  their  own  weight  or  with  that 
put  upon  them.  The  walls,  architraves,  and  vertical  supports  are 
always  far  stronger  than  the  mere  weight  of  the  roof  would  require. 


Dressed  Construction. 


109 


The  following  woodcut  shows  the  arrangement  of  supports, 
architraves  and  roof.  These  simple  arrangements  constitute  a 
complete  system  of  construction  which,  belonging  exclusively  to 
Egypt,  has  had  results  upon  which  we  cannot  too  strongly  insist. 
Both  roof  and  architraves  being  horizontal,  all  the  pressure  upon 


Fig.  73. — Egyptian  construction,  ei^itomized  by  Ch.  Chipiez. 

the  walls  is  vertical.     There  is  no  force  tending  to  thrust  the  walls 
outwards  nor  to  affect  the  immobility  of  the  supports. 

Consequently,  if  the  proportions  of  the  vertical  and  horizontal 
elements  of  a  building,  that  is  to  say,  its  sections,  have  been  skil- 
fullv  determined,  there  is  in  the  buildinof  itself  no  latent  cause  of 
disruption  ;   its   equilibrium  is  perfect,  and  can  only  be  destroyed 


Tio  A   History  of  Art  in  Ancient  Egypt. 

by  external  physical  causes,  by  long  exposure  to  the  weather,  by 
earthquakes,  or  by  the  hand  of  man.^ 

We  see  then  that  the  first  Impression  caused  by  the  external 
lines  of  the  architectural  monuments  of  Egypt  is  confirmed  and 
explained  by  further  study.  They  are  built,  as  said  the  Pharaohs 
themselves,  "for  eternity."  Stability,  in  a  word,  in  its  highest 
and  most  simple  form  is  the  distinguishing  characteristic,  the  true 
originality,  of  Egyptian  architecture. 

This  character  is  most  strongly  marked  in  stone  buildings,  but 
it  is  by  no  means  absent  from  those  built  of  materials  created  by 
human  industry.  Works  in  brick  form  the  transition  between  the 
construction  that  we  have  described  and  that  which  we  call 
compact.  A  stone  roof  is  not  often  found,  and  the  termination  is 
generally  a  terrace  in  which  wood  is  the  chief  element.  In  some 
cases  the  secondary  parts  of  such  edifices,  and  sometimes  the 
whole  of  them,  are  covered  in  by  brick  vaults,  and  maintained 
by  walls  of  a  sufficient  thickness. 

Although  the  use  of  monoliths  for  roofing  purposes  was  general 
in  Egypt,  it  must  not  be  thought  that  the  architects  of  that  country 
were  ionorant  of  the  art  of  coverinof  voids  with  materials  of  small 
size,  that  is  to  say,  of  building  vaults.  There  are  numerous 
examples  of  Egyptian  vaults,  some  of  them  of  great  aritiquity, 
and,  moreover,  the  Egyptian  builders  constructed  their  vaults  after 
a  method  of  their  own.  In  spite  of  the  facilities  which  they 
afforded,  they  played,  however,  but  a  secondary  7'dle  in  the 
development  of  art.  They  were  never  used  in  the  buildings  to 
which  greater  importance  was  attached  ;  they  are  introduced  chiefly 
in  out-of-the-way  corners  of  the  building,  and  in  the  substructures 
of  great  monumental  combinations.  This  method  of  construction, 
being  confined  within  such  narrow  limits,  never  resulted  in  Egypt 
in  an  architectural  system  ;  ^  neither  did  it  give  birth  to  any  ot 
those  accessory  forms  which  spring  from  its  use. 

-  The  vertical  support  and  the  architrave  form  the  two  vital  elements  of  an 
Egyptian  building,  which  is  therefore  enabled  to  dispense  with  those  buttresses  and 
other  lateral  supports  which  are  necessary  to  give  stability  to  the  edifices  of  many 
other  nations. 

2  We  may  here  remark  that  the  modest  dwellings  of  the  Egyptian  fellah  are  often 
covered  by  vaults  of  pise',  that  is  to  say,  of  compressed  and  kneaded  clay.  None 
of  the  ancient  monuments  of  Egypt  possess  such  vaults,  which  are  of  much  less 
durability  than  those  of  stone  or  brick.  We  are,  however,  disposed  to  believe  that 
they  were  used  in  antique  times. 


Dressed  Construction. 


1 1 1 


Egyptian   vaults    may    be    divided    into    two    great    categories, 
according;  to  the  method  of  their  construction. 

I.  Off-set  vaults.  These  vaults  are 
composed  of  courses  off-set  one  from 
another,  and  with  their  faces  hollowed 
to  the  segment  of  a  circle.      (Fig.  74,) 

If  the  face  of  those  stones  which,  in 
the  form  of  inverted  steps,  are  turned  to 
the    void    which    has  to    be 


Fig.  74. — Element  of  an  off-set 
arch. 


covered,  be  cut  into  the  line 
of  a  continuous  curve,  the 
superficial  appearance  of  a 
segmental  arch  or  barrel 
vault  will  be  obtained  ;  but 
this  appearance  will  be  no 
more  than  superficial,  the 
vault  will  be  in  fact  a  false 
one,  because,  in  such  a  con- 
struction, all  the  stones 
which  enframe  the  void  and 
ofter    to    the    eye    the    form. 


Fig.   75. — AiTangement  of  the  courses  in  an 
off-set  arch. 


■^,     f*       ^ .  -t'j  11''*'  -" 


Fig.  76. — Off-set  .'■emicircular  arcli. 


of    a    vault,    are    really   laid    horizontally  one  upon   another,   and 


I  12 


A   History  of  Art  in  Ancient  Egypt. 


their    lateral   joints    are    vertical.      (Fig.    76.)      When    the    units 
of   such     vaults    are    properly    proportioned    they    are    stable    in 

themselves,    and    they   have 
no  lateral  thrust. 

2.  Centred  vaults.     These 
are   true   vaults.      They    are 


Fig.  77. — Voussoir. 


composed  of  voussoirs,  whose 
lateral  joints  are  oblique,  and 
radiate  towards  one  centre  or 
more.  (Figs.  ^'].  78,  and  79.) 
This  method  of  construction  is  very  convenient  because  it 
enables   the    builder    to    utilize    constructive    units    of  very    small 


Fig.  78. — Arrangement  of  voussoirs. 


1  IG.  79. — Semicircular  vault. 


dimensions,    such    as  bricks.       But    this  advantage  has    a   corre- 
sponding drawback.     These  voussoirs  thrust  one  against  another 


Compact  Construction. 


113 


and  tend  towards  disintegration.  They  are  not  stable  in  them- 
selves, and  in  order  to  give  them  stability  they  must  be  kept  in 
place  by  surrounding  them  with  opposing  forces  which  will 
effectually  prevent  their  setting  up  any  movement  in  the  structure 
of  which  they  form  a  part.  This  function  is  fulfilled  by  the  wall 
in  Egyptian  architecture,  which  is  consequently  very  thick,  but  the 
radiating  arch  never  arrived  at  such  a  development- in  Egypt  as  to 
lead  to  the  adoption  of  any  contrivance  specially  charged  with  the 
maintenance  of  vaults  in  a  state  of  proper  rigidity.  The  Egyptians 
not  only  employed  the  semicircular  arch  ;  they  made  use,  in  a  few 
instances,  of  the  pointed  form,  and  many  of  their  underground 
buildings  have  roofs  cut  out  of  the  rock  in  the  form  of  a  segmental 
vault.  The  fact  that-  these  sepulchral  chambers  affected  the  aspect 
of  vaulted  halls,  can  only  be  explained  by  the  supposition  that  a 
similar  construction  was  common  in  the  dwellings  of  the  living.^ 


i^  5- 


-Compact  Construction. 


The  methods  employed  in  what  we  may  call  compact  construction 
permit  the  use,  in  considerable  quantities,  of  moulded  clay  mixed 
with  chopped  straw.  This  material  was  used  in  buildings  which 
were  homogeneous ;  it  was  poured  into  a  mould  formed  by  planks, 
which  was  raised  as  the  work  progressed 
and  the  mixture  dried.  But  the  material 
had  little  strength,  and  was  far  inferior  to 
those  modern  concretes  which  have  the 
density  and  durability  of  the  hardest  stone. 
The  Egyptians  do  not  seem  to  have  been 
acquainted  with  concrete  proper,  and  un- 
burnt  bricks  did  not  differ  essentially  from 
pise.     Such  bricks,  when  placed  one  upon 

another  after  being  imperfectly  dried,  combined,  under  the 
influence  of  the  weather  and  their  own  weight,  into  one  homo- 
geneous mass  so  that  the  separate  courses  became  undistinguish- 
able.     This  latter  fact  has  been  frequently  noticed  in  Assyria,  by 

^  Another  explanation  has  been  given  of  the  employment  of  the  vault  in  subter- 
ranean work.  Marietta  believed  the  arch  to  be  symbolic,  to  signify  the  canopy  of 
heaven,  the  heaven  of  Amen.  One  objection  to  this  is  the  fact  that  the  vault  was 
not  universal  in  tombs ;  some  of  those  at  Beni  Hassan  have  flat  ceilings,  others 
have  coves. 

VOL.  I.  '  Q 


Fig.  80. — Granaries,   from   a 
bas-relief. 


114 


A  History  of  Art  in  Ancient  Egypt. 


those  who  had  to  cut  through  the  thickness  of  walls  in  the  process 
of  excavation. 

If  voids  have  to  be  covered  in  pise,  one  of  those  self-supporting 
curves  which  we  have  described  under  the  name  of  vaults,  must 
be  made  use  of,  and  the  vault  must  be  constructed  over  a  centring 

of  wood.     But  we  have  no 
'v-  evidence  that  the  Egyptians 

could  carry  the  art  of  con- 
struction to  this  point  in  pise. 
On  the  contrary,  we  have 
fjood  reason  to  believe  that 
they  generally  made  use  of 
this  material  for  the  quiescent 
body  of  the  edifice  alone, 
and  that  voids  were  mostly 
covered  with  stone  or  wood. 
In  a  word,  the  Egyptians  did 
not  carry  the  use  of  artificial 
material  far  enough  to  form 
a  complete  system  based 
upon  it.  They  made  great  use  of  it,  but  only  in  a  strictly  limited 
fashion.  It  is  only  found  in  certain  well-defined  parts  of  buildings, 
which  were  never  of  any  very  great  interest  from  an  artistic  point 
of  view  (Fig.  80).  It  deserved  to  be  mentioned,  if  only  for  the 
frequency  of  its  use  in  Egypt,  in  the  private  architecture  of  both 
ancient  and  modern  times  (Fig.  81),  but  it  need  not  detain  us  longer. 


Fig.  Si. — Modern  pigeon  house,   Thebes. 


§  6. — Consh'uction  by  Assemblage. 

Carpentry,  or  construction  by  assemblage,  played  a  considerable 
part  in  ancient  Egypt,  but,  as  may  easily  be  understood,  few  traces 
of  it  are  to  be  found  in  our  day.  Those  edifices  which  were  con- 
structed of  wood  have,  of  course,  all  perished  ;  but,  in  spite  of  their 
disappearance,  we  can  form  a  very  good  idea  of  their  aspect  and  of 
the  principles  of  their  construction.  In  the  most  ancient  epoch  of 
Egyptian  art,  the  people  took  pleasure  in  copying,  in  their  stone 
buildino^s,  the  arranofements  which  had  characterised  their  work  in 
wood  ;    besides  which,  their  paintings  and  reliefs  often  represent 


Construction  by  Assemblage.  i  i  5 


buildings  of  the  less  durable  material.  The  constructive  principles 
which  we  have  next  to  notice,  have  thus  left  traces  behind  them 
which  will  enable  us  to  describe  them  with  almost  as  much 
accuracy  as  if  the  carpenters  of  Cheops  and  Rameses  were  working 
before  our  eyes. 

We  need  not  insist  upon  the  characteristics  which  distinguish 
assembled  construction  from  masonry  or  brickwork.  The  different 
parts  of  the  former  are,  of  course,  much  more  intimately  allied 
than  in  buildings  constructed  of  large  stones.  Supports  of  dressed 
stone  truly  fixed  with  the  plumb  line  are  perfectly  stable  of 
themselves. 

In  both  Egypt  and  Greece  we  often  come  upon  a  few  columns 
still  standing  upright  amid  their  desolate  surroundings,  and 
announcing  to  the  traveller  the  site  of  some  city  or  famous  temple 
which  has  been  long  destroyed.  But  wooden  supports  have  little 
thickness  in  comparison  with  their  height,  and  the  material  of 
which  they  are  formed,  being  far  less  dense  than  stone,  cannot 
maintain  itself  in  place  by  its  own  weight.  It  is  the  same  with 
wooden  architraves.  The  heaviest  beams  of  wood  will  not  keep 
their  places  when  simply  laid  one  upon  another,  and  are  in  that 
matter  far  inferior  to  those  well  dressed  stones  which,  in  so 
many  ancient  walls,  have  resisted  change  with  neither  tenons  nor 
cement  to  help  them. 

As  a  general  principle,  when  wood  has  to  be  employed  to  the 
best  advantage,  and  endowed  with  all  the  solidity  and  resisting 
power  of  which  it  is  capable,  the  separate  pieces  must  be  intro- 
duced one  into  another  (Fig.  82).  But  even  when  thus  combined 
and  held  in  place  by  mechanical  contrivances,  such  as  bolts  and 
nails,  they  will  never  form  a  homogeneous  and  impenetrable  mass 
like  brick  or  stone.  By  such  methods  an  open  structure  is 
obtained,  the  voids  of  which  have  afterwards  to  be  filled  up  by 
successive  additions,  and  these  additions  often  take  the  form  of 
what  we  call  panels. 

We  may  look  upon  the  different  faces  of  a  wooden  building  as 
separate  pieces  of  construction  which  should  be  put  together  upon 
the  ground  before  being  combined  with  each  other.  This  pro- 
cess, though  not  always  made  use  of  in  practice,  is  at  least  the 
most  logical  method  for  those  who  wish  to  make  the  best  use  of 
their  materials.  But  even  when  thus  put  together,  one  of  these 
single  faces  has  not  much  more  stability  than  each  of  its  constituent 


ii6 


A  History  of  Art  in  Ancient  Egypt. 


elements.  In  order  to  form  a  rigid  and  stable  whole,  the  several 
faces  must  be  allied  by  reciprocal  interpenetration  at  the  angles. 

It  was  necessary  to  call  attention  once  for  all  to  these  general 
characteristics  of  wooden  construction,  because  we  shall  hereafter 
have  occasion  to  examine  the  forms  and  motives  which  stone 
architecture  borrowed  from  wood  in  the  case  of  other  people 
besides  the  Egyptians.  We  must  now  determine  the  particular 
characteristics  offered  by  the  material  in  Egypt,  as  they  may  be 
learnt  in  the  representations  to  which  we  have  already  referred. 

When  a  wall  has  to  be  built  of  wood  so  as  neither  to  warp  nor 
give  way,  it  is  necessary  to  make  use  of  a  certain  number  of  oblique 


Fig.  82. — Elements  of  wooden  construction. 


members.  This  is  one  of  the  elementary  rules  of  the  carpenter's 
art,  and  to  form  an  idea  how  it  was  applied  in  our  own  country  it 
is  enough  to  cast  an  eye  over  any  of  the  wooden  buildings  of  the 
middle  ages  or  of  the  renaissance.  The  Egyptians  were  not  ignorant 
of  the  advantages  conferred  by  the  use  of  these  oblique  members 
because  they  employed  them  frequently  in  their  furniture  ;  but 
they  seem  never  to  have  introduced  them  into  the  construction  of 
their  buildings.  All  joints  are  there  made  at  a  right  angle.  They 
were  probably  led  to  reject  oblique  lines  by  their  unwillingness  to 
break  in  upon  the  simple  harmony  of  vertical  and  horizontal  lines 


Construction  by  Assemblage. 


1 1 


which  is  the  distinguishing  principle  of  all  their  architecture. 
Thus  self-deprived  of  a  valuable  resource,  they  were  driven  to  the 
discovery  of  some  other  means  of  giving  the  required  cohesion  and 
stability  to  their  walls.  This  requirement  they  thought  they  had 
fulfilled  in  exaggerating  the  points  of  connection  between  the 
vertical  and  horizontal  members,  which  were  thus  brought  into 
more  intimate  relation  than  would  in  these  days  be  thought 
necessary.^  The  consequence  of  this  was  that  theip  wooden 
buildings  presented  much  the  same  closed  appearance  (Fig.  83)  as 


Fig.  83. — Wooden  building  (first  system^  composed  by  Charles  Chipiez.- 


we  have  already  noticed  in  their  stone  constructions  ;  and,  more- 
over, as  every  joint  was  made  at  right  angles,  the  pyramidal  form 
was  entirely  absent. 

But  the  Egyptians  also  made  use  of  wood  for  buildings  very 

^  In  this  respect  there  is  a  striking  resemblance  between  Egyptian  carpentry  (see 
Fig.  83),  and  much  of  the  joinery  of  the  modern  Japanese. — Ed. 

-  In  this  figure  we  have  attempted  to  give  some  notion  of  what  a  wooden  building 
must  have  been  like  in  ancient  Egypt,  judging  from  the  imitations  of  assembled 
construction  which  have  been  found  in  tlie  tombs  and  sarcophagi  of  the  ancient 
emi^ire. 


ii8 


A  History  of  Art  in  Ancient  Egypt. 


different  from  those  to  which  we  have  hitherto  alluded.     Those 
were  closed  ;  but  we  have  now  to  speak  of  another  system,  of  one 


Fig.  84. — Wooden  building  (second  system),  composed  by  Charles  Chipiez. 

which,  by  contrast,  might  be  called  an  open  system  of  construction. 
The  edifices  upon  which  it  was  employed  were  generally  of  small 


Decoration.  119 


size,  and  in  this  respect  resembled  those  which  we  have  described, 
but  they  were  distinguished  by  a  different  system  of  carpentering. 
We  know  them  only  by  the  figured  representations  which  have 
come  down  to  us,  for  they  were  little  calculated  to  outlast  the 
centuries  (Fig.  84).  This  second  system  lends  itself  as  little  as 
the  first  to  pyramidal  and  kindred  forms  ;  horizontal  lines,  also, 
were  in  it  of  but  secondary  importance.  Composed  of  a  few 
vertical  members  bound  together  at  tlie  top,  such  a  building  was 
allied  to  the  portico  type  which  has  already  been  described.  This 
method  of  carpentry  seems  to  have  been  used  only  for  sub- 
ordinate buildings  ;  but  yet  it  should  not  be  passed  by  in  silence. 
It  was  frequently  used  for  the  construction  of  light  decorative 
pavilions,  and  it  had  a  set  of  principles  which  are  as  susceptible 
of  definition  as  those  of  the  most  ambitious  architecture. 

Metal  must  have  entered  into  the  construction  of  these  pavilions. 
It  may  have  furnished  either  the  horizontal  or  the  vertical 
members,  and  it  is  certain  that  it  was  partly  used  for  the  roofs. 

In  all  wooden  structures  the  roof  must  also  be  of  wood,  because 
the  light  walls  which  are  proper  to  the  material  could  not  support 
the  great  weight  of  a  flat  stone  covering,  still  less  could  they  stand 
up  against  the  combined  weight  and  thrust  of  a  stone  or  brick 
vault,  which  would  destroy  them  in  very  summary  fashion. 


§  7. — Decoration. 

We  have  hitherto  described  Egyptian  architecture  according  to 
the  general  character  of  its  forms  and  principles  of  construction  ; 
we  must  now  attempt  to  give  a  true  idea  of  its  method  of 
decoration.  This  may  be  described  in  a  very  few  words.  For  the 
decoration  of  the  vast  surfaces,  either  plain  or  curved,  which  their 
style  of  architecture  placed  at  their  disposal,  the  Egyptians  made 
use  of  paint.  They  overlaid  with  a  rich  system  of  colour  the 
whole  inside  and  outside  of  their  buildinfjs,  and  that  with  no  desire 
to  accentuate,  by  a  carefully  balanced  set  of  tones,  the  great 
constructive  lines,  contours  and  mouldings,  nor  with  any  wish  to 
produce  merely  a  complicated,  polychromatic  ornamentation. 
Groups  of  figures  borrowed  from  the  animal  and  vegetable 
kingdoms  form  its  chief  constituents.  In  these  picture  decora- 
tions, man  is  seen  in  every  attitude  or  vocation,  side  by  side  with 


I20  A  History  of  Art  in  Ancient  Egypt, 

birds,   fishes    and    quadrupeds,   and    with    those    composite    forms 
which  have  been  created  by  himself  to  represent  his  gods. 

Intaglio  and  bas=relief  often  lend  their  help  to  the  ornament. 
Images  and  explanatory  inscriptions  are  sometimes  cut  in  the 
stone,  sometimes  modelled  in  slight  relief;  but  in  either  case  all 
figures  are  distinguished  by  their  proper  colour  as  well  as  by  the 
carved  or  modelled  outlines. 

It  will  thus  be  seen  that  Egyptian  decoration  is  distinguished 
by  the  intimate  and  constant  alliance  of  two  elements  which  are 
often  separated  in  that  of  other  races.  The  first  is  the  employ- 
ment of  colour  to  give  variety  to  surfaces  and  to  distinguish 
different  members  of  the  architecture  by  the  opposition  of  tones. 
The  second  is  the  employment  of  colour  for  the  representation  of 
life,  for  which  purpose  every  surface  is  seized  upon,  whether  the 
face  of  a  wall,  or  the  round  shaft  of  a  column.  The  decorator  is 
not  satisfied  to  use  colour  to  g-ive  force  to  the  lines  of  a  buildino- 
and  to  increase  its  general  effect ;  he  also  makes  use  of  it  to 
interpret,  to  multiply,  and  to  immortalize  the  ideas  which  float 
through  his  own  brain.  A  building  thus  ornamented  presents  us 
with  a  series  of  pictures  embodied  in  its  own  constitution.  From 
cornice  to  foundation,  upon  wall  and  column,  it  is  covered  with  an 
unending  series  of  wall  paintings,  which,  like  a  gorgeous  tapestry, 
envelop  and  embellish  it  without  hiding  any  of  the  details  of  its 
construction. 

The  polychromatic  decoration  of  the  Egyptians  is  to  be 
explained,  like  that  of  the  Assyrians,  of  the  Greeks,  of  the  Italians, 
and  of  all  other  southern  nations,  by  the  quality  and  quantity  of 
their  daylight  and  the  way  in  which  it  affected  their  visual  organs. 
The  more  intense  the  light,  the  more  pleasure  does  the  eye  receive 
from  strength  and  variety  of  colour.  The  science  of  optics  gives 
us  an  explanation  of  this  fact,  but  at  present  we  are  concerned  only 
with  the  fact  itself,  which  is  a  matter  of  daily  experience.  It  is 
notorious  that  the  colours  of  birds  and  butterflies,  and  of  the  petals 
of  flowers,  become  brighter  and  gayer  in  exact  proportion  as  we 
near  the  equator  and  leave  the  pole ;  ^  the  same  rule  holds  good 

'  We  here  speak  of  the  fauna  as  a  whole,  disregarding  particular  genera  and 
species.  It  may  be  said  that  some  particular  plant  which  is  to  be  found  both  in 
France  and  Norway,  is  much  brighter  in  colour  when  it  grows  in  the  neighbourhood 
of  the  pole  than  in  our  temperate  climate,  but  this  apparent  exception  only  confirms 
the  rule  which  we  have  laid  down.     The  plant  whose  whole  season  of  bloom  is 


Decoration.  i  2 1 


with  the  habitations  of  mankind,  with  his  clothes  and  furniture, 
which  become  more  brilhant  in  colour,  and  more  audaciously 
abrupt  in  their  transitions  from  one  hue  to  another.  Delicate 
shades  of  difference  are  imperceptible  by  an  eye  blinded  with  the 
southern  sun  ;  it  sees  nothing  but  the  simplest,  strongest,  and 
frankest  colour  notes  to  the  exclusion  of  all  half-tint.^ 

Under  a  burning  and  never  clouded  sun,  objects  of  a  neutral 
colour  do  not  stand  out  against  their  background,  and  their 
shadows  lose  a  part  of  their  value,  "  comme  ddvordes  pa7^  la 
diffusion  et  la  rdverbdration  dune  incomparable  liiinierer  ^  In 
Egypt,  a  column,  a  minaret,  a  dome,  hardly  seem  to  be 
modelled  as  they  stand  against  the  depths  of  the  sky.  All  three 
seem  almost  flat.  The  warm  and  varied  hues  with  which 
polychromatic  decoration  endows  buildings  help  us  to  distinguish 
them  in  such  situations  from  the  ground  upon  which  they  stand, 
and  to  accentuate  their  different  planes.  They  also  compensate, 
in  some  degree,  for  the  absence  of  those  strong  shadows  which 
elsewhere  help  to  make  contours  visible.  Attention  is  drawn  to 
the  dominant  and  bounding  lines  of  an  architectural  composition 
by  contrasts  of  tint  which  also  serve  to  give  force  to  wall  paintings 
and  bas-reliefs. 

comprised  between  a  late  spring  and  an  early  autumn  develops  itself  much  more 
rapidly  than  with  us,  and,  granting  that  it  has  become  so  hardened  that  it  is  able  to 
resist  the  long  and  hard  frosts  of  winter,  it  receives,  during  the  short  summer,  much 
more  light  and  sun  than  its  French  or  German  sister.  During  those  fleeting  summers 
of  the  north,  whose  strange  charm  has  been  so  often  described,  the  sun  hardly 
descends  below  the  horizon  ;  the  nights  are  an  hour  long,  and  not  six  or  seven.  The 
colour  of  flowers  is  therefore  in  exact  proportion  to  the  amount  of  light  which 
they  receive. 

^  This  was  perceived  by  Goethe.  In  art,  as  in  natural  science,  he  divined  before- 
hand some  of  the  discoveries  of  our  century  by  the  innate  force  of  his  genius.  He 
was  not  surprised  by  the  discovery  that  the  temples  of  classic  Sicily  were  painted  in 
briUiant  tones,  which  concealed  the  surface  of  their  stone  and  accentuated  the 
leading  lines  of  their  architecture.  He  was  one  of  the  first  to  accept  the  views  of 
Hittorf  and  to  proclaim  that  the  architects  who  had  found  traces  of  colours  upon 
the  mouldings  of  Greek  buildings  were  not  deceiving  themselves  and  others. 

^  We  borrow  these  expressions  from  M.  Ch.  Blanx,  who,  when  in  Egypt,  was  very 
much  struck  with  this  phenomenon.  "  Those  villages  which  approach  in  colour. to 
that  Nile  mud  of  which  they  are  composed,  hardly  stand  out  at  all  against  the  back- 
ground, unless  that  be  the  sky  itself  or  those  sunny  rocks  which  reflect  the  light  in 
such  a  fashion  that  they  fatigue  the  most  accustomed  eyes.  I  notice  here,  as  I  did 
in  Greece,  at  Cape  Sunium,  that  cupolas  and  round  towers  have  their  modelling 
almost  destroyed  by  the  strong  reflections."  {Voyage  de  la  Haiilc  Egyptc,  1876, 
p.    114). 

VOL.   I.  R 


122  A  History  of  Art  in  Ancient  Egypt, 


Polychromy  is  thus  a  help  to  our  eyes  in  those  countries  where 
a  bhnding  light  would  otherwise  prevent  us  from  appreciating  the 
structural  beauties  of  their  architecture.  It  is  by  no  means 
peculiar  to  Egypt,  but  that  country  was  the  first  to  employ  it  upon 
rich  and  vast  undertakings,  she  employed  it  m.ore  constantly  and 
more  universally  than  any  other  people,  and  she  carried  it  to  its 
logical  conclusion  with  a  boldness  which  was  quite  unique. 

The  Egyptian  habit  of  sprinkling  figures  over  every  surface 
without  regard  to  its  shape,  its  functions,  or  those  of  the  mass  to 
which  it  belonged,  was  also  peculiar  to  themselves.  Upon  the 
round  shaft  of  the  column,  upon  the  bare  expanse  of  the  wall,  these 
figures  were  multiplied  and  developed  to  an  extent  which  was 
limited  only  by  the  length  of  the  wall  or  the  height  of  the  column. 
They  were  generally  painted  in  bands  of  equal  height,  separated 
one  from  another  by  a  narrow  fillet  which  indicated  the  plane  upon 
which  the  groups  of  figures  had  a  footing.  There  is  no  visible 
connection  between  the  bands  of  figures  and  the  structures  which 
they  ornament ;  right  and  left,  above  and  below,  they  spread  over 
every  surface  and  pay  no  attention  to  the  joints  and  other  struc- 
tural accidents  by  which  they  are  seamed  (Fig.  85  and  PI.  III.) 

It  may  be  said  that  these  joints  were  invisible  until  the  passage 
of  centuries  had  laid  them  bare  by  destroying  the  stucco  which, 
especially  w^here  sandstone  or  limestone  was  used,  once  veiled  the 
surface  of  the  bare  walls. ^  Doubtless  this  is  true  ;  but  even  in  a 
climate  such  as  that  of  Egypt,  the  architect  could  not  believe  that  a 

^  Wilkinson  thought  there  was  always  a  layer  of  stucco,  even  upon  the  beautiful 
granite  of  the  obelisks  {Manners  and  Customs  of  the  Ancioit  Egyptians^  2nd  ed., 
1878,  vol.  ii.  p.  286.)  His  statement  must  be  treated  with  great  respect.  During 
his  long  sojourn  in  Egypt  he  examined  the  remains  of  the  ancient  civilisation  with 
great  care  and  patience,  but  yet  we  think  his  opinion  upon  this  point  must  be 
accepted  with  some  reserve.  There  are  in  the  Louvre  certain  sarcophagi  and  other 
objects  in  hard  stone,  upon  which  traces  of  colour  are  clearly  visible  on  the  sunk 
beds  of  the  figures  and  hieroglyphics,  while  not  the  slightest  vestige  of  anything  of 
the  kind  is  to  be  found  upon  the  smooth  surface  around  those  carvings.  But  it  is 
certain  that  granite  was  often  stuccoed  over.  Mariette  has  verified  that  it 
was  so  on  the  obelisk  of  Hatasu  at  Thebes;  both  from  the  inscription  and  the 
appearance  of  the  monument  itself  he  came  to  the  conclusion  that  it  had  been 
gilded  from  top  to  bottom,  and  that  the  gold  had  been  laid  upon  a  coat  of  white 
stucco.  "The  plain  surface,"  he  says,  "alone  received  this  costly  decoration.  It 
had  been  left  slightly  rough,  but  the  hieroglyphs,  which  had  their  beds  most  care- 
fully polished,  preserved  the  colour  and  surface  of  the  granite."  {Itineraire,  p.  178.) 
As  for  buildings  of  limestone  or  sandstone,  like  the  temples  of  Thebes,  they  are 
always  coated 


Decoration.  123 


thin  coat  of  plaster  would  endure  as  long  as  the  massive  walls  upon 
which  he  laid  it.  We  have  here  a  great  contrast  in  principle 
between  the  decoration  and  the  architecture  of  Egypt.  In  the 
latter  the  chief,  if  not  the  only  aim,  seems  to  have  been  to  make 
sure  of  absolute  stability,  of  indefinite  duration  ;  and  yet  these 
eternal  walls  are  lined  with  a  rich  decoration  which  is  spoiled  by 
the  fall  of  a  piece  of  plaster,  which  is  injured  by  the  unavoidable 
settlings  of  the  masonry  and  destroyed  by  the  slightest  earthquake  ! 
Of  this  we  need  give  but  one  conclusive  instance.  Our  third  plate 
reproduces  that  admirable  portrait  of  Seti  I.,  which  is  the  wonder 
of  the  temple  at  Abydos,  This  beautiful  work  in  relief  is 
sculptured  upon  the  internal  faces  of  four  unequal  stones  in  the 
v^'all  of  one  of  the  rooms.  The  joints  may  be  distinguished,  but  as 
yet  they  have  not  opened  sufficiently  to  do  much  damage  to  the 
artistic  beauty  of  the  work  ;  but  it  cannot  be  denied  that  the  pre- 
servation of  the  royal  effigy  would  have  been  much  more  certainly 
assured  if  the  sculptor  had  chosen  a  single  stone  to  work  upon, 
instead  of  a  built-up  wall  which  so  many  causes  would  help  to 
destroy. 

When  Egyptian  buildings  were  new  and  their  colour  fresh,  this 
method  of  decoration  must  have  given  them  a  most  fascinating 
brilliancy.  Whether  the  pencil  alone  were  employed  to  trace  the 
designs  upon  the  smooth  walls,  or  whether  its  powers  were  sup- 
plemented by  the  work  of  the  chisel,  these  figures,  which  succeeded 
each  other  in  thousands  upon  every  wall  and  pillar,  mingled  with 
inscriptions  which  were  in  themselves  pictures,  and  dressed  in  the 
most  vivid  colours,  must  have  at  once  amused  the  eye  and  stirred 
the  brain  by  the  variety  of  their  tints  and  of  the  scenes  which  they 
represented.  But  in  spite  of  its  breadth  and  vivacity  the  system 
had  two  grave  defects. 

The  first  was  the  fragility  of  the  plaster  surface  upon  which  it 
was  displayed.  This  surface  may  be  compared  to  a  tapestry 
stretched  over  the  whole  interior  of  the  building,  and,  to  continue 
the  comparison,  when  once  any  portion  of  the  plaster  coat  became 
detached  from  the  wall,  there  was  nothing  left  but  the  ground  or 
reverse  of  the  stuff,^     The  design  and  colour  may  still  be  distin- 

1  Apropos  of  the  Temple  of  Khons,  Jollois  and  Devilliers  {Description  generale 
de  Thches^  cli.  ix.)  remark  :  "  It  was  upon  this  coat  that  the  hieroglyphs  and  figures 
were  sculptured.  .  .  .  The  contour  of  the  figures  is  sometimes  marked  upon  the  store 
beneath,  because  the  depth  of  the  cutting  is  greater  than  the  thickness  of  the  stucco." 


124 


A  History  of  Art  in  Ancient  Egypt. 


guished  or  divined,  but  there  is  a  great  difference  between  painted 
ornament  which  is  subject  to  such  damage  and  a  woven  hanging 
at  any  time  before  the  threads  of  the  woof  have  been  discoloured 
and  entirely  worn  out.  The  other  defect  in  the  system,  is  its 
uniformity.      It    is   monotonous    and  confused    in   spite  of  all  its 


Fig.   85. — Seti  I.   striking  prisoners  of  war  with  bis  mace.      Karnak,   Thebes.     (Champollion, 

PL  294.) 


richness.  It  suffers  from  the  absence  of  that  learned  balance 
between  plain  and  decorated  surface  which  the  Greeks  understood 
so  thoroughly.  In  the  Greek  temples,  sculptured  figures  had  the 
more  importance  in  that  the  eye  of  the  spectator  was  drawn 
forcibly  to  them  by  the  very  limitation  of  the  space  reserved  for 


JSxilpis  del  el  sc. 


KARNAK 
BAS    RELIEFS  IN  THE   GRANITE  CHAMBERS 


Decoration. 


125 


them.  They  were  cut  from  separate  blocks  of  marble,  which, 
though  carefully  and  skilfully  allied  with  the  architecture  which 
they  were  meant  to  adorn,  did  not  form  an  integral  part  of  it. 
Such  figures  ran  no  risk  of  being  cut  in  two  by  the  opening  of  the 
joints  between  the  stones.  Although  marvellously  well  adapted  to 
the  places  for  which  they  were  intended,  and  closely  allied  to  the 
architecture  by  their  subject  as  well  as  their  material  shape,  they 
yet  preserved  a  life  and  individuality  of  their  own.  To  take 
decorative  art  as  a  whole,  the  Greeks  did  not  make  use  of  so  many 
figures  as  the  Egyptians,  but  they  knew  better  how  to  economize 
the  sources  of  effect,  and  to  preserve  their  works  against  the 
destructive  action  of  time. 

To  Egypt,  then,  belongs  the  credit  of  having  been  the  first  to 
discover  the  obligation  imposed  upon  the  architect  by  the  sunlight 
of  the  south — to  accentuate  the  main  lines  of  his  edifice  by  means 
of  colour.  She  thoroughly  'understood  how  to  make  different 
tones  distinguish  between  the  various  parts  of  a  structure  and  de- 
fend its  contours  atjainst  the  effect  of  a  dazzling  ligrht.  On  the 
other  hand,  she  went  too  far  when  she  covered  every  surface, 
without  choice  or  stint,  with  her  endless  figure  processions.  Such 
a  decoration  was  only  rendered  possible  by  the  use  of  a  material 
which  compromised  its  durability  ;  and  that  is  not  her  only  short- 
coming. She  failed  to  understand  the  value  of  repose  and  the 
absolute  necessity  of  contrast  ;  she  failed  to  perceive  that  by 
multiplying  figures  to  infinity,  she  lessened  their  effect  and  made 
them  a  fatigue  to  the  eye  and  the  intellect. 


Mimi![iiiiii]iiJtiUif''i'^' 


CHAPTER   III. 

Sepulchral  Architecture. 

§   I. — The  Egyptian  Belief,  as  to  a  Future  Life  and  its  Influence 
npon  their  Sepulchral  Ajxhitecture. 

The  most  ancient  monuments  which  have  yet  been  discovered 
in  Egypt  are  the  tombs  ;  they  have'  therefore  a  right  to  the  first 
place  in  our  sketch  of  Egyptian  architecture. 

In  every  country  the  forms  and  characteristics  of  the  sepulchre 
are  determined  by  the  ideas  of  the  natives  as  to  the  fate  of  their 
bodies  and  souls  after  life  is  over.  In  order  to  understand  the 
Egyptian  arrangements,  we  must  begin  then  by  inquiring  into  their 
notions  upon  death  and  its  consequences  ;  we  must  ask  whether 
they  believed  in  another  life,  and  in  what  kind  of  life.  We  shall 
find  a  complete  answer  to  our  question  in  the  collation  of  written 
texts  with  figured  monuments. 

In  the  first  period  of  his  intellectual  development,  man  is  unable 
to  comprehend  any  life  but  that  which  he  experiences  in  his  own 
person.  He  is  as  yet  unable  to  observe,  to  analyse  or  to  generalize. 
He  does  not  perceive  the  characteristics  which  distinguish  him 
from  things  about  him,  and  he  sees  nothing  in  nature  but  a 
repetition  of  himself.  He  is  therefore  incapable  of  distinguishing 
between  life  such  as  he  leads  it  and  mere  existence.  He  dreams 
of  no  other  way  of  being  than  his  own.  As  such  is  the  tendency 
of  his  intellect,  nothing  could  be  more  natural  or  more  logical  than 
the  conception  to  which  it  leads  him  in  presence  of  the  problem 
offered  to  him  every  time  that  a  corpse  descends  into  the  grave. 
M.  Maspero  has  so  thoroughly  understood  the  originality  of  the 
solution  adopted  by  the  Egyptians  that  we  cannot  do  better,  in 
attempting  to  explain  the  hypothesis,  at  once  gross  and  subtle,  to 
which  they  had  recourse  for  consolation,  than  borrow  his  rendering 


iraiiiiii:iiiiiiiiiiimiiiiiiiir,Miaiiiiiiiiiiiiiin;:uiiiiiiiiiiiiiiir,.i  iiii,i!;ii!;,iiin'i\!iiiliiliiiii.!;iillllinKmiiiMi'.i;iiill'i|,'liiilillliiii!;i?; 


J.Sulpis  del  cL  sc- 


±iAS-KELlh,l'' AT  AHYDOS 


Sepulchral  Architecture.  127 

of  the  texts  which  throw  Hght  upon  this  subject,  together  with  some 
of  the  reflections  which  those  texts  suggested  to  him.^ 

Were  we  to  affirm  that  during  thousands  of  years  no  change 
took  place  in  the  ideas  of  the  Egyptians  upon  a  future  life,  we 
should  not  be  believed,  and,  as  a  fact,  those  ideas  underwent  a 
continual  process  of  refinement.  Under  the  eighteenth  and 
nineteenth  dynasties,  during  those  centuries  when  the  limits  of 
Egyptian  empire  and  Egyptian  thought  were  carried  farthest  afield, 
we  find  traces  of  doctrines  which  offer  notable  variations,  and  even, 
when  closely  examined,  actual  contradictions.  These  are  successive 
answers  made  during  a  lonor  course  of  time  to  the  eternal  and 
never-changing  enigma.  As  they  became  more  capable  of  philo- 
sophic speculation  the  Egyptians  modified  their  definition  of  the 
soul,  and,  by  a  necessary  consequence,  of  the  manner  in  which  its 
persistence  after  death  must  be  understood,  and  as  always  happens 
in  such  a  case,  these  successive  conceptions  are  super-imposed  one 
upon  another  ;  the  last  comer  did  not  dethrone  its  predecessor  but 
became  inextricably  blended  with  it  in  the  popular  imagination. 

We  refer  all  those  who  wish  to  follow  minutely  this  curious 
development  of  the  Egyptian  intellect  to  the  subtle  analysis  of 
M.  Maspero.  That  historian  has  applied  himself  to  the  apprehen- 
sion of  every  delicate  shade  of  meaning  in  a  system  of  thought 
which  has  to  be  grasped  through  the  veil  thrown  around  it  by 
extreme  difficulties  of  language  and  written  character,  but  at  the 
same  time  he  has  never  attempted  to  endow  it  with  a  precision  or 
logical  completeness  to  which  it  had  no  claim.  By  well  chosen 
comparisons  and  illustrations  he  enables  us  to  understand  how  the 
Egyptian  contented  himself  with  vague  notions,  and  how  he 
manao^ed  to  harmonize  ideas  which  seem  to  us  inconsistent. 

We  shall  not  enter  into  those  details.  We  shall  not  seek  to 
determine  the  sense  which  the  Egyptians  attached,  after  a  certain 
period,  to  the  word  bdii-  which  has  been  translated  soul,  nor  the 

^  "  Confcre7ice  siir  FHistoii-e  des  Ames  darts  V Egypte  aficieune,  d'aprcs  hs  Monuments 
du  Musee  du  Louvre,  in  the  Bulletin  hebdomadaire  de  F Association  scientifique  de 
France,  No.  594.  M.  Maspero  has  often  and  exhaustively  treated  this  subject, 
especially  in  his  numerous  lectures  at  the  College  de  France.  Those  lectures 
afforded  the  material  for  the  remarkable  paper  in  the  J^ounial  asiatique  entitled, 
"  Etude  sur  quelques  Peintures  et  sur  quelques  Textes  relatifs  aux  Funerailles  "  (numbers 
for  May,  June,  1878,  for  December-June,  and  November,  December,  1879,  and  May- 
June,  1880).  These  articles  have  been  republished  in  a  single  volume  with  important 
corrections  and  additions  (Maisonneuve,  1880).  "  Ox  ba. — Ed. 


128  A  History  of  Art  in  Ancient  Egypt. 


distinction  between  it  and  kJiou,  luminousness,  which  the  soul 
seems  to  have  enveloped  like  a  garment.  We  shall  not  follow  the 
soul  and  its  internal  light  in  its  subterranean  journey  across  Anient, 
the  Egyptian  Hades,  to  which  it  entered  by  a  cleft,  Pega,  to  the 
west  of  Abydos,  which  was  the  only  portal  to  the  kingdom  of  the 
shades  ;  nor  shall  we  accompany  them  in  the  successive  transfor- 
mations which  made  them  acquainted  with  every  corner  of  the 
earth  and  sky  in  the  infinite  series  of  their  becomes  (to  use  the 
Egyptian  expression)  ;  what  we  have  to  do  is  to  trace  out  the 
most  ancient  of  their  religious  conceptions,  the  conception  which, 
like  the  first  teachings  of  infancy,  was  so  deeply  engraved  upon  the 
soul  and  intellect  of  the  race  as  to  exercise  a  much  stronger 
influence  than  the  later  more  abstract  and  more  philosophical 
theories,  which  were  superimposed  upon  it.  In  this  primitive 
conception  we  ought  to  find  the  determining  cause  of  the  Egyptian 
form  of  tomb.  Its  constitution  was  already  settled  in  the  time  of 
the  ancient  Empire,  and,  from  the  Memphite  dynasties  until  the 
end,  it  remained  unchanged  in  principle.  In  this  constitution  we 
shall  find  embodied  the  essential  idea  adopted  by  the  Egyptians 
when  they  first  attempted  to  find  some  eternal  element  in  man,  or, 
at  least,  some  element  which  should  resist  the  annihilation  of  death 
for  a  period  much  longer  than  the  few  days  which  make  up  our 
mortal  life. 

The  Egyptians  called  that  which  does  not  perish  as  the  dying 
man  draws  his  last  sigh,  the  ka,  a  term  which  M.  Maspero  has 
rendered  as  the  double.  ''  This  double  was  a  duplicate  of  the  body 
in  a  matter  less  dense  than  that  of  the  body,  a  projection,  coloured 
but  aerial,  of  the  individual,  reproducing  him  feature  for  feature,  a 
child  if  coming  from  a  child,  a  woman  if  from  a  woman,  and  a  man 
if  from  a  man,"  ^ 

1  Cotiference,  p.  381.  Mr.  Herbert  Spencer,  in  the  first  chapters  of  his  Principles 
of  Sociology,  has  given  a  curious  and  plausible  explanation  of  how  this  conception 
of  a  double  was  formed.  He  finds  its  origin  chiefly  in  the  phenomena  of  sleep,  of 
dreams,  and  of  the  faintness  caused  by  wounds  or  illness.  He  shows  how  these 
more  or  less  transitory  suspensions  of  animation  led  men  to  suppose  that  death  was 
nothing  but  a  prolonged  interruption  of  life.  He  also  thinks  that  the  actual 
shadow  cast  by  a  man's  body  contributed  to  the  formation  of  that  belief.  But  had 
it  no  other  elements  which  belonged  to  the  general  disposition  of  humanity  in  those 
early  periods  of  intellectual  life  ?  Into  that  question  we  cannot  enter  here  further  than 
to  say  that  Mr.  Spencer's  pages  make  us  acquainted  with  numerous  facts  which 
prove  that  the  beliefs  in  question  were  not  confined  to  a  single  race,  but  were 
common  to  all  humanity. 


Sepulchral  Architecture.  129 

T^his  double  had  to  be  installed  in  a  lodging  suitable  to  its 
existence,  had  to  be  surrounded  by  objects  which  it  had  used  in  its 
former  state,  had  to  be  supplied  with  the  food  which  was  necessary 
for  the  sustenance  of  its  life.  And  all  these  things  it  obtained  from 
the  piety  of  its  relations,  who,  on  fixed  days,  brought  them  to  the 
threshold  of  the  good  diuelling  or  the  eternal  dwelli'ng,  which  were 
the  phrases  used  by  the  Egyptians.^  By  these  offerings  alone 
could  the  hungry  and  thirsty  phantom  which  had  replaced  the 
living  man  be  kept  alive.  The  first  duty  of  the  survivors  was 
to  take  care  that  this  dependent  existence  should  not  be  extin- 
guished by  their  neglect,  to  provide  food  and  drink  for  the  support, 
if  we  may  use  such  a  phrase,  of  the  precarious  life  of  the  dead,  who 
would  otherwise  be  irritated  against  them  and  use  the  almost  god- 
like power  attributed  to  his  mysterious  condition  for  the  punishment 
of  his  ungrateful  posterity.^ 

This  conception  is  not  peculiar  to  Egypt.  The  double  of  the 
Egyptian  sepulchral  records  corresponds  exactly  to  the  eiBcoXov  ^ 
of  the  Greeks  and  the  umbra  of  the  Latins.  Both  Greeks  and 
Latins  believed  that  when  the  funeral  rites  had  been  duly  accom- 
plished, this  image  or  shadow  entered  upon  the  possession  of 
a  subterranean  dwelling  and  began  a  life  which  was  no  more 
than  the  continuation  of  that  in  the  lig^ht.^  The  dead  thus 
remained  in  close  relation  with  the  living,  on  the  one  hand  by 
the  nourishment  which  they  received,  on  the  other  by  the  pro- 
tection which  they  afforded  ;  even  in  the  funeral  repast  they 
took  their  parts,  in  the  strictest  sense  of  the  word,  in  the  eating 
and  drinking.^     They  looked  impatiently  forward  to  these  supplies 

1  This  expression,  which  is  very  common  in  the  Egyptian  texts,  seems  to  have 
made  a  great  impression  upon  the  Greek  travellers.  The  following  passage  of 
DiODORUS  is  well  known  :  "  This  refers  to  the  beliefs  of  the  natives,  who  look  upon 
the  life  upon  earth  as  a  thing  of  minor  importance,  but  set  a  high  value  upon  those 
virtues  of  which  the  memory  is  perpetuated  after  death.  They  call  their  houses 
hotels,  in  view  of  the  short  time  they  have  to  spend  in  them,,  while  they  call  their 
tombs  their  eternal  dniellings  "  (i.  51). 

'•^  The  dead  were  put  under  the  protection  of,  and,  as  it  were,  combined  with, 
Osiris ;  they  talked  of  t/ie  Osiris  so  and  so  in  naming  one  who  was  dead. 

^  EiSwA.a  Ka/AWTiDv  (//.  xxiii.  72  ;   Od.  xi.  476 ;  xxiv.  14). 

*  This  belief  ii5  clearly  stated  in  a  passage  from  Cicero  quoted  by  Fustel :  "  Sub 
terra  censebant  reliquam  vitam  agi  mortuorum  "  [Ti/sc.  i.  16).  This  belief  was  so 
strong  that  it  subsisted  even  after  the  universal  establishment  of  the  custom  of 
burning  the  bodies  of  the  dead. 

•^  Texts  to  this  effect  abound.     Fustel  brought  the  more  remarkable  of  them 

VOL.  L  S 


A  History  of  Art  in  Ancient  Egypt. 


because,  for  a  moment,  they  awoke  their  dormant  thoughts  and 
feehngs  and  gave  them  ghmpses  of  the  true  Hfe,  the  hfe  above 
ground  and  in  the  sunshine.^  If  they  were  kept  waiting  too  long 
they  became  angry  and  revenged  themselves  upon  those  who 
had  caused  their  sufferings.  Woe  to  the  family  or  city  which 
was  not  careful  to  interest  the  dead  in  its  stability  and  thus  to 
associate  them  with  its  prosperity!  ^ 

These  beliefs  seem  to  have  been  common  to  all  ancient  peoples 
during  that  period  of  their  existence  which  is  lost  in  the  shadow 
of  prehistoric  times.  From  India  to  Italy  all  the  primitive  forms 
of  public  and  private  rights  betray  their  presence.  For  this 
fact  and  its  consequences  we  may  refer  our  readers  to  the  fine 
work  of  M.  Fustel  de  Coulange,  La  Cite  antique.^ 

together  in  his  Cite  antique  (p.  14).  We  shall  be  content  with  quoting  three  :  "Son 
of  Peleus,"  said  Neoptolemus,  "  take  this  drink  which  is  grateful  to  the  dead  ;  come 
and  drink  this  blood"  (Hecuba,  536).  Electra  says  when  she  pours  a  libation; 
"  This  drink  has  penetrated  the  earth  ;  my  father  has  received  it  "  (Choephoroj,  162). 
And  listen  to  the  prayer  of  Orestes  to  his  dead  father  :  "  Oh  my  father,  if  I  live  thou 
shalt  have  rich  banquets  ;  if  I  die  thou  wilt  have  no  portion  of  those  smoking  feasts 
which  nourish  the  dead"  (Choephoroe,  482-484).  Upon  the  strange  persistence  of 
this  belief,  traces  of  which  are  still  found  in  Eastern  Europe,  in  Albania,  in  Thessaly, 
and  Epirus,  the  works  of  Heuzey  {Missioji  archcologiqtie  de  Mackioine,  p.  156),  and 
Albert  Dumont  {/e  Balkan  et  l'  Adriatiqiie,  pp.  354-356),  maybe  consulted.  Some 
curious  details  relating  to  the  funeral  feasts  of  the  Chinese  are  to  be  found  in  the 
Comptes  rendus  de  FAcadhnie  des  Inscriptions,  1877,  p.  325.  There  are  some  striking 
points  of  resemblance  between  the  religion  of  China  and  that  of  ancient  Egypt ; 
in  both  one  and  the  other  the  same  want  of  power  to  develop  may  be  found. 
Taking  them  as  a  whole,  both  the  Chinese  and  the  Egyptians  failed  to  emerge  from 
the  condition  of  fetichism. 

1  In  the  eleventh  book  of  the  Odyssey  it  is  only  after  "  they  have  drunk  deep 
draughts  of  black  blood "  that  the  shades  are  capable  of  recognising  Ulysses,  of 
understanding  what  he  says  and  answering.  The  blood  they  swallowed  restored 
their  intelligence  and  powers  of  thought. 

^  The  speeches  of  the  Greek  orators  are  full  of  proofs  that  these  beliefs  had  a 
great  hold  upon  the  popular  mind,  even  as  late  as  the  time  of  Demosthenes.  In 
contested  cases  of  adoption  they  always  laid  great  stress  upon  the  dangers  which 
would  menace  the  city  if  a  family  was  allowed  to  become  extinct  for  want  of  pre- 
cautions against  the  failure  of  the  hereditary  line ;  there  would  then  be  some 
neglected  tomb  where  the  dead  never  received  the  visits  of  gift-bringing  friends,  a 
neglect  which  would  be  visited  upon  the  city  as  a  whole  as  the  accomplice  in  such 
abandonment.  Such  an  argument  and  others  like  it  may  not  seem  to  us  to  be  of 
great  judicial  value,  but  the  talent  of  an  Isjeus  understood  how  to  make  it  tell  with 
an  audience,  or  we  should  not  find  it  so  often  repeated  in  his  pleadings  (see 
G.  Perrot,  L Eloquence  politique  et  judiciare  a  Athenes.  Les  Precurseurs  de  Detnosthene, 

PP-  359-364)- 

2  Seventh  edition,  Hachette,  i8mo,  1879. 


Sepulchral  Architecture. 


131 


With  the  progress  of  centuries  and  the  development  of  reHgious 
thought,    more    elevated    ideas    prevailed.     The    growth    of    the 


!  1  i  I  I  '  '  '  ' 


!    i 


«=^innj 


r/fH^S/i44^^+ 


O/^l    '      1r    " — -'D=;Si.         dte  /f==^JJJj  /-^'yy^  \'---^_  n_.^s^  I 


1  = 

I 
I 


—  f         *  (C^ >       03= 


Fig.  86. — Stele  of  the  nth  dynasty.     Boulak.      Drawn  by  Bourgoin. 

scientific  spirit  tended  to  make  the  notion  of  a  being  suspended 
between    life    and    death    ever    more    strange    and    inadmissible. 


132  A   History  of  Art  in  Ancient  Egypt. 

Experience  accumulated  its  results  and  it  became  daily  more 
evident  that  death  not  only  put  an  end  to  the  activity  of  the 
organs,  but  that,  immediately  upon  its  occurrence,  it  began  to 
dissolve  and  decompose  their  tissues.  As  time  rolled  on  men 
must  have  found  it  very  difficult  to  believe  in  a  shadow  thus  placed 
outside  the  normal  conditions  of  life,  in  a  something  which  was 
not  a  spirit  and  yet  survived  the  destruction  of  its  organs. 

It  would  seem  then  that  observation  and  loo-ical  reflection 
should  soon  have  led  to  the  abandonment  of  a  theory  which 
now  appears  so  puerile ;  but,  even  in  these  scientific  times, 
those  whose  intellects  demand  well  defined  ideas  are  few  indeed.^ 
At  a  period  when  the  diffusion  of  intellectual  culture  and  the 
perfection  of  scientific  methods  add  daily  to  our  accumulations 
of  positive  knowledge,  most  men  allow  their  souls  to  be  stirred 
and  their  actions  to  be  prompted  by  the  vaguest  words  and  notions  ; 
how  much  greater  then  must  the  influence  of  those  confused 
beliefs  and  baseless  images  have  been  in  antiquity  when  but  a 
few  rare  minds,  and  those  ill  provided  with  means  of  research  and 
analysis,  attempted  to  think  with  originality,  clearness  and  freedom. 

The  prestige  of  this  illusion  was  increased  and  perpetuated 
by  its  intimate  connection  with  several  of  those  sentiments  which 
are  most  honourable  to  human  nature.  Such  a  worship  of  the 
dead  surprises  and  even  scandalizes  us  by  its  frank  materialism, 
but  if  we  seek  for  the  source  of  its  inspiration  and  its  primitive 
meaning,  we  find  them  in  the  remembrance  of  lost  objects  of 
affection,  in  feelings  broken  by  the  supreme  separation,  in  the 
gratitude  of  children  to  the  parents  who  gave  them  birth  and 
nourished  their  infancy,  in  the  recognition  by  the  living  of  the 
blessings  which  they  enjoy  through  the  long  and  laborious  efforts 
of  their  ancestors.  There  was  no  doubt  a  perishable  element 
in  the  funerary  ideas  of  Egypt,  an  element  which  the  progress  of 
reason  was  sure  to  destroy,  and  we  may  be  tempted  to  smile 
when  we  think    of  the    Greek    or   Egyptian  giving   himself  the 

'  Mr,  Herbert  Spencer,  in  his  ingenious  and  subtle  analysis  oi primitive  ideas  draws 
our  attention  to  their  frequent  inconsistencies  and  even  positive  contradictions ;  but  he 
shows  us  at  the  same  time  that  the  most  highly  civilised  races  in  these  modern  days 
admit  and  combine  ideas  which  are  logically  quite  as  irreconcilable  as  those  which 
seem  to  us  so  absurdly  inconsistent  when  we  think  of  the  beliefs  of  the  ancients  or 
of  savage  races.  Custom  renders  us  insensible  to  contradictions  which  we  should 
perceive  at  once  were  we  removed  to  a  distance  from  them.  (The  Principles  of 
Sociology,  vol.  i.  pp.  119,  185). 


Sepulchral  Architecturp:.  133 


trouble  to  feed  his  departed  ancestors  with  blood,  milk,  or  honey, 
but  with  all  their  simplicity,  both  one  and  the  other  were  alive  to  a 
truth  which  the  revolutionary  spirit  of  our  days,  with  its  childish 
and  brutal  contempt  for  the  past,  is  often  unable  to  grasp.  They 
realized  the  complete  solidarity  of  one  human  generation  with 
another.  Guided  by  their  hearts  alone  they  anticipated  the  results 
at  which  modern  thought  has  arrived  by  close  and  attentive  study 
of  history.  From  a  reasoned  out  conviction  of  this  truth  and 
its  consequences,  philosophy  now  draws  the  principles  of  a  high 
morality  !  but  long  before  our  days  this  Idea  and  the  tender, 
grateful,  sentiments  which  it  provoked  had  been  a  powerful  instru- 
ment in  the  moral  improvement  of  the  first-born  of  civilization  and 
a  bond  of  union  for  their  civil  and  domestic  life. 

We  have  thought  it  right  to  dwell  upon  this  worship  of  the 
dead  and  to  describe  its  character  at  some  length,  because  the 
beliefs  upon  which  it  was  based  are  not  to  be  found  so  clearly  set 
forth  in  the  art  of  any  other  people.  Their  most  complete,  clear, 
and  eloquent  expression,  in  a  plastic  form,  is  to  be  seen  in  the 
tombs  which  border  the  Nile.  And  why  is  this  so  ?  It  is  because 
the  Egyptian  industries  were  already  in  full  possession  of  their 
resources  at  the  period  when  those  beliefs  had  their  strongest 
hold  over  the  minds  and  feelings  of  the  people.  In  the  case 
of  Greece,  art  did  not  arrive  at  Its  full  development  until  the 
worship  of  the  dead  had  lost  Its  high  place  In  the  national 
conscience.  When  the  Greek  genius  had  arrived,  after  much 
striving,  at  its  complete  power  of  plastic  expression,  the  gods 
of  Olympus  had  been  created  for  several  centuries,  and  art  was 
called  upon  to  interpret  the  brilliant  polytheism  of  Homer  and 
Hesiod,  to  Q^ive  outward  imao^e  to  those  grods,  and  to  construct 
worthy  dwellings  for  their  habitation.  Sculptors,  painters,  and 
architects  still  worked  indeed,  at  the  decoration  of  the  tomb. 
They  strove  to  give  It  beauty  of  shape  and  arrangement,  and  to 
adorn  its  walls  with  bas-reliefs  and  pictures  ;  they  designed  for 
it  those  vases  and  terra  cottas  which,  in  our  own  day,  have  been 
found  in  thousands  in  the  cemeteries  of  Greece  and  Italy,  but  all 
this  was  only  a  subordinate  use  of  their  talent.  Their  ambition 
was  to  build  temples,  to  model  statues  of  Zeus,  Pallas,  and 
Apollo.  On  the  other  hand,  those  distant  ages  in  which  primi- 
tive and  childish  ideas  of  religion  prevailed,  had  no  art  in  which 
to  manifest  their  beliefs  with  clearness  and  precision. 


134  A   History  of  Art  in  Ancient  Egypt. 

It  was  otherwise  in  the  valley  of  the  Nile.  A  well  provided 
industry  and  an  experienced  art  laid  themselves  out  to  inter- 
pret the  popular  beliefs  and  to  defend  the  dead  against  final 
dissolution,  or  the  agonies  of  hunger  and  thirst.  Egypt  did  not 
differ  from  other  nations  in  its  opinions  upon  the  mystery  of 
death.  In  the  infancy  of  every  race  the  same  notions  on  this 
matter  are  to  be  found,  and  in  this  respect  the  only  difference 
between  the  Egyptians  and  the  rest  of  the  world  is  very  much  to 
the  credit  of  the  former  ;  they  rapidly  attained  to  a  degree  of 
civilization  which  was  only  reached  by  other  races  after  their 
religious  development  was  comparatively  mature.  Thanks  to 
this  advantage,  they  Vv^ere  enabled  to  push  their  ideas  to  con- 
sequences which  were  not  to  be  attained  by  tribes  which  were 
little  less  than  barbarous,  and  they  had  no  difficulty  in  expressing 
them  with  sufficient  force  and  precision. 

It  remains  for  us  to  show  the  use  which  the  Egyptians  made  of 
their  superiority  in  doing  more  honour  to  their  dead,  in  guarding 
them  more  safely  against  the  chances  which  might  shorten  the 
duration  or  destroy  the  happiness  of  their  life  in  the  tomb.  The 
fulfilment  of  this  duty  was,  as  the  Greek  travellers  rightly  affirmed, 
their  chief  pre-occupation.  Their  sepulchral  architecture  was,  of 
all  their  creations,  the  most  original  and  the  most  characteristic 
of  their  genius,  especially  in  the  forms  which  we  find  in  the 
cemeteries  of  the  Ancient  Empire.  In  the  time  of  the  New 
Empire,  at  Thebes,  it  is  less  complete  and  homogeneous.  In  the 
latter  the  arrangement  and  decoration  do  not  spring,  as  a  whole, 
from  a  unique  conception  ;  we  find  traces  in  it  of  new  hypotheses 
and  novel  forms  of  belief  These  do  not  supersede  the  primitive 
ideas ;  they  are  added  to  them,  and  they  bear  witness  to  the 
restless  efforts  made  by  human  thought  to  solve  the  problem  of 
human  destiny.  These  apparent  contradictions  and  hesitations 
are  of  great  interest  to  the  student  of  the  Egyptian  religion,  but 
from  the  art  point  of  view  the  Memphite -tombs  are  more  curious 
and  important  than  those  of  later  date.  They  have  the  great 
merit  of  being  complete  in  their  unity  both  of  artistic  form  and  of 
intellectual  conception.  They  are  the  offspring  of  a  single  growth, 
and  are  perfect  in  their  clear  logical  expression.  And  again,  they 
are  the  type  of  all  the  later  tombs,  of  those  at  Abydos,  at  Beni- 
Hassan,  and  at  Thebes.  Certain  details,  indeed,  are  modified, 
but   the  general  disposition  remains   the   same  to   the  end.     We 


Sepulchral  Architecture.  135 

m , 

shall,  therefore,  find  the  ruling  principle  of  Egyptian  sepulchral 
architecture  most  clearly  laid  down  in  the  cemeteries  of  Gizeh 
and  Sakkarah. 

The  first  and  most  obvious  necessity  for  the  obscure  form 
of  life  which  was  supposed  to  commence  as  soon  as  the  tomb 
had  received  its  inmate,  was  the  body.  No  pains,  therefore, 
were  spared  which  could  retard  its  dissolution  and  preserve  the 
organs  to  which  the  double  and  the  soul  might  one  day  return.^ 
Embalming,  practised  as  it  was  by  the  Egyptians,  rendered  a 
mummy  almost  indestructible,  so  long,  at  least,  as  it  remained  in 
the  dry  soil  of  Egypt.  On  the  warm  sands  of  Sakkarah  and  close 
to  the  excavations  from  which  the  fellahs  of  the  corvde  were 
returning  at  the  end  of  their  day's  labour,  we — my  travelling 
companions  and  myself — stripped  a  great  lady  of  the  time  of 
Ramses  of  the  linen  cloths  and  bandages  in  which  she  was  closely 
enveloped,  and  found  her  body  much  in  the  same  condition  as  it 
must  have  been  when  it  left  the  workshop  of  the  Memphite 
embalmer  !  Her  black  hair  was  plaited  into  fine  tresses ;  all  her 
teeth  were  in  place  between  the  slightly  contracted  lips ;  the 
almond-shaped    nails    of  her  feet  and   hands    were  stained  with 

^  The  texts  also  bear  witness  to  the  ideas  with  which  the  complicated  processes 
of  embalming  were  undertaken.  See  P.  Pierret,  Le  Dogme  de  la  Resurrection,  &c., 
p.  10.  "  It  was  necessary  that  no  member,  no  substance,  should  be  wanting  at  the 
final  summons  ;  resurrection  depended  on  that."  "  Thou  countest  thy  members  which 
are  cotnplete  and  intact."  (Egyptian  funerary  text.)  "  Arise  in  To-deser  (the  sacred 
region  in  which  the  renewal  of  life  is  prepared),  thou  august  and  coffined  mummy. 
Thy  bones  and  thy  substance  are  re-united  with  thy  flesh,  and  thy  flesh  is  again  in 
its  place  ;  thy  head  is  replaced  upon  thy  neck,  thy  heart  is  ready  for  thee." 
(Osirian  statue  in  the  Louvre.)  The  dead  took  care  to  demand  of  the  gods  "  that 
the  earth  should  not  bite  him,  that  the  soil  should  not  consume  himi"  (Mariette, 
Feuilles  d'Abydos.)  The  preservation  of  the  body  must  therefore  have  been  an 
object  of  solicitude  at  the  earliest  times,  but  the  art  of  embalming  did  not  attain 
perfection  until  the  Theban  period.  Under  the  ancient  empire  men  were  content 
with  comparatively  simple  methods.  Mariette  says  that  "  more  examples  would 
have  to  be  brought  together  than  he  had  been  able  to  discover  before  the  question 
of  mummification  under  the  ancient  empire  could  be  decided.  It  is  certain,  first, 
that  no  authentic  piece  of  mummy  cloth  from  that  period  is  now  extant ;  secondly, 
that  the  bones  found  in  the  sarcophagi  have  the  brownish  colour  and  the  bituminous 
smell  of  mummies. 

"  Not  more  than  five  or  six  inviolate  sarcophagi  have  been  found.  On  each  of 
these  occasions  the  corpse  has  been  discovered  in  the  skeleton  state.  And  as  for 
linen,  nothing  beyond  a  little  dust  upon  the  bottom  of  the  sarcophagus,  which 
might  be  the  debris  of  many  other  things  than  of  a  linen  shroud."  {Les  Tombes  de 
r  And  en  Empire^  p.  i6.) 


136  A   History  of  Art  in  Ancient  Egypt. 

^ 

henna.  The  Hmbs  were  flexible  and  the  graceful  shapes  but  little 
altered  under  the  still  firm  and  smooth  skin,  which,  moreover, 
seemed  to  be  still  supported  by  flesh  in  some  parts.  Had  it  not 
been  for  its  colour  of  tarred  linen  or  scorched  paper,  and  the  smell 
of  naphtha  which  arose  from  the  body  and  from  the  numberless 
bandages  which  were  strewn  about,  we  might  have  shared  the 
sentiment  attributed  to  Lord  Evandale  in  Theophile  Gautier's 
brilliant  Roman  de  la  Mojiiie  ;  with  an  effort  of  good-will  we  could 
almost  sympathise  with  those  emotions  of  tenderness  and  admi- 
ration which  were  excited  in  the  breast  of  the  young  Englishman 
at  the  sight  of  the  unveiled  charms  of  that  daughter  of  Egypt 
whose  perfect  beauty  had  once  troubled  the  heart  of  the  proudest 
of  the   Pharaohs.^ 

In  order  that  all  the  expense  of  embalming  should  not  be 
thrown  away,  the  mummy  had  to  be  so  placed  that  it  could  not  be 
reached  by  the  highest  inundations  of  the  river.  The  cemeteries 
were  therefore  established  either  upon  a  plateau  surrounded  by  the 
desert,  as  in  the  case  of  Memphis  and  Abydos,  or  in  the  sides  of 
the  mountain  ranges  and  in  the  ravines  by  which  they  were 
pierced,  as  at  Thebes  and  Beni  Hassan.  In  the  whole  valley  of 
the  Nile,  no  ancient  tomb  has  been  discovered  which  was  within 
reach  of  the  inundation  at  its  hio-hest.- 

The  corpse  was  thus  preserved  from  destruction,  first  by  careful 
and  scientific  embalming,  secondly  by  placing  its  dwelling  above 
the  highest  "  Nile."  Besides  this  we  shall  see  that  the  Egyptian 
architects  made  use  of  many  curious  artifices  of  construction  in 
order  to  conceal  the   entrance  to  the   tomb,  and  to  prevent  the 

^  Passalacqua  gives  the  following  description  of  the  mummy  of  a  young  woman 
which  he  discovered  at  Thebes :  "  Her  hair  and  the  rotundity  and  surprising 
regularity  of  her  form  showed  me  that  she  had  been  a  beauty  in  her  time,  and  that 
she  had  died  in  the  flower  of  her  youth."  He  then  gives  a  minute  description  of 
her  condition  and  ornaments,  and  concludes  by  saying  that  "the  peculiar  beauty  of 
the  proportions  of  this  mummy,  and  its  perfect  preservation,  had  so  greatly  im- 
pressed the  Arabs  themselves  that  they  had  exhumed  it  more  than  once  to  show  to 
their  wives  and  neighbours."  {^Catalogue  raisonne  et  historique  des  A?ifiquiics 
decoiivertes  en  Egypte,  8vo.  1826.) 

2  Rhind  describes  several  mummy-pits  in  the  necropolis  of  Thebes  which  receive 
the  water  of  the  Nile  by  infiltration  ;  but,  as  he  himself  remarks,  this  is  because 
those  who  dug  them  did  not  foresee  the  gradual  raising  of  the  valley,  and,  con- 
sequently, of  the  level  attained  in  recent  ages  by  the  waters  of  the  Nile.  It  is 
doubtless  only  within  the  last  few  centuries  that  the  water  has  penetrated  into  these 
tombs.     {Thebes,  its  Tombs  and  their  Tenants,  ^.  153.) 


Sepulchral  Architecture. 


137 


intrusion  of  any  one  coming  with  evil  intentions.  All  kinds  of 
obstacles  and  pitfalls  are  accumulated  in  the  path  of  the  unbidden 
visitor,  with  a  fertility  and  patient  ingenuity  of  invention  which 
has  often  carried  despair  into  the  minds  of  modern  explorers, 
especially  in  the  case  of  the  pyramids.  Mariette  was  fond  of 
saying  that  there  are  mummies  in  Egypt  which  will  never,  in  the 
strictest  sense  of  the  word,  be  brought 
to  Hcrht. 

But  in  spite  of  all  this  pious  and  subtle 
foresight,  it  sometimes  happened  that 
private  hate  or,  more  often,  the  greed  of 
gain,  upset  every  calculation.  Enemies 
might  succeed  in  penetrating  to  the  sepul- 
chres of  the  dead,  in  destroying  their 
bodies,  and  thus  inflicting  a  second  death 
worse  than  the  first ;  or  a  thief  might  drag 
the  corpse  from  its  resting  place,  and 
leave  it  naked  and  dishonoured  upon  the 
sands,  that  he  might,  with  the  greater  ease, 
possess  himself  of  the  gold  and  jewels  with 
which  it  had  been  adorned. 

The  liability  of  the  mummy  to  accident 
had  to  be  provided  against.  The  idea  of 
the  unhappy  condition  in  which  the  dojible 
would  find  itself  w^hen  its  mummy  had 
been  destroyed,  led  to  the  provision  of 
an  artificial  support  for  it  in  the  shape  of 
a  statue.  Art  was  sufficiently  advanced 
not  only  to  reproduce  the  costume  and 
ordinary  attitude  of  the  defunct  and  to 
mark  his  age  and  sex,  but  even  to  render 
the  individual  characteristics  of  his  physi- 
ognomy. It  aspired  to  portraiture ;  and 
the    development    of  writing  allowed   the 

name  and  qualities  of  the  deceased  to  be  inscribed  upon  his  statue. 
Thus  identified  by  its  resemblance  and  its  inscriptions  it  served  to 
perpetuate  the  life  of  the  double,  which  was  in  continual  danger  of 
dissolution  or  evaporation  in  the  absence  of  a  material  support. 

The  statues  were  more   solid   than    the   mummy,   and   nothing 
stood  in  the  way  of  their  multiplication.      The  body  gave  but  one 

VOL.  L  T 


Fig.    87. — Mummy    case   from 
the  1 8th  dynasty.     Boulak. 


138 


A  History  of  Art  ix  Ancient  Egypt. 


chance  of  duration  to  the  double ;  twenty  statues  represented 
twenty  chances  more.  Hence  the  astonishing  number  of  statues 
which  are  sometimes  found  in  a  single  tomb.  The  images  of  the 
dead  w^ere  multiplied  by  the  piety  of  surviving  relations,  and  con- 
sequently the  double  was  assured  a  duration  which  practically 
amounted  to  immortality.^ 

We  shall  see  that  a  special  recess  was  prepared  in  the  thickness 
of  the  built  up  portion  of  the  tomb  for  the  reception  of  wooden  or 
stone  statues,  so   that  they  might  be  kept  out  of  sight  and  safe 


Fig.  88. — Man  and  his  w  ife  in  the  style  of  the  5th  dynasty.     Calcareous  stone.      From  the  Louvre, 

from  all  indiscreet  curiosity.  Other  effigies  were  placed  in  the 
chambers  of  the  tomb  or  the  courts  in  front  of  it.  Finally,  we 
know  that  persons  of  consideration  obtained  from  the  king  per- 
mission to  erect  statues  in  the  temples,  where  they  were  protected 
by  the  sanctity  of  the  place  and  the  vigilance  of  the  priests."^ 

^  Maspero,  Confcrefice,  p.  381. 

■^  Maspero,  Notes  siir  differentes  Points  de  Grammaire  et  d' Histoire,  p.  155.  (In 
the  Recueil  de  Travaiix  relatifs  a  la  Philologie  et  a  F  Anheologie  Egyptieiuie  et 
Assyrienne,  vol.  i.) 


Sepulchral  Architecture. 


139 


From  the  point  of  view  of  the  ancient  Egyptians  such  precau- 
tions were  by  no  means  futile.  Many  of  these  effigies  have  come 
down  to  us  safely  through  fifty  or  sixty  centuries  and  have  found 
an  asylum  in  our  museums  where  they  have  nothing  to  fear  but 
the  slow  effects  of  climate  and  time.     Those  which  remain  intact 


;  J-C'LL-S.- 


FlG.    89. — bekhcin-ka,  his   wife  Ata,   and  his  son  Khneui,  in  the  style  of  the  5th  dynasty. 

Limestone.      From  the  Louvre. 


may  therefore  count  upon  immortality.  If  the  double  required 
nothing  to  preserve  it  from  annihilation  but  the  continued  existence 
of  the  image,  that  of  Chephren,  the  builder  of  the  second  great 
pyramid,  would  be  still  alive,  preserved  by  the  magnificent  statue  of 
diorite  which  is  the  glory  of  Boulak,  and  thanks  to  the  durability 


140 


A   History  of  Art  in  Ancient  Egypt. 


of  its  materia],  it  would  have  every  chance  of  lasting  as  long  as 
the  world  itself.  But,  unhappily  for  the  shade  of  Pharaoh,  this 
posthumous  existence  which  is  so  difficult  of  comprehension  to  us, 
was  only  to  be  prolonged  by  attention  to  conditions  most  of  which 
could  not  lonor  continue  to  be  observed. 


^ 


'    A: 


i 


li'i 

1 1 


IL 


7?     f  nii^-^i  V'  i/J     ill  .,1^1, 


^vlJ 


A,  ' 


f  \ 


hi  — 


11  "^  1 

V  ,'•11 


1 M.  "i; 


\JAJ"i\  / 


V 


c]^.  U:^  .^ 


Fig.  90. — Stele  of  Nefer-oun.     Boulak. 


It  was  entirely  a  material  life.  The  dead-alive  had  need  of  food 
and  drink,  which  he  obtained  from  supplies  placed  beside  him  in 
the   tomb,^   and  afterwards,   when  these  were  consumed,   by    the 

'  Jars,  which  seem  to  have  been  once  filled  with  water,  are  found  in  many  tombs 
of  all  epochs.     Different  kinds  of  dates  are  also  found,  together  with   the  fruit  of 


o 

'B 

o 


3 
O 


o 

C 

•  r-t 


o 

a; 


_bJO 

'C 


IS-, 

13 


bfl 


Si 

&, 


til 


Sepulchral  Architecture.  143 


repasts  which  took  place  periodically  in  the  tomb,  of  which  he 
had  his  share.  The  first  of  these  feasts  was  given  upon  the  con- 
clusion of  the  funeral  ceremonies,^  and  they  were  repeated  from 
year  to  year  on  days  fixed  by  tradition  and  sometimes  by  the 
expressed  wish  of  the  deceased. ^  An  open  and  public  chamber 
was  contrived  in  the  tomb  for  the  celebration  of  these  anni- 
versaries. It  was  a  kind  of  chapel,  or,  perhaps,  to  speak  more 
accurately,  a  saloon  in  w^hich  all  the  relations  and  friends  of 
the  deceased  could  find  room.  At  the  foot  of  the  stele  upon  which 
the  dead  man  was  represented  sacrificing  to  Osiris,  the  god  of  the 
dead,  was  placed  a  table  for  offerings,  upon  which  the  share 
intended  for  the  double  was  deposited  and  the  libations  poured. 
A  conduit  was  reserved  in  the  thickness  of  the  wall  by  which  the 
odour  of  the  roast  meats  and  perfumed  fruits  and  the  smoke  of 
the  incense  might  reach  the  concealed  statues.^ 

The  Egyptians  did  not  trust  only  to  the  piety  of  their  descend- 
ants to  preserve  them  from  a  final  death  by  inanition  in  their 
neglected  tombs.  At  the  end  of  a  few  generations  that  piety 
might  grow  cold  and  relax  its  care  ;  besides,  a  family  might  become 

the  sycamore,  corn,  cakes,  &:c.  See  the  Catalogue  oi  Passalacqua,  pp.  123,  157, 
and  elsewhere.  Quarters  of  meat  have  also  been  found  in  them,  which  are  easily 
recognised  by  their  well-preserved  bones. 

1  Maspero,  Etudes  sur  quelques  Peintures  funeraires,  in  the  Journal  Asiatique, 
May-June,  1880,  p.  387,  et  seq. 

^  In  one  of  the  great  inscriptions  at  Beni  Hassan,  recently  translated  anew  both 
by  M.  Maspero  and  Professor  Birch,  Chnoumhotep  speaks  thus:  "I  caused  to 
prosper  the  name  of  my  father.  I  completed  the  existing  temples  of  the  Ka.  I 
served  my  statues  at  the  great  temples.  I  sacrificed  to  them  their  food,  bread,  beer, 
water,  vegetables,  pure  herbs.  My  priest  has  verified  (I  chose  a  priest  for  the 
Ka, — Maspero).  I  procured  them  from  the  irrigation  of  my  work-people  (I  made  him 
master  of  fields  and  slaves, — Maspero).  I  ordered  the  sepulchral  offerings  of  bread, 
beer,  cattle,  fowl,  in  all  the  festivals  of  Karneter,  at  the  festivals  of  the  beginning  of 
the  year,  the  opening  of  the  year,  increase  of  the  year,  diminution  of  the 
year  (little  year, — Brugsch  and  Maspero),  close  of  the  year,  at  the  great  festival, 
at  the  festival  of  the  great  burning,  at  the  festival  of  the  lesser  burning,  the  five 
intercalary  days,  at  the  festival  of  bread  making  (of  the  entry  of  grain, — Maspero) 
at  the  twelve  monthly  and  half  monthly  festivals,  all  the  festivals  on  the  earth 
(plain),  terminating  on  the  hill  (of  Anubis).  But  should  my  sepulchral  priest  or  men 
conduct  them  wrongly  may  he  not  exist,  nor  his  son  in  his  place." — Birch,  Records 
of  the  Past,  vol.  xii.  p.  71.— Ed. 

^  In  each  opening  of  the  serdab  in  the  tomb  of  Ti,  at  Sakkarah,  people,  pro- 
bably relatives  of  the  deceased,  are  represented  in  the  act  of  burning  incense  in  a 
contrivance  which  resembles  in  form  the  6v/xLaTr]pLov  of  the  Greek  monuments. 
(Mariette,  Notice  des  priiicipaux  Mofiuiiicnis  de  Boulah,  p.  27,  note  i.) 


144 


A  History  of  Art  in  Ancient  Egypt. 


extinct.     All  those  who  could  afford  it  provided  against  such  con- 
tingencies as  these  by  giving  their   tombs   what  we   now  call   a 


Fig.  92. — Table  for  offerings,  Louvre. 

perpetual  foundation.     They  devoted  to  the  purpose  the  revenues 
of  some  part  of  their  property,  which  was  also  charged  with  the 

maintenance  of  the  priest  or 
priests  who  had  to  perform  the 
ceremonial  rites  which  we  have 
described.^  We  find  that,  even 
under  the  Ptolemies,  special 
ministers  were  attached  to  the 
sepulchral  chapel  of  Cheops,  the 
builder  of  the  great  pyramid.^ 
It  may  seem  difficult  to  believe 
that  a  "  foundation "  of  the 
ancient  empire  should  have  sur- 
vived so  many  changes  of  rc^^ime, 
but  the  honours  paid  to  the  early 
kines  had  become  one  of  the 
national  institutions  of  Egypt. 
Each  restoring  sovereign  made  it 
a  point  of  duty  to  give  renewed  life  to  the  worship  of  those  remote 
princes  who  represented  the  first  glories  of  the  national  history. 

1  See  the  paper  by  M.  Maspero  upon  the  great  inscription  at  Siout,  which  has 
preserved  for  us  a  contract  between  Prince  Hapi-Toufi  and  the  priests  of  Ap- 
Motennou,  by  which  offerings  should  be  regularly  made  to  the  prince's  statue, 
Ayhich  was  deposited  in  a  temple  at  Siout.  {Transactions  of  /he  Society  of  Biblical 
Archceology,  vol.  vii.  pp.  1-32.) 

2  It  was  the  same  in  the  case  of  a  still  older  king,  Seneferu,  the  founder  of  the 
fourth  dynasty.  (De  Rouge,  Recherches  sur  les  Monuments  que  I'on  pent  attrihier 
aux  six  premieres  Dynasties  de  Manethon,  p.  41.) 


Fig.   93. — Another  form  of  the  table  for 
offerings.     Boulak. 


Sepulchral  Architecture.  145 


Besides  which  there  were  priests  attached  to  each  necropolis,  who, 
for  certain  fees,  officiated  at  each  tomb  in  turn.  They  were 
identified  by  Mariette  upon  some  of  the  bas-rehefs  at  Sakkarah. 
Their  services  were  retained  much  in  the  same  way  as  masses  are 
bought  in  our  days.^ 

The  same  sentiment  led  to  the  burial  with  the  dead  of  all 
arms,  clothes,  jewels,  and  other  objects  of  which  they  might  have 
need  in  the  next  life.  We  know  what  treasures  of  this  kind  have 
been  obtained  from  the  Egyptian  tombs  and  how  they  fill  the  cases 
of  our  museums.  But  neither  was  this  habit  peculiar  to  Egypt. 
It  was  common  to  all  ancient  people  whether  civilized  or 
barbarous.  Traces  are  to  be  found  even  in  the  early  traditions  of 
the  Hellenic  race  of  a  time  when,  like  those  Scythians  described  by 
Herodotus,^  the  Greeks  sacrificed,  at  the  death  of  a  chief,  his 
wives  and  servants  that  they  might  accompany  him  to  the  next 
world.  When  she  began  to  reveal  herself  in  the  arts  Egypt 
was  already  too  far  civilized  for  such  practices  as  these  ;  thanks  to 
the  simultaneous  development  of  science,  art  and  religion,  she 
found  means  to  give  the  same  advantages  to  her  dead  without 
permitting  Scythian  cruelties.  Those  personal  attendants  and 
domestic  officers  whose  services  would  be  so  necessary  in  another 
life,  were  secured  to  them  at  a  small  expense  ;  instead  of  slaying 
them  at  the  door  of  the  tomb,  they  were  represented  upon  its  walls 
in  all  the  variety  of  their  occupations  and  in  the  actual  moment  of 
labour.  So  too  with  all  objects  of  luxury  or  necessity  which  the 
double  would  wish  to  have  at  hand,  as  for  instance  his  food  and 
drink. ^ 

A  custom  which  would  seem  to  have  established  itself  a  little 
later  may  be  referred  to  the  same  desire  ;  we  mean  the  habit  of 
placing  in  the  tomb  those  statuettes  which  we  meet  with  in  such 
vast  numbers  after  the  commencement  of  the  second  Theban 
Empire.^     Mariette    obtained    some    from   tombs    of  the    twelfth 

1  Tombes  de  T Ancien  Empire,  p. 87.  ^  Herodotus,  iv.  71,72. 

^  In  a  few  rare  cases  the  objects  destined  for  the  nourishment  of  the  double  are 
represented  in  the  round  instead  of  being  painted  upon  the  wall.  In  the  tomb  of 
the  personage  called  Atta,  a  wooden  table,  supporting  terra-cotta  vases  and  plucked 
geese  carved  in  calcareous  stone,  has  been  found.  (Mariette,  Tombes  de  FAncien 
Empire,  p.  17.)  The  vases  must  have  been  full  of  water  when  they  were  placed  in 
the  tomb ;  the  stone  geese  may  be  compared  to  the  papier-m&che  loaves  of  the 
modern  stage. 

*  All  Egyptian  collections  contain   coffers   of  painted  wood,   often   decorated  in 

VOL.    L  U 


146 


A  History  of  Art  in  Ancient  Egypt. 


dynasty,  and  the  sixth  chapter  of  the  Book  of  the  Dead,  which  is 
engraved  upon  them,  seems  to  be  one  of  the  most  ancient.  Egypt- 
tologists  are  now  inchned  to  beheve  that  the  essential  parts  of  this 
ritual  date  back  as  far  as  the  Memphite  period. 

These  statuettes  are  of  different  sizes  and  materials.  As  a  rule 
they  do  not  exceed  from  eight  to  twelve  inches,  but  there  are  a 
few  which  are  three  feet  or  more  in  height.  Some  are  in 
wood,  some  in  limestone,  and  some  in  granite,  but  as  a  rule 
they  are  made  of  that  kind  of  terra  cotta  which,  when  covered 
with  green  or  blue  enamel,  has  been  called  Egyptian  porcelain. 
They  are  like  a  mummy  in  appearance  ;  their  hands  are  crossed 


Fig.  94. — Labourers  heaping  up  ears  of  corn,   from  a  tomb  at  Gizeh.     {Description 

de  VEgyptf.) 


upon  the  breast  and  hold  instruments  of  agriculture  such  as  hoes 
and  picks,  and  a  sack  meant  for  grain  hangs  from  their  shoulders. 
The  meaning  of  all  this  is  to  be  sought  in  the  Egyptian  notions 
of  a  future  life  ;  it  is  also  explained  by  the  picture  in  chapter 
XC.  of  the  Ritual,  which  shows  us  the  dead  tilling,  sowing  and 
harvesting  In  the  fields  of  the  other  world.  The  texts  of  the  Ritual 
and  of  certain  inscriptions  call  these  little  figures  oushebti  or 
answerers  from  the  verb  ousJieb,  to  answer.      It  is  therefore  easy 

the  most  brilliant  fashion,  which  served  to  hold  these  statues  when  they  were  placed 
in  the  tomb.  The  size  and  the  richness  of  their  ornament  depended  upon  the 
wealth  of  the  deceased  for  whom  they  were  made. 


Sepulchral  Architecture 


147 


to  divine  the  part  attributed  to  them  by  the  popular  imagination. 
They  answered  to  the  name  traced  upon  the  tomb  and  acted  as  sub- 
stitute for  its  tenant  in  the  cultivation  of  the  subterranean  reeionsJ 
With  the  help  of  the  attendants  painted  and  sculptured  upon  the 


Figs.  95,  96. — Sepulchral  statuettes,  from  the  Louvre. 

walls  they  saved  him  from  fatigue  and  from   the  chance  of  want. 
This  is  another  branch  of  the  same  old  idea.      In  his  desire  to  take 

^  PiETSCHMAXN  {Dev  Egyptische  Fetisclidietist,  &c.,  p.  155),  has  well  grasped  the 
character  and  significance  of  these  statuettes.    Conf.  Pierret,  Dictiotinaire  cT Arche- 
ologie  egyptietine,  vol.  v.     See  also,  in  connection  with  the  personality  attributed  to 
them  and  to  the  services  which  were  expected  from  them,  a  note  by  M.  Maspero 
Siir  une  Tablette  appartenant  a  M.  Rogers.     {Reaieilde  TraTau\\  vol.  ii.  p.  12.) 


148  A  History  of  Art  in  Ancient  Egytt. 


every  precaution  against  the  misery  and  final  annihilation  which 
would  result  from  abandonment,  the  Egyptian  thought  he  could 
never  go  too  far  in  furnishing,  provisioning  and  peopling  his  tomb. 

The  ingenuity  of  their  contrivances  is  extraordinary.  Food  in 
its  natural  state  would  not  keep,  and  various  accidents,  might,  as 
we  have  shown,  lead  to  the  death  of  the  double  by  inanition.  It 
was  the  same  with  furniture  and  clothes  ;  the  narrow  dimensions 
of  the  tomb,  moreover,  would  forbid  the  accumulation  there  of 
everything  which  its  sombre  tenant  might  desire.  On  the  other 
hand  the  funerary  statuettes  were  made  of  the  most  indestructible 
materials  and  the  bas-reliefs  and  paintings  were  one  with  the 
thick  walls  of  stone  or  living  rock.  These  have  survived  practi- 
cally unaltered  until  our  day.  We  visited  the  tomb  of  Ti  a  short 
time  after  its  chambers  had  been  opened  and  cleared.  It  was 
marvellous  to  see  how  form  and  colour  had  been  preserved  intact 
and  fresh  under  the  sand,  and  this  work  which  was  four  or  five 
thousand  years  old  seemed  to  be  but  lately  finished.  By  the  bright- 
ness of  their  colours  and  the  sharp  precision  of  their  contours 
these  charming  reliefs  had  the  effect  of  a  newly  struck  medal. 
Such  scenes  from  the  daily  life  of  the  people  continued  to  be 
figured  upon  Egyptian  tombs  from  the  old  empire  to  the  new. 
When  their  study  and  comparison  were  first  begun  different 
explanations  were  put  forward.  Some  believed  that  they  were  an 
illustrated  biography  of  the  deceased,  a  representation  of  his 
achievements  or  of  those  over  which  he  had  presided  during  the 
course  of  his  mortal  life  ;  others  saw  in  them  an  illustration  of  his 
future  life,  a  setting  forth  of  the  joys  and  pleasures  of  the  Egyptian 
Elysium. 

Both  these  interpretations  have  had  to  give  way  before  the 
critical  examination  of  the  pictures  themselves  and  the  decipher- 
ment of  their  accompanying  inscriptions.  It  was  soon  perceived, 
through  comparisons  easily  made,  that  these  scenes  were  not 
anecdotic.  On  a  few  very  rare  occasions  they  seem  to  be 
connected  with  circumstances  peculiar  to  the  inhabitant  of  the 
tomb.  There  are  a  few  steles  and  tombs  upon  which  the  dead  man 
seems  to  have  caused  his  services  to  be  described,  with  the  object, 
no  doubt,  of  continuing  in  the  next  world  his  career  of  honour 
and  success  in  this.  Such  an  inscription  is  so  far  biographical, 
and  a  similar  spirit  may  extend  to  the  decorations  of  the  stele 
and    walls    of    the     tomb.      As    an     example     of    such     narrative 


ri 


o 

M 


o 


> 

O 


JO, 
ci 

c 
o 


ci 

6 
o 
u 


1^ 

o 

c 


Sepulchral  Architecture.  151 


epigraphs  we  may  cite  the  long  inscription  of  Ouna,  which 
Skives  us  the  Hfe  of  a  sort  of  grand-vizier  to  the  two  first  Kings 
of  the  sixth  dynasty  ;  ^  also  the  inscriptions  upon  the  tombs  of  those 
feudal  princes  who  were  buried  at  Beni-Hassan.  In  the  latter 
there  are  historical  representations  as  commentaries  upon  the  text. 
Among  these  is  the  often  reproduced  painting  of  a  band  of  Asiatic 
emigrants  bringing  presents  to  the  prince  and  demanding,  perhaps, 
a  supply  of  wheat  in  return,  like  the  Hebrews  in  the  time  of 
Jacob. 

But  all  this  is  exceptional.  As  a  rule  the  same  subjects  occur 
upon  the  tombs  again  and  again,  in  the  persistent  fashion  which 
characterizes  traditional  themes.  The  figures  by  which  the  flocks 
and  herds  and  other  possessions  of  the  deceased  were  numbered 
are  too  great  for  literal  truth. '^  On  the  other  hand  the  pictured 
tradesmen  and  artificers,  from  the  labourer,  the  baker,  and  the 
butcher  up  to  the  sculptor,  seem  to  apply  themselves  to  their  work 
with  an  energy  which  excludes  the  notion  of  ideal  felicity.  They, 
one  and  all,  labour  conscientiously,  and  we  feel  that  they  are 
carrying  out  a  task  which  has  been  imposed  upon  them  as  a  duty. 

For  whose  benefit  do  they  take  all  this  trouble  ?  If  we  attempt 
to  enter  into  the  minds  of  the  people  who  traced  these  images  and 
compare  the  pictured  representations  with  the  texts  which  accom- 
pany them,  we  shall  be  enabled  to  answer  that  question.  Let  us 
take  by  chance  any  one  of  the  inscriptions  which  accompany  the 
scenes  figured  upon  the  famous  tomb  of  Ti,  and  here  is  what  we 
find.  "  To  see  the  picking  and  pressing  of  the  grape  and  all  the 
labours  of  the  country."  Again,  "  To  see  the  picking  of  the  flax, 
the  reaping  of  the  corn,  the  transport  upon  donkeys,  the  stacking 
of  the  crops  of  the  tomb."  Again,  "  Ti  sees  the  stalls  of  the 
oxen  and  of  the  small  animals,  the  gutters  and  water-channels 
of  the  tomb." 

It  is  for  the  dead  that  the  vintage  takes  place,  that  the  flax 
is  picked,  that  the  wheat  is  threshed,  that  oxen  are  driven  into 
the  fields,  that  the  soil  is  ploughed  and  irrigated.  It  is  for  the 
supply  of  his  wants  that  all  these  sturdy  arms  are  employed. 

We  shall  leave  M.  Maspero  to  sum  up  the  ideas  which  presided 
at  the  construction  of  the  Egyptian  tomb,  but  first  we  must  draw 

1  De  RouGfe,  Memoire  sur  les  Monuments  dcs  six  premieres  Dynasties  (p.  80  et  seq.). 
Conf.  Maspero,  Histoire  Ancienne,  pp.  88-92. 

2  See  Mariette,  Tombes  de  VAncien  Empire,  p.  88. 


152 


A   History  of  Art  in  Ancient  Egypt, 


our  readers'  notice  to  the  fact  that  he,  more  than  once,  alludes  to  a 
conception  of  the  future  life  which  differs  somewhat  from  the  early 
Egyptian  notions,  and  belongs  rather  to  the  Second  Theban 
Empire  and  its  successors. 

"  The  scenes  chosen  for  the  decoration  of  tomb  walls  had  a  magic 
intention  ;  whether  drawn  from  civil  life  in  the  world  or  from  that 
of  Hades,  they  were   meant    to    preserve   the    dead  from    danger 

and   to   insure  him   a   happy   existence    beyond    the    tomb 

Their  reproduction  upon  the  walls  of  the  sepulchre  guaranteed 
the  performance  of  the  acts  represented.  The  dottble  shut  up  in 
his  avpi^^^  saw  himself  going  to  the  chase  upon  the  surrounding  walls 
and   he    went   to   the  chase  ;    eating  and   drinking   with   his  wife, 


Fig.  98. — Arrival  in  Egypt  of  a  company  of  Asiatic  emigrants  (Champollion,  pis.  362,  393) 


and  he  ate  and  drank  with  her  ;  crossing  in  safety  the  terrible 
gulfs  of  the  lower  world  in  the  barque  of  the  gods,  and  he  crossed 
them  in  safety.  The  tilling,  reaping,  and  housing  on  his 
walls  were  for  him  real  tilling,  reaping,  and  housing.  So,  too, 
the  statuettes  placed  in  his  tomb  carried  out  for  him  under  magic 

^  This  word,  a-vpiy^  (flute),  was  employed  by  the  Greeks  to  designate  those  long 
subterranean  galleries  cut  in  the  rock  of  the  necropolis  at  Thebes,  in  the  valley 
called  the  Valley  of  the  Kings;  modern  egyptologists  apply  it  in  a  more  general 
sense  to  all  tombs  cut  deeply  into  the  flanks  of  the  mountain.  For  the  reason 
which  led  the  Greeks  to  adopt  a  term  which  now  seems  rather  fantastic,  see  Pierret, 
Dictioiuiaire  d' Archeologie  egyptienne.  The  chief  passages  in  ancient  authors  in 
which  the  term  is  applied  either  to  the  subterranean  excavations  of  Egypt  or  to 
other  galleries  of  the  same  kind,  are  brought  together  by  Jomard  in  the  third 
volume  of  the  Description  {Antiquites,  vol.  iii.  pp.  12-14). 


Sepulchral  Architecture. 


153 


influence  all  the  work  of  the  fields,  and,  like  the  sorcerer's  pestle 
in  Goethe's  ballad,  drew  water  for  him  and  carried  grain.  The 
workmen  painted  in  his  papyri  made  shoes  for  him  and  cooked 
his  food  ;  they  carried  him  to  hunt  in  the  deserts  or  to  fish  in  the 
marshes.  And,  after  all,  the  world  of  vassals  upon  the  sides 
of  the  sepulchre  was  as  real  as  the  double  for  which  they  laboured  ; 
the  picture  of  a  slave  might  well  satisfy  the  shadow  of  a  master. 
The  Egyptian  thought  that  by  filling  his  tomb  with  pictures  he 
insured  the  reality  of  all  the  objects,  people,  and  scenes  represented 
in  another  world,  and  he  was  thus  encouraged  to  construct  his 
tomb  while  he  was  yet  alive.  Relations,  too,  thought  that  they 
were  doing  a  service  to  the  deceased  when  they  carried  out   all 


Fig.  98,  Contimced. — Arrival  in  Egypt  of  a  company  of  Asiatic  emigrants  (Champollion, 

pis.  362,  393). 


the  mysterious  ceremonies  which  accompanied  his  burial.  The 
certainty  that  they  had  been  the  cause  of  some  benefit  to  him 
consoled  and  supported  them  on  their  return  from  the  cemetery 
where  they  had  left  their  regretted  dead  in  possession  of  his 
imaginary  domain."  ^ 

Such  a  belief  is  astonishing  to  us  ;  it  demands  an  eftbrt  of  the 
imagination  to  which  we  moderns  are  in  no  way  equal.  We  have 
great  difficulty  in  realising  a  state  of  mind  so  different  from  what 
ours  has  become  after  centuries  of  progress  and  thought.  Those 
early  races  had  neither  a  long  enough  experience  of  things,  nor  a 


'  Journal  asiatique,  May-June,  18S0,  pp.  419,  420. 


VOL.    L 


X 


154 


A  History  of  Art  in  Ancient  Egypt. 


sufficiently  capable  power  of  reflection  to  enable  them  to  distinguish 
the  possible  from  the  impossible.  They  did  not  appreciate  the 
difference  between  living  things  and  those  which  we  call  inanimate. 
They  endowed  all  things  about  them  with  souls  like  their  own. 
They  found  no  more  difficulty  in  giving  life  to  their  carved  and 
painted  domestics,  than  to  the  mummy  or  statue  of  the  deceased,  or 
to  the  phantom  which  they  called  the  double.  Is  it  not  natural  to 
the  child  to  take  revenge  upon  the  table  against  which  he  hurts 
himself,  or  to  speak  tenderly  to  the  doll  which  he  holds  in  his 
arms  ? 


Fig.    99. — The  tomb  of  Ti  ;  wc/men,   representing  the  lands  of  the  deceased,  carrying  the 

funeral  gifts. 


This  power  to  endow  all  things  with  life  and  personality  is  now 
reserved  for  the  poet  and  the  infant,  but  in  the  primitive  days  of 
civilization  it  belonged  to  all  people  alike.  Imagination  had 
then  a  power  over  a  whole  race  which  in  our  days  is  the  gift  of 
great  poets  alone.  In  the  efforts  which  they  made  to  forestall  the 
wants  of  the  helpless  dead,  they  were  not  content  with  providing 
the  food  and  furniture  which  we  find  upon  the  walls.  They  had  a 
secret  impression  that  these  might  be  insufficient  for  wants  renewed 
through  eternity,  and  they  made  another  step  upon  the  way  upon 
which  they  had  embarked.  By  a  still  more  curious  and  still  bolder 
fiction  than  those  which  had  gone  before,  they  attributed  to  prayer 


Sepulchral  Architecture.  155 

the  power  of  multiplying,  by  the  use  of  a  few  magic  sentences,  all 
objects  of  the  first  necessity  to  the  inhabitants  of  the  tomb. 

Every  sepulchre  has  a  stele,  that  is  to  say,  an  upright  stone 
tablet  which  varied  in  form  and  place  in  different  epochs,  but  always 
served  the  same  purpose  and  had  the  same  general  character. 
Most  of  these  steles  were  adorned  with  painting  and  sculpture  ;  all 
of  them  had  more  or  less  complicated  inscriptions.^  In  the  semi- 
circle which  forms  the  upper  part  of  most  of  these  inscribed  slabs, 
the  dead  person,  accompanied  by  his  family,  presents  offerings  to  a 
god,  who  is  usually  Osiris.  Under  this  an  inscription  is  carved 
after  an  unchanging  formula  :  "  Offering  to  Osiris  (or  to  some  other 
deity,  as  the  case  may  be)  in  order  that  he  may  give  provision  of 
bread,  liquid,  beef,  geese,  milk,  wine,  beer,  clothes,  perfumes,  and 
all  good  and  pure  things  upon  which  the  god  subsists,  to  the  ka 
of  N  .  .  .,  son  of  M  .  .  .  ."  Below  this  the  defunct  is  often 
shown  in  the  act  of  himself  receiving  the  offerings  of  his  family. 
In  both  divisions  the  objects  figured  are  looked  upon  as  real,  as 
in  the  wall  decorations.  In  the  lower  division  they  are  offered 
directly  to  him  who  is  to  profit  by  them  ;  in  the  upper,  the  god  is 
charged  to  see  that  they  are  delivered  to  the  right  address.  The 
provisions  which  the  god  is  asked  to  pass  on  to  the  defunct  are 
first  presented  to  him  ;  by  the  intervention  of  Osiris  the  doubles  of 
bread,  meat  and  drink  pass  into  the  other  world  to  nourish  the 
double  of  man.  But  it  was  not  essential  for  the  gift  to  be  effective 
that  it  should  be  real,  or  even  qnasi-x^'sX  ;  that  its  image  should 
even  be  given  in  paint  or  stone.  The  first-comer  could  procure 
all  things  necessary  for  the  deceased  by  their  enumeration  in  the 
proper  form.  We  find  therefore  that  many  Egyptians  caused  the 
following  invocation  to  passing  strangers,  to  be  engraved  upon 
their   tombs  : 

"  Oh  you  who  still  exist  upon  the  earth,  whether  you  be  private 
individuals,  priests,  scribes,  or  ministers  entering  into  this  tomb,  if 
you  love  life  and  do  not  know  death,  if  you  wish  to  be  in  favour 
with  the  gods  of  your  cities  and  to  avoid  the  terrors  of  the  other 
world,  if  you  wish  to  be  entombed  in  your  own  sepulchres  and  to 
transmit  your  dignities  to  your  children,  you  must  if  you  be 
scribes,  recite  the  w^ords  inscribed  upon  this  stone,  or,  if  not,  you 
must  listen  to  their  recital :  say,  ofiering  to  Amen,  master  of 
Karnak,  that  he  may  give  thousands  of  loaves  of  bread,  thousands 

'  See  above,  Figs.  87  and  91. 


156  A  History  of  Art  in  Ancient    Egypt. 


of  jars  of  drink,  thousands  of  oxen,  thousands  of  geese,  thousands 
of  garments,  thousands  of  all  good  and  pure  things  to  the  ka,  or 
double,  of  the  prince  Entef."  ^ 

Thanks  to  all  these  subtle  precautions,  and  to  the  goodwill  with 
which  the  Egyptian  intellect  lent  itself  to  their  bold  fictions,  the 
tomb  deserved  the  name  it  received,  the  house  of  the  double.  The 
double,  when  thus  installed  in  a  dwelling  furnished  for  his  use, 
received  the  visits  and  offerings  of  his  friends  and  relations  ;  "  he 
had  priests  retained  and  paid  to  offer  sacrifices  to  him  ;  he  had 
slaves,  beasts  of  burden,  and  estates  charged  with  his  support.  He 
was  like  a  great  lord  sojourning  in  a  strange  country  and  having 
his  wants  attended  to  by  intermediary  officials  assigned  to  his 
service.    ^ 

This  analogy  between  the  house  and  the  tomb  is  so  complete 
that  it  embraces  details  which  do  not  seem  very  congruous.  Like 
the  house  of  the  living,  the  tomb  was  strictly  oriented,  but  after  a 
mystic  principle  of  its  own. 

As  soon  as  the  Egyptian  began  to  think  he  perceived  the  most 
obvious  of  the  similarities  between  the  sun's  career  and  that  of 
man.  Man  has  his  dawn  and  his  setting.  Man  grows  from  the 
early  glimmerings  of  infancy  to  the  apogee  of  his  wisdom  and 
strength  ;  he  then  begins  to  decline  and,  like  the  magnified  evening 
sun,  ends  by  disappearing  after  his  death  into  the  depths  of  the  soil. 

In  Egypt  the  sun  sets  every  evening  behind  the  Libyan  chain  ; 
thence  he  penetrates  into  those  subterranean  regions  of  Ament 
across  which  he  has  to  make  his  way  before  the  dawn  of  the  next 
day.  The  Egyptian  cemeteries  were  therefore  placed  on  the  left 
bank  of  the  Nile,  that  is,  in  the  west  of  the  country.  All  the 
known  pyramids  were  built  in  the  west,  and  there  we  find  all  the 
more  important  "cities  of  the  dead,"  the  necropolis  of  Memphis 
and  those  of  Abydos  and  Thebes.  A  few  unimportant  groups  of 
tombs  have  indeed  been  found  upon  the  eastern  bank  ;  but  these 
exceptions  to  a  general  rule  are  doubtless  to  be  explained  by  a 
question  of  distance.      For  any  city  placed  near  the  eastern  border 

1  We  borrow  the  translation  of  this  inscription,  as  well  as  the  reflections  which 
precede  it,  from  M.  Maspero  {Confirence,  p.  382).  According  to  M.  de  Rouge',  it 
dates  from  about  the  twelfth  dynasty.  An  invocation  of  the  same  kind  is  to  be 
found  in  another  epigraph  of  the  same  period,  the  inscription  of  Amoni-Amenem- 
hait.  hereditary  prince  of  the  nome  of  Meh,  at  Beni-Hassan.  See  Maspero,  La 
Grande  Inscription  de  Beni-Hassan,  p.  17  [  {Rtriieil  de  Travaux,  etc.,  vol.  i.  4to.). 

2  Maspero,  Conference,  p.  282. 


Sepulchral  Architecture.  157 

of  the  wider  parts  of  the  Nile  valley,  a-burying-place  in  the  Libyan 
chain  would  be  very  inconvenient  both  for  the  transport  of  the 
dead,  and  for  the  sepulchral  duties  of  the  survivors.^ 

Each  morning  sees  the  sun  rise  as  youthful  and  ardent  as  the 
morning  before.  Why  then  should  not  man,  after  completing  his 
subterranean  journey  and  triumphing  over  the  terrors  of  Ament, 
cast  off  the  darkness  of  the  tomb  and  again  see  the  light  of  day  ? 
This  undying  hope  was  revivified  at  each  dawn  as  by  a  new 
promise,  and  the  Egyptians  followed  out  the  analogy  by  the  way  in 
which  they  disposed  their  sepulchres.  They  were  placed  in  the 
west  of  their  country,  towards  the  setting  sun,  but  their  doors,  the 
openings  through  which  their  inmates  would  one  day  regain  the 
light,  were  turned  to  the  east.  In  the  necropolis  of  Memphis,  the 
door  of  nearly  every  tomb  is  turned  to  the  east,^  and  there  is  not 
a  single  stele  which  does  not  face  in  that  direction.^  In  the  ne- 
cropolis of  Abydos,  both  door  and  stele  are  more  often  turned 
towards  the  south,  that  is  towards  the  sun  at  its  zenith.'*  But 
neither  at  Memphis,  at  Abydos,  nor  at  Thebes  is  there  a  tomb 
which  is  lighted  from  the  west  or  presents  its  inscription  to  the 
setting  sun.^  Thus,  from  the  shadowy  depths  where  they  dwell, 
the  dead  have  their  eyes  turned  to  that  quarter  of  the  heavens 
where  the  life-giving  flame  is  each  day  rekindled,  and  seem  to  be 
waiting  for  the  ray  which  is  to  destroy  their  night  and  to  rouse 
them  from  their  long  repose.^ 

1  Among  the  cemeteries  of  the  right  bank  we  may  mention  that  of  Tell-el- 
Amarna;  where  the  tombs  would  have  been  too  far  from  the  city  had  they  been 
dug  in  the  Libyan  Chain.  The  cemeteries  of  Beni-Hassan  and  of  EiUthyia  {El-Kab) 
are  also  in  the  Arab  Chain.  In  spite  of  these  exceptions,  however,  the  west  was  the 
real  quarter  of  the  dead,  their  natural  habitation,  as  is  proved  by  the  tearful  funeral 
songs  translated  by  M.  Maspero  :  "  The  mourners  before  the  ever-to-be  praised 
Hor-Khom  say,  'O  chief,  as  thou  goest  toward  the  West,  the  gods  lament  thee.' 
The  friends  who  close  the  procession  repeat,  '  To  the  West,  to  the  West,  oh 
praiseworthy  one,  to  the  excellent  West !  '  "  Maspero, 'j5'/«^^  sur  quelques  Pemtures 
funeraires  (Jouj-nal  asiatique,  February-April,  1881,  p.  148). 

2  "  It  is  so,"  says  Mariette,  *'  four  times  out  of  five."  {Les  Tomhes  de  f  Ancien 
Empire,  in  the  Reinie  arch'eologique,  new  series,  vol.  xix.  p.  1 2). 

^  "In  the  further  wall  of  the  chamber,  and  invariably  facing  eastwards,  is  a 
stele."     {Ibidem,  p.  14.) 

*  Mariette,  Abydos,  vol.  ii.  p.  43 

^  The  tombs  in  the  Arab  Chain  form,  of  course,  an  exception  to  this  rule.  The 
unusual  circumstances  which  took  them  eastward  of  the  river  forced  them  also  to 
neglect  the  traditional  law. 

^  The  symbolic  connection  established  by  man   between  the  course  of  the  .sun 


8 


A  History  of  Art  in  Ancient  Egypt. 


The  ideas  and  beliefs  which  we  have  described  were  common 
to  all  Egyptians,  irrespective  of  class.  When  he  felt  his  last  hour 
approaching,  the  humble  peasant  or  boatman  on  the  Nile  was  as 
anxious  as  Pharaoh  himself  to  insure  the  survival  of  his  double 
and  to  guard  against  the  terrors  of  annihilation  : 

Mais,  jusqu'en  son  trepas, 

Le  riche  a  des  honneurs  que  le  pauvre  n'a  pas. 

Those  who,  when  alive,  had  to  be  content  with  a  hut  of  earth  or 
of  reeds,  could  not,  when  dead,  expect  to  have  a  tomb  of  stone  or 
brick,  a  habitation  for  eternity  ;  they  could  not  look  for  joys  in  the 
other  world  which  they  had  been  unable  to  procure  in  this.  So 
that  such  tombs  as  those  which  most  fully  embodied  the  ideas 
we  have  described  must  always  have  remained  the  exclusive  pri- 
vilege  more    or   less  of  the   governing  classes.     These  consisted 


Fig.  ioo. — Lid  of  the  coffin  of  Entef,  nth  dynasty.     Louvre. 

of  the  king,  the  princes  and  nobles,  the  priests,  the  military  chiefs, 
and  functionaries  of  every  kind  down  to  the  humblest  of  the 
scribes  attached  to  the  administration.  As  for  those  Egyptians 
who  did  not  belong  to  this  aristocracy,  they  had  to  be  content  with 
less  expensive  arrangements.  The  less  poor  among  them  at  least 
took  measures  to  be  embalmed  and  to  be  placed  in  a  coffin  of 
wood  or  papier-indchd,  accompanied  by  scarabs  and  other  charms 
to  protect  them  against  malignant  spirits.  The  painted  figures 
upon  the  coffin  also  helped  to  keep  off  evil  influences.  If  they 
could  afford  it  they  purchased  places  in  a  common  tomb,  where  the 
mummies  were  heaped  one  upon  the  other  and  confided  to  the 
care   of   priests   who    performed    the    funerary   rites    for    a    whole 


and  his  own  life  was  well  understood  by  Champollion,  who  used  it  to  explain  the 
paintings  in  the  royal  tombs  at  Thebes.  (See  his  remarks  on  the  tomb  of 
Rameses  V.  on  the  185th  and  following  pages  of  his  Lettres  d' Egypte,  &c.) 


Sepulchral  Architecture. 


159 


chamber  at  once.^      It  was  the  frequent  custom   to  put   with  the 
dead  those  pillows  of  wood  or  alabaster  which  the  Egyptians  seem 


Figs,  ioi,  102. — Scarabs.     Louvre. 


to  have  used  from  the  most  ancient  times  for  the  support  of  their 
heads    in   sleep.     This    contrivance,   which    does    away   with   the 


Figs.  103,  104. — Funerary  amulets.      Oudja  and  ta.     Louvre. 

necessity  for  continually  rearranging  their  complicated  head-dress, 
is  still  used  by  the  Nubians  and  Abyssinians. 

1  Upon  the  papyrus  known  as  the  Papyrus  Casatt,  mention  is  made  of  a  priest 
who  is  charged  to  watch  over  a  whole  collection  of  mummies. 

"  This  is  the  list  of  bodies  belonging  to  Osorvaris  : — 

"Imouth,  son  of  Petenefhotep,  his  wife  and  children  ; 

"  Medledk,  the  carpenter,  his  wife  and  children  ; 

"  Pipee,  his  wife  and  children,  from  Hermouth  ; 

"  The  father  of  Phratreou,  the  fuller  ; 

"  Aplou,  the  son  of  Petenhefhotep  the  boatman,  his  wife  and  children,  from 
Thebes  ; 

"  Psenmouth,  the  carpenter,  his  wife  and  children  ; 

"  Psenimonthis,  the  mason  ; 

"  Amenoth,  the  cowherd." 

There  are  many  more  lists  of  the  same  kind.  The  above  is  cited  from  M.  E.  Le 
Blant  [Tables  cgypticnnes  a  Inscriptions  grecques^  p.  6,  1875,  8vo.). 


i6o 


A   History  of  Art  in  Ancient  Egypt. 


But  those  who  could  procure  even  these  sHght  advantages  were 
still  among  the  favourites  of  fortune.  Many  were  unable  to  obtain 
even  this  minimum  of  funeral  honours.  On  the  confines  of  all  the 
great  cemeteries,  at  Thebes  as  well  as  at  Memphis,  corpses  are 
found  deposited  in  the  loose  sand  two  or  three  feet  from  the 
surface.  Some  of  these  are  packed  in  the  leaves  of  the  palm, 
others  are  roughly  enveloped  in  a  few  morsels  of  linen.  They 
have  been  hastily  dipped  in  a  bath  of  natron,  which  has  dirtied 
rather  than  embalmed  them.^  Sometimes  even  these  slender 
precautions  have  been  omitted.  Bodies  have  been  found  in  the 
earth  without  vestige  of  either  coffin  or  linen  swathes.  The  sand 
seems  to  have  been  intrusted  with  the  work  of  drying  them, 
and  they  have  been  found  in  our  days  in  the  condition  of 
skeletons. 


P'lG.  105. — Pillow,  Louvre. 

On  the  other  hand,  the  fortunate  ones  of  the  world,  those  who 
were  so  easy  in  their  circumstances  in  this  life  that  they  could 
place  themselves  in  the  same  happy  condition  in  the  next,  spared 
no  expense  in  anything  connected  with  their  burial.  They  never 
allowed  themselves  to  be  surprised  by  death,  as  we  so  often  do. 
Whether  kings  or  private  individuals,  they  made  their  preparations 
while  they  were  still  alive,  and  caused  their  tombs  to  be  constructed 
under  their  own  eyes.^     Their  forethought  when  living  and  the 

^  See  in  the  interesting  work  of  INIr.  H.  Rhind  {Thebes,  its  Tombs  and  their 
Tenants,  London,  1862,  8vo.),  the  chapter  headed  A  Burial-place  of  the  Poor. 

2  Mariette,  Tombes  de  V Ancien  Empire,  p.  83.  See  also  the  great  inscription  of 
Beni-Hassan,  the  first  hnes  of  which  run  thus:  "The  hereditary  chief.  .  .  . 
Khnumhotep  ....  has  made  a  monument  for  the  first  time  to  embeUish  his 
district ;  he  has  sculptured  his  name  for  ever  ;  he  has  embellished  it  for  ever  by  his 


Sepulchral  Architecture.  i6i 

piety  of  relations  spared  nothing  that  could  add  to  the  beauty  and 
convenience  of  dwellino-s  which  were  to  be  the  eternal  restingf 
places  of  their  inmates.  The  palaces  of  the  princes  and  rich  men 
of  Egypt  were  so  lightly  built  that  they  have  left  no  traces  upon  the 
soil  ;  but  many  of  their  tombs  have  subsisted  uninjured  to  our  day, 
and  it  is  from  them  that  we  have  obtained  our  treasures  of  Egyptian 
art.  All  the  other  nations  of  the  ancient  world  followed  the 
good  example  thus  set,  or  rather,  to  speak  more  accurately, 
being  all  penetrated  with  similar  ideas,  they  took  similar 
courses  without  borrowinof  one  from  the  other.  Whenever  we 
moderns  have  opened  any  of  those  ancient  tombs  which  have 
happily  remained  intact,  we  have  been  met  by  the  same  dis- 
coveries. Whether  it  be  in  Egypt  or  Phoenicia,  in  Asia  Minor, 
Cyprus  or  Greece,  in  Etruria  or  Campania,  the  same  astonishing 
sight  meets  our  eyes.  The  tombs  are  filled  with  precious  objects 
and  chefs  doettvres  of  art  which  their  depositors  had  intended  never 
again  to  see  the  light  of  day. 

In  modern  times,  when  piety  or  pride  stimulates  to  the 
decoration  of  a  tomb,  all  the  care  of  the  architect,  the  sculptor  and 
the  painter  is  given  to  the  outside,  to  the  edifice  which  surmounts 
the  actual  grave.  The  grave  or  other  receptacle  for  the  coffin  is 
as  plain  and  simple  in  the  most  sumptuous  monuments  of  our 
cemeteries  as  in  the  most  humble.  Our  funerary  architecture  is 
based  upon  our  belief  that  the  tomb  is  empty  ;  that  the  vital  part 
of  the  deposit  confided  to  it  has  escaped  to  rejoin  the  current  of 
eternal  life.  Under  such  conditions  the  tomb  becomes  above  all 
things  a  commemorative  structure,  a  more  or  less  sincere 
manifestation  of  the  grief  of  a  family  or  of  society  at  large  for  the 
loss  of  one  of  its  members.  As  for  the  narrow  pit  into  which  the 
"mortal  coil  "  is  lowered,  all  that  we  demand  of  it  is  that  it  should 

chamber  of  Karneter;  he  has  sculptured  the  names  of  his  household;  he  has 
assigned  their  place.  The  workmen,  those  attached  to  his  house,  he  has  reckoned 
amongst  his  dependants  of  all  ranks."  [Birch,  Records  of  the  Past,  vol.  xii. 
p.  67. — Ed.]  It  was,  no  doubt,  in  order  to  conform  to  the  Egyptian  custom  that 
Antony  and  Cleopatra  commenced  in  their  lifetime  that  tomb  which  Augustus 
ordered  to  be  finished  after  their  death  (Suetonius,  Augustus,  17).  "  To  be  laid  to 
rest  in  the  tomb  which  he  had  made  for  himself  and  furnished  with  every  necessary 
was  the  greatest  good  which  the  gods  could  insure  to  an  Egyptian.  In  Papyrus  IV. 
at  Boulak  we  find  the  following  phrases  :  '  Be  found  with  thy  dwelling  finished  in 
the  funerary  valley  :  in  every  enterprise  which  thou  meditatest  may  the  morning 
7vhen  thy  body  shall  he  hid  be  present  to  thee.'  "  (From  the  French  of  M.  Maspero, 
Jour7ial  asiatiqve,  7th  series,  t.  xv.  p.  165,  note  i). 

VOL.    L  Y 


1 62  A  History  of  Art  in  Ancient  Egypt. 

be  deep  enough  and  properly  closed.  Art  makes  no  attempt  to 
illumine  its  darkness.  She  leaves  to  workmen  the  task  of 
excavation  and  of  buildino-  its  walls  and  confines  herself  to  the 
visible  parts  of  the  tomb.  The  dead  within  furnishes  the  pretext 
for  her  activity,  but  it  is  the  admiration  of  the  living  that  is  her 
real  incentive. 

The  ideas  of  the  ancients  on  this  matter  were,  as  we  have  seen, 
very  different.  They  looked  upon  the  tomb  as  an  inhabited 
house  ;  as  a  house  in  which  the  dead  was  to  lead  some  kind  of 
existence.  Rich  men  wished  their  tombs  to  look  well  outside, 
even  to  the  distant  spectator,  but  it  was  to  the  inside  that  their 
chief  attention  was  turned.  They  washed  to  find  there  all  the 
necessaries,  the  comforts,  the  luxuries,  to  which  they  had  been 
accustomed  during  life.  So  we  find  that  the  Egyptians,  the 
Greeks  or  the  Etruscans,  were  willing  enough,  when  they  built 
their  own  tombs  or  those  of  their  relations,  to  throw  a  tumulus  of 
earth  above  it,  or,  later,  a  constructed  building  which  was  con- 
spicuous at  a  distance.  In  those  sepulchres  which  were  cut  out  of 
the  side  of  a  mountain,  the  fronts  were  carved  with  frieze, 
pediment  and  columns  into  the  shape  of  a  regularly  constructed 
portico  ;  but  the  chief  object  of  solicitude  to  the  proud  possessor 
of  such  a  tomb,  was  its  internal  furnishing  and  disposition.  For 
him  there  was  no  removal  should  he  be  discontented  with  his 
lodging.  When  a  man  is  condemned  by  illness  or  accident  to 
keep  his  room,  he  takes  care  to  surround  himself  with  everything 
that  he  may  want.  He  gathers  immediately  about  him  all  the 
comforts  and  luxuries  which  he  can  afford  ;  and  death  is  an  illness 
from  which  there  is  no  recovery. 

Impelled  by  such  ideas  as  these,  the  ancients  filled  their  tombs 
with  precious  objects  and  decorated  them  with  sumptuous  art,  all 
the  more  that  they  seemed  well  guarded  against  intrusion  for  the 
sake  of  gain.  Thus  the  Achseans  of  Mycenae  (if  that  be  the 
proper  name  of  those  mysterious  people)  buried,  in  the  sepulchres 
discovered  by  Dr.  Schliemann,  the  innumerable  objects  of  gold  and 
silver  which  now  fill  the  museum  of  Athens  ;  thus  the  tombs  of 
Boeotia  were  filled  with  those  marvels  of  crrace  and  delicate 
workmanship,  the  terra  cottas  of  Tanagra  ;  and  those  of  Etruria 
and  Campania  with  the  most  beautiful  painted  vases  ever 
produced  by  Greek  taste. 

Identity  of  religious  conception    thus   led,   from   end  to  end  of 


The  Tomb  under  the  Ancient  Empire.  163 


the  antique  world,  to  funerary  arrangements  which  bore  a  curious 
resemblance  one  to  another,  so  that  sepulchral  architecture  among 
the  ancients  had,  as  a  whole,  a  very  different  character  from  that 
of  the  moderns.  This  character  is  more  strongly  marked  in 
Egypt  than  anywhere  else,  and  therefore  we  have  studied  it  in 
detail.  The  general  observations  to  which  it  has  given  rise  have 
been  made  once  for  all,  and  we  shall  not  have  to  repeat  them 
when  we  describe  the  funeral  customs  of  other  ancient  peoples. 
We  shall  then  confine  ourselves  to  pointing  out  the  slight 
differences  which  naturally  spring  up  in  the  several  interpretations 
of  a  common  belief. 

We  have  still  to  show  how  the  varying  circumstances  of  time 
and  place  caused  the  Egyptian  tomb  to  pass  through  certain 
modifications  of  form  and  decoration,  which,  however,  were  never 
of  so  radical  a  nature  as  to  affect  its  general  appearance  and 
arrangement.  Until  Egypt  became  a  mere  geographical  expression 
and  her  venerable  civilization  lost  its  independence  and  originality, 
these  latter  remained  practically  unchanged. 


§  2. —  The  Tomb  under  the  Ancient  Einpire. 

Among  the  tombs  which  date  from  the  time  of  the  ancient 
empire,  the  most  interesting  to  the  traveller  are,  of  course,  the 
Pyramids.  Long  before  his  arrival  at  Cairo  he  sees  the  summits 
of  those  artificial  mountains  rising  into  the  air  above  the  vapours 
raised  by  the  sun,  and  above  the  dust  thrown  up  by  the  teeming 
population  of  the  city.  At  that  distance  their  peaks  seem  light 
and  slender  from  their  height  above  the  horizon  (Plate  I.  2). 

The  tourist's  first  visit  is  paid  to  the  Pyramids,  and  many  an 
European  leaves  Egypt  without  seeing  any  other  ancient  building. 
He  thinks  that  he  has  qualified  himself  to  discourse  upon 
Egyptian  architecture  because  a  few  shouting  Arabs  have  landed 
him,  exhausted,  upon  the  topmost  stone  of  the  pyramid  of  Cheops, 
and  have  painfully  dragged  and  thrust  him  along  those  passages  of 
the  interior  which  wull  ever  be  among  his  most  disagreeable  recol- 
lections. During  all  this  his  eyes  and  thoughts  are  entirely  given 
to  the  preservation  of  his  own  equilibrium,  and  he  sees  nothing  of 
the  real  constitution  of  the  structure  he  has  come  to  visit. 

In   spite  of  the  wonderful   panorama  which    repays  the  fatigues 


164  A   History  of  Art  in  Ancient  Egypt. 

of  the  ascent,  and  of  the  overpowering-  impression  made  upon  the 
mind  by  their  colossal  mass,  the  Pyramids,  as  we  see  them  to-day 
are  far  from  being  the  most  complete  and  interesting  of  the 
sepulchral  monuments  left  to  us  by  the  early  dynasties.  The 
largest  and  best  preserved  are  not  so  old  as  some  of  the  tombs  in 
the  necropolis  of  Memphis,  and,  royal  burying-places  as  they  are, 
their  arrangement  and  ornamentation  are  less  rich  and  expressive 
than  those  of  many  sepulchres  built  by  private  individuals.  Many 
of  the  latter,  in  their  comparatively  restricted  dimensions,  answer 
better  to  the  definition  of  a  tomb  suggested  to  us  by  our  study 
of  the  national  beliefs. 

We  shall,  therefore,  reserve  the  Pyramids  for  future  treatment, 
and  in  our  review  of  the  successive  forms  taken  by  sepulchral 
architecture,  we  shall  assign  the  first  place  to  those  private  tombs, 
dating  from  the  Ancient  Empire,  which  are  to  be  found  in  the 
necropolis  of  Memphis.  Notwithstanding  a  few  dififerences,  to 
which  we  shall  refer  hereafter,  these  tombs,  as  a  whole,  can  be 
traced  to  a  single  type,  of  which  Lepsius  was  the  first  to  perceive 
the  interest.^  This  type,  which  was  first  clearly  brought  to  light 
by  the  many  and  deep  excavations  carried  out  by  Mariette,  has 
been  known  for  some  years  past  by  the  Arab  term  mastaba,'^  which 
means  literally  a  bench,  a  bench  of  stone  or  wood.  This  name 
was  given  by  the  labourers  employed  upon  the  excavations,  and 
seemed  well  adapted  to  their  long  and  low  shapes,  which  bear 
some  resemblance  to  those  divans,  or  ottomans,  which  are  found 
in  every  room  of  an  oriental  dwelling.  Mariette  was  struck  by 
the  fitness  of  the  expression,  and  used  it  ever  after  to  designate 
that  particular  kind  of  tomb. 

Mariette  will  be  our  constant  guide  in  this  part  of  our  study. 
After  having  opened  many  hundreds  of  these  monuments, 
he  published  in  the  Revue  archeologiqite,  what  we  may  call  a 
theory  of  the  fjiastaba.^  In  all  essential  matters  we  shall  allow 
his  words  to  speak  for  themselves ;  when  he  enters  into  more 
detail  than  is  necessary  for  our  purpose,  w^e  shall  content  ourselves 
with  epitomizing  his  descriptions. 

^  Briefe  aus  ^Egypten,  p.  23  et  seg.  Before  the  Prussian  commission  left  Middle 
for  Upper  Egypt  they  had  studied  130  private  tombs,  of  wliich  the  principal  ones 
are  figured  in  the  Denk)nceler. 

2  Lexicographers  do  not  seem  to  know  the  origin  of  this  word  ;  they  believe  it 
to  be  foreign,  perhaps  Persian. 

^  Vol.  xix.  (1869),  pp.  1-22  and  Si-89. 


The  Tomb  under  the  Ancient  Empire.  i6  = 


THE  MASTABAS  OF  THE  NECROPOLIS  OF  MEMPHIS. 

The  space  over  which  the  monuments  which  we  propose  to 
describe  are  spread,  is  on  the  left  bank  of  the  Nile,  and  extends 
from  Abou-Roask  to  Dashoitr ;  it  is  thus,  in  all  probability,  the 
largest  cemetery  in  the  world,  being  more  than  fifteen  miles 
in  length,  and  of  an  averao-e  width  of  from  two  to  two  and 
a  half  miles.^  It  was,  in  a  word,  the  burial-place  for  Memphis 
and  its  suburbs,  and  Memphis  seems  to  have  been  the  largest 
city  of  Egypt,  and  to  have  boasted  an  antiquity  which  only  Thinis 
could  rival.  Excavations  have  failed,  apparently,  to  confirm  the 
assertion  of  Strabo,  who  describes  the  early  capital  of  Egypt  as 
reaching  to  the  foot  of  the  Libyan  chain.  On  the  contrary,  it 
seems  to  have  been  confined  between  the  canal  which  is  called  the 
BaJir  Yussef  and  the  Nile.  It  would  thus  have  formed  a  very 
long  and  rather  narrow  city,  close  upon  the  river,  of  which  the 
site  may  still  be  traced  by  the  more  or  less  barren  hillocks  strewn 
with  blocks  of  granite  and  fragments  of  walls,  which  crop  up  from 
the  plain  between  Gizeh  and  Chinbab.  For  forty  centuries 
there  was  a  continual  procession  of  corpses  from  Memphis  itself, 
and  probably  from  towns  on  the  other  side  of  the  Nile,  such  as 
Heliopolis,  to  the  plateau  which  lies  along  the  foot  of  the  Libyan 
chain.  The  formation  of  this  plateau  makes  it  peculiarly  well 
adapted  for  the  purpose  to  which  it  was  put.  It  consists  of  a  thick 
bed  of  soft  limestone,  covered  by  a  layer  of  sand  which  varies  in 
depth  from  many  yards  to  a  few  feet  according  to  the  inequalities 
of  the  ground  beneath  it. 

It  was  easy,  therefore,  either  to  lay  bare  the  rock  and  to 
construct  the  tomb  upon  it,  or  to  dig  the  mummy  pits  in  its 
substance,  and  the  winds  might  be  trusted  to  quickly  cover  the 
grave  with  sand  which  would  protect  it  when  made.  The  same 
sand  covered  the  coffinless  corpse  of  the  pauper  with  its  kindly 
particles.  Age  after  age  the  dead  were  interred  by  millions  in  this 
great  haven  of  rest.  At  first  there  was  plenty  of  room,  and  the 
corpses  were  strewn  somewhat  thinly  in  the  sand,^  but  with  time 

1  Ebers  {^4igypten,  p.  137)  gives  this  necropolis  a  length  of  more  than  forty-five 
miles,  but  in  making  it  extend  to  Meidoum  he  seems  to  be  exaggerating. 

-  Upon  the  plateau  which,  at  Sakkarah,  extends  westwards  of  the  stepped  pyramid 
the  manner  in  which  the  necropolis  was  developed  can  be  readily  seen.      In  walking 


1 66  A   History   of  Art  in  Ancient  Egypt. 

economy  of  space  had  to  be  practised,  until  at  last  bodies  were 
squeezed  into  the  narrowest  spaces  between  older  inhabitants. 
Sometimes  these  new  comers  even  intruded  into  the  tombs  of 
those  who  had  gone  before  them,  and  that  without  always 
troubling  themselves  to  conceal  their  usurpation  by  effacing  the 
name  of  the  rightful  owner.^   . 

The  number  of  tombs  was  increased  to  a  prodigious  extent 
by  the  non-employment  of  those  family  tombs  which,  as  we  shall 
see,  were  made  use  of  by  the  Phoenicians,  the  Greeks,  and  the 
Etruscans.  The  Egyptian  sepulchre  was  a  personal  appanage. 
The  husband  and  father  of  a  family  admitted  into  it  only  his  wife 
and  such  of  their  children  as  died  young.  The  son,  when  he  in 
turn  became  the  head  of  a  family,  built  a  tomb  for  himself.  Each 
generation,  each  human  couple  marked  their  passage  through  the 
world  by  the  erection  of  a  new  tomb. 

All  the  mastabas  belong  to  the  period  of  the  Memphite  empire. 
Those  who  built  them  were  able  to  give  free  play  to  their  fancies, 
and  to  develop  the  structure,  both  above  and  below  ground,  both 
in  arrangement  and  in  decoration,  to  any  extent  they  pleased.  We 
may  therefore  look  upon  them  as  the  freest,  the  most  spontaneous, 
and  the  most  complete  expression  of  the  ideas  formed  by  the 
men  of  that  remote  age  concerning  death  and  the  life  beyond 
the  grave. 

The  mastabas  of  Sakkarah  will  receive  most  of  our  attention, 
and  in  describing  them  we  shall  often  have  occasion  to  quote  the 
words  of  Mariette.^  Those  which  are  to  be  found  in  the  more 
northern  part  of  the  necropolis,  in  the  neighbourhood  of  the 
Great  Pyramid,  differ  only  in  unimportant  details  from  those  at 
Sakkarah.  The  general  appearance  now  presented  by  these 
monuments  may  be  guessed  from  the  sketch  which  ]\I.  Bourgoin 

eastwards,  that  is,  from  the  pyramid  towards  the  cultivated  land,  we  pass  a  first  zone 
of  tombs  which  date  from  the  Ancient  Empire,  a  second  which  possesses  sculptures 
of  the  twenty-sixth  dynasty,  and  a  third  which  dates  from  the  Greek  period. 

We  may  quote  as  an  interesting  example  of  such  usurpation  the  Theban  tomb 
first  opened  by  a  Scottish  traveller,  Henry  Rhind,  to  whose  interesting  work 
{Thebes,  its  Tombs  and  their  Tenants,  Ancient  and  Present,  with  a  Record  of  Excava- 
tions in  the  Necropolis,  Longman,  1862,  8vo.)  we  shall  often  have  to  refer.  This 
tomb  seems  to  have  been  made  in  the  reign  of  Amenophis  III.  by  a  brother  and 
sister  whose  statues  were  found  in  it,  but  it  also  contained  Sebau,  son  of  Menkara, 
a  high  official  of  the  time  of  the  Ptolemies,  with  his  wife  and  all  his  family  (c.  iv.). 

2  Mariette  {Voyage  dans  la  haute  Egypte,  p.  32)  thought  that  the  word  Sakkarah 
was  an  ancient  name  derived  from  Socharis,  a  Memphite  form  of  Osiris. 


The  Tomb  under  the  Ancient  Empire. 


167 


has    sent    us    of    the    tomb    of    Sabou   (Fig.    106).       The    other 
mastabas  figured  by   us  have  all   been   more  or  less  restored. 

"  The  mastaba  is  a  massive  structure,  rectangular  on  plan,  with 
four  faces  of  plain  walling,  each  being  inclined  at  a  stated  angle 
towards  their  common  centre.  This'  inclination  has  led  some 
people  to  assert  that  it  is  nothing  more  than  an  unfinished  pyramid. 
Such  an  idea  is  refuted,  however,  by  the  fact  that  the  divergence 
from  the  perpendicular  is  in  some  cases  so  slight  that,  were  the 
walls  prolonged  upwards,  their  ridges,  or  aretes,  would  not  meet 
for  some  eight   or  nine   hundred   yards.      The  mastaba   might  be 


Fig.  106. — Actual  condition  of  a  mastaba.     The  tomb  of  Sabou.     Drawn  by  Bourgoin. 


more  justly  compared  to  the  space  comprised  between  two  hori- 
zontal sections  of  an  obelisk,  supposing  the  obelisk  to  have  an 
oblong  base. 

"  The  major  axis  of  the  rectangle  upon  which  these  structures  are 
planned,  always  runs  due  north  and  south,  and  at  the  pyramids  of 
Gizeh,  the  necropolis  of  the  west,  they  are  arranged  upon  a  sym- 
metrical plan  so  as  to  resemble  a  chess  board  on  which  all  the 
squares  are  strictly  oriented.^  The  more  carefully  built  mastabas 
are  oriented  according  to  the  true  astronomical  north.  All  the 
others  show  the  same  intention,  and,  in  those  instances  where  an 
error  of  a  few   degrees  is  to   be  discovered,   it   is  to    be   clearly 

-  The  way  in  which  the  mastabas  were  arranged  with  respect  to  each  other  is  well 
shown  in  plates  xiv.  and  xviii.  of  Lepsius's  first  volume  (map  of  the  pyramids 
of  Gizeh  and  panorama  taken  from  the  summit  of  the  second  pyramid). 


i68 


A   History  of  Art  in  Ancient  Egypt. 


attributed  to  carelessness  on  the  part  of  the  builder,  a  common 
fault  in  these  tombs,  and  not  to  a  difference  of  intention.  We 
often  find  that  the  northern  face  is  not  strictly  parallel  to  the 
southern,  nor  that  on  the  west  to  that  on  the  east. 

"  Although  not  varying  much  from  true  orientation,  the  mastabas 
of  Sakkarah  are  not  arranged  with  the  symmetry  which  distin- 
guishes those  on  the  south  and  west  of  the  Great  Pyramid.^  They 
are  sprinkled  about  in  a  rather  haphazard  fashion.  Here  we  find 
them  well  interspaced  and  there  actually  placed  one  upon  another. 


KSii /^    /  \  7  ^-  I.    v::!i  /m  /._  r 


Fig.    107. — Three  mastabas  at  Gizeh.     Perspective  view,   after  the  plan  of  Lepsius. 

{Denkmcrlei-,  vol.  i.  pi.  24.) 

It  follows  that  the  chess-board  arranofement  which  is  so  con- 
spicuous  at  Gizeh  is  not  here  to  be  noticed.  Even  at  Sakkarah 
there   were  streets  between   the  rows  of  tombs,  but  they  are   so 

1  The  general  aspect  of  this  city  of  the  dead,  and  the  regularity  of  its  monuments, 
made  a  great  impression  upon  the  members  of  the  "  Institut  d'Egypte."  The  fol- 
lowing are  the  words  of  Jomard  {Description,  vol.  v.  p.  619) :  "  From  the  top  of  the 
building  one  sees  an  infinite  quantity  of  the  long  rectangular  structures  extending 
almost  to  the  Pyramids.  They  are  carefully  oriented,  and  exactly  aligned  one  with 
another.  I  counted  fourteen  rows  of  them,  in  each  direction,  on  the  west  of  the 
Great  Pyramid,  and  as  many  on  the  east,  making  nearly  four  hundred  in  all.  The 
sand  under  which  many  of  them  are  buried  leaves  their  forms  easily  distinguishable." 
Since  the  time  of  Jomard  many  of  the  mastabas  have  been  changed  by  the  excava- 
tions into  mere  formless  heaps  of  debris,  and  yet  the  general  arrangement  can 
still  be  clearly  followed. 


The  Tomb  under  the  Ancient  Empire, 


169 


irregularly  placed,  and  they  are  often  so  narrow,  many  of  them 
being  nothing  more  than  blind  alleys,  that  the  inexperienced  visitor 
may  well  fancy  himself  in  a  maze. 

"  The  Sakkarah  mastabas  are  built  either  of  stone  or  brick. 

"  The  mastabas  of  stone  are  of  two  kinds  :  those  of  a  very  hard 
blue  siliceous  limestone  and  those  of  a  softer  chalky  limestone 
which  is  found  upon  the  spot.  This  latter  stone  was  used  for  the 
Stepped  Pyramid.  The  tombs  upon  which  it  was  used  seem  to 
be  much  the  oldest  in  the  necropolis  ;  they  are  also  the  least  rich 
and  important. 


P"iG.  108. — Restoration  of  part  of  the  Necropolis  of  Gizeh. 

"  Our  general  notion  of  Egyptian  architecture  would  lead  us  to 
look  for  the  use  of  huge  stones  in  these  mastabas,  and,  in  fact, 
certain  important  monuments,  such  as  the  Mastabat-el-Fm^aotui, 
and  parts  of  important  monuments,  such  as  the  Temple  of  the 
Sphinx  and  the  passages  and  chambers  of  the  greater  pyramids, 
were  constructed  of  very  large  blocks.  But  the  Sakkarah  architects 
were  more  modest.  Apart  from  the  ceilings,  architraves  and  other 
places  where  big  stones  were  necessary,  the  blocks  are  of  an 
average  height  of  about  half  a  yard,  with  a  proportionate  length 
and  thickness. 

"  The  brick-built  sepulchres  are  of  two  kinds  also.      The  more 

VOL.     I.  Z 


I/O 


A  History  of  Art  in  Ancient  Egypt. 


elaborate  are  of  black  brick,  while  a  yellowish  brick  is  used  for  the 
others.  The  yellow  bricks  are  a  mixture  of  sand  and  pebbles  with 
a  little  clay  ;  the  black  bricks  are  of  earth  and  straw.  The  former 
are  always  small  {S'8  in.  x  4-4  in.  x  2*8  in.)  ;  the  latter  are 
comparatively  large  (15 "2  in.  x  7*2  in.  x  5-6  in.).  Both  kinds 
are  dried  simply  in  the  sun.  The  yellow  bricks  seem  to  be  the 
more  ancient.  Their  employment  begins  and  ends  with  the 
Ancient  Empire.  The  black  bricks,  on  the  other  hand,  appear  for 
the  first  time  about  midway  through  the  fourth  dynasty.  At  first 
they  were  rarely  employed,  but  under  the  eighteenth  dynasty  and 
those  which  followed  it,  they  came  to  be  exclusively  used." 

All  these  mastabas,  whether  of  brick  or  stone,  betray  an  amount 
of  negligence  in  their  construction  which  is  astonishing.  Con- 
sidering the  ideas  which  the  Egyptians  had  formed  of  a  future  life, 


Fig.  109. — The  Mastabat-el- Faraouu. 

the  chief  preoccupation  of  their  architects  should  have  been  to 
give  a  stability  to  their  sepulchres  which  would  have  insured  their 
perpetuity,  and,  with  it,  that  of  the  deposit  committed  to  their 
charge.  The  whole  of  our  description  will  be  pervaded  by 
accounts  of  the  minute  precautions  devised  to  that  end.  "  Now 
these  mastabas  are  constructed  with  care  on  their  outsides  alone. 
The  core  of  their  walls  is  composed  of  sand,  of  rubbish,  of  blocks 
of  stone  mingled  w^ith  the  flakes  struck  off  by  the  masons,  and  all 
this  in  most  cases  without  any  cement  to  give  it  coherence.  The 
mastabas  of  Sakkarah  are  not  homogeneous  constructions  of 
masonry  and  cement,  like  the  pyramids  and  most  of  the  mastabas 
of  Gizeh.  They  are  confused  heaps  of  ill  assorted  materials,  which 
would  collapse  but  for  the  retaining  strength  of  their  covering  of 
solid  stone. 


The  Tomr  under  the  Ancient  Empire. 


171 


"  At  Sakkarah  the  outward  faces  of  the  mastaba  are  not  smooth. 
Each  successive  course  is  sHghtly  set  back  from  the  one  below  it. 
At  Gizeh  the  walls  form  a  smooth  plane  gently  inclined  from  the 
perpendicular. 

"There  are  mastabas  of  all  sizes.  That  of  Sabou  measures  172 
feet  by  84 ;  that  of  Ha-ar,  149  by  74;  that  of  Ra-en-ma  169  by 
Si,  and  that  of  Hapi  no  more  than  25  ft.  6  in.  by  19  ft.  6  in.  In 
height  they  vary  less.  The  highest  are  not  more  than  from  26 
to  30  ft.  high,  the  smallest  about  12." 

The  roof  of  the  mastaba  is  a  plain  surface  without  irregularity 
of  any  kind  ;  but  the  soil  above  it  is  sprinkled  with  vases  buried 
at  a  slight  depth.     These  vases  are  pretty  evenly  distributed,  but 


Fig.  1 10. — Entrance  to  a  Mastaba  at  Sakkarah.     Marictte. 


they  are  rather  more  numerous  in  that  part  of  the  soil  which 
covers  the  ceilings  of  the  chambers,  a  circumstance  of  which 
Mariette  often  made  use  to  guide  him  in  his  excavations.  Like 
all  the  vases  of  this  epoch,  those  which  are  found  upon  the  roof  of 
the  mastabas  are  roughly  made,  pointed  at  the  bottom  and  without 
handles.  They  each  contain  a  thin  film  of  yellow  clay  deposited 
by  the  water  with  which  they  were  filled.  They  were  placed  in 
their  curious  position  under  the  notion  that  the  water  which  they 
contained  would  quench  the  thirst  of  the  dead  man  below,  l^he 
mouths  of  the  jars  were  covered  with  flit  stones,  and  the  water 
would  last  long  enough  to  satisfy  at  least  the  immediate  necessities 
of  the  inhabitant  of  the  tomb. 


I  72 


A   History  of  Art  in  Ancient  Egypt. 


"  The  principal  face  of  the  mastaba  is  turned  to  the  east.  In 
four  cases  out  of  five  the  entrance  to  its  chambers,  when  there  is 
one,  is  found  upon  this  face.  The  general  arrangement  is,  almost 
always,  as  follows  :  i.  At  a  few  metres  distance  from  the  north- 
eastern angle  we  come  upon  a  quadrangular  niche  or  recess,  very 
high  and  very  narrow,  in  the  depths  of  which  those  long  vertical 
grooves  which  distinguish  the  steles  of  this  epoch  are  carved  upon 
the  actual  masonry  of  the  tomb.  For  this  recess  an  unimportant 
stele,  with  or  without  inscription,  is  occasionally  substituted,  or  (2) 
we  find,  at  a  few  metres  distance  from  the  south-eastern  angle, 
either  a  deeper,  larger,  and  more  carefully  built  recess,  in  the 
depths  of  which  a  monolithic  stele  of  white  limestone  covered 
with  hieroglyphs  is  placed  ;  or  a  regular  architectural  facade  in 
miniature  with  a  door  in  the  centre.  When  the  recess  is  found 
near  the  southern  angle  of  the  eastern  face,  the  tomb  begins  and 


Fig.  III. — l-intel  of  the  tomb  of  Teta,  6th  dynasty.     Louvre. 


ends  there.  It  has  no  internal  chamber,  or  rather,  the  recess  acts 
as  substitute  for  one.  But  when,  instead  of  the  niche  or  recess, 
we  meet  with  a  door,  we  then  know  that  we  have  come  upon  a 
regularly  completed  tomb.  The  name  of  its  proprietor  is  often 
carved  upon  the  lintel.  Several  of  these  lintels,  of  a  peculiar 
shape,  are  to  be  seen  in  the  Louvre. 

"  Next  after  the  eastern  face,  in  relative  importance,  comes 
that  which  is  turned  to  the  north  When  the  entrance  is  in  the 
northern  wall  the  door  is  invariably  at  the  back  of  a  kind  of 
vestibule,  in  front  of  which  are  two  monolithic  columns,  without 
base  or  capital,  supporting  the  architrave  which,  in  turn,  supports 
the  roof 

"  Still  more  seldom  than  in  the  northern  face,  the  entrance  is 
occasionally  found  upon  that  which  is  turned  to  the  south.  This 
exceptional  arrangement  is,  in  most  instances,  caused  by  some 
local  circumstance   which   may  readily  be   perceived.      Wlien   the 


The  Tomb  under   the  Ancient  Empire.  173 

entrance   is  on  the  south  its  arrangement  is  sometimes  the  one, 
sometimes  the  other,  of  the  two  which  we  have  described. 

"  As  for  the  western  face,  we  have  no  evidence  that  it 
ever  played  any  more  ambitious  role  than  that  of  completing 
the  inclosure.  It  is  always  destitute  of  both  openings  and 
ornaments." 

We  have  thus  explored,  with  Mariette,  the  outside  of  the 
mastaba.  We  have  described  its  form  and  general  aspect,  we 
have  noticed  the  materials  of  which  it  was  constructed,  the 
principles  upon  which  it  was  oriented,  and  its  average  size. 
We  have  explained,  too,  how  this  single  type  of  sepulchre  was 
repeated  many  thousands  of  times  with  but  slight  variations,  until, 
upon  the  plateau  between  Memphis  and  the  desert  there  gradually 
arose  a  metropolis  of  the  dead  more  populous  than  that  of  the 
living.  It  remains  to  describe  the  contents  of  those  huge  blocks 
of  masonry.  We  shall  begin  by  visiting  the  chambers  planned  by 
the  architect  in  the  building  itself;  we  shall  afterwards  penetrate, 
by  the  paths  which  modern  curiosity  has  established  through  the 
ddbris  of  ages  and  the  depths  of  the  soil,  to  those  recesses  of  the 
tomb  which  were  meant  to  be  for  ever  inaccessible. 

The  interior  of  a  mastaba  is  composed  of  three  parts — the 
chamber,  the  serdab,  and  the  well.  The  last-named  of  the  three 
is  the  only  part  which  is  never  wanting.  Many  of  the  mastabas 
are,  in  fact,  solid.  In  them  the  chamber  is  in  a  very  rudimentary 
condition,  being  represented  merely  by  one  of  those  external 
niches  which  Mariette  has  described.  This  arrangement  was 
the  earliest,  and,  as  long  as  the  mastaba  continued  to  be  built, 
the  less  ambitious  tenants  of  the  necropolis  were  contented  to 
reproduce  it.  But  in  these  pages,  as  in  a  natural  history,  it  is 
important  to  study  the  species  when  fully  developed  and  provided 
with  all  its  organs.  When  we  have  clearly  established  a  general 
type  nothing  is  more  easy  than  to  recognise  and  point  out  its 
variations.  It  suffices  to  say  here  that  some  tombs  are  wanting  in 
one,  some  in  another  of  those  constituent  parts  whose  meaning 
and  uses  we  shall  attempt  to  determine,  and  that,  in  a  few,  they 
are  of  an  unaccustomed  importance. 

It  is  natural  that  we  should  first  turn  our  attention  to  the 
chamber.  This  was  a  kind  of  neutral  ground  upon  which  the 
quick  and  the  dead  could  meet,  the  former  to  present,  the  latter  to 
receive  the  funeral  offerings. 


174 


A   History  of  Art  in  Ancient  Egypt. 


"  The  interior  of  a  mastaba  may  be  divided  into  several 
'  chambers '  (there  are  three  in  the  tomb  of  Ti),  but  generally 
there  is  only  one.  It  is  entered  by  the  door  in  the  middle  of 
the  facade. 

"  These  chambers  have,  as  a  rule,  to  depend  upon  the  door  for 
light,  but  there  are  a  few  instances  in  which  they  are  lighted  from 
openings  in  the  roof.  A  remarkable  example  of  the  latter  arrange- 
ment is  to  be  seen  in  the  tomb  of  Ti,  where  the  innermost 
chamber,  which  otherwise  would  be  in  complete  darkness,  is 
lio-hted  from  the  roof. 

"  The  chamber  is  sometimes  quite  bare,  sometimes  covered 
with  sculptures  and  paintings  such  as  those  whose  character 
and  meaning  we  have  already  pointed  out.  At  its  further  end, 
and  always  facing  eastwards,  stands  the  inscribed  tablet  or  stele. 


^i^^ 


t 


P  D    O    D    □ 


paoD op 


Miao  0 


1 


Fig.  112.— Plan  of  the  tomb  of  Ti.  Figs.  113,  114.— Mastaba  at  Sakkarah,  from  Prisse. 

There  are  some  chambers  in  which  the  walls  are  bare  and  the 
stele  eno-raved,  but  there  are  none  where  the  walls  are  carved  and 
the  stele  plain." 

In  the  tomb  of  Ptah-Hotep,  of  which  we  reproduce  the 
principal  side,  the  stele  proper  is  on  the  left,  but  the  figures 
and  the  funerary  inscriptions  cover  all  the  central  part  of  the 
richly  decorated  wall  (Fig.  115). 

We  see,  then,  that  the  stele  is  the  one  indispensable  part  of 
this  complicated  whole.  It  was,  in  fact,  upon  the  formula  with 
which  it  was  inscribed,  that  the  Egyptians  depended  for  those 
mao-ical  ao-encies  by  which  Osiris  became  the  active  medium  of 
transmission  between  the  living  and  the  dead. 

"  At  the  foot  of  the  stele  there  was  often  a  table  for  offerings. 


M 


lijiHi  II  ill  II  II  ill  n  iH  Bi  ii  u  n  II  i!|  n  h  n  i\  ii1i 


v--^ 


I 


O 

a 
o 

d 

c3 


c3 

13 


Cl, 
<U 

O 

K 


.J3 

a 

c 


a 

-s 

<1) 


o 


The  Tomb  under  the  Ancient  Empire.  771 

in  granite,  alabaster,  or  limestone.     This  was  laid  flat  upon  the 
ground  (Fig.  92). 

"  As  a  rule  this  was  the  only  piece  of  furniture  in  the  chamber  ; 
but  occasionally  we  find,  on  each  side  of  the  stele  and  always 
placed  upon  the  ground,  either  two  small  limestone  obelisks,  or 
two  objects  in  that  material  resembling  table  legs  hollowed  out 
at  the  top  for  the  reception  of  offerings." 

This  chamber  was  left  open  to  every  comer.  The  entrance 
was  in  fact  left  without  a  door.  To  this  rule  Mariette  found 
but  two  exceptions  in  the  many  hundreds  of  tombs  which  he 
examined.^ 

"  Not  far  from  the  chamber,  oftener  on  the  south  than  the 
north,  and  oftener  on  the  north  than  the  west,  a  passage  in  the 
masonry,  high,  narrow,  and  built  of  very  large  stones,  is  found. 
The  workmen  employed  upon  the  excavations  christened  it  the 
serdad,  or  corridor,  and  their  name  has  been  generally  adopted."  ^ 
In  Figs.  1 16-1 19  we  give  the  plan  and  three  sections  of  a  mastaba 
at  Gizeh  which  has  four  serdabs. 

"  Sometimes  the  serdab  has  no  communication  with  the  other 
parts  of  the  mastaba,  it  is  entirely  walled  in,  but  in  other  instances 
there  is  a  narrow  quadrangular  opening,  a  sort  of  pipe  or  conduit, 
which  unites  the  serdab  with  the  chamber.  It  is  so  small  that  the 
hand  can  only  be  introduced  into  it  with  difficulty.^ 

"  The  use  of  the  serdab  is  revealed  by  the  objects  which  have 
been  found  in  it ;  it  was  to  hold  one  or  more  statues  of  the 
deceased.  The  Egyptians  believed  these  statues  to  be  the  most 
certain  guarantees,  always  with  the  exception  of  the  mummy  itself, 
of  a  future  life  for  the  dead.  Hidden  from  sight  in  their  dark 
prison,  they  were  protected  from  all  violence,  while  they  were 
separated    only   by   a  few    stones    from    the    chamber  where   the 

1  One  of  these  exceptions  is  furnished  by  the  tomb  of  Ti,  of  which  we  shall 
often  have  to  speak  (Fig.  114).  The  large  public  hall  near  the  entrance  to  the 
tomb  was  separated  from  the  two  chambers  farther  in  by  a  corridor  closed  at  two 
points  by  doors,  some  remains  of  which  were  found  in  place  when  the  tomb  was 
opened. 

^  This  is  a  word  of  Persian  origin  adopted  by  the  Arabs.  Its  strict  meaning  is  a 
dark  subterranean  opening,  cave,  or  passage. 

^  The  tomb  of  Ti  had  two  serdabs  as  well  as  three  chambers;  one  of  these  was 
close  to  the  door,  the  other  in  the  innermost  part  of  the  mastaba.  In  the  latter 
several  statues  of  Ti  were  found,  the  best  preserved  being  now  in  the  museum 
at  Boulak. 

VOL.    I.  A    A 


178 


A  History  of  Art  in  Ancient  Egypt. 


friends  and  relations  met  together,  and  the  conduit  by  which  the 
Intervening  wall  was  often  pierced,  allowed  the  smell  of  fruit  and 
incense  and  the  smoke  of  burnt  fat  to  come  to  their  nostrils.^ 

"No  inscriptions  have  been  found  In  a 
serdab  except  those  upon  the  statues.  And 
no  objects  other  than  statues  have  ever 
been  found  in  a  serdab."  So  that  the 
function  of  the  serdab  was  to  afford  a  safe 
and  final  asylum  to  the  statues.  These 
were,  no  doubt,  to  be  found  in  other 
situations  also,  because,  not  to  mention  the 
numerous  bas-reliefs  upon  which  the  figure 
of  the  deceased  appeared  in  the  chamber 
or  in  the  niche  which  sometimes  took  its 
place,  he  was  sometimes  portrayed  In  high 
relief,  and  of  full  life  size,  in  the  public  hall  of  the  tomb.^ 
Sometimes,   also,   we    find  a  statue  in  one   of  those  front  courts 


Fig. 


1 16. — Flan  of  a  mastaba 
with  four  Serdabs. 
(Lepsiu';,  i.,  pi,  24.) 


which,  especially    at    the    time    of   the    fourth    dynasty,    seem   to 


Fig    117. — Longitudinal  section  of  the  same  mastaba. 


have    been    in    great    favour.        But    this  court,    as    well    as    the 
chamber,  was  open  to  every  chance  passer   by,  and   the   statues 

1  In  a  Theban  tomb  described  by  M.  Maspero  {Etude  sur  quelqiies  Pehittires 
funcraires)  the  tenant,  Harmhabi,  is  made  to  speak  thus  :  "  I  have  come,  I  have 
received  my  bread  ;  joining  the  embalmed  offerings  to  my  members,  I  have  breathed 
the  scent  of  the  perfumes  and  incense."  It  is  also  possible  that  this  conduit  may 
have  been  intended  to  permit  of  the  free  circulation  of  the  double,  to  allow  it  to 
pass  from  its  supporting  statues  to  the  chapel  in  which  it  is  honoured.  This  curious 
idea,  that  the  spirit  of  the  dead  can  pass  through  a  very  small  hole,  but  that  it 
cannot  dispense  with  an  opening  altogether,  is  found  among  many  nations.  The 
Iroquois  contrived  an  opening  of  very  small  diameter  in  their  tombs,  through  which 
the  soul  of  the  dead  could  pass  and  repass.  See  Herbert  Spencer,  Principles  of 
Sociology,  vol.  i.  p.  192. 

2  There  is  an  example  of  this  in  a  mastaba  at  Gizeh  (Fig.  120).  See  No.  95  of 
Lepsius  {De7ikm(Eler,  vol.  i.  p.  29  ;  vol.  iii.  pi.  44). 


The  Tomb  under  the  Ancip:nt  Empire. 


179 


which  they  both  contained  were  in  continual  danger  from  care- 
less or  malicious  hands.  It  was  to  guard  against  such  chances 
as  these  that  the  inventive  architects  of  Egypt  contrived  a  safe 
retreat  in  the  heart  of  the  massive  structure  which  should  provide 
a  reserve  of  statues  against  every  contingency.  When  all  those 
which  were  exposed  to  accident  should  have  perished,  these  would 
still  survive  and  would  furnish  to  the  double  the  material  support, 


- , — .^-i/ 


Fig.  118. — Transverse  section  through  the  chamber. 


the  tangible  body,  to  which  that  phantom  was  obliged  to  attach 
himself  unless  he  wished  to  perish  entirely. 

These  precautions  were  not  ill  conceived.  The  serdab  kept 
efficient  guard  over  its  deposit ;  the  museum  of  Boulak  con- 
tains at  least  a  hundred  statues  from  the  ancient  empire  which 
were  found  at  Sakkarah,  and  nine-tenths  of  them  were  found  in 
the  serdabs. 

We  have  now  described  all 
those  parts  of  the  tomb  which 
were  above  ground.  We  have 
not  been  content  with  visiting 
the  chamber  only,  which  was 
freely  left  open,  we  have  pene- 
trated into  the  farthest  recesses, 
and  have  discovered  those 
secrets   of    the   massive    walls 

which  their  constructor  thought  to  hide  for  ever  from  the  eye 
of  man.  But  even  yet  we  have  not  arrived  at  the  actual  place 
of  burial ;  we  shall  reach  it,  however,  through  our  third  internal 
division,  the  well  or  pit. 

"  The  well  is  an  artificial  excavation,  square  or  rectangular  in 
plan,  never  round,  at  the  bottom  of  which  is  the  chamber  in  which 
the  mummy  is  deposited. 

"  To  arrive  at  the  opening  of  the  well,  we  must  mount  to  the 


F"iG.    119. — Transverse    section  through  the 
serdabs. 


i8o 


A  History  of  Art  in  Ancient  Egypt. 


platform,  or  roof,  of  the  mastaba  (Fig.  122).  As  there  was  never 
any  staircase  to  a  mastaba  either  within  or  without,  it  will  be  seen 
that  the  well  must  have  been  a  very  inaccessible  part  of  the  tomb." 
In  one  single  instance,  namely,  in  the  tomb  of  Ti,  the  well  is  sunk 
from  the  floor  of  the  largest  of  the  internal  chambers,  but  whether 
it  opened  upon  the  roof  or  upon  the  floor  of  the  chamber,  it  was 
always  closed  with  the  utmost  care  by  means  of  a  large  flat  stone. 


"  The  well  is  generally  situated  upon  the  major  axis  of  the 
mastaba,  and,  as  a  rule,  nearer  to  the  north  than  to  the  south.  Its 
depth  varies,  but,  on  an  average,  it  is  about  forty  feet.  Now 
and  then,  however,  it  has  a  depth  of  sixty-five  or  even  eighty 
feet.  As  the  well  begins  at  the  platform  and  ends  in  the  rock 
carved  mummy  chamber,  it  follows  that  it  passes  vertically  first 
through  the  mastaba,  secondly  through  the  rock  upon  which  the 
mastaba    is    founded.      The    built    part    of    the    well    is    carefully 


The  Tomb  under  the  Ancient  Emph^e. 


i8i 


constructed  of  large  and  perfect  stones,  and  in  this  we  find  one  of 
the  distinguishing  characteristics  of  the  tombs  of  the  ancient 
empire."  In  the  tomb  of  Ti  the  well  takes  the  form  of  an 
inclined  plain  like  a  passage  in  the  pyramids.  In  the  common 
form  of  well  the  mum.my  pit  could  only  be  reached  by  means  of 
ropes, 

"When   the   bottom   of  the   well   is  reached   a  gaping  passage 
is  seen  in  the  rock  which  forms  its  southern  wall.      This   passage, 
which  is  not  high  enough  to  allow  one  to  walk  upright,  does  not 
run    quite    parallel    to    the    axis    of   the 
mastaba.       It  is  directed    obliquely    to- 
wards the  south-east,  like  the  chamber 
above.      Suddenly  it  becomes  enlarged 
into  a  small  cavern,  which  is  the  mortuary 
chamber  properly  speaking,   that    is    to 
say,  the  room  with  a  view  to  which   the 
whole  structure   has   been   planned  and 
to    which    all    its    other   parts    are   but 
accessories. 

"  This  mortuary  chamber  is  vertically 
under  the  public  hall  above,  so  that  the 
survivors  who  came  together  in  the 
latter  for  the  funeral  ceremonies  had  the 
corpse  of  the  deceased  under  their  feet, 
at  a  distance  which  varied  according  to 
the  depth  of  the  well." 

The  mortuary  chambers  are  large  and 
carefully  built,  but  generally  without  ornament  or  inscription.  Of 
all  those  explored  by  him  Mariette  found  but  one  which  had  its 
walls  ornamented  ;  in  the  middle  of  its  decorations,  which  he 
does  not  describe,  he  contrived  to  make  out  a  few  phrases  which 
seemed  to  belong  to  the  Ritual  of  the  Dead. 

The  sarcophagus  was  placed  in  one  corner  of  the  chamber. 
It  was  generally  of  fine  limestone,  sometimes  of  red  granite, 
and  on  a  few  occasions  of  opaque  black  basalt.  It  was  rect- 
angular on  plan  with  a  round-topped  lid  squared   at  the  angles. 


Fig.  121. — The  upper  chamber, 
well,  and  mummy  chamber. ^ 


^  This  figure  is  a  composition  by  Mariette  for  the  purpose  of  showing  the  relation 
between  the  subterranean  and  constructed  parts  of  the  tomb.  {^Notice des  principaux 
Monuments,  p.  22.)  [It  shows,  however,  the  well  opening  from  the  floor  of  the 
upper  chamber,  an  arrangement  which  is  not  characteristic  of  the  mastaba. — Ed.] 


l82 


A  History  of  Art  in  Ancient  Egypt. 


Marietta  found  none  at  Sakkarah  with  inscriptions.  On  the 
other  hand  we  find  them  upon  the  sarcophagus  at  Khoo-foo-Ankh, 
which  was  discovered  at  Gizeh  and  belongs  to  the  fourth  dynasty 
(Figs.  123,  124). 

"  The  Egyptians  did  not  always  trust  to  the  mere  size  and 
weight  of  the  lid  for  the  secure  closing  of  the  sarcophagus.  The 
under-side  of  the  cover  is  made  with  a  rebate  at  its  edo^e  which  fits 
into  a  corresponding  groove  on  the  upper  edge  of  the  sarcophagus, 
and  the  two  edges  were  bound  still  more  tightly  together  by  a 


I'k;.  122. — Double  mastaba  at  Gizeh,  transverse  section  (from  Lepsius,  t.  i.,  pi.  22). 


very  hard  cement.  Finally,  as  if  all  these  precautions  were  not 
enoueh,  wooden  bolts  were  affixed  to  the  under-side  of  the  lid 
which  fitted  into  slots  in  the  sarcophagus  and  helped  to  render  the 
two  inseparable." 

So  far  as  we  can  judge  from  the  few  human  remains  which  have 
been  gathered  from  these  ancient  tombs,  the  process  of  embalm- 
ment was  then  carried  on  in  simple  and  elementary  fashion,  and  it 
was  this  imperfection  that  the  Egyptians  attempted  to  neutralize, 


The  Tomb  under  the  Ancient  Empire. 


183 


by  the  innumerable  and  complicated  precautions  which  they  took 
to  insure  that  the  corpse  should  not  be  disturbed  in  its  envelope  of 
stone.  In  later  times,  when  the  preparation  of  the  mummy  was 
better  understood,  they  were  not  so  careful  to  seal  up  the 
sarcophagus  from  the  outer  air. 

*'  The  furniture  of  the  mummy  chamber  comprised  neither 
statues,  nor  funerary  statuettes,  nor  amulets  of  any  kind.  Some- 
times a  few  ox  bones  bestrew  the  ground.  Two  or  three  large  and 
pointed  red  vases,  containing  nothing  but  a  thin  deposit  of  clay, 
rest  against  the  walls.  Within  the  sarcophagus  we  find  the  same 
sobriety  of  sepulchral  furniture.  Beyond  a  wooden  or  alabaster 
pillow  (Fig.  105)  and  half  a  dozen  little  drinking  cups  of  alabaster, 
nothing  has  been  found  there  but  the  mummy  itself." 


ro 


1«:iil 


u 


P7— ,    "f^'-ipi<r T 


i-''m 


t^W^ 


•si-irrrfTF'^-r 


>     I  '    !|| 


A" 


*    '. 


M 


''■■•A,. 
.M  -1-' 


':^:  :';^^ili!ii" 


■-T^ 

=^^ 


Fig.    123. — Sarcophagus  of  Khoo-f-io-Ankh.     Perspective  after  Bourgoin.     Red  granite. 

Height   I  "33  metres.     Boulak. 


These  beef  bones  must  be  the  remains  of  the  quarters  of 
meat  which  were  placed  in  the  tomb  for  the  nourishment  of  the 
dead.  No  scene  is  more  frequently  represented  upon  the  walls 
of  the  public  chamber  of  the  mastaba  than  the  killing  and  flaying 
of  victims  for  the  funeral  ceremonies  (Fig.  125).  Like  those  which 
are  found  upon  the  roof,  the  vases  must  have  held  water  for  the 
double.  The  pillow  was  placed  under  the  head  of  the  mummy, 
it  was  the  one  he  had  used  durino^  his  life.  As  for  the  drinking- 
cups,  their  use  has  not  yet  been  determined,  so  far  as  we  know. 

"As  soon  as  the  mummy  was  in  the  sarcophagus,  the  sarco- 
phagus sealed,  and  the  various  objects  which  we  have  described  in 


\ 


i84 


A  History  of  Art  in  Ancient  Egypt. 


place,  the  opening  at  the  bottom  of  the  well  was  walled  up  ;  the 
well  itself  was  filled  with   stones,  earth,  and  sand,  and  the  dead 


;Jl;'l^iill;,;fll:ii.',iy.^lHMi^;^'i'illurl'■iili^l^'ft^ 


mam 


mm 


mmm^^i0K^i^ifffiif^mm\mimmm 


EMra, 


"H!"";"" 


lll|IIJI|l|ltlllllU|IIIIJI||lllll|lllllllMI ■■ 


"^r^ 


p^^HBIRPaWBBBflfWB 


'l*i""ff! 


iii>ifniiiijiiijiniiiiiiiiiniiiiiiiii^i»iyajjaBm| 


i!iT^;'mirJ;:'i!iii"'!i!'iilli'i!^!!'i!i'"^ili!!-^^ 


•  HI 


t'/M 


■JWiiiiii|iiffMiiHiiiii|iti|H'iiiiHiiafiwgwwiBwiwaaw 


flflih  i  I 


I 


Fig.  124. — Detail-;  of  the  Sarcopl-ngus  of  Khoo-foi-Aiikb. 

was  left  to  his  eternal  sleep."  ^     These  precautions  make  it  no  easy 
thing  to  reach  the  mummy  chamber.      To   find  the  entrance  to  the 

'  The  broken  up  and  decayed  remains  of  wooden  boats  have  been  found  in  two 
or  three  mummy  pits  (Mariette,  Zes  Tomhes  de  V Aiicieji  Empire^  p.  17).  Tliey 
originally  formed  part,  perhaps,  of  the  boats  upon  which  the  corpse  was  transported 
across  the  Nile  to  the  nearest  point  of  the  western  bank  to  the  tomb.  There  can 
be  no  doubt  that,  in  placing  them  in  the  well,  the  survivors  believed  that  they  were 
serving  the  deceased.  Both  the  bas-reliefs  in  tlie  tomb  and  the  Ritual  contain 
many  representations  of  the  soul  navigating  the  regions  of  Ament  (see  the  upper 


The  Tomb  under  the  Ancient  Empire. 


185 


well  is  the  first  difficulty,  and  when  it  is  found,  many  hands  and  no 
little  time  are  required  to  remove  the  rubbish  with  which  it  is 
filled.  The  only  mechanical  helps  which  the  Egyptians  have  ever 
used  in  such  work  are  those  which  we  ourselves  have  seen  in 
the  hands  of  Mariette's  labourers,  namely,  the  wooden  shovel 
and  the  little  rush  basket  which  is  filled  with  a  few  handfuls  of 
sand  and  pebbles,  and  then  carried  on  the  head  to  be  emptied  at  a 
convenient  distance.  It  may  be  guessed  how  many  journeys 
to  and  fro  have  to  be  made  before  a  few  cubic  yards  of  ddbris  are 
cleared  by  such  means  as  this  ! 

We  have  so  far  followed  Mariette,  and  have  frequently  had  to 
make  use  of  his  ipsissima  verba.     To  his  pages  and  to  the  plates 


Fig, 


125.  —  Bas-relief  from  Sakkarah.      BouLik. 


of  the  great  work  of  Lepsius,  we  must  refer  those  readers  who 
are  not  contented  with  beinof  told  greneral  rules  but  wish  to 
know  the  exceptions  also.  We  shall  not  go  into  all  the  changes 
which  variety  of  taste  and  the  progress  of  art  introduced  into  the 
arrangement  and  decoration  of  Egyptian  buildings  ;  they  do  not 
affect  the  general  statements  which  we  have  made.  We  shall 
not  re-state  the  evidence  which  enabled  Mariette  to  apportion  the 
142   painted  and   sculptured    mastabas   explored  by  him  in  1869, 

section  of  Fig.  98).  In  certain  Theban  tombs,  models  of  fully  rigged  boats  have 
been  found  ;  there  are  some  of  them  in  the  Louvre  (Sa/k  Civile,  case  K).  [There 
are  two  in  the  British  Museum,  and  one,  a  very  fine  one,  in  the  museum  at 
Liverpool. — Ed.] 

VOL.    T.  B    B 


1 86  A   History  of  Art  Ix\  Ancient  Egypt. 

to  the  first  six  dynasties.  It  is  certain  that  those  monuments 
form  a  chronological  series  extending  over  a  space  of  from 
twelve  to  fifteen  centuries,  and  that  during  the  whole  of  that  long 
period,  the  general  character  of  Egyptian  sepulchral  architecture 
remained  unchanged. 

We  should  here,  perhaps,  in  order  to  make  our  description 
complete,  attempt  to  convey  a  true  idea  of  the  reliefs  which  cover 
the  sides  of  the  chamber,  and  of  the  statues  which  fill  the  serdab. 
We  should,  perhaps,  by  a  judicious  choice  of  examples,  endeavour 
to  estimate  their  style  and  composition  ;  but  we  shall  postpone 
all  such  examination  until  we  come  to  treat  of  sculpture,  and  of  the 
way  in  which  the  earliest  Egyptian  artists  treated  the  human  form. 
A  didactic  and  analytic  method  is  so  far  despotic  that  it 
compels  us,  in  order  to  marshal  our  facts  and  to  make  them 
easily  understood,  to  separate  phenomena  which  are  intimately 
connected,  and  to  destroy  the  unity  of  natural  groups.  We  have 
thus  been  driven  to  separate  the  figured  decorations  of  the 
tomb  from  the  architectural  arranofements  which  enframe  and 
support  them  ;  with  the  latter,  alone,  are  we  now  concerned. 

We  may  sum  up  the  foregoing  details  by  the  following  general 
description  of  the  Egyptian  tomb  as  it  was  established  in  the  early 
ages  of  the  national  life,  in  those  years  when  the  national 
civilization  put  on  the  form  and  colour  which  it  retained  until 
the  last  days  of   antiquity. 

This  tomb,  when  complete,  included  (i)  a  built  up  part  which, 
being  raised  well  above  the  surface  of  the  soil,  was  a  conspicuous 
object  in  the  landscape ;  and  (2)  a  subterranean  part  cut  in  the 
living  rock  which  was  never  more  than  a  few  feet  below  the 
surface  of  the  sand.  The  constructed  part  inclosed  a  chamber 
which  was  sometimes  internal  and  sometimes  external,  a  chamber 
in  which  the  relations  of  the  deceased  deposited  the  funeral 
offerings,  and  in  which  the  priests  officiated  before  the  stele,  to 
which  the  most  conspicuous  place  was  always  given.  Sometimes 
this  chamber  is  nothing  more  than  a  recess  in  the  facade,  a  mere 
frame  for  the  stele.  The  structure  also  contains  a  retreat  in  its 
thickness  where  the  statues  of  the  deceased  were  walled  up.  The 
subterranean  part  is  composed  of  the  well  and  the  mummy 
chamber.  The  well  is  sunk  from  different  parts  of  the  building ; 
usually  traversing  its  whole  depth  ;  it  leads  to  the  mummy 
chamber  which  is  found  at  varying  depths  in  the  bowels  of 
the  earth. 


The  Tomb  under  the  Ancient  Empire.  187 


\ 


Such  are  the  constituent  elements  of  the  mastaba,  that  is  to  say, 
of  those  private  tombs  which  were  contemporary  with  the 
Pyramids.  All  over  Egypt,  in  every  one  of  the  cemeteries,  no 
matter  where  they  are  situated  or  what  their  date,  the  same 
elements  are  to  be  found,  modified  in  certain  particulars  by  the  rank 
of  the  deceased,  by  the  nature  of  the  soil,  by  the  size  of  the  tomb, 
and  by  the  changes  of  fashion,  but  always  to  be  easily  recognized. 
Of  all  these  elements  there  is  but  one  which  does  not  persistently 
reappear  in  monuments  other  than  the  mastaba,  and  that  is  the 
serdab.  This  retreat  for  statues  has  not,  as  yet,  been  found  in  any 
of  the  royal  tombs  of  the  first  six  dynasties,  neither  has  it  been  met 
with  in  the  tombs  of  the  two  Theban  empires,  or  of  later  epochs. 
And  yet  it  was  connected  with  one  of  the  most  vital  hopes  of 
the  Egyptian  religion.  It  fulfilled  in  the  happiest  manner,  one  of 
the  conditions  imposed  upon  the  Egyptian  architect  by  the  strange 
conceptions  of  a  future  life  which  we  have  described.  Why  then 
do  we,  as  a  rule,  find  the  serdab  only  in  the  mastabas  of  the 
Memphite  necropolis  ?  Its  absence  under  the  Theban  princes 
is,  perhaps  to  be  explained  by  the  progress  made  in  the  science 
of  embalming.  The  heads  of  more  than  one  mummy  have 
now  been  exhibited  in  the  cases  of  European  museums  for  many 
years,  and,  in  spite  of  the  dampness  of  our  climates,  they  still 
preserve  their  skin,  their  teeth  and  their  hair  (Fig.  126).  When 
they  had  learnt  the  secret  of  preserving  the  body  from  corruption, 
so  that  after  a  long  series  of  centuries  it  should  be  pretty  much 
in  the  same  condition  as  on  the  day  after  death,  they  did  not 
indeed,  cease  to  make  those  images  which  were  supposed  to 
guard  the  dotible  from  annihilation,  but  they  attached  less 
importance  to  their  safety,  and  took  less  trouble  to  hide  them. 
They  considered  that  they  had  done  enough  for  their  preservation 
by  putting  them  in  the  precincts  of  their  tombs  and  temples,  and 
so  under  the  guardianship  of  their  venerated  religion. 

As  for  the  other  parts  of  the  tomb,  a  little  attention  will 
always  suffice  for  their  identification  even  in  those  sepulchres 
which  differ  most  from  the  mastaba.  In  some  instances  we  shall 
find  the  mummy  chamber  contrived  in  the  upper  structure,  in 
others  the  whole  tomb  is  cut  in  the  living  rock.  Sometimes  we 
find  the  chapel,  as  we  may  call  the  public  chamber  in  which  the 
miraculous  nourishment  of  the  double  took  place,  more  or  less 
distantly   separated    from    tlie    mummy   chamber ;    sometimes  the 


i88 


A  History  of  Art  in  Ancient  Egypt. 


well  almost  disappears,  sometimes  it  ceases  to  be  vertical  and 
becomes  a  long  corridor  with  but  a  gentle  slope.  As  a  rule  all 
these  variations  are  easily  explained,  and  may  be  connected 
without  difficulty  with  that  primitive  type  which  we  have 
attempted  to  define  by  its  most  wide-spread  and  constant  features. 
Another  method  of  sepulture  was  made  use  of  in  the  Ancient 
Empire,  a  method  which  afterwards  came  into  general  use  in 
Egypt,    we    mean    the    hypogejtm,    or    subterranean    tomb.      The 


Fio.  126. — Head  of  a  Mummy.      Louvre. 

Egyptian  Commission  has  described  several  rock-cut  tombs  in  the 
neighbourhood  of  the  Pyramids,  especially  some  which  face  the 
western  slope  of  the  Second  Pyramid.  Similar  tombs  are  to  be 
found  near  the  pyramid  of  Mycerinus.  Some  of  these  sepulchral 
grottos  declare  their  extreme  antiquity  by  their  imitations  of 
wooden  architecture  ;  ^  others    by   their    inscriptions    dating    from 

1  Description    de    PEgypte,    vol.    v.    p.    647,    and    Atlas,   Ant.   vol.    v.   pi.    16, 
Figs.  3,  4,  and  5. 


The  Tomb  under  the  Ancient  Empire.  189 

the  fourth  and  fifth  dynasties.  We  shall  not  dwell  long  upon 
these  rock-cut  tombs.  They  are  generally  composed  of  one  or  two 
small  sculptured  chambers,  upon  one  of  which  the  well  opens  which 
leads  to  the  mummy  chamber.  We  shall  postpone  their  study 
to  a  later  chapter,  as  the  time  of  the  Middle  Empire  affords 
us  richer  and  more  complete  examples  of  them  than  the  earlier 
period;  but,  indeed,  the  New  Empire  has  left  us  the  most 
important  examples  of  this  kind  of  sepulchre.  We  shall  here 
content  ourselves  with  pointing  out  that  the  architects  of  Mem- 
phis did  not  ignore  the  facilities  offered  by  the  easily  cut  limestone 
rocks,  not  only  for  construction  of  well  and  mummy  chamber,  but 
also  for  those  open  parts  of  the  tomb  where  the  funeral  rites 
and  the  ceremonies  of  the  Ritual  of  the  Dead  were  performed.  In 
the  whole  course  of  her  long  vitality  Egypt  did  little  more,  either 
in  art  or  religion,  than  develop,  with  variations,  the  themes 
presented  to  her  by  the  generations  which  were  ruled  by  her 
first  six  dynasties. 

THE  PYRAMIDS. 

The  mastaba  was  the  private  tomb  of  the  great  lord  or  rich 
citizen  of  primitive  Egypt ;  the  pyramid  was  the  royal  tomb 
for  the  same  epoch,  the  tomb  of  that  son  of  the  gods,  almost 
a  god  himself,  before  whom  all  foreheads  were  bowed  into 
the  dust.  As  his  head  towered  over  those  of  his  prostrate 
subjects  during  life,  so,  after  death,  should  his  sepulchre 
rise  high  above  the  comparatively  humble  tombs  of  his  proudest 
servants.  The  most  imposing  mastabas,  before  the  sand  had 
buried  them  to  the  summit,  must  have  looked  small  enough  beside 
those  prodigious  masses.  They  were  ant-hills  at  the  foot  of  a 
palace. 

It  may  seem  that  in  considering  the  mastaba  before  the 
pyramid  we  have  reversed  the  natural  order.  We  were  led  to 
do  so  by  the  fact  that  the  enormous  mass  of  the  pyramids  and 
their  peculiarities  of  construction  compelled  their  architects  to 
separate  elements  which  are  found  closely  allied  in  the  mastaba. 
In  consequence  of  this  separation  the  elements  in  question  have 
not  all  had  the  same  fate.  In  the  case  of  the  mastaba  all 
survived  or  perished  together,  but,  in  the  pyramids,  some  are  in 
a  marvellous  state  of  preservation,  while  others  have  disappeared 


190  A  History  of  Art  in  Ancient  Egypt. 


and  left  hardly  a  trace  behind.  We  are  therefore  obliged  to 
make  use  of  the  private  tomb  in  our  restoration  of  that  which 
was  peculiar  to  the  king. 

Philologists  have  attempted  to  trace  back  the  etymology  of 
the  word  Trvpa/xls  to  the  ancient  language  of  Egypt.  The  term 
was  first  employed  by  the  Greeks,  and  from  their  language  it 
has  been  adopted  into  that  of  every  civilized  nation,  with  a  meaning 
which  is  scientifically  exact.  Its  origin  has  been  sought  for  in 
the  Coptic  term  pi-rama,  height,  and  in  the  term  pir-aa,  which 
occurs  continually  in  Exodus,  and  was  used  by  Moses  to  signify  the 
reigning  Pharaoh.  But  egyptologists  now  seem  to  be  unanimous 
in  rejecting  both  these  derivations.  They  are,  we  are  told,  refuted 
by  the  fact  that  the  terms  which  are  supposed  to  have  meant  a 
pyramid  are  never  used  in  that  sense  in  any  of  the  texts. 
'  The  words  which  mean  a  royal  tomb  or  a  tomb  of  any  kind,  have 
not  the  remotest  likeness,'  says  Herr  Brugsch,^  '  to  the  term 
TTvpafits.  Each  royal  pyramid  had  its  own  name,  a  composite 
epithet  which  was  peculiar  to  itself.'  Thus  the  largest  of  them  all 
was  called  "the  brilliant  dwelling  of  Choufou  ;"  the  second,  "the 
great;"  the  third,  "that  which  is  on  high."  The  word  pyra^uin^ 
appears  therefore  to  be  a  purely  Greek  term,  derived  from  wvp, 
fire,  and  suggested  by  the  similarity  between  its  shape  and  that  of 
a  tong^ue  of   flame. 

We  shall  not  waste  our  time  in  noticing  and  refuting  those 
fantastic  explanations  of  the  pyramids  which  have  been  given  in 
modern  times.  We  shall  not  trouble  ourselves  to  prove  that 
they  were  not  observatories.  Those  sloping  tunnels,  at  the  bottom 
of  which  some  modern  writers  would  set  unlucky  astronomers 
to  watch  the  passage  of  stars  across  the  meridian,  were  her- 
metically sealed,  and  minute  precautions  were  taken  with  the  sole 
object  of  obstructing  and  concealing  their  entrance.  The  four 
slopes  of  the  pyramid  faced  to  the  cardinal  points,  simply 
because  the  orientation  of  the  tomb  was  habitual  with  the 
Egyptians ;  we  have  already  explained  its  meaning.  Still  less 
need  we  occupy  ourselves  with  the  theory,  which  made,  however, 
some  stir  in  its  time,  that  the  pyramids  were  bulwarks  by  which 
the  ancient  Egyptians  attempted  to  keep  back  the  sand  from  the 
fertile  valley  of  the  Nile.  The  science  of  M,  de  Persigny  was 
well  worthy  of  his  policy.  There  was  in  both,  the  same  turn  for 
^  History  of  Egypt  (English  version,  Murray,  1879),  vol.  i.  pp.  72,  73. 


The  Tomb  under  the  Ancient  Empire.  191 


fantastic  invention,  the  same  want  of  reflection  and  common  sense. 
If  such  a  costly  barrier  had  been  either  useful  or  necessary  it 
should  at  least  have  been  prolonged  from  one  end  of  Egypt  to  the 
other,  and  all  the  pyramids  would  not  have  been  found  assembled, 
with  but  few  exceptions,  in  the  neighbourhood  of   Memphis.^ 

No  one  in  our  day  thinks  of  either  starting  or  discussing  such 
theories  as  these.     There  are,  of  course,  several  obscure  points  in 
the  history  of  the  pyramids,  several  details  of  their  construction, 
which   stimulate   to  fresh  research  and  lend  themselves  to  many 
different  explanations  ;  but    there  can    be  no   doubt  as   to    their 
general  character.     Their  exploration  and    the  interpretations  of 
the  Egyptian  texts  have  confirmed  the  assertions  of  those  Greek 
writers  who  were  most  familiar  with  Egypt,  such  as  Herodotus,^ 
Diodorus   Siculus,^  and   Strabo.*     The  Pyramids  are   sepulchres. 
"  They  are  massive,  simply  conceived,  carefully  sealed  up  tombs. 
All  entrance  is  forbidden  even  to  their  most  carefully  built  corridors. 
They  are  tombs  without  windows,  without  doors,  without  exterior 
openings   of  any  kind.      They  are  the  gigantic  and  impregnable 
dwellings  of   the    mummy ;  .   .   .   their    colossal    dimensions    have 
been  invoked    to    bear  out  the   arguments  of    those  who   would 
attribute  to  them  some  other  destination,  but  they  are  in  fact  to  be 
found  of  all  sizes,  some   being  no  more  than  twenty  feet  high. 
Besides  this,  it  must  be  remembered  that  in  all  Egypt  no  pyramid, 
or  rather  group  of  pyramids,  is  to  be  found,  which  is  not  the  centre 
of  a  necropolis,  a  fact  which   is  enough  by  itself  to  indicate  their 
funerary  character."  ^      It  is  proved  still  more  definitely,  if  that  be 
possible,  by  the  sarcophagi  which  have  been  found   in  the  internal 
chambers,  empty  in  most  cases,  because  those  chambers  had  been 
entered  and  despoiled,  either  in  the  days  of  antiquity  or  in  those 
of  the  middle  ages,  but  sometimes  intact,  as  in  the   pyramid    of 
Mycerinus.     The  pyramids  were  hermetically  sealed.      Even  with- 
out direct  evidence  we  might  assert  that  it  was  so,  knowing  as  we 

1  FiALiN  DE  Persigny,  De  la  Destinatmi  et  de  /'  Utilitc  per7na7iente  des  Pyramides 
d' Egypte  et  de  Niibie  centre  les  Irruptions  sablonneiises  du  Desert,  Dh'eloppements  du 
Mcmoire  adresse  a  V Acadcmie  des  Sciences  /e  14  Juillet,  1844,  suivie  d^une  nouveUe 
inte7prttation  de  la  Fable  d Osiris  et  d'Isis.      Paris,  1845,  gr.  in-8. 

-  Herodotus,  ii.  127.  ^  Diodorus,  1.  64,  4. 

^  Strabo,  xvii.  p.  1161,  c. 

^  Mariette,  liineraire  de  la  Hatite-Egypte,  pp.  96,  97.  [An  excellent  translation 
of  this  work  into  English,  by  M.  Alphonse  Mariette,  has  been  published  (Trubner, 
1877,  8vo.)— Ed.] 


I 


192  A   History  of  Art  in  Ancient  Egypt. 

do  the  precautions  which  the  Egyptians  took  elsewhere  to  guard 
their  tombs  against  intrusion  ;  but  direct  proof  of  the  fact  is  not 
w^anting.  When  in  the  ninth  century  the  KaHph  Al-Mamoun 
wished  to  penetrate  into  the  Great  Pyramid  he  was  only  enabled  to 
do  so  by  breaking  into  it  violently,  near  the  centre  of  its  northern 
face,  and  thus  stumbling  accidentally  upon  the  descending  passage 
at  some  distance  from  its  mouth.  That  he  was  reduced  to  employ 
this  method  at  the  risk  of  meetings  with  nothing^  but  the  solid 
masonry  shows  that  no  external  indication  had  been  left  of  the 
opening  through  which  the  mummy  had  been  carried  in.  The 
casing  seems  to  have  been  then  complete  and  consequently  the 
four  sides  of  the  Pyramid  must  have  been  free  from  debris  and 
generally  uniform.  That  the  Arabs  should  have  chosen  the 
right  side  for  their  attack  was  perhaps  owing  to  the  survival  of 
some  ancient  tradition  indicating  the  northern  side  to  be  that  of  the 
entrance,  which,  as  a  fact,  it  has  been  found  to  be  in  all  the 
pyramids  as  yet  explored.  Perhaps  too  the  Arabs  may  have  been 
guided  by  the  traces  of  previous  attempts  made  either  in  the  time 
of  the  Persians  or  in  that  of  the  Romans.^  However  this  may  be 
it  is  very  certain  that  had  they  perceived  any  signs  of  an  original 
doorway,  they  would  have  directed  their  attentions  to  it.  Those 
who  seek  for  treasure  do  not,  like  archaeologists,  strike  out  lines  of 
exploration  in  all  directions  for  the  satisfaction  of  their  curiosity, 
they  go  straight  to  their  point. 

The  pyramid  includes  two  of  those  four  parts  into  which  we  have 
divided  the  typical  Egyptian  tomb  ;  it  contains  the  well  and  the 
mummy  chamber.  As  for  the  funerary  chapel,  there  were  obvious 
difficulties  in  the  way  of  including  it  in  the  thickness  of  the  monu- 
ment itself.  It  would  have  been  difficult  to  preserve  it  from  being 
crushed  by  the  immense  weight  above  it,  and  as  it  would  have  had 
to  be  lighted  from  the  door  alone,  it  must  always  have  been  of  the 
most  restricted  dimensions.  A  different  arrangement  had  therefore 
to  be  devised  from  that  adopted  in  the  case  of  the  mastaba.  The 
open  part  of  the  monument  was  separated  from  that  which  was 
destined  to  be  sealed  up  from  the  outer  world.  The  chapel  or 
temple,  in  which  the  successors  of  the  prince  buried  in  the  pyramid 

^  The  existence  of  the  passage  leading  to  the  mummy  chamber  was  not  unknown 
to  Strabo.  He  says  :  "  Very  nearly  at  the  middle  of  their  sides,  as  to  height,  the 
pyramids  had  a  stone  which  could  be  moved  away  \  when  this  is  done,  a  winding 
passage  appears,  which  leads  to  the  coffin"  (xvii.  p.  1161,  c). 


The  Tomb  under  the  Ancient  Empire.  193 

and  the  priests  told  off  for  its  service  performed  the  prescribed 
rites,  was  erected  at  some  distance  from  the  eastern  face.  The 
remains  of  such  buildings  have  been  found  to  the  east  of  both 
the  second  and  third  pyramids.  That  of  Cheops  has  not 
been  discovered,  but  we  may  assert  with  confidence  that  it  has 
either  been  destroyed  by  the  hand  of  man,  or  that  it  still  lies  under 
the  veil  of  sand.  Were  there  any  serdabs  concealed  in  the  thick- 
ness of  the  temple  walls  ?  That  question  cannot  be  answered. 
The  remains  of  those  buildings  are  in  such  a  condition  that  all 
traces  of  such  an  arrangement  would  have  vanished  had  there  been 
any.  The  walls  have  disappeared.  The  lower  courses  of 
masonry  are  still  in  place,  and  allow  us  to  follow  the  very  simple 
plan  upon  which  these  chapels  were  erected  ;  and  that  is  all.  It  is 
possible,  however,  that  the  Egyptians  depended  solely  upon  the 
profound  respect  which  was  felt  for  the  royal  person,  combined  with 
the  sacredness  of  the  spot  and   the   vigilance  of  the  established 


V  34   so       mo       i:*""*- 
Fin.  127.—  Plans  of  the  temples  belonging  to  the  second  and  third  pyramids  ;  from  Perring. 

priesthood  of  the  necropolis,  to  preserve  fhe  august  images  of  the 
sovereign  from  insult  or  destruction.  The  seven  or  eight  statues 
of  Chephren  which  were  found  at  the  bottom  of  a  pit  in  what  is 
called  the  Temple  of  the  Sphinx,  were  all  more  or  less  mutilated, 
proving  that  this  want  of  precaution  was  sometimes  disastrous. 
In  the  course  of  so  many  centuries,  during  which  the  Hyksos,  the 
Ethiopians,  the  Assyrians,  and  the  Persians  overran  the  country  by 
turns,  such  statues  as  were  not  sheltered  in  some  well  dissembled 
retreat  must  more  than  once  have  been  struck  off  their  pedestals 
and  broken,  or,  like  those  of  the  unlucky  Chephren,  thrown  head- 
foremost into  the  depths  of  the  earth. 

As  such  vast  importance  was  attached  to  the  preservation  of 
the  portrait  statues  upon  which  the  prolongation  of  life  after 
death  was  made  so  largely  to  depend,  is  it  not  probable  that  the 
idea   of  hiding   some   of  them    in    the   innermost   recesses  of   the 

VOL.    I.  C    C 


194  ^^  History  of  Art  in  Ancient  Egypt. 

pyramids  themselves  may  have  occurred  to  those  who  caused 
those  monuments  to  be  built  ?  It  is  obvious  that  no  hiding-place 
could  be  more  secure.  No  such  retreats  have  yet  been  discovered 
in  any  of  the  galleries  which  have  been  exj^lored  by  modern 
curiosity,  but  it  does  not  follow  that  they  do  not  exist  in  some 
corner  which  has  not  yet  been  reached,  which  will  perhaps  never 
be  reached  by  the  most  persevering  explorer.  Quite  lately  M. 
Maspero  believed  that  he  recognized  a  serdab  in  a  subterranean 
chamber  with  three  niches  which  he  found  near  the  mummy 
chamber  in  the  Pyramid  of  Ounas,  the  last  king  of  the  fifth 
dynasty.^  Before  we  could  say  that  such  an  arrangement  does  not 
exist  elsewhere,  we  should  have  to  take  some  pyramid  to  pieces 
from  the  first  stone  to  the  last.  It  might,  however,  be  asserted 
that  the  images  of  the  deceased  would,  if  hidden  in  the  pyramid, 
be  too  far  removed  from  that  public  hall  to  which  his  relations 
brought  their  offerings  and  their  pious  homage.  At  such  a  distance 
they  would  not  have  heard  the  friendly  voices  or  the  magic  chants  ; 
nor  would  the  scent  of  the  incense  have  reached  their  nostrils.  In 
a  word,  they  would  have  been  ill  placed  for  the  fulfilment  of  the 
office  assigned  to  them  by  the  Egyptian  faith. 

We  have  hitherto  spoken  only  of  the  social  purposes  of  the 
pyramid,  of  its  office  as  the  sepulchre  of  the  ancient  kings  of 
Egypt,  or  rather  as  the  part  of  that  sepulchre  that  corresponded 
to  the  least  interesting  parts  of  private  tombs.  In  the  plants  of 
our  gardens  and  orchards,  we  see  cultivation  develop  certain 
organs  at  the  expense  of  others.  We  find  stamens  changed  into 
petals,  giving  us  double  flowers,  and  the  envelope  of  the  seeds 
thickened  and  made  to  shed  perfume.  We  see  the  same  process  of 
development  in  the  tombs  of  the  early  Egyptian  monarchs.  Under 
the  influence  of  their  pride  of  station,  and  as  a  consequence  of  the 
effort  which  they  made  to  perpetuate  their  rank  even  after  death, 
the  stone  hiding-place  which  protected  the  mummy  took  a  size 
which  is  oppressive  to  the  imagination,  while  the  funerary  chapel 
remained  of  modest  dimensions.  This  disproportion  is  to  be  easily 
explained.      The  simple  method  of  construction  which  distinguishes 

1  This  pyramid  was  opened  on  February  28,  1881.  Circumstantial  accounts  of 
the  discoveries  to  which  it  led  have  not  yet  been  published.  The  Aloiiitcur  Egyptien 
of  March  15,  1S81,  contains  a  short  account  of  the  opening.  [Since  this  note  was 
written,  a  full  account  of  the  entrance  and  exploration  of  this  pyramid,  together  with 
the  texts  discovered,  has  been  published  by  M.  Maspero  in  the  Recueil  de  Travaux, 
vol.  iii.  liv.  3  and  4,  18S2. — Ed.] 


The  Tomb  under  the  Ancient  Empire.  195 

the  pyramid  permitted  almost  indefinite  extension,  while  archi- 
tecture, properly  speaking,  was  not  yet  sufficiently  advanced  to 
make  use  of  those  grandiose  orders  which  distinguish  the  porticos 
and  hypostyle  temples  of  the  Theban  period. 

We  have  now  to  consider  the  pyramids  from  another  point  of 
view,  from  that  of  their  probable  origin,  of  their  variety  of  form, 
and  of  the  materials  of  which  they  are  composed.  Descriptions  of 
these  monuments,  such  as  those  contained  in  the  great  works  of 
Vyse^  and  Perring,  works  which  gave  to  the  world  the  accumulated 
results  of  long  and  costly  explorations,  must  not  be  looked  for  in 
these  volumes.  We  do  not  think  it  necessary  that  we  should  give 
even  a  succinct  account  of  the  more  important  pyramids,  such  as 
that  given  by  Baedeker  or  Isambert.  Such  a  proceeding  would  be 
a  mere  duplication  of  those  excellent  manuals,  and  would  moreover, 
be  foreign  to  the  purpose  which  we  have  before  us.  We  take  the 
pyramids  as  known.  The  two  books  just  mentioned  are  within 
the  reach  of  all.  Thanks  to  the  precise  information  and  the 
numerous  figures  which  they  contain,  we  may  content  ourselves 
with  making  a  few  greneral  observations.  Some  of  these  observa- 
tions  will  refer  to  the  pyramids  as  a  whole,  some  to  the  peculiarities 
of  construction  which  distinguish  a  few,  peculiarities  which  do  not 
affect  that  general  type  which  seems  to  be  as  old  as  the  Egyptian 
monarchy  itself. 

As  soon  as  a  society  had  sprung  up  on  the  banks  of  the  Nile 
which  attempted  to  organize  itself  under  the  directing  lead  of  chiefs 
or  headmen,  the  latter  seem  to  have  been  stung  by  the  desire  to 
make  known  their  final  resting-place  by  some  conspicuous  sign. 
The  most  simple  way  of  arriving  at  the  desired  result  was  to  heap 
up  the  earth  over  the  corpse,  so  as  to  form  an  artificial  hillock 
which  should  be  visible  from  a  distance  over  the  level  plain.  This 
was  the  origin  of  that  funerary  mound  which  modern  archaeologists 
call  a  tumulus.  The  tumulus  is  to  be  found  in  most  districts  of 
the  ancient  as  well  as  of  the  modern  world.  But  to  confine  our- 
selves to  our  own  province,  it  is  to  be  found  in  pre-Christian  times 

^  Vyse  (Howard),  Operations  carried  on  at  the  Pyramids  of  Gizeh  in  1837,  7vit]i 
an  Account  of  a  Voyage  i?ito  Upper  Egypt,  and  an  Appendix.  (London,  1840, 
3  vols.  Svo.) 

Perring  (J.  L.),  The  Pyramids  of  Gizeh,  fvm  Actual  Survey  and  Admeasure- 
ment, illustrated  by  Notes  and  References  to  the  Several  Plans,  toith  Sketches  taken  on 
the  Spot  by  J.  Andrews.     (3  parts,  large  oblong  folio.      London,  1S39-42.) 


196  A   History  of  Art  in  Ancient  Egypt. 

among  the  Scythians  of  Herodotus  and  our  ancestors  the  Gauls,  as 
well  as  amongf  the  Greeks  of  the  heroic  aofe.  We  all  know  the 
frequent  expression  of  Homer,  o-zj/xa  xeiveti^,  which  is  literally  to 
display  a  signal,  that  is  to  say,  to  accumulate  over  the  corpse  of  a 
warrior  a  sufficient  number  of  spadefuls  of  earth  to  signalize  it,  for 
the  worship  and  admiration  of  posterity.  Tradition  ascribes  those 
tumuli  which  are  yet  to  be  seen  on  the  plain  of  Troy  to  the 
observance  of  this  custom. 

The  funerary  architecture  of  Egypt  commenced  in  the  same 
fashion,  in  those  distant  ages  which  were  called  by  the  Egyptians 
themselves  the  times  of  Hor-schesou  or  slaves  of  Horus.  We 
cannot  doubt  that  the  pyramid  sprang  from  the  mound.  Its  birth 
must  have  taken  place  after  Menes  had,  by  uniting  the  various 
tribes  under  his  own  sceptre,  caused  the  whole  race  to  take  a  distinct 
step  onwards  in  civilization.  The  pyramid  is  but  a  built  mound. 
It  is  a  tumulus  in  which  brick  and  stone  take  the  place  of  earth. 
This  substitution  adds  very  greatly  to  its  chances  of  duration,  and 
makes  it  a  much  safer  place  of  deposit  and  a  much  more  lasting 
monument  for  the  body  committed  to  its  charge.  The  Nile  mud, 
when  moulded  and  dried  in  the  sun,  gave  bricks  which  still  remain 
good  ;  their  manufacture  and  their  constructive  use  seem  to  have 
been  understood  by  the  Egyptians  as  soon  as  they  emerged  from 
primitive  barbarism.  Thanks  to  the  facilities  thus  afforded,  they 
were  enabled  to  build  monuments  upon  the  graves  of  their  rulers 
which  could  offer  a  better  resistance  to  injuries  of  time  and  human 
enemies  than  a  few  handfuls  of  earth  and  grass.  They  began, 
perhaps,  by  placing  a  few  blocks  of  stone  upon  their  mounds,  so  as 
to  fix  them  more  securely,  or  by  covering  them  v;ith  a  thin  coat  of 
brickwork.  But,  after  a  few  experiments  in  that  direction,  they 
found  it  better  to  construct  the  whole  body  of  the  tumulus  in  the 
harder  material.  Its  size  increased  with  the  constructive  skill  and 
material  appliances  of  its  builders,  until  it  became  first  a  hillock  and 
finally  a  mountain  of  stone,  with  the  impenetrable  rock  for  its  base 
and  flanks  of  solid  masonry. 

The  built-up  tumulus  of  masonry  took  a  form  very  different,  in 
its  definite  lines,  from  the  rounded  slopes  of  the  mound.  The 
squared  forms  of  brick  or  cut  stone  infallibly  give  to  the  edifice 
upon  which  they  are  employed  one  of  those  more  or  less  rigid 
forms  which  are  defined  by  geometry.  When  they  leave  the  hands 
of  the  builder  they  are  either  cubes  or  parallelopipeds,  pyramids 


The  Tomb  under  the  Ancient  Empire.  197 

or  prisms,  cylinders  or  cones.  They  present  the  general  appear- 
ance, they  possess  the  essential  properties,  of  one  of  those  forms. 
We  may  say  that  architecture  was  born  on  the  day  when  man 
began  to  use  the  unyielding  materials  by  which  definite  geome- 
trical forms  can  alone  be  given.  As  soon  as  this  early  development 
was  reached  he  set  to  work  to  combine  those  elementary  forms  in 
different  proportions  and  to  add  to  their  effect  by  elegance  and 
richness  of  decoration,  and  so  in  the  end  to  form  national 
architectures. 

When  the  first  pyramid  was  built  upon  the  borders  of  the  desert 
man  was  on  the  threshold  of  the  movement  to  which  we  have 
referred.  The  form  adopted  for  the  royal  tomb  was  one  of  the 
most  simple  which  could  be  chosen  for  a  building.  The  most 
simple  of  all  would  have  been  the  tetrahedron,  or  pyramid  built 
upon  a  triangular  base.  But  not  a  single  pyramid  of  that  kind  has 
been  discovered  in  Egypt.  The  whole  of  the  pyramids,  large  or 
small,  are  built  upon  a  right-angled  base,  and  in  most  instances 
upon  one  with  sides  practically  equal. ^  Mystic  reasons  for  this 
shape  have  been  given.  It  has  been  said  that  each  face  was 
dedicated  to  one  of  the  four  powers  of  Amen,  which  corresponded 
to  the  cardinal  points  of  heaven.^  We  are  not  yet  sufficiently  well 
acquainted  with  the  genesis  of  the  Egyptian  religion  to  be  able  to 
decide  how  far  into  the  past  the  four  powers  of  Amen  may  be 
traced  :  but  it  is  quite  possible  that  they  were  derived  from  the 
four  faces  of  the  strictly  oriented  pyramids.  Were  we  inclined  to 
enter  into  this  discussion  we  should  rather,  perhaps,  attribute  the 
shape  of  the  pyramid  to  the  prevailing  Egyptian  desire  to  turn  one 
face  of  their  tombs  towards  the  west,  the  abode  of  the  dead,  and 
another  to  the  east,  whence  the  hoped-for  resurrection  was  to 
come.  The  three-sided  pyramid  would  not  have  lent  itself  to  such 
an  arrangement. 

There  is  also  something  unpleasant  to  the  eye  in  the  sharp 
angles  which  form  the  three  arites  of  the  tetrahedron  ;  it  looks  as  if 
there  had  been  a  lack  of  material,  and  as  if  the  structure  would 
suffer  in  consequence.  The  four-sided  pyramid  has  more  dignity 
and  more  amplitude  ;  its   four  faces,  placed  back  to  back  in  pairs, 

^  The  base  of  the  great  pyramid  at  Sakkarah  is  a  rectangle,  measuring  390  feet 
from  north  to  south,  and  347  from  east  to  west.  The  three  great  pyramids  at  Gizeh 
like  most  of  these  structures,  are  built  upon  a  base  which  is  practically  square. 

2  Mariette,  Itineraire  de  la  Haute-Egypte,  p.  96. 


198 


A   History  of  Art  in  Ancient  Egypt, 


seem  to  counterpoise  and  sustain  each  other  in  a  fashion  which  is 
impossible  in  the  case  of  the  tetrahedron. 

The  one  characteristic  possessed  in  common  by  those  rehcs  of 
the  Ancient  Empire  which  we  call  pyramids,  is  their  four-sided- 
ness.  To  an  attentive  observer  these  buildings  offer  more 
diversities  than  would  at  first  shAit  be  believed.  From  Meidoum 
in  the  south  to  Abou-Roash  in  the  north  is  a  distance  of  43^ 
miles  as   the  crow  flies.      Between  these  two  points,  which  may 


MmMM&m 


M 

y 


ii]ii||i  I  "I,  IP'   I  I.  :•  I'A'.f  ill,  I 

III  -    i-'i  :.  K|•i''> 
mi  1  m 

llll-^  .    'I'    >  <    •.(\\ 

■;■;/.■■ 

I  :    .1  "--^  C'l    I 


Fjg.  128. — Plan  of  the  Pyramid  of  Cheops. 


be  called  the  northern  and  southern  boundaries  of  the  pyramid 
field,  about  one  hundred  have  been  discovered,  sixty-seven  of 
which  have  been  examined  by  Lepsius.  Now,  in  this  whole 
number  there  are  not  two  which  resemble  each  other  in  all 
particulars,  or  which  seem  to  be  copies  of  one  model.  We  do  not 
refer  only  to  their  height,  which  differs  in  an  extreme  degree. 
The   three  large  pyramids  at    Gizeh   are    482,   454,  and  218  feet 


The  Tomb  under  the  Ancient  Empire. 


199 


high  respectively,  while  at  their  feet  are  several  little  pyramids 
which  hardly  exceed  from  50  to  70  feet  of  vertical  height. 
Between  these  two  extremes  many  of  intermediate  sizes  may  be 
inserted.  The  Stepped  Pyramid,  near  Sakkarah,  is  about  190 
feet  high;  the  largest  of  those  at  Abousir  is  about  165  ;  one  of 
those  at  Dashour  is  not  quite  100  feet.  These  differences  in 
height  are  easily  explained  by  one  of  those  national  habits  to  which 
we  have  already  alluded.  Every  Egyptian,  as  soon  as  he  arrived 
at  years  of  discretion,  set  about  building  his  own  tomb.  He  dug 
the  well  and  the  mummy  chamber,  he  caused  the  sarcophagus  to 
be  carved  and  the  funerary  chapel  to  be  built.      It  often  happened 


fi/SSSaSSiSiicn-C 


S'-'t^ 


Fig.    129. — The  great  pyramid  and  the  small  pyramids  at  its  foot  ;  from  Perriiig. 

that  those  who  had  ordered  such  works  died  long  before  they  were 
finished,  and  it  would  seem  that  their  heirs  were  content  with 
doing  no  more  than  was  strictly  necessary.  They  placed  the 
mummy  in  its  grave  with  the  prescribed  ceremonies,  they  filled  up 
the  well  and  sealed  the  private  parts  of  the  tomb  ;  but  being 
occupied  with  the  preparations  for  their  own  funeral,  they  did  not 
continue  the  decoration  of  the  chapel,  which  thenceforward 
remained  in  statu  quo.  Thus  only  can  we  explain  the  state  in 
which  several  important  tombs  have  been  discovered  both  at 
Memphis  and  at  Thebes.  On  one  wall  we  find  paintings  and 
sculptures  carried  out  with  the  greatest  care   and  finish,   while  on 


200  A  History  of  Art  in  Ancient  Egypt. 

another  nothing  Is  to  be  seen  but  the  first  rough  outhne,  in  red 
paint,  by  the  artist  charged  with  the  undertaking.  The  comple- 
tion of  the  work  must  have  been  suddenly  arrested  by  the  death 
of  the  destined  inhabitant  of  the  tomb. 

It  was  the  same  with  the  sepulchres  of  the  kings.  Each  sovereign 
began  the  construction  of  his  pyramid  as  soon  as  he  found  himself 
upon  the  throne.  But,  in  case  his  life  and  his  reign  should  be  cut 
short,  he  began  with  those  constituents  of  the  tomb  which  were 
absolutely  necessary.  He  pressed  on  the  work  until  he  had  raised 
a  pyramid  of  moderate  size  with  its  mummy  chamber.  When  this 
point  had  been  reached,  his  immediate  anxiety  came  to  an  end  ; 
but  that  was  no  reason  for  interrupting  the  course  of  the  work. 
The  higher  and  wider  his  pyramid,  the  more  efficient  a  guardian 
of  his  body  would  it  be,  and  the  more  impressive  would  be  the 
message  carried  down  by  it  to  posterity  as  to  the  power  of  its 
builder.  Year  after  year,  therefore,  he  employed  crowds  of  im- 
pressed workmen  to  clothe  it  in  layer  after  layer  of  dressed  stone 
or  brick,  so  that  the  edifice  raised  in  comparative  haste  at  the 
beeinnino;  of  his  reio^n,  became  in  time  nothino-  but  the  nucleus  or 
kernel  of  one  many  times  its  size.^  The  construction  was  thus 
begun  in  the  centre  and  was  developed  outwards,  like  the  timber 
of  a  tree  in  successive  years.  As  the  pyramid  grew  in  extent  and 
height,  each  successive  coat,  so  to  speak,  required  more  hands  and 
more  time.  We  have  no  reason  to  believe  that  each  coat  had  to 
be  finished  within  a  certain  period,  and  so  it  would  be  futile  to 
attempt  to  found  any  calculation  as  to  the  duration  of  the  different 
reigns  upon  the  number  of  these  concentric  layers ;  but  we  may 
assert  in  a  general  way  that  the  highest  pyramids  correspond  to 
the  longest  reigns.  We  know,  by  the  witness  of  ancient  authors, 
that  the  kings  who  built  the  three  great  pyramids  at  Gizeh,  namely, 
Cheops,  Chephren,  and  Mycerinus,  each  reigned  about  sixty  years. 
History  thus  confirms  the  truth  of  the  induction  which  arises  from 

'  This  method  of  construction  may  be  easily  recognized  in  the  Pyramid  of 
Meidoum.  That  curious  structure  was  built  in  concentric  layers  round  a  nucleus. 
These  layers  are  by  no  means  equal  in  the  excellence  either  of  the  workmanship  or 
of  the  materials  employed.  Some  show  supreme  negligence  ;  in  others  we  find  the 
builders  of  the  Ancient  Empire  and  their  materials  both  at  their  best.  The  same 
fact  has  been  observed  in  regard  to  the  Stepped  Pyramid  and  the  pyramids  at 
Abousir.  It  would  seem  that  the  work  was  assigned  in  sections  to  different  corvees, 
whose  consciences  varied  greatly  in  elasticity.  (Mariette,  Voyage  de  la  Haute- 
Egypte,  p.  45-) 


The  Tomb  under  the  Ancient  Empire. 


20I 


the  study  of  those  monuments  and  from  a  comparison  of  the  con- 
structive processes  made  use  of  by  the  architects  of  the  pyramids.^ 
The  author  of  Baedeker's  Guide  has  not  been  content  with 
beheving,  hke  Perring,  Lepsius,  and  Mariette,  that  the  pyramid 
grew  by  tlie  appHcation  of  successive  envelopes  of  stone  round  the 
central  mass,  either  in  horizontal  courses  or  in  courses  sloping 
towards  the  axis  of  the  buildino^.  He  has  brouo^ht  forward  an 
elaborate  theory  of  construction,  which,  though  very  ingenious, 
encounters  several  grave  objections.  We  shall  point  out  those 
objections  while  we  endeavour  to  explain  the  system  itself  by  the 
help  of  special  illustrations  drawn  for  us  by  the  author  of  the 
Guide  in  question.^ 


Fig.  1 30. — The  three  great  pyramids  ;  from  the  south. 

When  Cheops  first  began  to  think  about  building  his  tomb,  he 
could  not  have  counted  upon  giving  it  the  colossal  dimensions 
which  it  presents  even  in  its  actual  injured  condition.  The  area 
of  the  great  pyramid  is  more  than  double  that  of  Saint  Peter's  at 
Rome.  If  we  deduct  from  its  total  volume  the  core  of  rock 
which  it  incloses  ^  and  the  openings  v/hich  it  contains,  the  masonry 


1  Lepsius,  Briefe  aus  Aigypten,  pp.  41,  42  (in  speaking  of  the  Pyramid  of 
Meidoum,  from  which  he  received  the  first  hint  of  this  explanation).  See  also  his 
paper  entitled  Ueber  den  Ban  der  Fyramiden,  in  the  Monatsbericht  of  the  Berlin 
Academy,  1843,  pp.  177-203. 

^  /Egypt en.,  First  part,  1878,  p.  341. 

^  It  has  been  suggested  by  Mr.  Cope  Whitehouse  that  the  nucleus  of  rock  under 
the  great  pyramids  was  originally  much  more  important  than  is  commonly  supposed. 
During  his  expedition  in  March,  1882,  he  ascertained  that  a  profile  from  the 
Mokattam  across  the  Nile  valley  into  the  western  desert  would  present  the  contours 

VOL.   I.  D    D 


202 


A   History  of  Art  in  Ancient  Egypt. 


in  its  primitive  integrity  must  have  amounted  to  a  total  of 
3,479,600  cubic  yards.  Even  now,  when  so  much  of  its  sub- 
stances has  been  detached  and  carried  away,  there  still  remains  the 
enormous  mass  of  3,246,600  cubic  yards.  Supposing  that,  two 
or  three  years  after  the  commencement  of  a  work  upon  this  colossal 
scale,  death  had  carried  off  its  projector,  can  we  believe  that  any 
successor,  even  a  son  who  was  sincerely  devoted  to  the  memory  of 
his  father,  would  have  burthened  himself  with  the  continuation  and 
completion  of  such  an  enterprise  ?  The  new  sovereign  would  have 
enough  to  do  in  commencing  and  carrying  on  the  erection  of  his 
own  tomb,  and,  moreover,  would  be  irresistibly  tempted  to  make 
use,  for  its  construction,  of  the  accumulated  material  and  collected 
labour  of  his  predecessor. 

Even  four  or  five  thousand  years  before  our  era,  men  were  too 
sagacious  to  reckon  upon  the  piety  or  gratitude  of  an  heir.  For 
the  closing  and  final  sealing  up  of  the  pyramid,  its  builder  and 
destined  inhabitant  was  obliged  to  depend  upon  his  survivors,  he 
could  not  do  it  himself.  Moreover,  the  external  completion, 
which,  in  the  case  of  the  greater  monuments,  must  have  been  a 
long  and  costly  matter,  had  to  be  entrusted  to  the  same  hands. 
The  reigning  king,  so  long  as  he  was  not  too  sternly  reminded  of 
the  end  by  disease  or  the  infirmities  of  age,  must  have  felt  great 
reluctance  to  order  the  cessation  of  the  work  which  had  gone  on 
under  his  own  eye  for  so  many  years,  or  to  arrest  that  course  of 

shown  in  the  annexed  woodcut.      He  concludes  that  a  large  part  of  the  material  of 
those  pyramids  wa^  obtained  upon  their  sites,  and  quarried  above  the  level  at  which 


aoo  WADV    —     rADHI 

00 

'rump 


KOM-CL-KASHAB      aBU-ROASH        '"^'^^^""^^ 


the  stones  were  finally  placed.     He  cites  Herodotus  (ii.  125)  as  conveying  in  an 
imperfect  form  the  tradition  that  the  pyramids  were  "constructed  from  above." 


The  Tomb  under  the  Ancient  EiMpire.  203 

development  which,  after  being  a  continual  source  of  pride  and 
pleasure  to  himself,  might  end  in  gi'C^ing  him  a  monument  surpass- 
ing those  of  his  famous  predecessors.  He  was,  therefore,  very 
likely  to  be  surprised  by  death  with  his  tomb  still  unfinished,  with 
the  final  cope-stone  still  upon  the  ground,  or,  even  when  that  had 
been  put  in  place  so  as  to  show  the  total  height,  with  the  casing  of 
polished  stone  which  was  destined  to  hide  the  inner  courses  of  the 
masonry  and  the  entrances,  still  incomplete.  Upon  two-thirds  or 
three-quarters  of  each  face,  his  pyramid  would  still  present  the 
aspect  which  necessarily  belonged  to  it  during  the  period  of  its 
construction  ;  an  aspect  which  has  again  distinguished  the  great 
pyramid  since  it  was  despoiled  of  its  casing.  As  each  course  was 
set  back  from  that  upon  which  it  was  placed,  the  final  ensemble 
looked  like  an  enormous  staircase  with  steps  gradually  dimanishing 
in  length  as  they  neared  the  summit. 

There  were  many  of  the  Egyptian  princes  who  from  want  of 
patience  or  zeal,  or  from  some  other  motive,  failed  to  carry  on  the 
enterprise  of  their  predecessors  to  its  destined  conclusion.  We 
are  ignorant  as  to  the  condition  of  the  three  great  pyramids  of 
Gizeh  at  the  death  of  their  projectors.  But  they  appear  to 
have  been  finished  in  most  of  their  details  with  a  care  which  would 
seem  to  indicate  that  Cheops,  Chephren,  and  Mycerinus,  must  have 
been  there  to  overlook  the  smallest  details  of  their  execution. 
Other  pyramids,  on  the  other  hand,  seem  to  have  been  left  in  a 
comparatively  imperfect  state. 

These  observations  furnish  us  with  an  initial  objection  to  the 
theory  to  which  we  have  referred.  Some  may  refuse  to  believe 
that  Cheops  intended  from  the  beginning  that  his  pyramid  should 
have  the  dimensions  and  the  internal  arrangements  which  we  now 
see.  But  why  should  he  not  have  done  so  ?  If  he  had  died  at 
the  end  of  a  few  years,  his  pyramid  would,  perhaps,  have  presented 
to  us  a  shape  like  that  of  some  other  edifices  of  the  same  kind,  a 
large  base  which  had  never  received  either  its  cope-stone  or  its 
casing.  So  too  with  those  of  Mycerinus  and  Chephren.  Have 
not  absolute  monarchs  existed  at  all  times,  whose  infinite  power 
seems  to  have  made  them  forsret  the  eternal  limits  of  time 
and  space  ?  Sometimes  Fortune  has  been  cruel  to  them  ; 
but  often  she  seems  to  have  placed  herself  entirely  at  their 
disposal. 

Amone  the  causes  which  combine  to  make  the  roval   tombs  of 


204  A   History  of  Art  in  Ancient  Egypt. 

the  first  six  dynasties  so  unequal  in  height  and  appearance,  the 
very  unequal  length  of  the  reigns  is  the  most  important.  If  we 
were  better  acquainted  with  the  condition  of  Egypt  in  those 
remote  epochs,  we  should,  no  doubt,  be  enabled  to  give  other 
reasons  for  their  want  of  uniformity.  The  chances  of  completion 
and  even  of  preservation  in  its  complete  state  enjoyed  by  a 
pyramid  must  have  greatly  depended  upon  the  descent  of  the 
crown.  Wlien  king  succeeded  king  in  one  family  those  chances 
were  much  better  than  when  dynasty  succeeded  dynasty,  whether 
the  break  were  due  to  internal  revolution  or  to  the  failure  of  the 
legitimate  line.  It  is  even  possible  that  some  of  those  pyramids 
which  are  now  to  outward  appearance  mere  heaps  of  debris 
never  received  the  mimimy  for  whose  reception  they  were 
designed  and  built. 

The  pyramids  differ  also  in  the  materials  employed.  The 
great  pyramids  at  Gizeh  are  built  of  fine  limestone  from  Mokattam 
and  Toura  ;  the  chief  one  at  Sakkarah  of  a  bad  clavish  limestone 
from  the  neighbouring  rocks  ;  at  Dashour  and  Abou-Roash  there 
are  pyramids  of  unburnt  brick.  Finally  there  are  pyramids  built 
chiefly  of  stone  which  is  kept  in  place  by  a  carefully  constructed 
skeleton,  so  to  speak,  of  brick.  This  construction  is  to  be  found 
in  the  pyramid  of  Illahoun,  at  the  entrance  to  the  Fayoum 
(Fig.  131). 

There  is  the  same  variety  in  the  position  of  the  mummy 
chamber.  Sometimes  this  is  within  the  sides  of  the  pyramid  itself, 
as  in  that  of  Cheops ;  sometimes,  after  the  example  of  the  mastaba, 
it  is  cut  out  of  the  living  rock  upon  which  the  pyramid  stands. 
This  arrangement  is  to  be  found,  for  instance,  in  the  pyramid  of 
Mycerinus,  where  the  roof  of  the  mummy  chamber  is  about  ^i^^i 
feet  below  the  lowest  course  of  the  pyramid  itself.  So  too  in 
the  Stepped  Pyramid,  where  the  whole  complicated  system  of 
corridors  and  cells,  which  distinguishes  that  edifice,  is  cut  in  the 
rock,  so  that  the  building  itself  is  absolutely  solid.  Most  of  the 
pyramids  have  no  more  than  one  or  two  entrances,  giving  access 
to  narrow  galleries,  sometimes  ascending,  sometimes  descending, 
which  lead  to  one  or  two  chambers  of  very  small  dimensions  when 
compared  to  the  enormous  mass  which  rises  above  and  around 
them  (Fig.  132).  In  the  subterranean  part  of  the  Stepped  Pyramid 
the  proportion  of  voids  to  solids  is  far  less  insignificant.  This 
pyramid,  which   is   not   nearly  so  carefully  oriented  as  the  others. 


The  ToMi;  under  the  Ancient  Empire. 


20: 


has  four  entrances  and  a  series  of  internal  passages,  horizontal 
galleries,  staircases  and  cells,  which  make  it  little  else  than  a 
subterranean  labyrinth.  It  is  singular  also  in  having,  upon  its 
central  axis  and  at  the  point  upon  which,  at  various  heights,  all  its 
galleries  converge,  a  sort  of  large  well,  a  chamber  about  twenty 
feet  square  and  eighty  feet  high,  in  the  pavement  of  which  a  huge 
block  of  granite  cut  into  the  shape  of  a  cork  or  plug  was  so  placed 
as  to  open  at  will  ^  and  leave  a  free  passage  for  the  descent  into  a 
second  chamber,  the  purpose  of  which  is  more  than  obscure,  as  it 
is  too  small  even  to  have  contained  a  sarcophagus.-     The   end  of 


>C^-^3-^ — ;^-^j 


Fig.    131. — The  pyramid  of  Illahoun,  horizontal  section  in  perspective;  from  the  plan 

of  Perripe, 


the  long  passage  which  leads  to  the  thirty  chambers  which  have 
been  counted  beneath  this  pyramid  has  been  found  in  the 
neighbouring  sands   (Fig.    134). 

Another  point  of  difference  :  most  of  the  pyramids  are  built 
round  a  core  of  living  rock,  which  is  embraced  by  the  lower 
courses  of  their  masonry.  But  the  pyramid  of  M^'cerinus  is  just 
the  reverse  of  this.  It  is  built  over  a  hollow  in  the  rock  which  is 
filled  up  with  masonry.  The  inequalities  of  the  surface  were 
usually  taken  advantage  of  so  as  to  economize  material,  and  make 

'  The  weight  of  this  stopper  is  about  four  tons,  and  it  lias  long  been  a  puzzle  to 
egyptologists  how  it,  and  others  like  it,  could  be  raised  and  lowered.  M.  Perrot's 
words  must  not,  therefore,  be  taken  too  literally. — Ed. 

-  Arthur  Rhonk,  L Egypte  a  pe'ites  Journces,  p.  259. 


2o6 


A   History  of  Art  in  Ancient  Egypt. 


a  greater  show  with  less  labour.  Mycerinus,  however,  did  not 
fear  to  increase  his  task  by  rearing  his  pyramid  over  a  depression 
in  the  plateau. 

There  is  no  less  diversity  in  the  external  aspects  of  the 
pyramids.  We  are  most  familiar  with  the  shapes  of  the  great 
pyramids  at  Gizeh  (Fig.  130  and  PI.  I,  2);  their  images  have 
been  multiplied  to  infinity  by  engraving  and  photography,  but  we 
make  a  great  mistake  when  we  imagine  all  the  royal  tombs  at 
Memphis  to  be  built  upon  this  one  model.  They  do  not  all 
present  the  same  simplicity  of  form,  the  same  regular  slope  from 
summit  to  base,  or  the  smooth  and  polished  casing  which  distin- 
guished   those   great    monuments    when    they    were    in    complete 


0       ic      i2 


-TT-* ' — ±r 


FiG.  132. — Section  of  the  pyramid  of  Cheops  ;  from  Perring. 


preservation.  The  southern  pyramid  of  Dashour  otters  us  one 
of  the  most  curious  variations  upon  the  original  theme  (Fig.  133). 
Its  anofle-ridofes  are  not  unbroken  straight  lines  from  base  to 
summit.  The  slope  of  its  faces  becomes  less  steep  at  about 
half  their  height.  The  lower  part  of  its  sides  make  angles  of 
54°  41''  with  the  horizon,  while  above  they  suddenly  fall  back  to 
an  angle  of  42°  59'.  This  latter  slope  does  not  greatly  differ  from 
the  43°  36'  of  the  other  pyramid  in  the  same  neighbourhood.  No 
indication  has  vet  been  discovered  as  to  the  builder  of  this 
pyramid. 

A  second  variation,   still   more  unlike  the   Gizeh  type,   is  to   be 


The  Tomb  under  the  Ancient  Empire. 


!07 


found  in  the  great  pyramid  of  Sakkarah,  the  Stepped  Pyramid, 
which  was  considered  by  Mariette  as  the  oldest  of  them  all. 
Taking  a  passage  from  Manetho  as  his  authority,  he  thought 
himself  justified  in  attributing  it  to  the  fourth  king  of  the  first 
dynasty,  Ouenephes  or  Ata,  and  he  was  inclined   to  see   in  it  the 


Fig.  133. — The  southern  pyramid  of  Dashour  ;  from  the  measurements  of  Perring. 

Serapeum,  or  Apis  tomb,  of  the  Ancient  Empire.  Its  present 
elevation  is  about  190  feet.  Each  of  its  sides  is  divided  hori- 
zontally into  six  large  steps  with  inclined  faces.  The  height  of 
these  steps  decreases  progressively,  from  the  base  to   the  summit, 


I-    !i-'- 


Fig.  134. — Section  of  the  Stepped  Pyramid  ;  from  Perring. 


from  38  feet  2  inches  to  29  feet  6  inches.  The  width  of  each  step 
is  nearly  7  feet.  It  will  be  seen,  therefore,  that  this  building 
rather  tends  to  the  pyramidal  form  than  achieves  it  ;  it  is  a  rough 
sketch  for  a  pyramid. 

Does  this    want    of  completion   result    from    accidental    causes, 


2oS  A   History  of  Art  in  Ancient  Egypt. 

or  must  it  be  referred  to  ignorance  of  the  full  beauties  of  the 
pyramidal  form  on  the  part  of  its  builders  ?  If  the  conjecture 
of  Mariette  is  well  founded,  the  Stepped  Pyramid  is  not  only  the 
most  ancient  building  in  Egypt  but  in  the  whole  world  ;  and 
in  the  remote  century  which  witnessed  its  construction  men  may 
not  yet  have  learnt  to  fill  up  the  angles  left  in  their  masonry,  they 
may  have  been  quite  satisfied  to  leave  their  work  in  a  condition 
which  to  us  seems  imperfect. 

The  Germans  have  evolved  a  complicated  system  of  con- 
struction from  notes  made  by  Lepsius  upon  the  details  of  the 
masonry  in  different  pyramids.  In  order  that  this  system  may 
be  more  easily  understood,  we  give,  on  the  opposite  page,  a  series 


Fig.  135. — The  Stepped  Pyramid  ;  restored  from  the  measurements  of  Perring. 

of  representations  of  such  a  pyramid  in  different  stages  of  com- 
pletion (Figs.  136  to  142).  A  commencement  was  made  by 
erecting  a  very  narrow  and  perpendicular  pyramid  crowned  by  a 
pyramidion,  like  a  stumpy  obelisk  (Fig.  136).  This  finished, 
sloping  masses  were  erected  against  it  so  as  to  form,  with  the 
pyramidion  of  the  first  mass,  a  second  pyramid.  The  apex  of  this 
pyramid,  a  pyramidion  of  a  single  stone,  might  be  put  in  place  and 
the  work  considered  finished  (Fig.  137)  ;  or,  if  the  builder  were 
sanguine  as  to  time,  he  might  seek  to  push  on  still  farther. 
Then,  at  the  line  where  the  slopes  of  the  pyramid  left  the  earth,  four 
perpendicular  walls  were  erected  to  the  height  of  the  pyramidion. 
.The  space  between  the  sides  of  the  pyramid  and  the  inner  faces 


The  Tomb  under  the  Ancient  Empire. 


209 


of  these  walls  was  filled  in,  and  thus  a  kind  of  terrace,  or  huge 
rectangular  block,  was  obtained  (Fig.  138),  which  served  as  the 
core  for  a  new  pyramid  (Fig.  139).  This  again  disappeared  under 
a  pyramid  of  larger  section  and  gentler  slope  (Fig.  140),  whose 
sides  reached  the  ground  far  beyond  the  foundations  of  the  terrace. 
In  the  case  of  a  long  reign  this  operation  might  be  repeated  over 


Fig.  136, 


Fig.  137. 


Fig.  nS. 


Fig.  139. 


Fig.  140. 


Fig.  141. 


Fig.  142. 


Figs.    136—142. — Successive  states   of  a   pyramid,    according  to   the   system  advocated   in 

Btedeker's   Guide. 


and  over  again  (Figs.  140  and  142).  A  large  pyramid  would 
thus  be  composed  of  a  series  of  pyramidal  envelopes  placed  one 
upon  another.  The  mummy-chamber  was  either  cut  in  the  rock 
before  the  laying  of  the  first  course  of  stone,  or  it  was  contrived 
in  the  thickness  of  the  masonrv  itself;  as  the  casingr  of  stone  went 
on   increasing  in  thickness,  galleries  were   left  for  ventilation   and 


VOL.   I. 


E    E 


2IO  A   History  of  Art  in  Ancient  Egypt. 


for  the  introduction  of  the  sarcophagus  and  the  mummy.  The 
mummy-chamber  is  always  found  either  upon  the  axis  of  the 
pyramid,  or  in  its  immediate  neighbourhood,  and  always  nearer 
the  base  than  the  summit. 

We  are  told  that  the  system  of  construction  here  set  forth  is 
rendered  almost  certain  by  the  fact  that  "  the  deeper  we  penetrate 
into  the  pyramid  the  more  careful  do  we  find  the  construction, 
which  becomes  more  and  more  careless  as  the  exterior  is 
approached.  In  fact,  as  each  new  envelope  was  commenced, 
the  chances  of  its  being  completed  became  less."  The  mass  of 
stone  to  be  worked  and  placed  was  greater,  while  the  king,  upon 
whose  life  the  whole  operation  depended,  was  older  and  nearer 
his  death.  The  builders  became  less  sure  of  the  morrow  ;  they 
pressed  on  so  as  to  increase,  at  all  hazards,  the  size  of  the  monu- 
ment, and  trusted  to  the  final  casing  to  conceal  all  defects  of 
workmanship. 

This  system  of  pyramid  building  would  explain  the  curious 
shapes  which  we  have  noticed  in  the  Stepped  Pyramid  and  the 
southern  pyramid  of  Dashour.^  Both  of  those  erections  would 
thus  be  unfinished  pyramids.  At  Sakkarah,  the  angles  left  by  the 
successive  stages  would  be  waitinof  for  their  fillinsf  in  ;  at  Dashour, 
the  upper  part  of  a  pyramid  of  gentle  slope  would  have  been 
constructed  upon  the  nucleus  which  was  first  erected,  but  the 
continuation  of  the  slope  to  the  ground  would  have  been  pre- 
vented by  the  stoppage  of  the  works  at  the  point  of  intersection  of 
the  upper  pyramid  and  its  provisional  substructure.  Hence  the 
broken  slope  which  has  such  a  strange  efiect,  an  effect  which  could 
not  have  entered  into  the  orio-inal  calculations  of  the  architect. 

But  although  this  theory  seems  satisfactorily  to  explain  some 
puzzling  appearances,  it  also,  when  tested  by  facts,  encounters 
some  very  grave  objections.  The  explorers  of  the  pyramids  have 
more  than  once,  in  their  search  for  lost  galleries  and  hidden 
chambers,  cut  for  themselves  a  passage  through  the  masonry,  but 
neither  in  these  breaches  made  by  violence,  nor  in  the  ancient  and 

1  There  are  other  stepped  pyramids  besides  that  at  Sakkarah.  Jomard  describes 
one  of  crude  and  much  crumbled  brick  at  Dashour.  It  is,  he  says,  about  140  feet 
high.  Its  height  is  divided  into  five  stages,  each  being  set  back  about  1 1  feet 
behind  the  one  below.  These  steps  are  often  found,  he  adds,  among  the  southern 
pyramids,  and  there  is  one  example  of  such  construction  at  Gizeh.  {Descriptioii  de 
r Egypte,  vol.  x.,  p.  5.)  At  Matarieh,  between  Sakkarah  and  Meidoum,  there  is  a 
pyramid  with  a  double  slope  like  that  at  Dashour. 


The  Tomb  under  the  Ancient  Empire.  2 1 1 

carefully  constructed  passages   to  which   they  were  the  means  of 
giving    access,    have   any    signs    been,    discovered,    or    at    least, 
reported,  of  the  junctions   of   different  surfaces  and   slopes  which 
must  have  existed  according  to  the  theory  which  we  arc  noticing. 
We  should  expect,   at   least,   to  find    the  nearly  upright  sides   of 
the  cubic  mass  with  which  the  pyramid  began,  contrasting  with  the 
comparatively  gentle  slopes  which  were  built   against   it.      These 
different  parts  of  the   pyramid,  we   are   told,   were  built  and    fin- 
ished separately,   a  proceeding  which,  if  the  later  parts  were  to  be 
properly  fitted  to  the  earlier  and  the  final  stability  of  the  monu- 
ment assured,  would  have  demanded  a  minute  and  scrupulous  care 
which  was  not  common  with   Egyptian  workmen.      How,  without 
numerous  throuo-h  bondinof-stones,  could  those  slides  and   settle- 
ments    be    prevented   to   which  the   want  of  homogeneity  in  the 
structure  would  otherwise  be  sure  to  lead  '^.      But  we  are  not  told 
that  any  such  junctions  of  old  and  new  work  are  to  be  found  even 
in  those  points  where  they  would  be  most  conspicuous,  namely,  in 
the  galleries   leading  to  the    internal  chambers,  where  a  practised 
eye  could  hardly  fail  to  note  the  transition.     We  do  not  say  that 
there  are   no  such  transitions,  but  we  think  the  advocates  of  the 
new  theory  should  have  begun  by  pointing  them  out  if  they  exist. 
There  is  another  difficulty  in  their  way.      How  is  their  system 
to  explain  the  position  of   the  mummy-chamber   in  certain  pyra- 
mids ?     Let  us  take  that  of  Cheops  as  an  example.      If  its  internal 
arrangements  had   been  fixed  from  the  beginning,  if  the  intention 
had  been  from  the  first   to  place  the  mummy-chamber  where  we 
now  find  it,  at  about  one-third  of  the  whole  height,  why  should  the 
builders  have  complicated  their  task  by  imposing  upon  themselves 
these  ever  difficult  junctions  ?      Would  it  not  have  been  far  better 
to  build  the  pyramid  at  once  to  the  required  height,  leaving  in  its 
thickness  the  necessary  galleries  ?     The  same  observation  applies 
to  the  discharging  chambers  above  the   mummy-chamber.      The 
whole  of  these  arranQ[-ements,  the  \\\<y\\  vestibule  with  its  wonderful 
masonry,    the    chambers    and    the    structural    voids    above    them, 
appear  to  have  been  conceived  and  carried  out  at  one  time,  and 
by  the  same  brains  and  hands.      Not  a  sign  is  to  be  found  of  those 
more  or  less  well-veiled  transitions  which  are  never  absent  when 
the  work  of  one  time  and  one  set  of  hands  has  to  be  united  with 
that  of  another.      Or  are  we  to  believe  that,  they  commenced  by 
building  a  hill  of  stone  composed  of  those  different  pyramids  one 


2  12  A   History  of  Art  in  Ancient  Egypt. 

within  another,  and  that  they  afterwards  carved  the  necessary 
chambers  and  corridors  out  of  its  mass  ?  One  of  the  heroes  of 
Hoffmann,  the  fantastic  Crespel,  made  use  indeed  of  some  such 
method  in  giving  doors  and  windows  to  his  newly-built  house,  but 
we  may  be  sure  that  no  architect,  either  in  Egypt  or  elsewhere, 
ever  thought  of  employing  it.  The  disintegration  to  which  it 
would  lead  may  easily  be  imagined. 

We  may  here  call  attention  to  a  circumstance  which  justifies  all 
our  reserves.  There  is  but  one  pyramid  which  seems  to  have 
been  built  upon  a  system  which,  though  much  less  complicated, 
resembled  that  which  we  are  noticing  in  some  degree,  v/e  mean 
the  Stepped  Pyramid  of  Sakkarah.  Now  we  find  that  the  whole 
of  the  complicated  net-work  of  chambers  and  passages  in  that 
pyramid  is  cut  out  of  the  living  rock  beneath  its  base,  and  that 
they  are  approached  from  without  by  subterranean  passages. 
The  difficulty  of  deciding  upon  the  position  of  the  chambers  in 
advance,  and  of  constructing  the  galleries  through  the  various 
slopes  of  the  concentric  masses  which  were  to  form  the  pyramid, 
was  thus  avoided,  and  the  builder  was  able  to  devote  all  his 
attention  to  increasing  the  size  of  the  monument,  by  multiplying 
those  parallel  wedges  disposed  around  a  central  core  of  which 
it  is  composed. 

The  observations  made  by  Lepsius  in  the  Stepped  Pyramid  and 
in  one  at  Abousir  seem  to  prove  that  some  pyramids  were  con- 
structed in  this  manner.  In  both  of  those  buildings  all  necessary 
precautions  were  taken  to  guard  against  the  weaknesses  of  such 
a  system.  It  is  difficult  to  understand  how  separate  slices  of 
masonry,  placed  one  upon  the  other  in  the  fashion  shown  by  the 
section  which  we  have  borrowed  from  Perring's  work  (Fig.  134), 
could  have  had  sufficient  adherence  one  to  another.  Lepsius 
made  a  breach  in  the  southern  face  of  this  pyramid,  and  the 
examination  which  he  was  thus  enabled  to  institute  led  him  to 
suggest  a  rather  more  probable  system  of  construction.  Upon 
the  external  sloping  face  of  each  step  he  found  two  casing-walls, 
but  these  did  not  extend  from  the  ground  to  the  apex  of  the 
monument,  they  reached  no  higher  than  the  single  step,  so  that 
they  found  a  true  resisting  base  in  the  flat  mass  (see  Fig.  143) 
upon  which  they  rested.  Moreover,  the  architect  provided  for 
the  lateral  tying  of  the  different  sections  of  his  work,  as  Lepsius 
proves   to   us    by   a    partial  section    of   the   pyramid   of   Abousir. 


The  Tomb  under  the  Ancient  Empire. 


2  I 


Two  walls  of  fine  limestone  blocks  inclose  a  fillins;-  in  of  rubble, 
to  which  they  are  bound  by  perpend  stones  which  penetrate  its 
substance.  This  method  of  construction  has  its  faults,  but  it  is 
so  rapid  that  its  employment  is  not  to  be  wondered  at. 


JJ        ZO       JO      *o       So 


Fig.  143.— Section  of  the  Stepped  Pyramid  at  Sakkarah  ;  from  Lepsius.^ 

Do  these  parallel  walls  reach  from  top  to  bottom  ?  A  detail 
discovered  by  Minutoli  would  seem  to  indicate  that  a  base  was 
first  constructed  of  sufficient  extent  for  the  whole  monument.  In 
the  lower  part  of  the  Stepped  Pyramid  Minutoli"^  shows  concave 


Fig,  144. — Construction  of  the  Pyramid  of  Abousir  in  parallel  layers;  transverse  section  in 
perspective  from  the  geometrical  section  of  Lepsius.^ 

courses  of  stone  laid   out  to   the    segment  of  a    circle.      These 
courses  formed  a  kind  of  inverted  vault,  abutting,  at  its  edges,  upon 

^  Fig.  5  of  his  paper,  Ueber  den  Ban  der  Pyramiden. 

2  Voyage  an   Temple  de  Jupiter  Ammo7i  et  dans  la  Haute- Egyte.     (Berlin,  1824, 
4to.  and  folio;  pi.  xxvii.  fig.  3.) 

3  Fig.  8  of  his  paper,  Ueber  den  Ban  der  Pyramiden. 


2  14  A   History  of  Art  in  Ancient  Egypt. 

the  rock.  This  curious  arrangement  should  be  studied  upon  the 
spot  by  some  competent  observer.  As  we  do  not  know  whether 
these  curves  exist  upon  each  face  or  not,  or  whether  they  meet 
each  other  and  penetrate  deeply  into  the  structure  or  not,  we  cannot 
say  what  their  purpose  may  have  been.  But  however  this  may 
be,  they  afford  another  argument  against  the  notion  that  all  the 
great  pyramids  were  built  round  such  a  pyramidoid  core  as  that 
represented  by  our  Fig.  136.  We  fear  that  this  system  must  be 
regarded  merely  as  an  intellectual  plaything.  The  views  of 
Lepsius  as  to  the  enlargement  of  the  pyramid  by  the  addition  of 
parallel  slices  are  worthy  of  more  respect,  and  their  truth  seems  to 
be  demonstrated  in  the  case  of  some  pyramids.  But  these  all 
belong  to   that  category  of  monuments  which  have  subterranean 


s         10         /J        sj/C 
_j 1 1 1 


Fig.  145. — Partial  section  of  the  Stepped  Pyramid  ;  from  Minutoli. 

chambers  only.  We  have  yet  to  learn  that  they  were  ever  made 
use  of  in  those  pyramids  which  inclose  the  mummy-chamber  and 
its  avenues  in  their  own  substance.  Variety  is  universal  in 
that  ,  Egypt  which  has  so  often  been  described  as  the  land  of 
uniformity  and  immobility — no  two  of  the  pyramids  resemble 
each  other  exactly. 

We  have  yet  to  speak  of  two  ancient  monuments  in  which  some 
would  recognize  unfinished  pyramids,  namely,  the  Pyramid  of 
Meidoum  and  the  Mastabat-el-Faraoun.  W^e  do  not  agree  with 
this  opinion,  which  has,  however,  been  lately  put  forward,  so 
far,  at  least,  as  the   former  monument  is  concerned.^     These  two 

'  B^DEKER,    Egypt,    part    i.    1878.     The  pages  dealing   with    tlie    nionuniental 
remains  were  edited  in  great  part  by  Professor  Ebers. 


The  Tomb  under  the  Ancient  Empire. 


215 


sepulchres  seem  to  us  to  represent  a  different  type  of  funerary 
architecture,  a  type  created  by  the  ancient  empire,  and  meriting 
special  notice  at  our  hands. 

The  monument  which  rises  so  conspicuously  from  the  plain  near 
the  village  of  Meidoum  on  the  road  to  the  Fayoum,  is  called  by 
Arabs  the  Haram-el-Kabbab,  or  "the  false  pyramid."  It  is,  in 
fact,  not  so  much  a  pyramid,  strictly  speaking,  as  a  mass  formed 
of  three  square  towers  with  slightly  inclined  sides  superimposed 
one  upon  the  other,  the  second  being  less  in  area  than  the  first, 
and  the  third  than  the  second.  The  remains  of  a  fourth  story 
may  be  distinguished  on  the  summit  of  the  third  ;  some  see  in 
them  the  remains  of  a  small  pyramid;  others  those  of  a  cone. 
Judging  from   the    names   found    in    the    neighbouring  mastabas, 


FiG.  146, — The  Pyramid  of  Meidoum  ;  from  Perrinsj. 


which  were  opened  and  examined   by  Mariette,  this  is  the  tomb 
of  Snefrou  I.,  one  of  the  greatest  kings  of  the  third  dynasty.^ 

The  Mastabat-el-Faraoun  or  "  Seat  of  Pharaoh,"  as  the  Arabs 
call  it,  is  a  huge  rectangular  mass  with  sloping  sides ;  it  is  about 
66  feet  high,  340  long,  and  240  deep.  It  is  oriented  like  the 
pyramids.  It  is  a  royal  tomb  with  internal  arrangements  which 
resemble  those  in  the  pyramid  of  Mycerinus  ;  the  same  sloping 
galleries,  the  same  chambers,  the  same  great  lateral  niches. 
Upon  a  block  lying  at  the  foot  of  the  structure  of  which  it 
had  once  formed  a  part,  Mariette  found  a  quarry-mark  traced  in 
red  ochre  which  seemed  to  him  to  form  part  of  the  name  of 
Ounas,  one  of  the  last  kings  of  the  fifth  dynasty  (Figs.  109 
and  147). 

^    P(?iY74,'-t'  (fans  la  Hai/fc-Egypfe,  vol.  i.  p.  45. 


2l6 


A   History  of  Art  in  Ancient  Egypt. 


Upon  the  platform  of  the  Mastabat-el-Faraoun  certain  blocks 
are  to  be  found  which,  from  their  position,  must  have  been 
bonding-stones.  They  seem  to  hint,  therefore,  either  that  the 
structure  was  never  finished,  or  that  it  has  lost  its  former  crown. 
The  latter  hypothesis  is  the  more  probable.     Among  the  titles  of 


r'' 


r^^^^i^^^r-s^r^.  vcJ5^^--i^^^'%i^F;:^:^:5^ 


-2.  ■-'  -:'"  c.'^«:Tiari=^sSs?^i~i^  ?^  .z^-r'r. 


L--^  -C^*--^ 


Fig.  147. — The  Mastabat-el-Faraouii ;  from  Lepsius. 


people  buried  in  the  necropolis  at  Sakkarah,  we  often  come  upon 
those  of  priests  attached  to  the  service  of  some  monument  with  a 
form  similar  to  that  represented  by  our  Fig.  148.  Who  can  say 
asks  Mariette,  that  it  is  not  the  Mastabat-el-Faraoun  itself?^ 

M.  Mariette  cites,  in  support  of  this 
conjecture,  certain  other  structures  of  a 
similar  character,  such  as  the  large  tomb 
situated  near  the  south-eastern  angle  of 
the  second  pyramid  at  Gizeh,  and  the 
little  monument  which  is  called  the 
Pyramid  of  Righa.  From  these  he 
concludes  that  the  principles  of  the 
mastaba  and  the  pyramid  were  sometimes 
combined  under  the  ancient  empire. 
The  royal  tombs  in  the  Memphite  region  were  not  always 
pyramids,  they  were  sometimes  composed  of  a  mastaba  and  of  one 
or  more  high  square  tower-like  erections  upon  it,  the  whole  ending 
in  one  of  those  small  pyramids  which  we  call  pyramidions.     This 


Fig.    148. — Funerary   monument 
represented  in  the  inscriptions. 


Voyage  dans  la  Hai/te-Egypfe,  vol.  i.  p.  34. 


The  Tomb  under  the  Ancient  Empire.  217 


type  allowed  of  numerous  combinations,  many  of  which  are  to  be 
discovered  in  the  monuments  of  a  later  period. 

The  pyramid  was  employed  as  a  terminal  form  throughout  the 
whole  of  Egyptian  history.  Both  Thebes  and  Abydos  offer  us 
many  examples  of  its  use,  either  in  those  sepulchral  edifices  which 
are  still  extant,  or  in  the  representations  of  them  upon  bas-reliefs. 
But  the  pyramid  properly  speaking  was  confined  to  the  Memphite 
period.  The  princes  of  the  twelfth  dynasty  seem  to  have 
constructed  some  in  the  Fayoum.  The  pyramids  of  Hawara  and 
Illahoon  correspond  to  those  which,  we  are  told,  were  built  in 
connection  with  the  labyrinth  and  upon  the  islands  of  Lake  Moeris 
respectively.  These,  so  far  as  we  can  judge,  were  the  last  of  the 
pyramids.  There  are,  indeed,  in  the  necropolis  of  Thebes,  upon 
the  rocks  of  Drah-abou'1-neggah,  a  few  pyramids  of  crude  brick, 
some  of  which  seem  to  belong  to  Entefs  of  the  eleventh  dynasty  ; 
but  they  are  small  and  carelessly  constructed.^  When  the  art  of 
Egypt  had  arrived  at  its  full  development,  such  purely  geometrical 
forms  would  seem  unworthy  of  its  powers,  as  they  did  not  allow 
of  those  varied  beauties  of  construction  and  decoration  which  its 
architects  had  gradually  mastered. 

The  pyramids  have  never  failed  to  impress  the  imaginations  of 
those  foreign  travellers  who  have  visited  Egypt.  Their  venerable 
antiquity  ;  the  memories,  partly  fable,  partly  history,  which  were 
attached  to  them  by  popular  tradition  ;  their  colossal  mass  and  the 
vast  space  of  ground  which  they  covered,  at  the  very  gates  of  the 
capital  and  upon  the  boundary  between  the  desert  and  the 
cultivated  land,  all  combined  to  heighten  their  effect.  Those 
nations  who  came  under  the  living  influence  of  Egypt  could 
hardly,  then,  escape  from  the  desire  to  imitate  her  pyramids  in 
their  own  manner.  We  shall  find  the  pyramidal  form  employed 
to  crown  buildings  in  Phoenicia,  Judaea,  and  elsewhere.  But  the 
kingdom  of  Ethiopia,  the  southern  annexe  of  Egypt  and  the 
copyist  of  her  civilization,  was  the  chief  reproducer  of  the 
Egyptian  pyramid  as  it  was  created  by  the  kings  of  the  ancient 
empire.  Napata,  Meroe,  and  other  places  have  pyramids  which 
may  be  counted  by  dozens.  Like  those  in  Egypt,  they  are 
the  tombs  of  the   native  monarchs. 

We  shall  not  attempt  any  study  of  these  remains.      Like  all  the 

'   Lepsius,   Deiikmceler,   part  i.   pi.    94.     Rhind,    Thebes^    its    Tombs  and  their 
Tenants,  p.  45.     Mariette,  Voyage  dans  la  Haute-Egypte,  vol.  ii.  i).  80. 
VOL.    I.  F    F 


2i8  A  History  of  Art  in  Ancient  Egypt. 

other  products  of  Ethiopian  art,  they  are  neither  original  nor 
interesting.  The  people  who  inhabited  the  region  which  we  now 
call  by  the  names  of  Nubia  and  the  Soudan,  had,  indeed, 
reconquered  their  political  independence  a  thousand  years  before 
our  era,  but  they  were  not  gifted  by  nature  with  power  to 
assimilate  the  lessons  of  their  former  masters.  Even  durinof  the 
short  period  when  the  Ethiopian  monarchs  reigned  over  a 
divided  and  weakened  Egypt,  Ethiopia  remained  the  clumsy 
pupil  and  imitator  of  the  northern  people.  She  never  learnt  to 
give  to  her  royal  pyramids  the  air  of  grandeur  which  distin- 
guishes the  great  structures  of  Memphis.  The  Ethiopian 
pyramids  were  generally  so  narrow  and  steep  in  slope  that 
their  whole  character  was  different  from  those  of  Egypt.  In 
Egypt  the  base  line  was  always  greater  than  the  vertical  height, 
upon  the  Upper  Nile  the  proportions  were  reversed.^  The 
latter  edifices  thus  lost  some  of  that  appearance  of  indestructible 
solidity  which  is  their  natural  expression.  They  remind  one 
at  once  of  the  obelisk  and  the  pyramid.  Add  to  this  that  an 
unintelligent  taste  overspread  them  with  ill-devised  decoration. 
Thus  the  upper  part  of  their  eastern  faces,  for  they  are 
oriented,  generally  bears  a  false  window  surmounted  by  a 
cornice,  about  as  incongruous  an  ornament  as  could  well  be 
conceived,  and  one  which  expresses  nothing  either  to  the  eye 
or  the  mind. 

We  shall,  therefore,  take  no  further  notice  of  these  more  or 
less  ill-designed  variations  upon  the  type  which  was  created 
by  the  Egyptians  in  the  early  days  of  their  civilization  and  fully 
understood  by  themselves  alone. 

We  must  return,  however,  to  that  type  for  a  moment,  in 
order  to  show,  in  as  few  words  as  possible,  how  far  the  art  of 
workinof  and  fixinof  stone  had  advanced  even  at  the  time  of 
the  first  dynasties. 

The  Great  Pyramid  affords  us  a  curious  example  of  the 
elaborate  precautions  taken  against  the  violation  of  the  royal 
tomb    (Fig.     132).      At   the    point    where    the    ascending    gallery 

'  Thus  the  Great  Pyramid  was  482  feet  high,  while  the  length  of  one  side  at  the 
base  is  764  feet  On  the  other  hand,  the  "  third  pyramid  "  at  Gebel-Barkal  (Napata) 
is  35  feet  square  at  the  base  and  60  feet  high  ;  the  "  fifth  "  is  39  feet  square  at  the 
base  and  nearly  50  feet  high.  Their  proportions  are  not  constant,  but  the  height  of 
the  Nubian  pyramids  is  always  far  greater  than  tlie  length  of  one  side  at  its  base. 


The  Tomb  under  the  Ancient  Empire. 


219 


branched  off  from  that  descending  corridor  which  was  the  only 
entrance  to  the  pyramid,  the  mouth  of  the  former  was  closed 
by  a  block  of  granite  which  exactly  fitted  it.  This  block  was  so 
heavy  and  so  well  adjusted,  that  entrance  could  only  be  obtained 
by  cutting  a  passage  through  the  surrounding  masonry,  which, 
being  of  limestone,  did  not  offer  such  an  unyielding  resistance  to  the 
tools  brought  against  it.  Formerly  the  mouth  of  a  gallery,  which 
seemed  to  be  the  continuation  of  the  entrance  corridor,  remained 
open,  and,  when  followed  to  the  end,  led  to  an  unfinished 
chamber  cut  in  the  rock  at  about  the  level  of  the  Nile.  If 
this  had  been  finished  the  waters  would  perhaps  have  invaded 
it  by  infiltration.  This  seems  to  have  been  intended  by  the 
constructor,  because  Herodotus,  who  no  doubt  thought  the  work 
had    been    completed,   tells    us    of   a   subterranean   conduit  which 


Fig.  149. — Plan  and  elevation  of  a  pyramid  at  Meroe  ;  from  Prisse. 


admitted  the  waters  of  the  Nile.^  The  violaters  of  the  tomb 
would  believe  the  corpse  to  be  in  this  unsuspected  reservoir,  and 
would  search  no  farther,  or  if  they  guessed  the  deception  and 
persevered  till  they  found  the  entrance  to  the  ascending  gallery, 
they  would  find  another  obstacle  to  their  success  which  would 
be  likely  to  arrest  them  longer  than  the  first.  The  upper 
extremity  of  the  great  gallery,  at  which  we  suppose  them  arrived, 
opens  upon  a  small  vestibule  which  would  still  separate  them 
from  the  sarcophagus-chamber  itself.  Four  flat  blocks  of 
granite,  sliding  in  grooves,  masked  the  entrance  to  the  latter ; 
Figs.  150  and  151  show  the  arrangement  of  these  portcullis 
stones.  The  narrow  passage  leading  to  the  discharging  chambers 
above  the  mummy-chamber,  would  be  likely  to  lead  our  sup- 
posed   robbers     into     the    upper    part     of    the    pyramid.       The 

'   Hfrodotus,  ii.  124. 


220 


A   History  of  Art  in  Ancient   Egypt. 


entrance  to  this  passage  is  high  up  in  the  end  wall  of  the  grand 
gallery  ;  it  was  left  open.  The  unbidden  visitors  would  thus 
have  explored  the   interior  of  the  pyramid  high  and  low  without 


/    ^„///  /////  // '//////////////  V/'  ^ 


'Mw<^y' 


^mm 


Fro.  150. — Method   of  closing  a  gallery  by  a  stone   portcullis;  from  the  fouthern  pyramid  of 
Dashour.     Drawn  in  perspective  from  the  plans  and  elevations  of  Perring. 

result,  and  even  supposing  that  they  expended  considerable  time 
and  trouble  in  the  search,  they  might  easily  have  failed  to 
penetrate  into  the   mummy-chamber   itself^ 


Fig.  151. — Portcullis  closed. 

Another  incfenious  arrangement  which  demands  our  notice  is 
that  of  those  discharging  chambers  to  which  we  have  already 
alluded.     These    chambers    were    explored,    not   without   trouble, 

Du  Barry  de  Merval,  Etiides  sur  T Architechire  Egyptiemie,  ])p.  129,  130. 


The  Tomb  under  the  Ancient  Empire. 


22  I 


by  Colonel  Howard  Vyse  and  J.  L.  Perring,  who  at  once  compre- 
hended their  use.  The  roof  of  the  sarcophagus- chamber  consists 
of  nins  slabs  of  fine  red  granite,  like  those  which  form  the  walls 
of  the  same  chamber.  They  are  i8  feet  9  inches  long  and  their 
ends  rest  upon  the  side  walls  of  the  chamber.  In  spite  of  their 
thickness  and  of  the  hard  nature  of  the  rock  of  which  they  are 
composed,  it  was  feared  that  they  might  give  way  under  the 
enormous    weight    of  the    masonry   above,   for  the    floor   of  the 


Fig.  1 52. ^Transverse  section,  in  perspective,  through  the  sarcophagus-chamber  and  the 
discharging  chambers  ;  from  the  elevation  of  Perring. 


chamber  is  still  nearly    340    feet   below  the    actual    apex    of  the 
pyramid.      This  danger  was  met  in  the  fashion  figured  above. 

As  the  structure  grew  above  the  roof  of  the  mummy-chamber, 
five  small  chambers  were  left,  one  above  the  other,  to  a  total 
height  of  56  feet,  which  would  relieve  the  fiat  ceiling  of  the 
mummy-chamber  of  the  weight  to  be  placed  above  it.  The 
first  four  of  these  chambers  were  of  similar  shape  and  had  flat 
roofs,    but    the    roof  of   the    fifth    was    formed    of   sloping    slabs, 


222 


A   History  of  Art  in  Ancient  Egypt. 


meeting  In  a  ridge,  and  giving  the  chamber  a  triangular  section 
(see  Fig.  152).  Thanks  to  this  succession  of  voids  immediately 
over  the  main  chamber,  and  to  the  pointed  arch  which  surmounts 
them,  the  vertical  pressure  of  the  superstructure  is  discharged 
from  the  chamber  itself  and  distributed  over  the  lateral  parts 
of  the  pyramid.  These  precautions  have  been  quite  effectual. 
Not  a  stone  has  been  stirred  either  by  the  inward  thrust  or  by 
the  crushing  of  their  substance  ;  not  a  block  is  out  of  place  but 
those  which  have  been  disturbed  by  the  violence  of  man  ;  and, 
moreover,  the  whole  structure  is  so  well  bonded  and  so  well 
balanced  that  even  his  violent  attacks  have  led  neither  to  dis- 
ruption nor  settlement  in  the  apartment  of  Cheops  or  in  the 
galleries  which  lead  to  it.^ 


Fig.    153. — Longitudinal  section  through  the  hnver  chambers  ;  perspective  after  Pening, 


The  glory  of  the  workmen  who  built  the  Great  Pyramid  is 
the  masonry  of  the  Grand  Gallery,  the  gallery  which  opens 
immediately  into  the  vestibule  of  the  King's  Chamber.  As  this 
corridor  is  28  feet  high  and  7  feet  wide,  the  visitor  can  breathe 
more  freely  than  in  the  low  and  narrow  passages  which  lead  to 

^  The  discovery  of  these  chambers  Avas  interesting  from  another  point  of  view. 
The  name  of  Choufou  was  found  continually  repeated  upon  the  blocks  of  which 
they  are  formed.  It  was  written  in  red  ochre,  and,  in  places,  it  was  upside  down, 
thus  proving  that  it  must  have  been  written  before  the  stones  were  put  in  place. 
It  cannot  therefore  have  been  traced  after  the  tradition  which  assigned  the 
pyramid  to  Cheops,  that  is,  to  Khoufou,  arose  ;  and  so  it  affords  conclusive 
corroboration  of  the  statements  of  Herodotus. 


The  Tomb  under  the  Ancient  Empire.  223 

it,  and  can  examine  at  his  ease  the  beautiful  blocks  of  limestone 
from  Mokattam  of  which  its  polished  -walls  are  composed.  The 
faces  of  these  blocks  have  been  dressed  with  a  care  which  is  not 
to  be  surpassed  even  by  the  most  perfect  examples  of  Hellenic 
architecture  on  the  Acropolis  at  Athens.  The  internal  faces 
must  have  been  worked  with  equal  care.  No  cement  has  been 
employed  in  the  fixing,  and  the  adherence  is  so  perfect  that, 
in  the  words  of  Abd-ul-Latif  "  not  a  needle,  not  even  a  hair, 
can  be  introduced  into  the  joints."  ^  These  joints  are  not  even 
to  be  distinguished  without  careful  examination.  The  roof 
of  this  gallery  is  built  with  no  less  care.'-^  Each  of  the  upper 
courses  is  slightly  set  off  from  the  one  below  it,  so  that  in  time 
they  come  so  near  together  that  the  opening  may  be  closed  by 
a  single  stone,  or  rather,  row  of  stones.  These,  being  held 
between  the  two  upper  courses  of  a  quasi  vault,  play  the  part 
of  key  stones.  This  method  of  vaulting  has  been  employed  in 
other  parts  of  the  pyramid,  especially  in  what  is  called  the  Queeiis 
Chamber,  which  is  almost  directly  beneath  the  king's,  or  sarcophagus- 
chamber.  The  same  care  is  conspicuous  in  those  linings  of  red 
granite  which  form  the  walls  of  the  two  chambers.  Even  the 
fine  limestone  used  for  the  walls  of  the  Grand  Gallery  was  not 
considered  rich  and  solid  enough  for  the  walls  of  the  apartment 
in  which  the  prince  in  whose  honour  the  whole  of  the  colossal 
edifice  was  reared  would  repose ;  and  it  was  determined  to  use 
the  richest  and  most  costly  material  of  which  the  Egyptian 
architect  could  dispose.^  The  plain  sarcophagus,  without  either 
inscription  or  ornament,  which  is  still  in  the  King's  Chamber,  is 
also  of  red  granite. 

The  external  casing  of  the  pyramid  has  entirely  disappeared,  as 
we   have   already   said.     On   account  of  their  moderate  size   the 

^  This  is  no  exaggeration.  Jomard  expresses  himself  to  the  same  effect  ahiiost 
in  the  same  terms.     {^Description  de  I' Egypfe,  vol.  v.  ]).  628.) 

^  The  extremity  of  this  gallery  appears  on  the  right  of  Fig.  152. 

^  The  presence  of  this  lining  in  the  "  Queen's  Chamber"  also  led  to  its  being 
dubbed  a  funerary  chamber,  for  no  trace  of  a  sarcophagus  was  found  in  it.  If  we 
had  any  reason  to  believe  that  the  pyramid  was  built  in  successive  wedges,  we  should 
look  upon  this  as  a  provisional  chamber,  made  before  it  was  certain  that  the 
pyramid  would  attain  its  present  dimensions.  As  the  work  went  on,  it  would  be 
decided  that  another,  larger,  and  better  defended  chamber  should  be  built.  In  this 
case  the  first  may  never  liave  been  used,  and  may  always  have  been  as  empty  as 
it  is  now. 


224  ^   History  of  Art  in  Ancient  Egypt. 

stones  of  which  it  was  composed  would  seem  to  be  especially  well 
fitted  for  use  in  building"  those  great  cities  which,  after  the  collapse 
of  the  ancient  civilization,  succeeded  each  other,  under  different 
names,  in  the  neighbourhood  of  the  Memphite  necropolis.  This 
casinof  seems  to  have  been  made  of  more  than  one  kind  of  stone,  if 
we  may  believe  an  ancient  text  which  has  been  interpreted  by 
Letronne  with  the  skill  and  sagacity  of  which  he  has  given  so 
many  proofs.^ 

The  author,  named  Philo,  of  a  treatise  upon  the  Seven 
Wonders  of  the  World,  tells  us  that  the  Egyptians  employed  upon 
this  work  "  the  most  brilliant  and  varied  stones,  which  were 
carefully  fixed."  He  mentions  as  contributing  to  the  splendid 
result  white  marble,  basalt,  porphyry,  and  a  green  breccia  from 
Arabia,  which  must  have  been  what  is  called  verde  antique.  And 
as  for  his  white  marble,  it  must  have  been  the  white  limestone 
from  Mokattam,  which,  in  its  best  strata,  is  almost  as  white  and 
fine  in  grain  as  marble.  Marble,  properly  speaking,  was  only 
introduced  into  Egypt  by  the  Greeks,  and  that  in  very  small 
quantities,  for  the  use  of  sculptors.  Philo  says  nothing  of  granite, 
but  its  use  was  so  general  that  it  must  have  found  a  place  in  the 
scheme  of  decoration.^ 

The  various  kinds  of  stone  must  have  been  so  placed  as  to  form 
zones,  and  perhaps  patterns,  of  different  colours,  white,  red,  black, 
rose,  green,  and  so  on.  To  form  an  idea  of  the  effect  we  must 
think  of  Giotto's  campanile  at  Florence  and  various  other  Italian 
buildino-s  of  the  same  kind. 

It  has  been  questioned  whether  the  testimony  of  this  Philo  is 
to  be  depended  upon,  as  few  of  those  who  have  busied  themselves 
with  the  pyramids  seem  to  have  laid  much  stress  upon  it.  It 
seems  to  us  to  be  worthy  of  great  respect.     We   do  not  know 

^  These  observations  are  to  be  found  in  one  of  the  early  works  of  Letronne. 
Their  presence  is  in  no  way  hinted  at  by  the  title,  which  is  :  Recherches  Geographiques 
et  Critiques  sur  le  Livre  '■  De  mensura  orbis  terrce^  (8vo.  1844).  The  treatise,  Ilept 
TMv  eiTTa  Oeafidrow,  may  have  been  written  either  by  Philo  of  Heraclea  or  Philo  of 
Byzantium.  They  both  belonged  to  the  third  century  before  our  era,  but  the 
bombastic  style  and  numerous  errors  incline  us  to  believe  that  the  little  work  must 
have  been  from  the  pen  of  some  unknown  rhetorician  of  a  later  date. 

2  These  are  the  words  of  Philo,  which  we  have  translated  rather  freely : — IIoi/ctAat 

8e  Kol  7rop(f)vpai  Xl6wv  (^j'crcts  dA.A,r;A.ats  €7ri8c8d/x€ra(.,  kol  ra  jxev  kcmv  y  Trerpa  XevKr,  Kat 
liapfjiapLTYj';'  rrj  Se  AWlottik-tj  kol  fxeXatva  /cat  /xera  Tavrrjv  6  KaAoi'/x,0'Os  aljxaTLTrjs  XiOo^' 
etra  ttoikiAos  Kat  Sta^^Xcopos  airb  ti}s   Apa/3ta?  KeKO/xtcr/xeVo?,  p.   2,259?  A. 


The  Tomb  under  the  Ancient  Empire.  225 

when  Philo  lived,  but  we  know  that  the  casino-  of  the  pyramid  was 
still  in  place,  at  least  in  part,  during  the  ^liddle  Ages,  because  in 
the  time  of  Abd-ul-Latif  it  had  almost  its  original  height,  and  its 
ascent  was  still  ver}'  difficult.^  On  the  other  hand  we  have  proofs 
that,  although  the  author  of  the  Seven  Wonders  of  the  World  may 
have  written  more  in  the  tone  of  a  rhetorician  than  of  an  eye- 
witness of  the  wonders  which  he  describes,  he  took  some  of  his 
information  from  excellent  sources.  In  fact  with  the  exception  of 
Pliny,  he  is  the  only  ancient  writer  who  gives  us  an  approxi- 
mately true  statement  of  the  length  of  the  base  line  of  Cheops' 
Pyramid.  While  the  measurements  of  other  writers  are  very  far 
from  accurate,  the  figure  given  by  Philo  is  only  16  feet  6  inches  in 
excess  of  the  truth.      The  idea  of  decorating  such  an  expanse  of 


^  According  to  the  calculations  of  Letronne.  the  Great  Pyramid  must  have  been 
482  feet  high  when  it  was  complete.  In  the  time  of  Diodorus  it  was  slightly  over 
480  feet;  in  that  of  Abd-ul-Latif  it  measured  477  feet  3  inches.  In  1795  it  ^^'^^ 
only  456  feet  and  a  few  inches,  so  that  it  lost  about  24  feet  in  the  course  of  eighteen 
centuries.  This  lowering  of  the  summit  was  mainly  caused  by  the  destruction  and 
removal  of  the  outer  casing.  Since  it  disappeared  the  Arabs  have  been  in  the 
habit  of  loosening  the  stones  on  the  top  and  launching  them  down  the  sides  for  the 
amusement  of  travellers  ;  the  smooth  casing  alone  could  prevent  such  outrage  as 
this.  The  common  idea  that  the  PvTamid  of  Cheops  is  the  highest  building  in  the 
world  is  erroneous.  Even  if  we  take  its  height  when  complete,  it  is  surpassed  by 
at  least  two  modern  buildings,  as  may  be  seen  by  the  following  table  of  the 
most  lofty  buildings  now  existing  : — 

Feet. 

Spires  of  Cologne  Cathedral 533 

Fleche  of  the  Cathedral  at  Rouen 500 

Spire  of  St.  Nicholas,  Hamburg 480 

Dome  of  St.  Peter's,  Rome 476 

Spire  of  Strasbourg  Cathedral 473 

Pyramid  of  Cheops 456 

Spire  of  St.  Stephen's,  Vienna 450 

Spire  of  St.  Martin's,  Land  shut 443 

Spire  of  the  Cathedral  of  Freiburg,  Breisgau 417 

Spire  of  Antwerp  Cathedral,  not  including  the  cross 411 

Spire  of  Salisbury  Cathedral 404 

Dome  of  Cathedral  at  Florence 396 

Dome  of  St  Paul's,  London 371 

Fleche  of  Milan  Cathedral 363 

Tower  of  Magdeburg  Cathedral 344 

Victoria  Tower,  Westminster 336 

Rathhaus  Tower,  Berlin 293 

Spire  of  Trinity  Church,  New  York 287 

Pantheon,  Paris 266 

Towers  of  Notre  Dame,  Paris 226. 

VOL.    I.  G    G 


226  A   History  of  Art  in  Ancient  Egypt. 

surface  with  varied  colour  was  quite  in  accordance  with  Egyptian 
taste.  They  loved  polychromatic  ornaments  ;  they  covered  every 
available  surface  with  the  gayest  hues ;  they  delighted  in  the 
juxtaposition  of  the  most  brilliant  tones.  They  could  hardly 
think  of  covering  such  an  immense  surface  with  paint,  and  as  it 
was  necessary,  in  any  case,  to  cover  it  with  a  smooth  casing,  it 
would  be  no  more  difficult  to  employ  many  kinds  of  stone  than 
one.  They  would  thus  obtain  a  kind  of  gigantic  mosaic  which 
may  perhaps  have  been  heightened  in  effect  by  the  use  of  gold. 
We  know  that  the  pyramidion  of  an  obelisk  was  frequently  gilded, 
and  it  is  probable  enough  that  similar  means  were  sometimes 
taken,  in  the  case  of  the  more  magnificent  and  carefully  finished 
pyramids,  to  draw  the  eye  to  their  topmost  stone  and  thus  to  add 
to  the  impression  made  by  their  height.  No  more  fitting  adorn- 
ment could  be  imagined  for  the  sharp  peak  of  a  pyramid  rising 
nearly  five  hundred  feet  into  the  pure  blue  of  an  Egyptian  sky. 

But  this  is  a  conjecture  which  can  never  be  verified.  Even  if 
the  topmost  stone  were  still  in  place  upon  any  of  the  pyramids  it 
would,  after  all  these  ages,  have  lost  all  traces  of  gilding ;  but  the 
whole  of  those  edifices  have  their  apex  more  or  less  truncated. 
Even  before  our  era,  Diodorus  found  the  Great  Pyramid  crowned 
by  a  plateau  six  cubits  square. 

It  has  sometimes  been  supposed  that  the  pyramids,  when 
complete,  were  terminated  by  such  a  plateau  as  that  described  by 
Diodorus,  and  that  it  bore  a  statue  of  the  king  whose  mummy 
rested  below.  This  hypothesis  is  founded  upon  the  passage  of 
Herodotus  which  treats  of  the  Lake  Mceris.  "  There  are,"  he 
says,  "  in  the  middle  of  the  lake,  two  pyramids,  each  fifty  fathoms 

high    {309    feet) each    of    them    is    .'surmounted    by  a 

colossal  stone  statue  seated  upon  a  throne."^  Herodotus  insists 
so  often  upon  having  seen  the  Labyrinth  and  Lake  Mceris  with 
his  own  eyes,  that  we  cannot  affect  to  doubt  his  assertions  ;  we 
shall  therefore  confine  ourselves  to  a  few  observations  upon 
them. 

In  the  descriptions  which  he  gives  of  the  three  great  pyramids, 
and  among  his  comments  upon  the  methods  employed  in  their 
construction,  Herodotus  does  not  say  a  word  which  can  be 
construed  into  the  most  distant  allusion  to  statues  upon  their 
summits.  If  he  had  seen  colossi  perched  upon  those  lofty 
1  Diodorus,  i.  63,  64.  -  Herodotus,  ii.  49. 


The  Tomb  under  the  Ancient  Empire.  227 

pedestals,  or  if  he  had  heard  from  his  dragomans  — ■  whose 
exaggerations  he  has  elsewhere  so  naively  reproduced — that  they 
had  formerly  existed,  would  he  not  have  made  some  allusion  to 
them  in  that  passage,  at  least,  where  he  explains  how  they  raised 
such  huge  stones  to  so  great  a  height,  and  describes  the  successive 
stages  in  the  construction  of  a  pyramid  ?  ^  Would  he  not  have 
found  room,  in  the  elaborate  antithetical  passage  in  which  he 
contrasts  the  virtues  of  Mycerinus  with  the  imaginary  wickedness 
of  Cheops  and  Chephren,  for  moral  and  critical  reflections  called 
up  by  the  sight  of  their  statues  upon  their  respective  pyramids  ; 
still  more  if  one  of  them  had  happened  to  be  missing  ?  Would 
he  not  have  attempted,  through  some  popular  tradition,  to  have  ac- 
counted for  the  presence  of  one  statue  and  the  absence  of  another  ? 
It  is  evident,  therefore,  that  Herodotus  neither  saw  any  statues  upon 
the  Pyramids  of  Memphis  nor  had  he  any  reason  to  suppose  those 
structures  had  ever  been  crowned  in  such  a  fashion.  He  lays 
stress  upon  the  seated  statues  of  the  pyramids  in  Lake  Mneris 
because  they  were  new  to  him,  because  he  had  seen  nothing  of  the 
same  kind  in  the  neighbourhood  of  the  ancient"  capital. 

Unless  we  are  very  much  mistaken,  this  superposition  of  a 
colossus  upon  a  pyramid  was  a  novelty  devised  by  the  architects 
of  the  middle  empire,  when,  under  the  Ousourtesens  and 
Amenemhats,  it  was  proposed  to  revive  the  pyramidal  form  of 
tomb  with  which  the  early  Pharaohs  had  obtained  such  imposing 
results.  Although  most  conservative  on  the  whole,  the  art  of 
Egypt  attempted,  at  each  period  of  renascence,  to  introduce  new 
combinations  into  the  details,  at  least,  of  the  ancient  forms,  and 
this  was  one  of  the  number. 

Another  innovation  of  the  same  kind  is  to  be  found  in  the 
decoration  which  covered,  again  according  to  Herodotus,^  another 

^  M.  Maspero  has  given  in  the  Annuaire  de  F Association  pour  F Encouragement  des 
Etudes  Grecques  and  elsewhere,  several  extracts  from  a  commentary  upon  the  second 
book  of  Herodotus,  which  we  should  like  to  see  published  in  its  entirety.  We  may 
point  out  more  particularly  his  remarks  upon  the  text  of  the  Greek  historian  in  the 
matter  of  the  1,600  talents  of  silver  which,  he  says,  was  the  value  of  the  onions, 
radishes,  and  garlic  consumed  by  the  workmen  employed  upon  the  Great  Pyramid 
(ii.  125).  He  has  no  difficulty  in  showing  that  Herodotus  made  a  mistake,  for 
which  he  gives  an  ingenious  and  probable  explanation.      {Annuaire  de  1875,  p.  16.) 

2  Herodotus,  ii.  148.  Diodorus  (1.  89)  speaks  of  the  same  pyramid,  and 
Strabo,  who  also  appears  to  have  seen  it,  asserts  its  funerary  character  (p.  11 65,  c). 
He  says  it  was  four  plethra  (393  feet)  both  in  width  and  height.    This  last  dimension 


228  A   History  of  Art  in  Ancient  Egypt. 

pyramid  constructed  at  about  the  same  time,  namely,  that  which 

formed  one  side  of  the   Labyrinth..     "  It  had,"  says  the  historian, 

"  forty  fathoms,  and  it  was   sculptured  with  animals  of  large  size. 

The  entrance  was  by  a  subterranean  passage."     From  the  Greek 

word  used   {cyytyXu-rrTai)   we   see  that    Herodotus   means   that  the 

faces,   or  perhaps  only  the  principal    face,   of  this  pyramid  about 

two  hundred  and   fifty  feet  high,    were    covered   with   bas-reliefs. 

There  is  in  Egypt  no  other  example  of  a  pyramid  so  decorated. 

The    architectural    works    of    this    period    have    almost    entirely 

vanished,    but   we   may,    perhaps,    look   upon   it    as    one    of   their 

characteristics  that  the  bareness  which    they  had    inherited    from 

the  early  creators  of  Egyptian  art,  was  relieved  and  adorned  by 

the  intervention  of  the  sculptor. 

It  was  the  desire  for  such  ornament   that  made  them   convert 

,their    pyramids    into    gigantic    pedestals    for    statues.      According 

to    all   the   analogies   afforded   by   later  ages,   these    statues   must 

have  been  those  of  the  princes  who  built  the  pyramids  in  question. 

We  have  no  reason  to  suppose  that  any  of  the  kings  of  the  first 

six  dynasties  erected  any  colossal   figures  like  those  which  were 

set    up    in   such    numbers    by    the    Theban    dynasties ;    with    the 

single  exception  of  the  Sphinx,  none  of  the   statues  left  to  us  by 

the    ancient   empire   greatly  exceed   the   natural    size.      But   it    is 

evident  that  such   figures  as  would  be  fit  to  crown  the  pyramids 

of  Cheops  and   Chephren  would  have  to  be  of  extravagant  size 

even   if  no   more   than   their   oreneral   outlines   were   to   be   visible 

from  below.      Seen   from   a  point   nearly   500  feet  below,   and    in 

consequence    of   the    inclination    of   the    pyramid    faces,    at    some 

considerable    distance    laterally,    even    a    statue    fifty    feet    high, 

like   the  two  colossi  of  Amenophis    HI.  on  the  plain  of  Thebes, 

would    appear    small   enough   to   a  spectator.      Its  artistic   results 

would    be    very    slender,    and     yet    its    erection     would    require 

prodigious  mechanical   efforts.      It    would    have    required    all    the 

multitudes   of  labourers,   the    patience,   and    the    time,   which   the 

Egyptians   alone  dared  to  expend   upon   their    monuments.      But 

perhaps  it   may   be   said   that   these   colossi  were  statues  built-up 

of   comparatively    small    stones.      To    this    we   must    answer    that 

every  colossus    as    yet   discovered    in    Egypt    is  a    monolith.      A 

statue,    of   whatever    size,    made    in   different    pieces   would    form 

is  obviously  exaggerated,  because  in  all  the  Egyptian  pyramids  that  are  known  to  us 
the  shortest  diameter  of  the  base  is  far  in  excess  of  the  height. 


The  Tomb  under  the  Ancient  Empire.  2  2< 


an  exception  to  the  whole  practice  of  Egyptian  sculpture  as 
we  know  it.  Until  such  works  are  proved  to  exist  we  decline 
to  believe  in  them. 

The  problem  was  a  much  simpler  one  in  the  cases  of  the  pyra- 
mids in  Lake  Moeris.  They  were  not  nearly  so  lofty.  Accord- 
ing to  Herodotus  they  were  about  309  feet  high,  doubtless 
including  their  statues.  Situated  as  they  were  in  the  middle  of 
the  lake,  Herodotus  could  not  himself  have  measured  them, 
and  his  statement  that  they  sank  as  far  below  the  level  of  the 
water  as  they  rose  above  it  is  an  obvious  exaggeration.  When 
the  bed  of  the  lake  was  formed,  two  masses  of  rock  were  no  doubt 
reserved,  as  in  the  cases  of  the  other  pyramids,  to  form  the 
core  of  the  projected  edifices,  and  therefore  it  is  likely  enough 
that  the  lowest  courses  of  the  constructions  themselves  dipped 
but  little  below  the  surface  of  the  lake.^  In  his  amazement  at  the 
scale  upon  which  the  Egyptian  buildings  were  conceived, 
Herodotus  has  too  often  attributed  excessive  dimensions  to 
them  ;  thus  he  says  that  the  height  of  the  Great  Pyramid  was 
eight  plethra,  or  about  820  feet,  nearly  340  feet  in  excess  of 
the  truth.  It  is  therefore  probable  that  the  figures  which  he 
gives  for  the  lake  pyramids  are  also  exaggerated.  These 
pyramids  were,  on  account  of  their  comparatively  modest  dimen- 
sions, much  better  adapted  to  the  ideas  of  the  Ousourtesens 
and  Amenemhats  than  the  gigantic  piles  of  Cheops,  Chephren, 
and  Mycerinus. 

Finally  there  is  not  a  text  to  be  found,  outside  the  pages  of 
Herodotus,  which  mentions  pyramids  surmounted  by  statues, 
and  upon  none  of  those  monuments  which  in  one  way  or  another 
bear  representations  of  the  pyramids  are  they  shown  in  any  other 
way  than  with  pointed  summits.  Thus  do  we  find  them  in  the 
papyri,    upon    those    steles    of    the    Memphite    necropolis    which 

^  If  the  passage  in  which  Herodotus  makes  the  statement  here  referred  to  be 
taken  in  connection  with  the  remarks  of  Diodorus,  a  probable  explanation  of  the 
old  historian's  assertion  may  be  arrived  at.     Diodorus   says   that  the  king  opi'rrwv 

TavT-qv  {XLfjLvrjv  sub.)  KareXiTrev  iv  [xio-rj  to'ttoi',  iv  w  rdcfiov  WKoSofiTjcre  kol  t)vo  7ri;pa/Atoa?, 
Tr]v  jjikv  kavTov,  rrjv  8e  t-^s  yvvatKos,  oraStaia?  to  tJi/^o?.  By  this  it  would  appear  that,  ill 
excavating  the  bed,  or  a  part  of  the  bed,  of  the  famous  lake,  a  mass  of  earth  was 
left  in  order  to  bear  future  witness  to  the  depth  of  the  excavation  and  the  general 
magnitude  of  the  work.  This  mass  would  probably  be  reveted  with  stone,  and,  in 
order  that  even  when  surrounded  and  almost  hidden  by  water,  its  significance  should 
not  be  lost,  the  pyramids  raised  upon  it  were  made  exactly-  equal  to  it  in  height. 
—Ed. 


230 


A  History  of  Art  in  Ancient  Egypt. 


commemorate  the  j^riests  devoted  to  their  service,  and  in  those 
tombs  at  Memphis,  Abydos,  and  Thebes  where  the  pyramid, 
placed  upon  rectangidar  figures  of  various  heights,  is  used  as  a 
terminal  element.  Neither  in  the  small  number  of  pyramids 
which  have  come  down  to  us  comparatively  intact,  nor  in  those 
which  are  represented  in  reliefs,  is  there  the  smallest  sign  of  a  trun- 
cated summit  or  of  any  platform  which  could  by  any  possibility 
have  borne  a  statue. 

We  may  say  the  same  of  those  small  pyramidions  which  have 
been  found  in  such  great  numbers  in  tombs  and  which  fill  our 
museums.      It  is   Well   known   that   these  are   votive   offerings   in 


Fig.  154. — Pyramidlou  ;   Louvre. 

connection  with  the  worship  of  the  sun.  "  The  principal  figure," 
says  M.  de  Rouge,  "  is  generally  shown  in  a  posture  of  adoration, 
with  his  face  turned  to  the  sun.  On  his  left  hand  is  the  invocation 
to  the  rising,  and  on  his  right  that  to  the  setting  sun.  These 
arrangements  are  modified  in  various  ways,  but  they  are  always 
upon  the  same  genera  lines  as  the  orientation  of  the  tombs 
themselves."  ^  These  minute  pyramids  also  end  in  a  point 
whether  they  be  of  basalt,  granite,  or  calcareous  stone,  and  it  is 
natural  that  we  should  look  upon  them  as  the  faithful  reproductions 

^  Notice  sonunaire  des    Alonuments  Egyptiens  exposes  dans  Ics  Galeries  du  Loui've 
(4th  edition,  1865,  p.  56). 


The  Tomb  under  the  Anch^nt  Empire.  231 

upon  a  small  scale  of  those  great  funerary  monuments  which 
furnished  a  type,  consecrated  by  the  most  venerable  of  the  national 
traditions,  of  that  structure  facing  the  four  cardinal  points  which 
we  may  call  the  normal  Egyptian  tomb. 

We  may  believe,  then,  that  the  pyramid  of  the  ancient  empire 
terminated  in  a  pyramidion.  This  apex  once  fixed  in  place,  the 
workmen  charged  with  the  final  completion  of  the  edifice  worked 
downwards  from  one  course  to  another,  covering  the  immense 
steps  which  each  face  displayed  five  or  six  thousand  years  ago 
and  now  displays  again,  with  the  final  casing  which  protected 
ihem  for  so  many  centuries.  Even  Herodotus  saw  that  this  must 
have  been  the  method  of  completion.^  Any  other  way  of 
proceeding  would  have  been  too  dangerous  after  the  slope  of  the 
sides  had  been  made  smooth  and  continuous  by  the  completion  of 
the  casing  of  polished  granite.  Workmen  could  only  have  kept 
their  footing  upon  such  a  surface,  with  its  51  or  52  degrees  of 
elevation,  by  means  of  a  complicated  arrangement  of  ropes  and 
ladders.  And  again,  points  of  resistance  could  not  have  been 
obtained  for  the  elevation  of  the  materials  to  ever  increasing 
heights  without  cuttincj  or  leavino-  holes  in  the  casino^,  which 
would  afterwards  have  to  be  filled  up.  These  difficulties  would 
have  unnecessarily  complicated  an  operation  which  was  a  simple 
matter  when  begun  from  the  top.  The  masons  could  then  make 
use  of  the  steps  for  their  own  locomotion,  and  when  the  stones 
were  too  large  to  be  lifted  from  hand  to  hand,  nothing  could 
be  easier  than  to  fix  windlasses  by  which  the  largest  blocks 
could  be  raised  with  facility. 

As  the  workmen  approached  the  base  they  left  above  them  an 
ever  increasing  extent  of  polished  surface,  sloping  at  such  an  angle 
that  no  foot  could  rest  upon  it,  and  forming  the  only  safeguard 
against  the  degradation  of  the  pyramid  by  removing  its  copestone 
or  its  violation  by  breaking  into  the  passages  which  led  to  the 
mummy-chamber.  The  casing  gave  to  the  pyramid  those 
continuous  lines  which  were  necessary  to  make  its  beauty  complete, 
and,  if  the  materials  employed  were  varied  in  the  way  suggested, 
it  furnished  colour  effects  which  had  their  beauty  also.  But, 
above  all,  it  was  a  protection,  a  defensive  armour.  So  long  as 
the  pyramid  preserved   its  cuirass  intact,  it  was  difficult  for  those 

'¥lt€TroLrj6rj    owr   ra    di  o'jrara    avTrj<i    TrpfnTa,    /xctu    8e    to.    CTro/xera    tovtwv    i$e7roLevv 
.    .    .   (ii.   ^2s). 


232  A    History  of  Art  in  Ancient  Egypt. 

who  meditated  violence  to  know  where  to  betjin  their  attack. 
But  this  obstacle  once  pierced  it  was  comparatively  easy  to  learn 
all  the  secrets  of  the  building.  The  inner  mass  was  much  less 
carefully  built  than  the  casing  ;  the  joints  were  comparatively  open, 
and  the  stones  were  soft  and  easily  cut.  Hence  we  see  that  some 
pyramids,  especially  those  which  were  built  of  bricks,  have  been 
reduced  by  the  action  of  time  into  heaps  of  debris,  in  which  the 
pyramidal  form  is  hardly  to  be  recognised. 

Philo,  who  seems  to  be  so  well  informed,  tells  us  with  what 
extreme  care  the  casing  was  put  in  place.  "  The  whole  work," 
he  says,  "  is  so  well  adjusted,  and  so  thoroughly  polished,  that  the 
whole  envelope  seems  but  one  block  of  stone."  ^  The  pyramid  of 
Cheops  has  been  entirely  despoiled  of  its  outer  covering,  and  it  is 
to  that  of  Mycerinus  that  we  must  now  turn  if  we  wish  to  have 
some  idea  of  the  care  with  which  the  work  was  done.  The  lower 
part  of  this  pyramid  is  still  covered  with  long  blocks  of  the  finest 
granite,  fixed  and  polished  in  the  most  perfect  manner.  At  the 
foot  of  the  Great  Pyramid  several  blocks  have  been  found  which 
seem  to  have  formed  part  of  the  casing  of  that  edifice.-  They 
are  trapezoidal  in  form,  and  they  show,  as  Letronne  ^  long  ago 
remarked,  that  the  casing  stones  were  placed  one  upon  another, 
and  adjusted  by  their  external  faces.  They  were  not,  as  was  at 
first  supposed,  sunk  into  the  upper  face  of  the  course  below  by 
mortices  which  would  correspond  to  the  trench  in  the  living  rock 
in  which  the  first  course  was  fixed.  As  to  whether  the  external 
faces  of  these  blocks  were  dressed  to  the  required  angle  before 
they  left  the  quarry,  or  whether  the  work  were  done  after  they 
were  in  place  we  cannot  say  with  any  certainty,  but  it  is  most 
likely  that  the  methods  of  proceeding  changed  with  the  progress 
of  time  and  the  succession  of  architects.  In  such  a  matter  we 
should   find,   if  we  entered   into  details,   diversity  similar  to  that 

^vvapfiov  8e  kol  KaTc^ea/xevov  to  ttSv  epyov,  wcrre  Sokciv  okov  tov  KaTacrKevdcrfj.aTO<i 
fitav  eTvac  TTcVpas  av/jLcjiVLav,  p.  2259,  A.  So,  too,  the  elder  Pliny,  though  with 
rather  less  precision:  "Est  autem  saxo  natural!  elaborata  et  lubrica "  [IVaf. 
Hist,  xxxvi.   12). 

"  According  to  Jomard,  the  casing  stones  of  the  Great  Pyramid  were  "a  compact 
grey  limestone,  harder  and  more  homogeneous  than  those  of  the  body  of  the 
building "  (^Description  de  V Egypte,  t.  v.  p.  640)  ;  but  according  to  Philo,  this 
casing  was  formed,  as  we  have  already  said,  of  various  materials,  so  we  need  feel 
no  surprise  if  blocks  of  granite  or  other  rock  are  shown  to  have  formed  part  of  it. 

"  Journal  des  Savants,  August,   1841. 


The  Tomb  under  the  Ancient  Empire. 


^33 


which  we  have  already  shown  to  have  characterized  the  forms  of 
the  pyramids,  their  internal  arrangements,  and  the  materials  of 
which  they  were  composed. 

Thus  some  triangular  prisms  of  granite  have  been  found  at  the 
foot  of  the  pyramid  of  Chephren,  which  seem  to  have  formed  part 
of  its  lower  casing.^  Such  a  section  seems,  upon  paper,  the 
simplest  that  could  be  adopted  for  the  filling  in  of  the  angle 
between  two  of  the  steps,  but  it  is  far  inferior  in  solidity  to  the 
trapezoidal  section.  The  prisms  had  no  alliance  one  with  another  ; 
they  had  to  depend  for  their  security  entirely  upon  their  adherence 
to  the  faces  of  the  graded  core,  so  that  they  could  easily  be  carried 
off,  or  become  dislocated  from  natural  causes.  This  system,  unlike 
the  first  described,  did  not  give  a  homogeneous  envelope  with  a 
thickness  of  its  own,  and  partly  independent  of  the  monument 
which  it  protected.^ 


i^sS^SCas^ 


iL^iliVWi. 


Fig.  155. — The  casing  of  the  pyramids  ;  drawn  in  perspective  from  the  elevation  of  Perring. 

The  casing  of  the  Second  Pyramid,  moreover,  does  not  seem  to 
have  been  carried  out  on  the  same  principle  from  top  to  bottom. 


*  B.^DEKER,  Egypt,  part  i.  p.  338  (ed.  of  1S78).  Herodotus  (ii.  127)  says  that 
the  first  course  of  the  Great  Pyramid  was  built  of  a  parti-coloured  Ethiopian  stone 
({iTroSet/Aa?  tov  ivpitiTov  So/xor  XlOov  AWcottlkov  ttolklXov).  By  Ethiopian  stone  we  must 
understand,  as  several  illustrations  prove,  the  granite  of  Sytne.  The  Greek 
historian  seems  to  have  thought  that  the  whole  of  the  first  course,  throughout  the 
thickness  of  the  pyramid,  was  of  this  stone.  His  mistake  was  a  natural  one.  In 
his  time  the  pyramid  was  in  a  good  state  of  preservation,  and  he  never  thought  of 
asking  whether  or  no  the  core  was  of  the  same  material  as  the  outer  case. 

2  On  the  other  hand,  tliese  awkwardly  shaped  prisms  offered  less  inducement  to 
those  who  looked  upon  the  pyramids  as  open  quarries  than  the  easily  squared  blocks 
of  Cheops,  while  their  position  in  the  angles  of  the  internal  masonry  enabled  them 
to  keep  their  places  independently  of  the  lower  courses  of  the  casing. — Ed. 
VOL.     I.  H    H 


234  A   History  of  Art  in  Ancient  Egypt. 

The  upper  part,  which  still  remains  in  place,  is  composed  of  a 
hard  cement  formed  of  chalk,  gypsum,  and  pieces  of  burnt  brick. 
They  may  have  wished  to  obtain  the  parti-coloured  effect  of  which 
Philo  speaks,  by  making  simultaneous  use  of  granite  and  concrete, 
and  it  is  quite  possible  that  yet  other  materials  entered  into  the 
composition  of  the  casing.^ 

In  other  pyramids  we  find  different  combinations  again.  In 
the  double-sloped  erection  at  Dashour,  the  courses  of  casing 
stones  are  vertical  instead  of  horizontal,-  while  a  brick  pyramid 
- — the  most  northern — in  the  same  locality,  was  covered  with 
slabs  of  limestone,  fixfed,  no  doubt,  with  mortar. 

Sometimes  we  find  the  revetment  in  a  state  of  semi-completion  ; 
the  blocks  in  place,  and  cut  to  the  proper  angle,  but  without  their 
final  polish.  Such  is  the  case  with  the  Second  Pyramid,  upon 
which  blocks  of  granite  are  to  be  found  which  are  still  rough  in 
face.  It  would  seem  that  the  patience  required  for  the  minute 
completion  of  such  a  terribly  long  and  tedious  piece  of  work  was 
not  forthcoming.  But  we  ought  in  fact  to  be  surprised,  not  so  much 
at  the  unfinished  state  of  a  pyramid  here  and  there,  but  rather  that 
they  should  ever  have  been  completed. 

The  variety  which  is  so  conspicuous  in  the  architectural  con- 
struction of  the  pyramids  is  also  to  be  found  in  their  epigraphy. 
The  first  explorers  of  the  Pyramids  of  Gizeh  were  surprised  at  the 
absence  of  all  inscriptions  beyond  the  masons'  marks  ;  the  silence 
of  those  enormous  structures  seemed  amazing  ;  but  soon  Colonel 
Vyse  discovered  in  the  pyramid  of  Mycerinus  the  sarcophagus  of 
that  king,  and  the  mummy  case,  now  in  the  British  Museum, 
which  bears  an  inscription  of  some  length.  Recent  discoveries, 
too,  of  which  full  details  are  yet  wanting,  prove  that  some  of  the 
pyramids  contained  long  texts,  which  contain  the  names  of  kings 
and  other  information  which  is  of  sfreat  imoortance  to  the  historian 
of  the  Egyptian  religion.  In  1879  and  1880,  Mariette  caused 
three  pyramids  at  Sakkarah  to  be  opened,  which   until   then   had 

'  The  determination  to  use  a  concrete  such  as  that  described  affords  a  good 
reason  for  the  prismatic  shape  of  the  granite  blocks  used  in  the  lower  courses.  It 
would  evidently  be  easy  enough  to  cover  the  pyramid  with  a  coat  of  cement — 
working  downwards — if  its  surface  did  not  greatly  overpass  the  salient  angles  of  the 
steps,  while  the  difficulty  would  be  enormously  increased  if  the  coat  were  to  have  a 
considerable  thickness  of  its  own  independently  of  the  pyramid,  like  the  casing 
shown  in  Fig.  155. — Ed. 

-  Descripiion  de  fEgyp/c,  Antiquith,  vol.  v.  p.  7. 


The  Tomb  under  the  Ancient  Empire.  2 


OD 


remained  unexplored.  One  of  them  was  silent  and  empty,  but  in 
the  others  the  inscriptions  and  sarcophagi  of  two  kings  of  the 
sixth  dynasty,  Papi  and  his  son  Merenzi,  were  found.  Frag- 
ments of  a  Ritual  of  the  Dead  were  recognized  among  them. 
Pleasure  at  this  discovery,  the  last  which  he  was  destined  to 
make  in  the  soil  of  Egypt,  brightened  the  last  days  of  Mariette.^ 

In  March  1881,  M.  Maspero,  the  successor  of  M.  Mariette 
as  director  of  the  excavations,  opened  a  pyramid  belonging  to 
a  different  group,  which  turned  out  to  be  the  tomb  of  Ounas, 
the  last  Pharaoh  of  the  fifth  dynasty.  In  this  pyramid,  portcullis 
stones  similar  to  those  which  have  already  been  figured  were 
found.  When  these  obstacles  were  passed  "  the  continuation 
of  the  passage  was  found,  the  first  part  of  polished  granite, 
the  second  of  the  close-grained  limestone  of  Tourah.  The  side 
walls  are  covered  with  fine  hieroglyphs  painted  green,  the  roof 
sprinkled  with  stars  of  the  same  hue.  The  passage  finally  opens 
into  a  chamber  half  filled  with  debris,  upon  the  walls  of  which  the 
inscription  Is  continued.  .  .  .  The  mummy-chamber,  like  that 
which  precedes  It,  Is  covered  with  hieroglyphs,  with  the  exception 
of  the  wall  opposite  to  the  entrance.  This  wall  is  of  the  finest 
alabaster,  and  is  effectively  decorated  with  painted  ornaments. 
The  sarcophagus  is  of  black  basalt,  without  inscription.  .  .  .  The 
text  of  the  inscription  which  covers  the  walls  is  almost  identical 
with  that  In  the  tomb  of  Papi,  but  it  has  the  advantage  of  being 
complete.  M.  Maspero,  whom  Mariette  had  previously  entrusted 
with  taking  squeezes  from  the  Inscription  In  the  tomb  of  Papi, 
recognised  certain  formulae  and  phrases  which  had  already  struck 
him  in  another  place.  .  .  .  These  texts  make  up  a  composition 
analogous  to  one  which  covers  the  walls  of  certain  little. known 
Theban  tombs.  Without  presenting  any  very  considerable 
difficulties,  they  demand  careful  examination  from  those  who  would 
comprehend  their  meaning. 

"  M.  Maspero,  encouraged  by  this  first  success,  ordered  a  second 
pyramid  to  be  opened.  He  wished  to  verify,  upon  the  spot,  a 
theory  which  he  had  long  upheld  In  spite  of  the  adverse  opinions 
of  the  majority  of  egyptologists.  It  Is  well  known  that  between 
the  sixth  and  the  tenth  dynasties  a  great  gap  exists,  so  far  as 
monumental  remains  are  concerned.  M.  Maspero  has  always 
believed  that  there  Is  no  such  gap.      He  has  observed    that    the 

'  (t.  Charmes,  in  the  Journal  dcs  Dclmts,  February  8,  iS8i. 


236  A  History  or  Art  in  Ancient  Egypt. 

pyramids  are,  so  to  speak,  grouped  chronologically  from  north  to 
south ;  those  of  the  fourth  dynasty  at  Gizeh,  those  of  the  fifth  at 
Abooseer,  those  of  the  twelfth  in  the  Fayoum.  The  excavations 
of  Mariette  as  well  as  his  own  showed  the  tombs  of  the  fifth  and 
sixth  dynasties  to  have  been  at  Sakkarah.  Hence  M.  Maspero 
thinks  that  the  pyramids  erected  by  the  sovereigns  of  the 
seventh,  eighth,  ninth,  and  tenth  dynasties  are  those  between 
Sakkarah  and  the  Fayoum.  The  future  will  show  whether  he 
is  right  or  wrong.  In  any  case  science  will  profit  by  the  new 
excavations  which  he  is  about  to  undertake."  ^ 

When  cross-examined  by  such  questioners  as  M.  Maspero  the 
pyramids  will  tell  us  much.  Hitherto  they  have  attracted  but  little 
of  that  examination  which  discovers  the  most  curious  secrets,  but 
their  size  and  the  beauty  of  their  masonry  will  ever  make  the 
three  great  pyramids  of  Gizeh  the  most  striking  objects  to  the 
traveller  and  to  the  historian  of  art. 

Considering  their  age,  these  three  pyramids  are  wonderfully  well 
preserved.  In  their  presence,  even  in  their  actual  state  of  partial 
ruin,  the  oriental  hyperbolism  of  Abd-uI-Latif,  an  Arab  writer  of 
the  thirteenth  century,  seems  no  more  than  natural.  "  All  things 
fear  Time,"  he  cries,  "  but  Time  fears  the  Pyramids !  "  And  yet 
time  has  done  its  work  during  the  last  few  hundreds  of  years. 
The  summits  of  the  great  structures  have  been  slightly  lowered  ; 
the  gaping  breaches  in  their  flanks  have  been  gradually  widened  ; 
and  although  in  spite  of  their  stripped  flanks  and  open  wounds 
they  still  rear  their  heads  proudly  into  the  Egyptian  sky,  all  those 
accessory  structures  which  surrounded  them,  and  fulfilled  their  own 
well-defined  offices  in  the  general  monumental  ensemble,  have 
either  been  destroyed  by  the  violence  of  man  or  engulfed  by  the 
encroaching  sand.  Where,  for  example,  are  those  wide  and 
substantial  causeways,  whose  large  and  carefully  adjusted  blocks 
excited  the  wonder  of  Herodotus.^  After  having  afforded  an 
unyielding  roadway  for  the  transport  of  so  many  heavy  materials, 
they  formed  truly  regal  avenues  by  which  the  funeral  processions 
of  the  Egyptians  reached  the  centre  of  the  necropolis  as  long  as 

'  Moniteur  Egyptien,  March  15,  1881. 

2  The  causeway  which  led  to  the  Pyramid  of  Cheops  still  exists  for  some  400 
yards  of  its  length ;  here  and  there  it  rises  as  much  as  eighty-six  feet  above  the 
surface  of  the  plateau,  A  similar  causeway  is  to  be  distinguished  on  the  eastern  side 
of  the  Third  Pyramid.  At  Abou-Roash,  at  Abousir,  and  elsewhere,  similar  remains 
are  to  be  found. 


The  Tomb  under  the  Ancient  Empire. 


237 


their  civilization  lasted.  In  the  plain  they  were  above  the  level  of 
the  highest  inundations,  and  their  gentle  slope  gave  easy  access  to 
the  western  plateau.  The  great  Sphinx,  the  image  of  Harmachis, 
or  the  Rising  Sun,  was  placed  at  the  threshold  of  the  plateau. 
Immovable  among  the  dead  of  the  vast  cemetery,  he  personified 
the  idea  of  the  resurrection,  of  that  eternal  life  which,  like  the 
morning  sun,  is  ever  destined  to  triumph  over  darkness  and  death. 


Fig.  156. — Plan  of  the  Pyramids  of  Gizeh  and  of  that  part  of  the  necropolis  which  immediately 

surrounds  them. 


His  head  alone  now  rises  above  the  sand,  but  in  the  days  of 
Herodotus  his  vast  bulk,  cut  from  a  rock  nearly  70  feet  high, 
was  well  calculated  to  prepare  the  eye  of  the  traveller  for  the  still 
more  colossal  masses  of  the  pyramids.  His  features  have  now 
been  disfigured  by  all  kinds  of  outrage,  but  in  the  thirteenth 
centurv,  althouofh  even  then   he   had   been  mutilated,  Abd-ul-Latif 


238 


A   History  of  Art  IxNT  Ancient  Egypt, 


was  able  to  admire  his  serene  smile,  his  head  enframed  in  a  richly 
carved  wig  which  added  to  its  size  and  dignity.  His  body  was 
never  more  than  roughly  blocked  out,  but  a  painted  decoration, 
of  which  traces  may  still  be  found,  compensated  in  some  degree  for 
the  deficiencies  in  the  modelling. 

The  soil  around  each  pyramid  was  carefully  levelled  and  paved 
with  dressed  limestone  slabs.  Upon  this  pavement  rested  the 
foundations  of  the  stylobate  surrounding  the  pyramid.  Both 
stylobate  and  pavement  are  now  in  almost  every  case  concealed  by 
sand  and  ddbris,  but  at  the  pyramid  of  Chephren,  which  is  less 


Fig.  157. — The  Sphinx. 


banked  up  than  the  others,  traces  of  them  have  been  proved  to 
exist.  They  added  somewhat  to  the  imposing  effect  of  those 
monuments  upon  the  eye,  and  gave  additional  definition  to  their 
bases.i  The  area  thus  paved  was  inclosed  with  a  wall,  which  had 
an  opening  towards  the  east,  in  front  of  which  the  temple,  or 
funerary  chapel  of  the  pyramid,  was  raised.     The  latter,  no  doubt, 

■  Description  de  VEgypte,  vol.  v.  p.  643.  See  also  in  the  plates,  Antiqiiites,  vol.  v. 
pi.  xvi.  fig.  2.  According  to  Jomard,  the  surbase  of  the  second  pyramid  was 
in  two  parts — a  stylobate,  10  feet  high  and  5  feet  thick,  and  a  plinth  about 
3  feet  high. 


The  Tomb  under  the  Ancient  EMpn<E. 


239 


was  magnificently  decorated.  At  the  foot  of  the  mountains  of 
'stone  under  which  reposed  the  ashes  of  the  Pharaohs  themselves, 
smaller  pyramids  were  raised  for  their  wives  and  children.  Of 
these  some  half  dozen  still  exist  upon  the  plateau  of  Gizeh.  One 
of  them  has  been  recognized  as  the  tomb  of  that  daughter  of 
Cheops,  about  whom  Herodotus  tells  one  of  those  absurd  stories 
invented  by  the  Egyptians  of  the  decadence,  with  which  his 
dragomans  took  such  delight  in  imposing  upon  his  simple  faith. ^ 
Around  the  space  which  was  thus  consecrated  to  the  adoration  of 
the  dead  monarch,  the  long  rows  of  mastabas  stretched  away  for 
miles  through  the  vast  necropolis. 

The  great  ones  of  Egypt,  all  those  who  had  been  near  the 
Pharaoh  and  had  received  some  of  his  reflected  glory,  grouped 
their  tombs  as  closely  as  possible  about  his.  Distributed  thus  by 
reigns,  the  private  tombs  were  erected  in  close  juxtaposition  one 
with  another,  each  being  provided  with  a  stele,  or  sepulchral  tablet 


.  y^>7<.'-*^nj:\'  v.^  •>.^-<l>>v.t».'VT- 


■^^x 


T^.-zv      i.jvje-%-f-vjy:.rr<-iTT.'.TfrT7Wr;7 


s-=5== 


b- 


Fig.  158. — Pj^ramid  with  its  inclosure,  Abousir  ;  from  Perring. 

upon  which  the  name  of  the  deceased  was  inscribed,  most  of  them 
being  adorned  with  painted  bas-reliefs,  and  a  few  with  statues 
placed  upon  their  fagades.  Upon  the  causeways  which  connected 
Memphis  with  the  necropolis,  upon  the  esplanades  erected  by  the 
Pharaohs  to  the  memory  and  for  the  adoration  of  their  ancestors, 
in  the  countless  streets,  lanes,  and  blind  alleys  which  gave  access 
to  the  private  tombs,  advanced  endless  processions  of  mourners, 
driving  before  them  the  bleatinof  and  lowinof  victims  for  the  funeral 
rites.  Priests  in  white  linen,  friends  and  relations  of  the  dead 
with  their  hands  full  of  fruit  and  flowers,  flitted  hither  and  thither. 
On  the  days  appointed  for  the  commemoration  of  the  dead,  all  this 
must  have  afforded  a  curiously  animated  scene.  The  city  of  the 
dead  had  its  peculiar  life,  we  might  almost  say  its  festivals,  like 
that  of  the  living.  But  amid  the  coming  and  going,  amid  all  the 
bustle  of  the   Egyptian  jour  des  morfs,  it  was   the  giant  forms  of 

'  Herodotus,  ii.  126. 


240  A    History  of  Art  in  Ancient  Egypt. 

the  pyramids,  with  their  poHshed  slopes  ^  and  their  long  shadows 
turning  with  the  sun,  that  gave  the  scene  a  peculiar  solemnity  and 
a  character  of  its  own.  Morning  and  evening  this  shadow  passed 
over  hundreds  of  tombs,  and  thus,  in  a  fashion,  symbolized  the 
royal  dignity  and  the  almost  superhuman  majesty  of  the  kingly 
office. 

Of  all  this  harmonious  conception  but  a  few  fragments  remain. 
The  necropolis  is  almost  as  empty  and  deserted  as  the  desert 
which  it  adjoins.  The  silence  is  only  broken  by  the  cry  of  the 
jackal,  by  the  footsteps  of  a  few  casual  visitors  hurrying  along 
its  deserted  avenues,  and  by  the  harsh  voices  of  the  Bedouins 
who  have  taken  possession  of  the  Pyramid  of  Cheojos,  and,  in  their 
own  fashion,  do  its  honours  to  the  curious  visitor.  But  despoiled 
though  they  be  of  their  ornaments  and  of  their  proper  surroun- 
dings, the  pyramids  are  yet  among  those  monuments  of  the  world 
which  are  sure  to  impress  all  who  possess  sensibility  or  powers 
of  reflection.  In  a  remarkable  passage  in  the  Descinptio7i g^nh'ale 
de  Memphis  et  des  Pyramides,  Jomard  has  well  defined  the  effect 
which  they  produce  upon  the  traveller  and  the  impressions  which 
they  leave  behind  :  "  The  general  effect  produced  by  the 
pyramids  is  very  curious.  Their  summits,  when  seen  from  a 
distance,  look  like  those  of  higrh  mountains  standing  out  aeainst 
the  sky.  As  we  approach  them  this  effect  diminishes  ;  but  when 
we  arrive  within  a  very  short  distance  of  their  sides  a  totally 
different  impression  succeeds ;  we  begin  to  be  amazed,  to  be 
oppressed,  almost  to  be  stupefied  by  their  size.  When  quite 
close  to  them  their  summits  and  ansrles  can  no  longer  be  seen. 
7'he  wonder  which  they  cause  is  not  like  that  caused  by  a  great 
work  of  art.  It  is  the  sense  of  their  simple  grandeur  of  form 
and  of  the  disproportion  between  the  individual  power  and  stature 
of  man  and  these  colossal  creations  of  his  hands.  The  eye  can 
hardly  embrace  them,  nor  the  imagination  grasp  their  mass. 
We  then  begin  to  form  some  idea  of  the  prodigious  quantity 
of  dressed  stone  which  goes  to  make  up  their  height.  We  see 
hundreds  of  stones  each  containino-  two  hundred  cubic  feet 
and  weighing  some  thirty  tons,  and  thousands  of  others  which 
are  but  little  less.     We  touch  them  with  our  hands  and  endeavour 

'  Jomard  remarks  that  the  upper  part  of  the  second  pyramid  still  reflects  the  rays 
of  the  sun.  "  It  still  possesses,"  he  says,  "a  portion  of  its  polished  casing,  which 
relects  the  rays  of  the  sun  and  declares  its  identity  to  people  at  a  vast  distance." 


The  Tomb  under  the  Middle  Empire.  241 


to  realize  the  power  which  must  have  been  required  to  quarry- 
dress,  carry,  and  fix  such  a  number  of  colossal  blocks,  how  many 
men  must  have  been  employed  on  the  work,  what  machines  they 
used,  and  how  many  years  it  must  have  taken  ;  and  the  less 
we  are  able  to  understand  all  these  things,  the  greater  is  our 
admiration  for  the  patience  and  power  which  overcame  such 
obstacles."  ^ 


§  3.    The    Tomb  under   the  Middle  Efnpire. 

We  have  shown  how  the  mastaba,  that  is  to  say,  the  most 
ancient  form  of  tomb  in  the  necropolis  of  Memphis,  was  an 
expression,  both  in  arrangement  and  in  decoration,  of  the  ideas 
of  the  Egyptians  as  to  a  future  life.  In  literature  and  in  art  the 
works  created  by  a  people  in  its  infancy,  or  at  least  in  its  youth, 
are  the  most  interesting  to  the  historian,  because  they  are  the 
results  of  the  sincere  and  unfettered  expansion  of  vital  forces  ;  this 
Is  especially  the  case  when  there  is  no  possibility  of  a  desire 
to  Imitate  forelofn  models.  The  mastaba  deserved  therefore  to  be 
very  carefully  studied.  No  other  race  has  given  birth  in  Its 
funerary  architecture,  to  a  type  so  pure,  a  type  which  may  be 
explained  in  every  detail  by  a  master-idea  at  once  original  and 
well  defined.  We  therefore  dwelt  upon  it  at  some  length  and 
described  it  with  the  care  which  It  demanded.  We  found  it 
again  in  the  pyramids,  the  royal  tombs  of  the  Ancient  Empire, 
which  though  sensibly  modified  by  the  great  change  in  proportion, 
by  the  colossal  dimensions  which  the  pride  of  the  Pharaohs  gave 
to  one  part  of  their  tomb,  are  yet  penetrated  by  the  same  spirit. 
We  have  yet  to  follow  the  development  of  the  same  idea  through 
the  later  years  of  Egyptian  civilization,  and  in  localities  more  or 
less  removed  from  that  in  which  she  gave  her  first  tokens  of  power. 
In  one  place  we  shall  find  it  modified  by  the  nature  of  the  soil  to 
which  the  corpse  had  to  be  committed,  in  another  by  the  inevitable 
progress  of  ideas,  by  the  development  of  art,  and  by  the  caprices 
of  fashion,  which  was  no  more  stationary  in  Egypt  than  elsewhere. 

The   most  important   necropolis   of   the    First   Theban   Empire 
was  that  of  Abydos  in   Upper  Egypt,  upon  the  left  bank  of  the 

'  Description  de  I'Egypte,  Antiqtiiies,  vol.  v.  j).  597. 
VOL.    I.  I    I 


242  A  History  of  Art  in  Ancient  Egypt. 

river.  The  great  number  of  sepultures  which  took  place  in 
it,  from  the  first  years  of  the  monarchy  until  the  end  of  the 
ancient  civilization,  is  to  be  explained  by  the  peculiarly  sacred 
character  of  the  city  of  Abydos,  and  by  the  great  popularity, 
from  one  end  of  the  Nile  valley  to  the  other,  of  the  myths  which 
centred  in  it.  According  to  the  Egyptian  belief,  the  opening 
through  which  the  setting  sun  sank  into  the  bowels  of  the  earth 
for  its  nightly  transit,  was  situated  to  the  west  of  Abydos.  We 
know  how  the  Egyptian  intellect  had  established  an  analogy 
between  the  career  of  the  sun  and  that  of  man  ;  we  may  therefore 
conclude  that  in  choosing  a  final  resting-place  as  near  as  possible 
to  the  spot  where  the  great  luminary  seemed  to  make  its  nightly 
plunge,  they  believed  they  were  making  more  completely  sure 
of  triumphing,  like  him,  over  darkness  and  death. 

The  sun  is  not  extinguished,  he  is  but  hidden  for  a  moment 
from  the  eyes  of  man.  This  sun  of  the  infernal  regions  is  Osiris, 
who,  of  all  the  Egyptian  gods,  was  most  universally  adored. 
Although  many  Egyptian  towns  could  show  tombs  in  which  the 
members  of  Osiris,  which  had  been  dispersed  by  Set,  were 
re-united  by  Isis  and  Nephthys,  none  of  them  were  so  famous, 
or  the  object  of  such  deep  devotion,  as  that  at  Abydos.  It  was, 
if  we  may  be  permitted  to  use  such  a  phrase,  the  Holy  Sepulchre 
of  Egypt.  As,  in  the  early  centuries  of  Christianity,  the  faithful 
laid  great  stress  upon  burial  in  the  neighbourhood  of  some 
holy  martyr,  "  The  richest  and  most  influential  Egyptians,"  says 
a  well  informed  Greek  writer,  "  were  ambitious  of  a  common 
tomb  with  Osiris."  ^ 

Under  such  conditions  it  may  readily  be  understood  why 
Mariette  should  have  concentrated  so  much  of  his  attention  upon 

1  Pseudo-Plutarch,  On  Isis  and  Osiris,  c.  xx.  M.  Maspero  finds,  however,  no 
confirmation  of  this  statement  in  the  monuments  themselves.  "  All  the  tombs 
which  have  yet  been  discovered  at  Abydos,"  he  says  {Revue  Critique,  January  31, 
t88i),  "are  those  of  Egyptians  domiciled  at  Abydos.  But  the  author  from  whom 
this  Plutarch  derived  his  inspiration  must  have  known  the  ancient  fiction  according 
to  which  the  soul  could  only  pass  into  the  next  world  by  betaking  itself  to  Abydos, 
and  thence  through  the  opening  to  the  w^est  of  that  town  which  gave  access  to  the 
regions  of  Ament.  Hence  the  voyage  of  the  dead  to  Abydos  which  we  find  so 
often  represented  on  tombs  ;  an  imaginary  voyage,  as  the  mummy  would  be  reposing 
safely  at  Thebes  or  Memphis  (Fig.  159).  At  all  events,  the  family,  after  the  death 
of  its  head,  or  any  Egyptian  during  his  own  life,  could  deposit  upon  the  ladder  of 
Osiris  a  stele,  upon  which  the  tomb  actually  containing  his  body  could  be  represented 
and  unmistakably  identified  with  its  original  by  the  formula  inscribed  upon  it." 


The  Tomb  under  the  Middle  Empire. 


24: 


Abydos.  In  spite  of  all  his  researches  he  did  not  succeed  in 
discovering  the  tomb  of  Osiris  itself,  but  yet  his  digging 
campaigns  afforded  results  which  are  most  interesting  and 
important  from  every  point  of  view.^ 

One  district  of  this  necropolis  is  made  up  by  a  vast  number  of 
tombs  dating  from  the  time  of  the  ancient  empire,  and  particularly 
from  the  sixth  dynasty.  Arrangements  similar  to  those  of  the 
mastabas  at  Sakkarah  are  found,  but  on  a  smaller  scale — the 
same  funerary  chambers,  the  same  wells,  sometimes  vertical, 
sometimes  horizontal  as  in  the  tomb  of  Ti  and  the  pyramids, 
the    same   materials.     The  situation  of   this   tomb-district,   which 


Fig.  159. — The  river  transport  of  the   Mummy.     (Champollion,  pi.  173.) 


Marlette  calls  the  central  cemetery,  has  allowed  arrangements 
to  be  adopted  similar  to  those  on  the  plateau  of  Memphis,  where 
the  sand  is  the  only  covering  to  a  stratum  of  living  rock  in  which 
it  was  easy  to  cut  the  well  and  the  mummy-chamber. 

In    the    remainder   of  the    space    occupied  by    the    tombs    the 
subsoil  is  of  a  very  different  nature.     "  The  hard  and  impenetrable 

^  Mariette,  Abydos^  Description  des  Foidlles  exccutees  si/r  F Eviplaceincnt  de  cette 
Vilie,  folio,  vol.  i.  1869  ;  vol.  ii.  1880.  Mariette  thought  that  the  sacred  tomb  was 
probably  in  the  immediate  neighbourhood  of  the  artificial  mound  called  Koum-es- 
Soultan,  which  may  cover  its  very  site.  In  the  article  which  we  quote  above, 
M.  Maspero  has  set  forth  the  considerations  which  lead  him  to  think  that  the 
staircase  of  Osiris,  upon  which  the  consecrated  steles  were  placed,  was  the  flight  of 
steps  which  led  up  to  the  temple  of  that  god.  Consequently  the  tomb  of  Osiris,  at 
Abydos  as  at  Denderah,  would  be  upon  the  roof  of  his  temple. 


244 


A  History  of  Art  in  Ancient  Egypt. 


rock  is  there  covered  with  a  sandstone  in  course  of  formation  ; 
this  is  friable  at  some  points,  at  others  so  soft  that  but  few 
mummies  have  been  entrusted  to  it,"  ^  This  formation  extends 
over  nearly  the  whole  of  the  ground  upon  which  the  tombs  of  the 
eleventh,    twelfth,   and  especially  of  the   thirteenth,   dynasties,  are 


Fig.  i6o. — Tomb  at  Abydos  ;  drawn  in  perspective  from  the  elevation  of  Mariette. 

packed  closely  together.  This  Mariette  calls  the  northern 
cemetery.  "  The  tombs  of  Abydos  have  no  subterranean  story, 
properly  speaking.  Well,  mummy-chamber,  and  funerary  chapel 
are  all  constructed,  not  dug.  In  the  few  instances  in  which  the 
ground  has  been  excavated  down  to  the  friable  sandstone  which 

over-lies  the  hard  rock,  the  opening  has 
been  lined  w^ith  rubble. 

"  Hence  the  peculiar  aspect  which  the 
necropolis  of  Abydos  must  have  pre- 
sented when  intact.  Imagine  a  multi- 
tude of  small  pyramids  live  or  six  metres 
high,  carelessly  oriented  or  not  at  all,  and 
uniformly  built  of  crude  brick.  These 
pyramids  always  stand  upon  a  plinth, 
they  are  hollow,  and  within  they  are 
formed  into  a  clumsy  cupola  by  means 
of  roughly  built  off-sets.  The  pyramid 
stands  directly  over  a  chamber  in  its  foundations  which  shelters 
the  mummy.  As  soon  as  the  latter  was  in  place,  the  door  of 
its  chamber  was  closed  by  masonry."  ^  An  exterior  chamber 
was   often   built   in    front  of   the  pyramid,    and   being   always  left 

'  Mariette,  Voyage  dans  la  Haute- Egypte,  vol.  i.  1879.  2  Ji^idem. 


Fig.   161.- 


-Section  of  the  above 
tomb. 


The  Tomb  under  the  Middle  Empire. 


245 


open,  served  for  the  performance  of  the  sepulchral  rites ;  but 
sometimes  this  chamber  was  absent  and  then  those  rites  were 
carried  through  in  the  open  air,  before  the  stele  of  the  deceased. 
This  latter  was  sometimes  erected  upon  the  plinth,  sometimes  let 
into  its  face.     A  little  cube  of  masonry  is  sometimes  found  at  the 


Fig.  162. — Tomb  at  Abydos  ;  di-awn  in  perspective  from  the  elevation  of  Marietta. 

foot  of  the  stele,  destined,  no  doubt,  for  funeral  offerings.  Sometimes 
the  tomb  had  a  surrounding  wall  of  the  same  height  as  its  plinth  ; 
this  served  to  mark  out  the  ground  which  belonged  to  it,  and  when 
the  friends  of  the  deceased  met  to  do  him  honour,  the  entrance 
could  be  closed,  and  compara- 
tive privacy  assured  even  in  the 
absence  of  a  funerary  chapel. 

These  tombs,  which  were 
generally  constructed  with  no 
great  care,  were  for  the  most 
part  without  casing.  The  pyra- 
midal form  was  given  by  setting 
each  course  of  bricks  slightly 
back  from  the  one  below  it. 
When  this  part  of  the  work  was 
finished,  each  face   was  covered, 

as  a  rule,  with  a  coat  of  rough  concrete,  which,  in  its  turn,  was 
hidden  under  a  layer  of  white  stucco.  This  multitude  of  little 
monuments,  all  of  the  same  shape  and  of  much  the  same  size, 
must,  when  complete,  have  looked  like  the  tents  of  an  encamped 
army. 


■t" 


Fig.  163. — Section  of  the  above  tomb. 


246 


A  History  of  Art  in  Ancient  Egypt. 


As  these  tombs  were  all  upon  the  surface  of  the  ground  they 
have  suffered  more  than  any  others  from  the  attacks  of  man. 
Those  which  are  reproduced  among  these  lines  of  text  were 
only  recovered  by  Mariette  by  dint  of  patient  excavation.     And 


[pTiwrTT^ii  liiyj  n^ii  Niiiiiiteii.vTrni,ii!!!;it:N-mii  imii   iin 


ig?^sslS27SS?*^5S3-ASsr^ifM^!M^t 


^^m3^sM^W^^^mTSi3£E(i^-ki 


Fig.  164. — Stele  of  the  eleventh  dynasty,  Abydos.     Drawn  by  Bourgoin.     (Boulak.) 


although  these  ill  constructed  edifices,  so  far  as  their  materials 
are  concerned,  are  still  standing,  they  will  soon  follow  the  many 
thousands  which  once  stood  in  serried  ranks  round  the  sepulchre 
of  Osiris.     The  only  remains  of  this  necropolis  which  are  likely  to 


Fig.  165.— Stele  of  Pinahsi,  priest  of  Ma  :  Abydos.    New  Empire.   Drawn  by  Bourgoin.   (Boulak.) 


The  Tomb  under  the  Middle   Emph<e.  249 


be  preserved  are  the  numberless  steles  which  Mariette  rescued 
from  its  ddbj'is.  They  form  about  four-fifths  of  the  total  number 
of  those  monuments  now  preserved  in  the  museum  at  Boulak.^ 
We  figure  two  of  them,  one  belonging  to  the  Middle,  the  other 
to  the  New  Empire  (Figs.  164  and  165). 

Whenever  religious  motives  did  not  affect  their  choice,  the 
Egyptians  preferred,  during  the  period  we  are  now  considering, 
to  cut  their  tombs  horizontally  out  of  some  rocky  eminence. 
Such  a  tomb  was  called  a  aiveos  by  the  Greeks.  The  most 
interesting  examples  of  these  constructions  are  offered  by  the 
tombs  of  the  twelfth  dynasty  at  Beni- Hassan  and  at  Siout,  both 
situated  between  Memphis  and  Abydos. 

Champollion  was  the  first  to  appreciate  the  importance  of  the 
grottos  of  Beni-Hassan.  Ever  since  his  time  they  have  received, 
for  various  reasons,  much  of  the  attention  of  egyptologists.  We 
have  already  referred  to  their  inscriptions,  which  are  as  interesting 
to  the  historian  of  ideas  as  to  the  student  of  political  and  social 
organizations.  We  have  alluded  above  to  the  varied  scenes  which 
cover  the  walls  of  their  chambers,  the  most  important  of  which  have 
been  reproduced  by  Champollion,  Lepsius,  and  Prisse  d' Avenues  ; 
we  have  finally  to  speak  of  those  famous  protodoric  columns,  as 
they  are  called,  in  which  some  have  thought  they  saw  the  original 
model  of  the  oldest  and  most  beautiful  of  the  Grecian  orders. 
We  are  at  present  concerned,  however,  with  the  arrangement  of 
the  tombs  themselves.  These  are  the  same,  with  but  slight 
variations,  for  the  smallest  and  most  simple  tombs  as  for  those 
which   are  largest  and  most  elaborately  decorated. 

These  facades  are  cut  into  the  cliff-like  sides  of  the  hills  of  the 
Arab  Chain,  about  half-way  up  their  total  height.  They  are, 
therefore,  hioh  above  the  surface  of  the  river.  When  the  cuttinof 
was  made,  two  or  three  columns  were  left  to  form  a  portico, 
the  deep  shadows  of  which  stand  out  strongly  against  the  white- 
ness of  the  rock.  This  portico  leads  to  a  chamber  which  is 
lighted  only  from  the  door.  Its  ceiling  is  often  cut  into  the  form 
of  a  vault.  A  deep  square  niche  is  cut,  sometimes  opposite  to 
the  door,  sometimes  In  one  of  the  angles.  It  once  contained 
the  statue  of  the  deceased.      Most  of   the   tombs   have   but   one 

^  All  these  steles  are  figured  in  the  last  work  published  by  Mariette,  the  Catalogue 
general  des  Motntmeuts  d' Abydos,  dcco^iverts  pendant  les  Foiiilles  dc  ecttc  Mile,  i  vol. 
4to.     Paris,  1880. 

VOL.    I.  K    K 


250 


A  History  of  Art  in  Ancient  Egypt. 


chamber,  but  a  few  have  two  or  three.  In  a  corner  either  of  the 
only  chamber  or  of  that  which  is  farthest  from  the  door,  the 
"opening  of  a  square  well  is  found ;  this  leads  to  the  mummy- 
chamber,  which  is  excavated  at  a  lower  level. 


Fig.  i66. — Fa9ade  of  a  tomb  at  Beni-Hassaii. 


The  chamber  upon  which  the  portico  opens  is  the  funerary 
chapel,  the  place  of  reunion  for  the  friends  and  relations  of  the 
dead.  As  Mariette  very  truly  remarks,  from  the  first  step  which 
the  traveller  makes  in  the  tomb  of  Numhotep  at  Beni- Hassan, 
he    perceives    that,    in   spite    of   all    differences    of   situation,   the 


The  Tomb  under  the  Middle  Empire. 


251 


traditions  of  the  Ancient  Empire  are  still  full  of  vitality.  "  The 
spirit  which  governed  the  decorators  of  the  tomb  of  Ti  at 
Sakkarah  still  inspired  the  painters  who  covered  the  walls  of  the 
tomb  of  Numhotep  at  Beni-Hassan.  The  defunct  is  at  home 
among  his  own  possessions  ;  he  fishes  and  hunts,  his  cattle  defile 


Fig.  167. — Facade  of  a  tomb  at  Beni-Hassan,  showing  some  of  the  adjoining  tombs. 


before  him,  his  people  build  boats,  cut  down  trees,  cultivate  the 
vine  and  gather  the  grapes,  till  the  earth,  or  give  themselves 
up  to  gymnastics  or  to  games  of  skill  and  chance,  and  among 
them  the  figure  of  the  dead  is  carried  hither  and  thither  in  a 
palanquin.  We  have  already  found  pictures  like  these  in  the 
mastabas  of  the   Ancient    Empire,  and  here   we   find   them   again. 


252 


A   History  of  Art  in  Ancient  Egypt. 


But    at    Beni-Hassan     this    painted    decoration     becomes  more 

personal    to    the    occupant    of   the    tomb,    the    inscriptions  enter 

into  precise    and    copious    biographical    details,   which   are  never 
found  elsewhere."  ^ 


l'"iG.    i68. — Interior  of  a  toiiib  at  Beni-Ha'-san.     Drawn  in  perspective  from  the  elevation  of 

Lepsius  (i.  pi.  6e). 


The  necropolis  of  Siout,  in  the  Libyan  chain,  offers  the  same 
general  characteristics.  The  tomb  of  Hapi-Tefa,  a  feudal  prince 
of  the  twelfth  dynasty,  and  consequently  a  contemporary  of  those 
princes   of   the  nome  of    Meh   who   are   buried   at    Beni-Hassan, 

is  the  most  remarkable.  It  is  composed 
of  three  large  chambers  communicating 
one  with  another,  and  with  the  external 
air  by  a  wide  portico.  The  mummy-pit 
is  reached  from  the  innermost  of  these 
chambers. 

Neither    statues,    mummies,    nor    any 
other  movable  objects  have  been  found 
in  these  grottos.     When  their  accessible 
situation  and  their  conspicuous  appear- 
ance is  remembered,  this  should  not  cause 
surprise.     Many  centuries  ago  the  acacia 
doors,  which  are  mentioned  in  one  of  the 
texts  at  Beni-  Hassan,  disappeared,  and, 
in  spite  of  the  accumulation  of  sand,  the  mouths  of  the  wells  could 
be  found  so  easily,  and  could  so  readily  be  cleared,  that  all  objects 
of  value  and  interest  must  have  been  abstracted  from  the  mummy- 
^  Mariktte,  Voyai:;e  dans  la  Haute-Egvpte,  vol.  i.  p.  51. 


U--1 1 1 1 1 1— J — 1 


Fig.  169. — Plan  of  the  above  tomb. 


The  Tomb  under  the  Middle  Empire. 


253 


chambers  in  very  remote  times,  perhaps  before  the  fall  of  the  antique 
civilization.  The  inscriptions  and  the  painted  walls  alone  remained 
practically  intact  down  to  the  commencement  of  the  present 
century.  The  dryness  of  the  climate,  and  the  difficulty  of 
detaching  them  from  the  wall  contributed  to  their  preservation, 
which  was  nowhere  more  complete  than  at  Beni- Hassan.  But 
since  travelling  in  Egypt  became  the  fashion  their  sufferings  have 
begun.  The  mania  for  carving  names  upon  every  surface,  and  for 
preserving  souvenirs  of  all  places  of  interest,  has  destroyed  the 
whole  of  one  wall.  The  smoke  of  torches  has  also  done  its  work 
in  reducing  the  brilliant  tones  and  blunting  the  delicate  contours. 
Happily,  the  more  interesting  examples  are  all  reproduced  in  those 
great  works  to  which  we  have  already  had  such  frequent  occasion 
to  refer. 


^^.^^'^ji^^^^^.j'^s.-C^^j'jy' s    ,* 


Fig.  170. — Chess  players,  Beni-Hassan.      (Champollion,  pi.  369.) 

The  rich  necropolis  of  Thebes  has  not  preserved  any  monu- 
ments from  this  period  in  such  good  condition  as  those  of  Abydos, 
Beni-Hassan,  or  Siout.  M.  Maspero  has  discovered,  however,  in 
the  district  known  as  the  Drak-Aboid-N'co-^-ak,  some  remains  of 
the  royal  tombs  of  the  eleventh  dynasty.  Several  of  these  tombs 
resemble  in  their  general  arrangements  those  of  the  feudal  princes 
of  Meh  and  Siout.  Thus  the  sepulchre  of  the  King  Ra-Anoub- 
KJioper-Eiitef  is  what  the  Greeks  called  a  hemi-speos,  that  is,  it  was 
partly  built  and  partly  hollowed  out  of  the  living  rock.  Before 
the  facade  thus  built  against  the  mountain,  two  obelisks  were 
reared.  The  tombs  of  the  other  princes  belonging  to  the  family 
of  Entef  were  built  upon  the  open  plain.     They  were  structures 


2  54  ^  History  of  Art  in  Ancient  Egypt. 

in  masonry,  and  seem  at  one  time  to  liave  been  crowned  by 
pyramids.  Some  idea  of  their  shape  may  be  obtained  from  our 
ilkistrations  of  the  tombs  at  Abydos.^ 

To    complete    our    observations    upon   the   tombs  of    the   first 
Theban   Empire,   it  will    be    sufficient    to    recall   what    we    have 
already  said  about  the  pyramids  in  the   Fayoum,  which  were  the 
work  of  the  thirteenth  dynasty.      It  is  difficult  to  form  an  accurate 
idea   of   the    appearance    of  those    monuments    when    complete. 
Time  has  treated  them  w^th  great  severity,  and  in  their  present 
state  it  is  impossible  to  verify  the  assertions  of  Herodotus  as  to 
the  peculiarities  of  their  casing  and  crowning  ornaments.      But  it 
is  quite  certain  that  the  Middle  Empire  made  no  original  inventions 
in    the   matter    of   sepulchral    architecture.      It    appears    to    have 
discontinued  some  of  the  ancient  arrangements,  but  in  those  which 
it    preserved    its    efforts  were    confined   to  putting  old    elements 
together   in  a  new  fashion   and  with  new  proportions.     It  made 
frequent  use  of  one  mode  of  sepulture  which  had  previously  been 
quite  exceptional.      No  mastaba  is  known  which  dates  from  this 
epoch,  but  the  kings  had  not  ceased  to  confide  their  mummies  and 
the  perpetuation   of  their  glory  to  pyramids,  but  these  were    no 
longer  of  such  colossal  dimensions  as  under  the  Ancient  Empire, 
while  their  character  was  complicated,  to  some  extent,  by  the  colossi 
with  which  they  are  said  to  have  been  surmounted,  and  the  figured 
decoration  of  their  walls.     Finally,  they  were  often  employed,  not 
as  self-contained   monuments   in  themselves,  but    merely  as    the 
culminating  points  in  a  more  complex  ensemble.     They  were  built 
upon  a  rectangular  platform  or  tower  with  walls  slightly  inclined  from 
the  perpendicular. 

It  would  seem  that  the  idea  of  this  arrangement  had  occurred 
to  the  primitive  Egyptians.  So,  too,  had  that  of  the  spcos  or  rock- 
cut  tomb  ;  but  the  Memphite  architects  have  left  nothing  which 
at  all  resembles  the  orrottos  in  the  mountain  sides  of  Beni-Hassan 
and  Siout.  Neither  in  the  neighbourhood  of  the  pyramids  nor 
in  any  other  district  where  the  tombs  of  the  early  epoch  are 
found,  has  any  sepulchre  been  discovered  which  shows  the 
monumental  facades,  the  large  internal  development,  and  the 
simple  and  dignified  lines  of  the  artificial  chambers  in  the  Arab 
and  Libyan  chains. 

1  Maspero,  Rapport  sur  une  Mission  en  Italic  (in   the  Reateil  de  Travaux^  vol.  ii. 
p.  1 66).     The  Abbott  Papyrus  gives  a  Ust  of  these  Httle  pyramids. 


The  Tomb  under  the  New  Empire.  255 


§  4..  The   Tomb  under  the  New  Empire. 

The  subterranean  tombs  for  which  the  first  Theban  Empire 
had  shown  so  marked  a  preference,  became  firmly  seated  in  pubHc 
favour  during  the  succeeding  centuries.  We  do  not  know  what 
the  funeral  customs  may  have  been  during  those  centuries  when 
the  Hyksos,  or  shepherd  kings,  were  masters  of  Egypt ;  but,  after 
their  expulsion,  the  great  Theban  dynasties,  the  eighteenth,  the 
nineteenth,  and  the  twentieth,  by  whom  the  glory  of  Egyptian 
arms  and  culture  was  spread  so  widely,  hardly  made  use  of  any 
sepulchre  but  the  chamber  hollowed  laboriously  in  the  rocky  sides 
of  that  part  of  the  Libyan  chain  which  lies  to  the  west  of  Thebes. 
Every  traveller  visits  the  royal  tombs  which  lie  in  the  gloomy 
ravine  called  the  Bab-el-AIolouk,  or  the  Gate  of  the  Kings.  The 
valley  is  about  three  miles  in  length  and  has  a  mean  width  of 
about  eleven  hundred  yards  ;  its  sides  are  riddled  with  galleries 
penetrating  more  or  less  deeply  into  the  mountains,  and  starting 
sometimes  from  the  slopes,  sometimes  from  the  base  of  the  clift's, 
which  here  and  there  attain  a  height  of  400  feet.  The  word 
speos  seeming  to  the  Greeks  to  give  an  inadequate  idea  of  the 
depth  of  these  excavations  and  of  their  narrow  proportions,  they 
called  them  avpijyes,  or  pipes  ;  and  modern  archaeologists  have 
often  employed  the  same  picturesque  term  in  speaking  of  the 
Theban  tombs.  Five-and-twenty  of  these  tombs  are  royal ;  the 
rest  belong  to  wealthy  subjects,  priests,  warriors,  and  high  officers 
of  state.  In  extent  and  richness  of  ornament  some  of  the  latter 
are  in  no  way  inferior  to  the  tombs  of  the  sovereigns. 

Our  studies  must  first,  however,  be  directed  to  the  royal 
tombs,  because  in  them  we  find  the  most  original  types,  the 
most  important  variations  upon  those  which  have  gone  before. 
In  them  the  art  of  the  New  Empire  gives  a  clearer  indication  of 
all  the  changes  which  the  progress  of  ideas  had  brought  about  in  the 
Egyptian  conception  of  a  future  life.  In  them,  too,  the  Egyptian 
taste  for  ample  dimensions  and  luxuriant  decoration  is  more 
freely  indulged.  The  architects  of  Seti  and  Rameses  had 
resources  at  command  far  beyond  those  of  which  their  early 
rivals  could  dispose.  They  were,  therefore,  enabled  to  indulge 
their  employers'  tastes   for  magnificence,   and   to  give  to  certain 


256  A   History  of  Art  in  Ancient  Egypt. 

parts  of  the  tomb  a  splendour  which  had  been  previously 
unknown.  And  such  parts  were  never  those  upon  which  the 
pyramid   builders  had  lavished   most   of  their   attention. 

Nothing-  could  be  more  simple  than  the  course  of  proceeding 
of  the  earlier  architects,  whose  services  and  high  social  position 
are  indicated  for  us  by  more  than  one  stele  from  the  Ancient 
Empire.  They  had  to  distinguish  the  royal  from  the  private  tomb, 
and  no  means  to  such  an  end  could  be  more  obvious  than  to 
make  use  of  a  form  of  construction  which  allowed  the  height 
and  extent  to  be  added  to  ad  injinitimi  without  compromising 
the  stability  of  the  monument.  Their  one  idea,  therefore,  was 
to  push  the  apex  of  the  pyramid  as  far  up  into  the  sky  as  they 
could.  The  height  grew  as  the  flanks  swelled,  so  that  it  became, 
by  one  process  of  accretion,  ever  more  imposing  and  better  fitted 
to  safeguard  the  precious  deposit  hidden  within  it.  In  such 
a  system  the  important  point  was  this  envelope  of  the  mummy- 
chamber,  an  envelope  composed  of  thousands  of  the  most  care- 
fully dressed  and  fixed  blocks  of  stone,  which,  in  their  turn,  were 
covered  with  a  cuirass  of  still  harder  and  more  durable  materials. 
In  order  that  all  access  to  the  sarcophagus  might  be  more  safely 
guarded  against,  the  funerary  chapel  was  separated  from  the 
mountain  of  hewn  stone  which  inclosed  the  mummy-chamber. 
Supposing  the  latter  to  be  decorated  with  all  the  taste  and 
richness  which  we  find  in  the  tomb  of  Ti,  it  would  still  be 
comparatively  small  and  unimportant  beside  the  colossal  mass 
which  overshadowed  it,  and  to  which  it  belonged.  The  dis- 
proportion is  easily  explained.  When  the  pyramids  were  built, 
the  workman,  the  actual  mason,  had  little  more  to  learn.  He 
was  a  thorough  master  of  the  dressing  and  fixing  of  stone  and 
other  materials  ;  but,  on  the  other  hand,  the  art  of  architecture 
was  yet  in  its  infancy.  It  had  no  suspicion  of  those  rich  and 
varied  effects  which  the  later  Egyptians  were  to  obtain  by  the 
majesty  of  their  orders  and  the  variety  of  their  capitals.  It  was 
not  till  much  later  that  it  learnt  to  raise  the  pylon  before  the 
sacred  inclosures,  to  throw  solemn  colonnades  about  their  courts, 
and  to  greet  the  visitor  to  the  temples  with  long  naves  clothed 
in  all  the  glory  of  colour. 

Two  periods  of  national  renascence,  in  the  thirteenth  and 
eighteenth  dynasties,  had  to  intervene  before  these  marvels  could 
be    realized.      The    earlier    of    these   two   periods   is  only   known 


c 
O 


N^li.if'/cO/' 


S' 


''^. 

^ 


**--  .  .  t 


VOL.    I. 


L    L 


The  Tomb  under  the  New  Empire.  259 

to  us  by  a  few  works  of  sculpture  in  our  museums.  We  are 
forced  to  guess  at  its  architecture,  as  we  have  nothing  but 
descriptions,  which  are  at  once  incomplete  and  exaggerated,  to 
euide  our  imao-inations. 

The  second  Theban  Empire  may  be  studied  under  very  different 
conditions.  The  architects  of  that  epoch  excelled  all  their 
predecessors  in  the  skill  with  which  they  used  their  materials, 
and  the  artistic  ability  with  which  they  laid  their  plans.  In  a 
word,  they  realized  the  ideal  towards  which  Egyptian  builders 
had  been  tending  for  many  centuries,  and  their  genius  is  still 
to  be  seen  in  buildings  which,  even  in  their  ruin,  charm  by  the 
grandeur  of  their  conception  and  the  finish  of  their  execution. 

In  the  century  which  saw  the  construction  of  the  great  temples 
of  Abydos,  of  Karnak,  and  of  Luxor,  the  architect  who  was 
charged  with  the  building  of  the  royal  tombs  could  dispose  of 
all  the  resources  of  an  empire  which  stretched  from  the  southern 
boundaries  of  Ethiopia  to  Damascus  and  Nineveh.  He  would 
have  fulfilled  the  wishes  of  neither  prince  nor  people  had  he  not 
found  means  to  give  an  amplitude  and  a  beauty  to  those  tombs 
which  should  stand  a  comparison  with  the  sumptuous  edifices 
which  the  same  kings  had  erected,  in  another  part  of  the  city, 
in  honour  of  the  great  deities  of  the  country. 

The  simple  and  massive  forms  of  the  pyramid  did  not  lend 
themselves  to  success  in  such  an  enterprise.  They  afforded  no 
opportunity  for  the  happy  combinations  of  horizontal  and  vertical 
lines,  for  the  contrasts  of  light  and  shadow  and  splendour  of 
decoration  which  distinguished  the  epoch.  The  experience  of  the 
Middle  Empire  proved  that  it  was  better  to  make  a  fresh  departure 
than  to  attempt  to  foist  upon  the  pyramid  a  class  of  ornament 
which  was  destructive  to  the  simplicity  in  which  so  much  of  its 
grandeur  consisted.  The  highest  expression  of  the  new  form  of 
art  was  in  the  temple,  the  development  of  which  was  rather  in  a 
horizontal  than  in  a  vertical  direction  ;  in  the  long  avenues  of 
sphinxes,  in  the  pylons  and  colossal  statues  of  the  kings,  in 
porticoes  and  forests  of  columns.  The  problem,  therefore,  was  to 
embody  some  of  these  elements  in  the  design  of  the  tomb.  For 
this  purpose  it  was  necessary  to  give  increased  dimensions  and 
greater  importance  to  a  part  of  the  royal  sepulchre  which  had  been 
hitherto  comparatively  neglected.  The  funerary  chapel  had  to  be 
expanded   into  a  temple   in   miniature,    into   a  temple   where  the 


26o  A   History  of  Art  in  Ancient  Egypt. 

king,  who  had  rejoined  the  deities  from  whom  he  was  descended, 
could  receive  the  homage  and  worship  of  his  people. 

The  exploits  of  these  princes,  which  were  greater  than  anything 
of  which  Egypt  had  to  boast  in  the  whole  of  its  glorious  past, 
must  have  helped  to  suggest  the  temple  form  of  their  tombs.  To 
this  form  the  general  movement  of  the  national  art  also  pointed, 
and  to  give  it  effect  nothing  more  was  required  than  the  separa- 
tion of  the  chapel  from  the  tomb  proper,  to  which  previous 
tradition  had  so  closely  allied  it.  The  situation  of  the  sepulchre, 
after  Thebes  had  become  a  populous  city  and  the  real  capital  of 
Egypt,  was  no  longer  a  matter  of  question.  It  was  in  the  rocky 
flanks  of  the  Libyan  chain  that  all  its  inhabitants  sought  that 
asylum  for  their  dead  which  the  inhabitants  of  Memphis  found 
upon  the  eastern  edge  of  the  desert.  The  Libyan  chain  to  the 
west  of  Thebes  offers  no  platform  like  that  of  the  necropolis  of 
Memphis.  Its  cliffs  and  intersecting  ravines  offer  no  sites  for 
constructed  works  ;  hence  the  ordinary  form  of  Theban  tomb  is 
the  speos,  or  the  pipe,  which  is  but  an  exaggerated  form  of  the 
speos. 

Like  the  chief  men  among  his  subjects,  the  sovereign  loved  to 
take  his  last  repose  in  the  immediate  neighbourhood  of  the  city  in 
which  he  had  dwelt  during  his  life,  in  which  the  streets  had  so 
often  resounded  to  the  cries  of  triumph  which  greeted  his  return 
from  some  successful  campaign,  or  had  seen  him  pass  in  some  of 
those  long  processions  which  are  figured  upon  the  walls  of 
Medinet-Abou  (Fig.  172).^  His  tomb  was  a  cavern  like  that  of 
his  subjects.  Fashion  and  the  physical  conditions  of  the  country 
governed  him  as  well  as  his  inferiors  in  rank.  From  the  reign  of 
Seti  I.  onwards,  the  kings  chose  for  their  place  of  sepulture  the 
wild  and  deserted  valley  in  which  Belzoni  found  the  tomb  of  that 
conqueror.  In  the  time  of  the  Ptolemies  it  contained  the  bones 
of  no  less  than  forty  Egyptian  monarchs.  It  would  be  difficult  to 
Imagine  any  site  better  calculated  for  the  isolation  and  conceal- 
ment of  the  mummy  than  this  valley,  where  the  rocks  split  and 
crumble  under  the  sun,  and  the  sand  blown  hither  and  thither  by 
the    winds   from   the   desert   fills   up    every   crevice   in    the    cliffs. 

^  Fig.  172  reproduces  only  a  part  of  the  long  plate  given  in  Wilkinson.  In  order 
to  bring  the  more  important  groups  within  the  scope  of  one  page,  we  have  been 
compelled  to  omit  the  central  portion,  which  consists  principally  of  columns 
of  hieroglyphs. 


^ 

c*ri 

la, 


The  Tomb  under  the  New  Empire.  263 

Nothing  could  be  easier  than  to  mask  the  entrance  in  such  a  place, 
but,  on  the  other  hand,  no  constructed  building  of  any  importance 
was  possible  except  at  a  great  expenditure  of  time  and  labour.^ 

But  the  plain  at  the  foot  of  the  range  offered  all  that  the 
architect  could  wish.  It  was  still  within  the  district  consecrated 
to  the  dead,  and  yet  its  level  surface  presented  no  obstacles  to  an 
unlimited  extension  of  any  buildings  which  might  be  placed 
upon  it. 

We  have  here  then  the  facts  which  determined  the  course  of  the 
Egyptian  architects. 

In  the  space  inclosed  between  the  left  bank  of  the  river  and  the 
first  slopes  of  the  Libyan  chain,  certain  edifices  were  raised  which 
are  still,  in  great  part,  extant.  Their  funerary  signification  was 
never  completely  understood,  in  spite  of  the  confused  hints  to  that 
effect  given  by  the  Greek  writers,  until  within  the  last  few  years. 
To  Mariette  belongs  the  credit  of  having  at  last  removed  all  doubt 
on  the  subject.  It  is  probable  enough  that  the  number  of  these 
buildings  was  formerly  much  greater  than  it  is  at  present,  but 
those  which  have  come  down  to  us  are  still  in  sufficiently  good 
preservation  to  enable  us  to  discern  and  define  their  true  character, 
a  character  which  was  doubtless  common  to  all  the  temples  on  the 
left  bank  of  the  Nile. 

They  were  certainly  temples.  Their  general  arrangements  do 
not  differ  from  those  of  other  religious  edifices,  both  in  Thebes 
itself,  and  elsewhere  in  Egypt.  There  is,  however,  a  difference 
which  was  not  perceived  until  the  texts  which  contained  the 
history  of  each  temple  and  of  the  prince  who  claimed  the  credit  of 
its  erection  were  deciphered.  The  famous  buildings  at  Luxor  and 
Karnak  may  be  taken  as  typical  examples  of  the  temple,  properly 
speaking,  in  its  richest  and  most  complete  development.  The 
translation  of  the  inscriptions  and  royal  ovals  which  cover  their 
walls  has  sufficed  to  show  that  they  were  national  monuments, 
public  sanctuaries  consecrated  by  the  king,  as  the  representative  of 
the  people,  to  the  worship  of  those  great  deities  who  were  at  once 
the  principles  of  life  and  the  faithful  protectors  of  the  Egyptian 
race.  Century  after  century  they  never  ceased  to  found  such 
temples,  to  increase  and  to  embellish  them.  From  the  princes  of 
the    twelfth    dynasty  down    to    the   Ptolemies,  and    even   to    the 

1  See  the  description  of  the  Valley  of  the  Kings  in   the  Lettres  cT Egypte  d  de 
Ntibie  of  Champollion  (p.  183  of  the  second  edition). 


264  A   History  of  Art  in  Ancient  Egyrt. 

Roman  emperors,  every  successive  family  which  occupied  the 
throne  held  it  a  point  of  honour  to  add  to  the  creations  of  its 
predecessors.  One  prince  built  a  hypostyle  hall,  or  a  court 
surrounded  by  a  colonnade  ;  another  added  to  the  long  rows  of 
human  or  ram-headed  sphinxes  which  lined  the  approaches ;  a 
third  added  a  pylon,  and  a  fourth  a  laboriously  chiselled  obelisk. 
Some  kings,  who  reigned  in  periods  of  recuperation  after  civil  war 
or  barbaric  invasion,  set  themselves  to  repair  the  damage  caused 
by  time  and  the  violence  of  man.  They  strengthened  foundations, 
they  lifted  fallen  columns,  they  restored  the  faded  colours  of  the 
painted  decorations.  The  foreign  conquerors  themselves,  whether 
Ethiopians,  Persians,  or  Greeks,  as  soon  as  they  believed  them- 
selves to  have  a  firm  hold  upon  the  country,  set  themselves  with 
zeal  to  obliterate  the  traces  of  their  own  violence.  Each  of  these 
sovereigns,  whether  his  contribution  to  any  work  had  been  great 
or  small,  took  care  to  inscribe  his  own  name  upon  it,  and  thus 
to  call  upon  both  posterity  and  his  own  contemporaries  to  bear 
witness  to  his  piety. 

The  temple  as  we  see  it  at  Karnak  and  Luxor  is  the  collective 
and  successive  work  of  many  generations.  Such,  too,  was  the 
character  of  the  great  buildings  at  Memphis  which  were  con- 
secrated to  Ptah  and  Neith. 

But  on  the  left  bank  of  the  Nile,  and  in  the  neighbourhood  of 
the  Theban  necropolis,  we  find  a  group  of  temples  whose 
physiognomy  is  peculiar  to  themselves.  Nothing  exactly  like 
them  is  to  be  found  elsewhere  ;  ^  and  they  all  belong  to  one 
period,  that  of  the  three  great  Theban  dynasties,  the  eighteenth, 
nineteenth,  and  twentieth.^  "  These  temples  are  monuments 
raised  by  the  kings  themselves  to  their  own  glory.  They  are  not, 
like  the  temples  at  Karnak  and  Luxor,  the  accumulated  results  of 

^  Ebeks,  indeed,  found  something  of  the  same  kind  in  the  temple  of  Abydos.  He 
found  there  a  cenotaph  consecrated  to  his  own  memory  by  Seti  I.  This  cenotaph 
was  near  the  tomb  of  Osiris,  while  the  king  himself  was  buried  in  the  Theban 
necropolis.     {/Egypten,  pp.  234,  235.) 

2  The  beautiful  little  temple  of  Dayr-el-Medinet,  begun  by  Ptolemy  Philopator 
and  finished  by  his  successors,  especially  by  Physco,  has  often  been  considered  a 
funerary  monument.  It  is  alleged  that  the  situation  of  the  temple  in  the  necropolis, 
and  the  nature  of  the  subjects  represented  in  the  interior,  particularly  in  the  Western 
Chamber,  prove  that  it  was  so.  If  we  accept  this  opinion,  we  must  look  upon  the 
temple  as  a  mere  freak  of  fancy,  suggested  to  Ptolemy  Philopator  by  a  journey  to 
Thebes.  The  Greek  prince  was  interred  far  from  it,  and  it  could  have  formed  no 
l^art  of  his  tomb. 


The  Tomb  under  the  New  Empire. 


265 


several  generations.  Each  temple  was  begun  and  finished  by  the 
king  who  planned  it,  so  far  at  least  as  construction  was  concerned. 
In  those  cases  where  the  decoration  was  left  incomplete  at  the 
death  of  the  royal  builder,  his  successor  finished  it  in  his  name. 
In  these  decorations  the  founder  of  the  temple  was  represented 
either  worshipping  the  gods,  or  in  the  eventful  moments  of  his 
military  career,  or  in  his  great  hunts  ;  and  thus,  while  yet  alive,  he 
laid  the  foundations  of  an  edifice  destined  to  carry  the  memory  of 
his  glory  and  piety  down  to  the  latest  posterity."  ^ 


i'j  /  'ill 


.5»44lAii:^'^ji^'^>^  '^  -  ? 


''i ' 


■i     r 


Fig.  173. — Rameses  III.  hunting  ;  from  Medinet-Abou. 

Surrounded  on  all  sides  by  tombs  and  packed  Into  a  com- 
paratively narrow  space,  these  temples  are  separated  from  the 
Bab-el-Molouk  only  by  the  slopes  of  the  surrounding  El-Assassif. 
The  oldest  of  them  is  that  at  Dayr-el-Bahari.  It  was  built  by  the 
regent  Hatasu,  of  whose  career  we  know  enough  to  strongly 
excite,   but   too   little    to    satisfy,   our    curiosity.      We    know    that 

1  Mariette,  Deir-el-Bahari^  §   i.     (Atlas,  folio,  Leipsic,    1877,   with   40  pages 
letterpress,  4to.). 

VOL.    I.  M    M 


2  66  A  History  of  Art  in  Ancient  Egypt. 

Hatasu,  the  wife  and  sister  of  Thothmes  II.,  governed  Egypt  with 
skill  and  energy  for  seventeen  years,  in  trust  for  her  brother, 
Thothmes  III.  Where  does  her  mummy  repose  ?  Is  it  in  that 
ravine  on  the  south-west  of  the  Bab-el-Molouk  which  is  called 
the  Valley  of  the  Queens,  because  the  tombs  of  many  Theban 
princesses  have  been  found  in  it  ?  Or  is  it  in  the  slopes  of  the 
mountain  behind  the  temple  itself?  Numerous  sepulchral 
excavations  have  been  found  there,  and  many  mummies  have  been 
drawn  from  their  recesses.  The  artists  to  whom  the  decoration  of 
the  temple  was  committed,  were  charged  to  represent  the  chief 
actions  of  Hatasu  as  regent,  and,  although  their  works  do  not  give 
us  a  detailed  history  of  her  eminently  successful  administration, 
they  deal  at  length  with  the  enterprise  of  which  the  regent  herself 
seems  to  have  been  most  proud,  namely,  the  maritime  expedition 
against  Ptint,  a  distant  region  which  must  have  been  either 
southern  Arabia,  the  country  of  the  Somalis,  or  the  eastern  coast 
of  Africa. 

Next  in  point  of  age  to  the  building  of  Queen   Hatasu  is   that 
which  is  called  the  Ra7nesse2iin  ;  this  is  no  other,  as  the  members  of 
the  Institut  d'Egypte  have  clearly  proved,  than  the  so-called  Tomb 
of  Osyinandias   which   is  described   at  such  length  by  Diodorus.^ 
Erroneous  though  it  be,  this  latter   designation  is   by  no   means 
without  interest,  as  it  proves  that,  at  the  time  of  Diodorus,  per- 
sistent tradition  ascribed  a   funerary  origin  to   the   edifice.     The 
whole  temple,  inside  and   out,  recalled    Rameses    II.  ;  the   great 
conqueror    seemed    to    live    and    breathe    on    every    stone ;    here 
majestic   and    calm,   like    force    in    repose,    there    menacing    and 
terrible,  with  his  threatening  hand  raised  over  the  heads  of  his 
conquered    enemies.      His  seated  statue,    fifty-six  feet    in    height, 
was    raised    in    the    courtyard ;    to-day    it    lies    broken    upon    the 
ground.      Battle  scenes  are  to  be  distinguished  upon  the  remains 
of  the  walls.     An  episode  of  the  war  against  the  Khetas  may  be 
recognized,  which  seems  to  have  made  a  great  impression  upon 
the  kinsf  and  his  comrades   in  arms.      It   deals  with   that   battle 
fought  upon  the  bank  of  the  Orontes,   in  which   Rameses,  when 
surrounded   by  the   enemy,   won   safety  for  himself  by  his    own 
personal  valour  and   presence   of  mind.      His  prowess  was  cele- 
brated by  Pentaour,  a  contemporary  poet,  in  an  epic  canto  which 
has  survived  to  our  day.     Rameses  is  there  made  to  ascribe  his 

Diodorus,  i.  §§  47-49. 


The  Tomb  under  the  New  Empire.  267 

safety  and  all  the  honour  of  his  victory  to  his  father  Amen,  who 
heard  his  appeal  for  help  and,  precipitating  himself  into  the  7neUe, 
snatched  him  from  the  very  hands  of  his  enemies. 

Medinet-Abou,  which  might  be  called  The  Second  Ramesseum,  is 
to  Rameses  III.  what  the  pretended  tomb  of  Osymandias  is  to 
Rameses  II.  His  presence  pervades  both  the  temple  itself  and 
its  adjoining  pavilion.  Its  bas-reliefs  represent  one  of  the  greatest 
events  in  Egyptian,  we  might  almost  say,  in  ancient,  history , 
namely,  the  victory  won  by  Rameses  over  a  confederation  of  the 
nations  of  the  north  and  west,  of  those  who  were  called  the 
maritime  races.  This  victory  was  mainly  instrumental  in  driving 
westwards  certain  peoples  who  were  destined,  in  more  recent 
times,  to  play  a  great  part  in  the  politics  of  the  Mediterranean. 

Each  of  the  buildings  which  we  have  just  noticed  had  but  a 
single  proprietor.  They  were  each  dedicated  to  the  memory  of 
some  one  individual ;  but  there  was  nothing  to  prevent  the 
association  in  a  single  temple  of  two  sovereigns  who  might  happen 
to  be  united  by  strong  ties  of  blood,  and  this  course  was  taken  in 
the  temple  of  Gournah,  situated  in  the  same  district  of  Thebes. 
It  was  commenced  by  Rameses  I.,  the  founder  of  the  nineteenth 
dynasty,  continued  by  his  son,  Seti  I.,  and  finished  by  his  grand- 
son, Rameses  II.  The  first  Rameses  and  Seti  figure  in  it  with 
the  attributes  of  Osiris.  The  inscriptions  enumerate  the  sources 
of  revenue  set  aside  by  the  king,  in  each  nome,  for  the  service  of 
the  annual  sacrificial  celebrations  ;  and  thus  the  building  reveals 
itself  as  a  temple  to  the  perpetual  honour  of  the  two  first  princes 
of  a  race  which  did  so  much  to  add  to  the  greatness  and  prosperity 
of  Egypt. 

The  famous  colossi  of  Amenophis  III.,  known  to  the  ancients 
as  the  Statues  of  Memnon,  no  doubt  formed  part  of  a  similar 
building  (fig.  18  and  pi.  vi).  The  temple  built  by  this  prince 
near  the  site  of  the  Ramesseum  has  almost  entirely  disappeared, 
but  the  slight  traces  which  still  exist  cover  a  vast  space,  and 
suggest  that  the  building  must  have  been  one  of  rare  magni- 
ficence.^ 

1  This  must  have  been  the  structure  which  Strabo  calls  the  Memnonium,  and  near 
to  which  he  seems  to  place  the  two  colossi  (xvii.  p.  816).  The  true  name  of  the 
author  of  both  temple  and  colossi  might  easily  be  confused  with  that  of  the  mythical 
Greek  personage  which  the  Hellenic  imagination  persisted  in  discovering  everywhere 
in  Egypt,  and  the  similarity  of  sound  must  have  helped  to  perpetuate  the  mistake 
among  all  the  foreign  travellers  who  visited   the  country.     A  curious  passage  in 


268  A  History  of  Art  in  Ancient  Egypt. 


Only  one  of  those  Theban  temples  which  rise  upon  the  left 
bank  of  the  river  is  free  from  all  trace  of  a  funerary  or 
commemorative  purpose,  namely,  the  temple  at  Medinet-Abou 
which  bears  the  ovals  of  Thothmes  II.  and  Thothmes  III.  It 
shows  signs,  moreover,  of  having  been  frequently  enlarged  and 
added  to,  some  of  the  additions  having  been  made  as  recently 
as  in  the  time  of  the  Roman  Emperors.  In  this  respect  it 
resembles,  in  spite  of  its  comparatively  small  size,  the  great 
temples  upon  the  right  bank  of  the  river.  Like  them,  its  creation 
was  a  gradual  and  impersonal  matter  Every  century  added 
its  stone,  and  each  successive  king  engraved  his  name  upon  its 
walls.  How  to  account  for  its  exceptional  situation  we  do  not 
know.  It  is  possible  that  those  funerary  temples  of  which  we 
have  spoken  w^ere  an  original  invention  of  the  successors  of 
Thothmes ;  perhaps  that  constructed  by  Hatasu  at  Dayr-el- 
Bahari   was   the   first  of  the  series. 

However  this  may  have  been,  the  new  type  became  a  success 
as  soon  as  it  was  invented ;  all  the  other  temples  in  the  district 
may  be  more  or  less  immediately  referred  to  it.  The  great 
deities  of  Egypt,  and  more  particularly  those  of  Thebes,  are 
never  forgotten  in  them.  They  contain  numerous  representations 
of  the  princes  in  whose  honour  they  were  erected  performing 
acts  of  worship  before  Amen-Ra,  the  Theban  god  par  excllence, 
who  is  often  accompanied  by  Mout  and  Khons,  the  other  two 
members  of  the  Theban  triad.  These  temples  were  therefore 
consecrated,  like  almost  all  the  other  sacred  buildings  of  Thebes, 
to  those  local  deities  which,  after  the  establishment  of  that  city 
as  the  capital  of  the  \\  hole  country,  became  the  supreme  national 
gods.     Those  gods  were  as   much    at    home    in    the    temples    of 

Pausanias  {Attica,  42)  shows  us,  however,  that  the  Egyptian  scholars  of  his  time 
knew  how  properly  to  convey  the  name  of  the  prince  represented  in  the  colossi  to 
foreigners  :  "  I  was  less  struck  by  that  marvel,"  he  says,  in  speaking  of  some  sonorous 
stone  which  was  shown  to  him  at  Megara,  "  than  by  a  colossal  statue  which  I  saAV 
beyond  the  Nile  in  Egypt,  not  far  from  the  pipes.  This  colossus  is  a  statue  of 
the  sun,  or  of  Memnon,  according  to  the  common  tradition.  It  is  said  that  Memnon 
came  from  Ethiopia  into  Egypt,  and  that  he  penetrated  as  far  as  Susa.  But  the 
Thebans  themselves   deny  that  it  is    Memnon.      They  declare  that  it  represents 

Phamenoph  ($a/>tevo^),   who  was  born  in  their  own  country "     The  story 

told  by  Philostratus  (^Life  of  Apollonius,  1.  vi.  p.  232)  of  the  visit  of  the  sorcerer  to 
Memnon,  shows  that  in  his  time  the  colossus  was  surrounded  by  nothing  but  ruins, 
such  as  broken  columns  and  architraves,  fragmentary  walls  and  shattered  statues. 
Even  then  the  monumental  completeness  of  the  "  Amenophium  "  had  vanished. 


The  Tomb  under  the  New  Empire.  269 

which  we  are  speaking  as  in  their  own  pecuHar  sanctuaries  on 
the  right  bank  of  the  river.  In  both  places  they  received  the 
same  homage  and  sacrifices,  but  in  the  funerary  temples  of  the 
left  bank  they  found  themselves  associated,  paredral  as  the 
Greeks  would  say,  with  the  princes  to  whose  memory  the  temples 
were  raised.  These  princes  were  represented  with  the  attributes 
of  Osiris,  both  in  the  statues  which  were  placed  against  the  piers 
in  the  courtyard  and  in  the  bas-reliefs  upon  the  long  flat 
surfaces  of  the  walls.  By  these  attributes  they  became  more 
closely  allied  with  the  great  deity  who  was  the  common 
protector  of  the  dead  and  the  guarantor  of  their  future 
resurrection.  In  this  capacity  the  deceased  prince  was  wor- 
shipped as  a  god  by  his  own  family.  Thus,  in  the  temple 
of  Gournah,  we  find  Rameses  I.  seated  in  a  naos  and  receiving 
the  homage  of  his  grandson,  Rameses  II.  ;  and,  again,  the  latter 
worshipping  Amen-Ra,  Khons,  and  Rameses  I.  at  one  and 
the  same  time.  This  presentation  of  offerings  to  the  deified 
king,  as  represented  in  the  chambers  of  these  temples,  recalls 
the  scene  which  is  carved  upon  almost  all  the  steles,  and  with 
greater  variety  and  more  detail,  in  the  bas-reliefs  on  the  internal 
walls  of  the  mastabas. 

The  analogy  which  we  are  endeavouring  to  establish  between 
the  western  temples  at  Thebes  and  the  funerary  chambers  of 
the  private  tombs,  is  completed  by  the  biographical  nature  ot 
the  pictures  which  form  almost  the  sole  decoration  of  those 
temples.  The  images  presented  to  our  gaze  by  the  chamber  walls 
of  the  mastaba  are  not,  indeed,  so  personal  and  anecdotical  as 
those  of  the  temples,  but  they  contain  an  epitomized  representa- 
tion of  the  every-day  life,  of  the  pleasures  and  the  more  serious 
occupations  of  a  rich  Egyptian.  It  is  easy  to  understand  how, 
with  the  progress  of  civilization,  the  more  historic  incidents  in 
the  life  of  an  individual,  and  especially  when  that  individual 
was  a  king,  came  to  be  figured  in  preference  to  those  which  were 
more  general  in  their  application.  To  embellish  the  tomb  of  a 
conqueror  with  pictures  of  his  battles  and  victories  was  to 
surround  him  after  death  with  the  images,  at  least,  of  those  things 
which  made  his  happiness  or  his  honour  while  alive.  Pictures 
of  some  famous  feat  of  arms  would  give  joy  to  the  dottblc  of  him 
who  had  performed  them,  and  would  help  to  relieve  the  ennui 
of  the  monotonous  life  after  death.      Hence  the  tendency  which 


270  A  History  of  Art  in  Ancient  Egypt. 

is  so  marked  in  the  bas-reliefs  of  the  first  Theban  Empire, 
especially  in  those  at  Beni-Hassan.  The  constant  and  universal 
themes  which  sufficed  for  the  early  centuries  of  the  Egyptian 
monarchy  were  not  abandoned  ;  scenes  similar  to  those  of  the 
mastabas,  are,  indeed,  frequently  met  with  in  the  Theban  tombs  ; 
but  it  is  evident  that  in  many  cases  scenes  were  sought  out 
for  reproduction  which  would  have  a  more  particular  application  ; 
there  is  an  evident  desire  to  hand  down  to  future  generations 
concrete  presentments  of  any  political  or  other  events  which 
might  appear  worthy  of  remembrance.  "History  and  biography 
thus  came  in  time  to  play  an  important  part  in  sepulchral  decora- 
tion, especially  w^hen  kings  or  other  royal  personages  were 
concerned. 

Similar  pictures  are  to  be  met  with  here  and  there  in  temples 
proper,  as  may  be  seen  by  a  glance  at  the  bas-relief  figured 
upon  the  opposite  page  (Fig.  174),  but  in  such  cases  they  are 
invariably  on  the  outer  walls.  At  Luxor,  for  instance,  the 
campaigns  of  Rameses  II.  against  the  peoples  of  Syria  are 
thus  displayed  ;  and  at  Karnak  it  is  upon  the  external  walls  of 
the  hypostyle  hall  that  the  victories  of  Seti  I.  and  Rameses  H. 
are  sculptured.  In  the  interiors  of  all  these  courts  and  halls  we 
hardly  find  any  subjects  treated  but  those  which  are  purely 
religious  ;  such  as  female  deities  assisting  at  the  birth  of  a 
king,  or  taking  him  upon  their  knees  and  nourishing  him  from 
their  breasts,  a  theme  which  is  also  found  in  royal  tombs  (Fig.  175); 
or  one  god  presenting  the  king  to  another  (Fig.  2)3)  ',  or  the  king 
paying  homage  to  sometimes  one,  sometimes  another,  of  his 
divine  protectors  (Figs.  14  and  176).  We  find  such  religious 
motives  as  these  continually  repeated,  upon  wall  and  column 
alike,  from  the  first  Theban  kings  to  the  epoch  of  the  Ptolemies. 
On  the  right  bank  of  the  river  pictures  of  a  mystically  religious 
character  are  universal ;  on  the  left  bank  those  with  an  histori- 
cal aim  are  more  frequent. 

It  will  be  seen  that  the  difterence  between  the  two  kinds  of 
temple,  between  that  of  the  necropolis  and  that  of  the  city,  is 
not  so  striking  and  conspicuous  as  to  be  readily  perceived  by  the 
first  comer  who  crosses  from  the  one  bank  of  the  river  to  the 
other  ;  but  the  variations  are  quite  sufficiently  marked  to  justify 
the  distinction  propounded  by  Mariette.  According  to  him  the 
temples   in  the   necropolis  are  funerary   chapels   which   owe  their 


CO 


Oh 
C 

_o 

'o 
&, 

S 

cS 

u 


o 

h-1 


£ 


The  Tomb  under  the  New  Empire. 


273 


increased  size  and  the  richness  of  their  decoration  to  the  Qreneral 
magnificence  and  highly  developed  taste  of  the  century  In  which 
they  were  built.  But  it  is  enough  for  our  present  purpose  to  have 
indicated  the  places  which  they  occupied  in  the  vast  architectural 
compositions  which   formed  the   tombs   of  a  Seti  or   a    Rameses. 


Eu;.    175. — Painting  in   a  royal   tomb  at   Gournab.     Amenop.bis  11.  upon  tbe  lap  of  a  goddess. 

(Cbampollion,  pi.  160.) 


They  had  each  a  double  function  to  fulfil.  They  were  foundations 
made  to  the  perpetual  honour  of  a  deceased  king,  chapels  in  which 
his  fete-day  could  be  kept  and  the  memory  of  his  achievements 
renewed  ;  but  they  were  at  the  same  time  temples  in  which  the 
national  gods  were  worshipped  by  himself  and  his  descendants,  in 
which   those  gods  were  perpetually  adored  for  the  services  which 

VOL.     I.  ■  X    X 


274 


A  History  of  Art  in  Ancient  Egypt. 


they  had  done  him  while  ahve  and  for  those  which  they  might 
still  do  him  when  dead.  In  their  latter  capacity  these  buildings 
have  a  right  to  be  considered  temples,  and  we  shall  defer  the 
consideration   of   their    architectural    arrangements,    which     differ 

only  in  details  from  those  of 
the  purely  religious  buildings, 
until  we  come  to  speak  of  the 
religious  architecture  of  Egypt. 
We  shall  here  content  our- 
selves with  remarking  that  the 
separation  of  the  tomb  and  the 
funerary  chapel  by  some  mile 
or  mile  and  a  half  was  a  novelty 
in  Egypt.  The  different  parts 
of  the  royal  tomb  were  closely 
connected  under  the  Memphite 
Empire,  and  the  change  in 
arrangement  must  have  been  a 
consequence  of  some  modifica- 
tion in  the  Egyptian  notions  as 
to  a  second  life. 

In  the  mastaba  the  double 
had  everything  within  reach  of 
his  hand.  Without  trouble  to 
himself  he  could  make  use  of 
all  of  the  matters  which  had 
been  provided  for  the  support 
of  his  precarious  existence :  the 
corpse  in  the  mummy  pit,  the 
statues  in  the  serdab,  the  por- 
traits in  bas-relief  upon  the 
walls  of  the  public  chamber. 
Through  the  chinks  between 
the  pieces  of  broken  stone  by 
which  the  well  was  filled  up, 
and  through  the  conduits  con- 
trived in  the  thickness  of  the  walls,  the  magic  formulae  of  the 
funerary  prayers,  the  grateful  scent  of  the  incense,  and  of  the 
burnt  fat  of  the  victims  (Fig.  177),  reached  his  attentive  senses. 
Brought  thus  into  juxtaposition  one  with  another,   the  elements 


Fig.  176. — Amenophis  III.  presenting  an  offering 
to  Amen.  Decoration  of  a  pier  at  Thebes  ; 
from  Prisse. 


The  Tomb  under  the  New  Empire. 


275 


of  ihe  tomb  were  mutually  helpful.  They  lent  themselves  to 
that  intermittent  act'  of  condensation,  so  to  speak,  which  from 
time  to  time  gave  renewed  substance  and  consistency  to  the 
phantom  upon  which  the  future  life  of  the  deceased  depended. 
This  concentration  of  all  the  acts  and  objects,  which  had  for 
their  aim  the  preservation  of  the  deceased  for  a  second  term 
of  life,  was  obviously  destroyed  as  soon  as  the  division  of  the 
tomb  into  two  parts  took  place.  The  mummy,  hidden  away 
in    the    depths    of    those    horizontal    wells    in    the    flank    of   the 

■  \-   ^-'rr'-  :'■■  ■.'■  ■■;^"m'S| 


Western  Range  of  which  w^e  have  spoken,  would  seem  to  be 
in  danoer  of  losincr  the  benefit  of  the  services  held  in  its  honour 
Upon  the  Theban  plain.  At  such  a  distance  it  would  neither  hear 
the  prayers  nor  catch  the  scent  of  the  offerings.  And  the  dotid/e  ? 
Is  it  to  be  supposed  that  he  oscillated  between  the  colossi  in  the 
temple  where  the  funerary  sites  were  celebrated,  and  the  chamber 
in  which  the  corpse  reposed  ? 

Before  they  could  have  accepted  this  division  of  the  tomb  into 
two  parts  the  Egyptians  must  have  arrived  at  some  less  childish 
conception  of  the  future  life  than  that  of  their  early  civilization. 


2/6  A   History  of  Art  in  Ancient  Egypt, 


That  primitive  conception  was  not  entirely  banished  from  their 
minds;  evidence  of  its  persistency  is,  indeed,  plentiful,  but  a  more 
intelligent  and  less  material  notion  gradually  superimposed  itself 
upon  the  ancient  belief.  The  indescribable  being  which  was 
the  representative  of  the  deceased  after  death  became  gradually 
less  material  and  more  spiritual ;  in  time  it  escaped  from  its 
enforced  sojourn  in  the  tomb  and  approached  more  nearly  to 
that  which  we  call  the  soul.  This  soul,  like  the  nocturnal  sun, 
passed  a  period  of  probation  and  purgation  in  the  under  world, 
and,  thanks  to  the  protection  of  Osiris  and  the  other  deities  of  the 
shades,  was  at  last  enabled  to  return  to  earth  and  rejoin  the  body 
which  it  had  formerly  inhabited.  The  problem  of  death  and 
a  future  life  was  resolved  in  much  the  same  way  by  the  Greeks 
and  by  all  other  races  who  drew  much  of  their  inspiration  from  the 
Egyptians.  They  all  looked  upon  the  corpse  as  still  alive  when 
they  expressed  their  hopes  that  the  earth  upon  which  they  poured 
out  wine  and  milk  would  like  lie  lightly  upon  it.  After  a  time 
they  added  Tartarus  and  the  Elysian  Fields  to  their  beliefs,  they 
introduced  the  heroic  fathers  of  their  race  into  the  councils  of 
the  gods,  and  they  described  and  figured  the  joys  which  awaited 
the  just  upon  the  Happy  Islands. 

These  various  hypotheses  are  contradictory  enough  from  a 
logical  point  of  view  ;  they  exclude  and  destroy  one  another. 
But  when  it  is  a  question  of  notions  which  are  essentially  incapable 
of  being  strictly  defined,  the  human  intelligence  is  singularly 
content  to  rest  in  vasfue  o-eneralities.  Contradictions  do  not 
embarrass  it ;  its  adaptability  is  practically  infinite. 

The  beliefs  which  we  have  just  described  tended  for  many 
centuries  to  become  more  and  more  general.  They  were  taught 
in  that  Ritual  of  the  Dead  which,  although  certain  of  its  parts  date 
from  the  most  ancient  times,  did  not  take  its  complete  and  definite 
form  until  the  Theban  epoch.  Being  more  spiritual  and  less 
material,  they  were  less  opposed  to  the  subdivision  of  the  sepulchre 
than  the  more  primitive  idea  ;  and  this  subdivision  was  necessary 
if  the  public  and  commemorative  part  of  the  tomb  were  to  receive 
a  splendour  and  amplitude  befitting  the  exploits  of  a  Thothmes, 
a  Seti,  or  a  Rameses.  Dayr-el-Bahari  proves  that  the  change  had 
already  been  made  under  the  eighteenth  dynasty,  but  it  was  not 
until  the  nineteenth  that  it  became  definitely  adopted.  The 
progress  of  ideas  and  of  art  had  then  advanced  so  far,  that  more 


The  Tomb  under  the  New  Empire. 


2/7 


ambitious  desires  could  be  satisfied,  and  the  country  filled  with 
magnificent  edifices,  which,  like  the  temples  of  the  two  Rameses, 
were  original  in  so  far  as  they  belonged  at  one  and  the  same  time 
to  religious  and  funerary  architecture.  We  should  call  them 
cenotaphs,  were  it  not  that  the  Egyptians,  like  all  the  other  races 
of  antiquity,  believed  in  the  real  presence  of  their  dead  in  the 
buildings  erected  in  their  honour. 


■  "''V^1\(»?^Sii 


\  m^^M& 


mm 


I  '■  ■  ^ 

'^1 

1  .' '    ■ 

■'■i^ 

;'^.- 

"'mm 


ip.: 


Fig.  178. — Entrance  to  a  royal  tomb.     {DcscripUon  de  TEgyptc,  ii.,  \A.  79.) 


The  other  division  of  the  tomb  is  that  wliich  contains  the  well 
and  the  mummy-chamber,  the  eternal  dwelling-place  of  the 
illustrious  dead.  The  second  half  of  the  royal  sepulchre  had 
,to  be  as  sumptuous  and  luxurious  in  its  way  as  the  first,  but 
the  problem  placed  before  the  architect  was  diametrically  opposed 
to  that  which  he  had  to  solve  in  the  other  part  of  his  task.  In  con- 
structing and  decorating  the  funerary  temple  upon  the  plain,  he 
was  working  before  the  eyes  of  the  pubKc,  for  their  benefit  and 
for  that  of  the  remotest  posterity. 


2/8  A   History  of  Art  in  Ancient   Egypt. 


But  the  task  of  hewing  out  the  tomb  was  a  very  different  one. 
For  long  years  together  he  pursued  his  enterprise  in  the  mystery 
and  shadow  of  a  subterranean  workshop,  to  which  all  access  was  no 
doubt  forbidden  to  the  curious.  He  and  his  assistants  cut  and 
carved  the  living  rock  by  the  light  of  torches,  and  his  best  ingenuity 
was  taxed  to  devise  means  for  preserving  from  the  sight  of 
all  future  generations  those  works  of  the  best  artists  of  Egypt 
with  which  the  walls  were  to  be  covered.  Those  prodigies  of 
patience  and  skill  were  executed  for  the  benefit  of  the  deceased 
alone.  Important  though  it  was  that  the  sepulchre  of  a  great  man 
should  be  ornamented  to  the  greatest  extent  possible,  it  was  of  still 
greater  moment  that  his  last  resting-place  should  not  be  troubled 
by  the  visits  of  the  living  ;  and  the  more  completely  the  mummy 
was  concealed,  the  greater  were  the  deserts  of  the  faithful  servant 
upon  whom  the  task  had  been  placed. 

In  order  that  this  blessing  of  undisturbed  peace  in  his  eternal 
dwelling  should  be  secured,  the  royal  tomb  seems  to  have  been  con- 
structed without  any  such  external  show  as  would  call  attention  to 
its  situation.  The  tombs  of  private  individuals  usually  had  a 
walled  courtyard  In  front  of  them  to  which  access  was  obtained  by 
a  kind  of  porch,  or  tower,  with  inclined  sides  and  crowned  by 
a  small  pyramid.  But  the  explorers,  Belzoni,  Bruce  and  others, 
who  disengaged  the  entrances  to  the  royal  tombs,  found  them 
without  propylsea  of  any  kind.^  The  doorway,  cut  vertically  in 
the  rock,  is  of  the  utmost  simplicity,  and  we  have  every  reason  to 
suppose  that,  after  the  introduction  of  the  mummy,  it  was  carefully 
masked  with  sand  and  rocky  debris?'  ^ 

The  existence  of  the  temples  in  the  plain  made  it  unnecessary 
that    the    tombs    themselves    should   be    entered    after    that    final 

1  In  many  cases  the  sites  are  wanting  for  such  external  constructions.  The  fine 
tomb  of  Seti  I.,  for  instance,  opens  upon  a  ravine  which  is  filled  with  the  waters 
of  a  mountain  torrent  at  certain  seasons. 

2  When  Belzoni's  workmen  found  the  entrance  to  the  tomb  of  Seti,  they  declared 
that  they  could  not  advance  any  farther,  because  the  passage  was  blocked  with 
big  stones  to  such  an  extent  as  to  be  impracticable  {^Narrative  of  the  Operations 
and  Recent  Discoveries  within  the  Pyramids,  d^r.,  /;/  -Egypt  and  Nubia,  1820,  4to). 
Mariette  also  believed  that  as  soon  as  the  mummy  was  in  place,  the  external  door 
was  closed  and  earth  heaped  against  it  in  such  a  way  as  effectually  to  conceal 
it.  It  is  thus  that  the  clashing  between  the  tomb  of  Rameses  III.  and  another 
is  accounted  for.  The  workmen  did  not  see  the  entrance  of  the  latter,  and  were,  in 
fact,  unaware  of  its  existence  until  they  encountered  it  in  the  bowels  of  the  rock. 
[Voyage  dans  la  Haute-Egypte,  t.  ii.  p.  81.) 


The  Tomb  under  the  New  Empire.  279 

operation  had  been  performed.  Some  words  of  Diodorus  are 
significant  in  this  direction.  "  The  priests  say  that  their  registers 
attest  the  existence  of  forty-seven  royal  tombs,  but  that  at  the 
time  of  Ptolemy  the  son  of  Lagus,  only  seventeen  remained." 
This  assertion  cannot  be  accepted  literally,  because  twenty-one 
tombs  have  already  been  discovered  in  the  Bab-el-Molouk,  some  of 
them  in  a  state  of  semi-completion,  besides  four  in  the  ravine 
which  is  called  the  Valley  of  the  West,  which  makes  twenty-five  in 
all.  What  the  priests  meant  when  they  spoke  to  Diodorus  was 
no  doubt,  that  at  the  time  of  the  Ptolemies,  no  more  than  seventeen 
of  their  entrances  had  been  discovered.  If  through  the  plans  made 
for  their  construction  and  preserved  in  the  national  archives  there 
were  some  who  knew  their  situation,  they  preserved  the  secret. 
We  know,  by  the  inscriptions  upon  their  walls,  that  fifteen  of  the 
tombs  which  are  now  accessible,  were  open  in  the  time  of  the 
Ptolemies ;  several  of  them  seem  to  have  been  shown,  to  the 
Roman  and  other  travellers  who  visited  Egypt,  as  national  objects 
of  interest.^ 

The  precautions  taken  to  hide  and  obstruct  the  openings  of  the 
royal  tombs  were  thus  successful  in  many  cases.  Some  of  these 
have  only  been  discovered  in  our  own  times,  through  the  ardour 
and  patience  w4iich  characterize  modern  research,  and  we  have 
still  good  reason  to  suppose  that  there  are  others  which  yet 
remain  to  be  found.  In  1872  Professor  Ebers  discovered  a 
beautiful  private  tomb,  that  of  Anemenheb,  which,  although 
situated  close  to  one  of  the  most  frequented  paths  in  the 
necropolis,  had  been  previously  unknown.  It  was  open,  but  the 
opening  had  been  carefully  concealed  with  rough  pieces  of  rock 
and  general  rubbish  by  the  fellahs,  who  used  the  tomb  as  a  hiding- 
place  from  the  recruiting  officers  of  the  viceroy.  They  would 
remain  concealed  in  it  for  weeks  at. a  time  until  the  officers  had 
left  their  village.  The  royal  cemetery  of  the  Ramessides  has 
possibly  much  more  to  tell  us  before  its   secrets   are  exhausted. 

The  entrance  to  the  tomb  always  ran  a  certain  chance  of  being 
discovered  and  freed  from  its  obstacles.  It  was  difficult,  of 
course,    to   prevent    the    survival    of    some    tradition    as    to    the 

^  Diodorus,  i.  46. 

-  "Above  the  Memnonium,"  says  Strabo  (xvii.  46),  "there  are  royal  tombs  cut 
in  the  living  rock  to  the  number  of  forty ;  their  workmanship  is  excellent  and  well 
worthy  of  attention." 


28o  A   History  of  Art  in  Ancient  Egypt. 

whereabouts  of  the  burial-places  of  those  great  sovereigns  whose 
memory  w^as  a  consolation  to  Egyptian  pride  in  the  clays  of 
national  abasement  and  decay.  Provision  had  to  be  made,  as  in 
the  case  of  the  pyramids,  against  a  forced  entry  into  the  gallery 
either  by  an  enemy  or  by  some  robber  in  search  of  treasure,  and 
we  find  that  the  precautions  adopted  were  similar  to  those  which 
we  have  described  in  noticing  the  royal  tombs  at  Memphis.  Let 
us  take  as  an  example  the  finest  and  most  complete  of  all  the 
tombs  of  the  Ramessides,  that  of  Seti  I.  After  descending  two 
flights  of  steps,  and  traversing  two  long  and  richly  decorated 
corridors,  Belzoni  arrived,  without  discovering  either  sarcophagus 
or  anything  that  looked  like  the  site  of  a  sarcophagus,  at  an 
oblong  chamber  13  feet  6  inches  by  12  feet.  A  wide  and  deep 
well,  which  here  barred  the  passage,  seemed  to  indicate  that  the 
extremity  of  the  excavation  had  been  reached.  Belzoni  caused 
himself  to  be  lowered  into  the  well.  The  walls  were  everywhere 
hard  and  firm,  and  without  resonance,  and  there  was  no  sign  of  a 
passage,  either  open  or  concealed,  by  which  access  to  a  lateral 
chamber,  or  to  a  second  series  of  cralleries  mio-ht  be  obtained. 
But  Belzoni  was  too  old  an  explorer  to  be  deceived  by  such 
appearances.  On  his  first  arrival  at  the  edge  of  the  well  he  had 
perceived  in  the  wall  on  the  farther  side  of  it  a  small  opening, 
about  two  feet  wide,  and  two  feet  and  a  half  high.  This  had  been 
made,  at  some  unknown  period,  in  a  wall  covered  with  stucco  and 
painted  decorations.  Across  the  well  a  beam  was  still  lying,  which 
had  served  the  purpose  of  some  previous  visitor  to  the  tomb.  A 
cord  hung  from  this  beam,  and  it  was  after  discovering  that  the 
well  ended  in  nothing  that  the  screen  of  masonry  on  the  other 
side  had  been  pierced.  Belzoni  had  therefore  only  to  follow  the 
road  opened  for  him  by  earlier  explorers.  A  plank  bridge  was 
thrown  across  the  well,  the  opening  was  enlarged,  and  a  new 
series  of  galleries  and  chambers  was  reached,  which  led  at  last  to 
the  sarcophagus-chamber  itself^ 

Belzoni  remarked  that  throughout  the  whole  course  of  the 
excavation  the  doors  of  the  chambers  showed  evidence  of  having 
been  walled  up,  and  that  upon  the  first  steps  of  one  of  the  stair- 
cases a  heap  of  stone  rubbish  had  been  collected,  as  if  to  dis- 
courage any  one  who  might  penetrate  beyond  the  well  and  pierce 
the  barrier  beyond  its  gaping  mouth.  It  seems  likely  that  the 
^   Belzonf,  Narrative  of  Operations,  G>r.,  pp.  233  r/. ?<-</. 


The  Tomb  under  the  New  EiMpire.  281 

first  violator  of  the  tomb  knew  the  secret  of  all  these  arranee- 
ments,  and  consequently  that  its  first  opening  took  place  in 
very  ancient  times,  and  was  the  work  of  some  native  Egyptian 
robber. 

In  the  sarcophagus-chamber  Belzoni  discovered  a  contrivance 
of  the  same  kind  as  that  which  had  failed  to  stop  him  almost  upon 
the  threshold  of  the  tomb.  The  sarcophagus  of  oriental  alabaster 
was  in  place,  but  empty ;  the  lid  had  been  raised  and  broken. ^ 
From  the  sound  given  out  by  the  floor  when  struck  the  explorer 
perceived  that  there  must  be  a  hollow  space  under  the  base  of  the 
sarcophagus.  He  cut  a  hole  and  brought  to  light  the  first  steps  of 
a  staircase,  which  led  to  an  inclined  plane  by  which  the  interior  of 
the  mountain  was  deeply  penetrated.  A  wall  had  been  raised  at 
the  foot  of  these  steps,  beyond  which  a  settlement  of  the  super- 
incumbent rock  put  an  end  to  all  advance  after  a  distance  of  fifty- 
one  yards  had  been  traversed.  Is  it  not  possible  that  Belzoni 
only  discovered  a  false  sarcophagus,  placed  to  deceive  unbidden 
visitors  like  himself,  and  that  the  mummy  was  deposited,  and  still 
lies,  in  a  chamber  at  the  end  of  this  corridor  ?  The  point  at  which 
the  fallen  rock  arrested  his  progress  is  four  hundred  and  eighty-three 
feet  from  the  external  opening,  and  about  one  hundred  and  eighty 
below  the  level  of  the  valley.  At  such  a  depth,  in  these  narrow 
and  heated  galleries,  where  there  is  no  ventilation  and  where  the 
smoke  of  the  torches  rapidly  becomes  stifling,  it  is  not  astonishing 
that,  in  spite  of  his  admirable  perseverance,  Belzoni  held  his  hand 
before  completing  the  exploration.^ 

These  subterranean  tombs  are  hardly  less  astonishing  than  the 
colossal  masses  of  the  pyramids  for  the  sustained  effort  which  they 
imply  ;  if  we  take  the  trouble  to  reflect  upon  the  peculiarly  difficult 
conditions  under  which  they  were  constructed,  they  may  even 
impress  our  imaginations  more  profoundly  than  the  artificial 
mountains  of  Cheops  and  Chephren.  We  have  already  men- 
tioned a  figure  which  gives  some  idea  of  the  surprising  length  of 
their  passages  ;  and  although  no  one  of  the  other  tombs  quite 
equals  that  of  Seti,  many  approach  it  in  dimensions.  The  tomb 
of  Rameses    III.    is  416  feet  long,   that  of  Siptah   370   feet,  and 

^  This  beautiful   sarcophagus  is  now  in  the  Soane  Museum. — Ed. 

2  Belzoni  believed  that  this  passage  led  again  into  the  open  air;  that  it  was,  in 
fact,  another  entrance  to  the  tomb.  "  I  have,"  he  says,  "reasons  to  think  so  ;"  but 
he  does  not  give  his  reasons. 

VOL.    I.  00 


!82 


A   History  of  Art  in  Ancient  Egypt. 


others  varied  between  200  and  270  feet.  For  the  construction 
of  such  places  an  enormous  number  of  cubic  yards  of  rocky  debris 
had  to  be  cut  from  the  interior  of  the  mountain,  and  carried  up  by 
narrow  and  steep  corridors  to  be  "  shot "  in  the  open  air.  Still 
more  surprising  is  the  elegance  and  completeness  of  the  decora- 
tion. In  the  tombs  of  Seti  and  of  Rameses  III.  there  is  not  a 
single  surface,  whether  of  avails,  piers,  or  ceilings,  which  is  not 
covered  with  the  work  of  the  chisel  and  the  brush,  with  orna- 
mental designs,  with  the  figures   of  gods   and   genii,  of  men  and 


^/////y.>y.yyy//.^/////M///.f/^^^ 


yy/y/////////////: 
Fig.  179. — Plan  of  the  tomb  of  Rameses  IT.  ;  froEL  Prisse. 


, L 


10         -.S         ^0  tS  i!- 

FiG.  180. — Horizontal  section  of  the  same  tomb  ;  from  Prisse.^ 


animals.  These  figures  are  far  too  numerous  to  count.  They 
swarm  like  ants  in  an  anthill  ;  a  single  chamber  often  contains 
many  hundreds.  Colour  is  everywhere  ;  here  it  is  used  to  give 
salience  to  the  delicate  contours  of  the  figures  in  relief,  there  it  is 

1  The  tomb  of  Seti  having  been  so  often  reproduced,  we  have  thought  it  better 
to  give  the  plan  and  section  of  that  of  Rameses  II.,  which  is  less  generally  known. 
The  general  arrangements  are  pretty  much  the  same  as  those  of  Seti's  tomb,  but 
the  plan  is  a  little  more  complicated. 


The   Tomb  under  the  New  Empire. 


283 


laid  flat  upon  the  carefully-prepared  surfaces  of  white  stucco.  In 
these  sealed-up  caverns,  in  which  the  air  is  constantly  warm  and 
dry,  the  pictures  have  preserved  their  freshness  of  tint  in  the  most 
startling  fashion.  And  to  obtain  all  this  harmonious  effect  no 
light  but  an  artificial  one  was  available.  It  was  by  the  smoky 
glare  of  torches,  or  by  the  flickering  flame  of  little  terra-cotta 
lamps,  suspended  from  the  roof  by  metal  threads,  that  the  patient 
artists  of  Egypt  drew  these  masterly  contours,  and  elaborated  the 
exquisite    harmony    of   their    colour   compositions.      Egyptian   art 


Fig.    181. — The  smaller   sarcophagus-chamber   in   the   tomb   of  Rameses   VI.       (From 

Horeau,  pi.  21.)  ^ 


never  reached  greater  perfection  than  in  these  characteristic 
productions  of  its  genius,  and  yet  no  human  eye  was  to  enjoy 
them  after  that  day  upon  which  the  final  touch  was  to  be  given  to 
their  beauties,  upon  which  they  were  to  be  inclosed  in  a  night 
which,  it  was  hoped,  would  be  eternal. 

But  yet  all  this  work  was  not  labour  lost.  These  pictures, 
in  which  the  details  change  continually  from  one  tomb*  to  another, 
were  all  inspired  by  a  single  desire,  and  all  tended   to   the  same 

1  Panorama  de  PEgypte  et  de  la  Nubie,  folio. 


284 


A  History  of  Art  in  Ancient  Egypt. 


end.  Like  those  which  we  have  found  in  the  tombs  of  the 
Ancient  Empire,  they  had  a  sort  of  magic  virtue,  a  sovereign 
power  to  save  and  redeem.  The  personages  and  articles  of  food 
represented  on  the  mastabas  were  shadows  of  people  and  shadows 
of  material  sustenance,  destined  for  the  service  and  the  food  of 
a  shadow,  the  double  of  the  defunct  proprietor  of  the  tomb. 
The  all-powerful  influence  of  prayer  and  faith,  working  through 
Osiris,  turned  these  shadows  into  realities. 


Fig.  182. — Entrance  to  tlie  tomb  of  Rameses  III.     (From  Horeau,  pi.  21.) 


Representations  of  this  kind  are  common  enough  in  the  royal 
tombs  of  Thebes.  It  will  suffice  if  we  notice  those  which  are 
still  to  be  seen  in  the  sepulchre  of  Rameses  III.,  in  the  series 
of  small  chambers  in  the  first  two  passages.  Like  the  hunting 
scene  which  we  take  from  the  walls  of  a  private  tomb  (Fig.  183), 
these  pictures  have,  beyond  a  doubt,  the  same  meaning  and 
value  as  those  in  the  mastaba.  But  in  the  Theban  tombs  their 
significance  'is  only  secondary.  Ideas  had  progressed  to  some 
purpose  since  the  days  of  the  Memphite  kings.  Both  in  its 
general   arrangement   and    in    the    details    of    its    ornamentation, 


The  Tomb  under  the  New  Empire.  285 

the  sepulchres  in  the  Bab-el-Molouk  gave  expression  to  the  new, 
more  philosophical,  and  more  moral  conception  which  had  come 
to  overlie  the  primitive  beliefs. 

The  first  conception  was  that  of  the  double,  inhabiting  the 
tomb,  and  kept  alive  in  it  by  sacrifice  and  prayer.  But  in  time 
the  Egyptians  would  appear  to  have  realized  that  the  double 
was  not  the  only  thing  that  remained  after  the  death  of  a  human 
unit.  Their  powers  of  apprehension  were  quickened,  in  all 
probability,  by  that  high  moral  instinct  of  which  the  oldest 
pages  of  their  literature  give  evidence.  Good  or  bad,  every 
man  had  a  double,  the  continuance  and  prosperity  of  which 
depended  in  no  way  upon  his  merits  or  demerits.  Unless  the 
just  and  the  unjust  were  to  come  to  one  and  the  same  end, 
something  more  was  wanting.  This  something  was  the  soul 
iba).  Instead  of  vegetating  in  the  interior  of  the  tomb,  this 
soul  had  to  perform  a  long  and  difficult  subterranean  journey — ■ 
in  imitation  of  the  sun  and  almost  upon  his  footsteps — during 
which  it  had  to  undergo  certain  tests  and  penances.  From  this 
period  of  trial  it  would  emerge  with  more  or  less  honour, 
according  to  its  conduct  during  the  few  short  years  passed  by 
it  on  earth  and  in  company  with  the  body  to  which  it  had 
belonged.  It  had  to  appear  before  the  tribunal  of  Osiris-Khent- 
Ament,  the  Sun  of  Night,  around  whose  seat  the  forty-two 
members  of  the  infernal  jury  were  assembled.^ 

^  This  belief  in  the  appearance  of  the  dead  before  Osiris  and  his  assessors  gave 
rise  to  one  of  the  most  curious  errors  made  by  the  Greeks  in  speaking  of  Egypt. 
The  scene  in  question  is  figured  upon  many  of  the  tombs  visited  by  the  Greek 
travellers,  and  in  many  of  the  illustrated  papyri  which  were  unrolled  for  their 
gratification.  In  the  fragments  of  some  funerary  inscription  or  of  some  of  these 
manuscripts,  hastily  translated  for  them  by  the  accompanying  priests,  they  found 
frequent  allusions  to  this  act  of  trial  and  judgment.  They  were  greatly  struck  by 
the  importance  attached  by  the  Egyptians  to  the  sentence  of  this  tribunal,  but, 
always  in  a  hurry,  and  sometimes  not  especially  intelligent,  they  do  not  seem  to  have 
always  understood  what  the  dragoman,  without  whom  they  could  not  stir  from  the 
frontier,  told  them  as  to  this  matter.  They  believed  that  the  judges  in  question 
were  living  men,  and  their  tribunal  an  earthly  one,  and  that  they  were  charged  to 
decide  whether  sepulture  should  be  granted  to  the  dead  or  not.  One  of  the  early 
travellers,  we  do  not  know  which,  gave  currency  to  this  belief,  and  we  know  how  it 
has  served  as  the  foundation  for  much  fine  writing,  from  the  time  of  Diodorus  to 
that  of  Bossuet.  We  can  find  nothing  either  in  the  figured  monuments  or  in  the 
written  texts  which  hints  at  the  existence  of  such  a  custom.  Ever  since  the  key 
to  the  hieroglyphics  was  found,  egyptologists  have  been  agreed  upon  this  point. 
Every  Egyptian  was  placed  in  a  sepulchre  befitting  his  station  and  fortune;  his 


286 


A   History  of  Art  in  Ancient  Egypt. 


There,  before  the  "  Lords  of  truth  and  justice,"  the  soul  had 
to  plead  its  cause,  and  there  it  had  to  repeat,  with  an  amount  of 
assurance  and  success  which  would  depend  upon  its  conduct  in 
the  light,  that  negative  confession  which  we  read  in  chapter 
cxxv.  of  the  "  Book  of  the  Dead','  which  contains  an  epitome 
of  Egyptian  morality.-^  But  those  incorruptible  judges  were  not 
guided  solely  by  the  testimony  of  the  ba  in  its  own  favour. 
They  weighed  its  actions  in  a  pair  of  scales  and  gave  judgment 
according  to  their  weight.^  The  impious  soul  was  flogged,  was 
delivered  to  storm  and  tempest,  and,  after  centuries  of  suffering, 
underwent  a  second  death,  the  death  of  annihilation.  The  just 
soul,  on  the  other  hand,  had  to  conquer  in  many  a  combat  before 


Fig.  183. — Hunting  scene  upon  a  tomb  at  Gournah.     (ChampoUion,  pi.  171.) 


it  was  admitted  to  contemplate  the  supreme  verities.  During 
its  transit  across  the  infernal  regions,  hideous  forms  of  evil  sprang 
up  before  it  and  did  their  best  to  arrest  its  progress  by  terrifying 

relations  and  friends  had  to  ask  no  permission  before  they  placed  him  in  it ;  it  was 
in  the  other  world  that  he  was  brought  up  for  judgment,  and  had  to  fear  the  sentence 
of  an  august  tribunal. 

1  Maspero  gives  a  translation  of  it  into  French  in  his  Histoire  A?idenne, 
pp.  44  and  45. 

2  This  weighing  of  the  actions  of  the  deceased  was  represented  in  the  illustrated 
specimens  of  the  Ritual  of  the  Dead  and  upon  the  walls  of  the  tombs,  and 
perhaps  upon  those  monuments  decorated  with  Egyptian  motives  which  were 
sprinkled  by  the  Phoenicians  over  the  whole  basin  of  the  jMediterranean.  Coming 
under  the  eyes  of  the  Greeks,  it  was  modified  by  their  lively  imaginations  into  tliat 
ij/vxoa-Taa-La,  or  zi'eig/n'//g 0/ sou/s,  which  we  find  in  the  Iliad  (xxii.  208-212),  where 
success  in  a  combat  between  two  heroes  depends  upon  the  result  of  that  operation. 
(See  Alfred  Maury,  Revue  archeologique,  1844,  pp.  235-249,  291-307  ;  1845,  pp. 
707-717,  and  De  U'itte,  ibide?n,  1844,  pp.  647-656.) 


The  Tomb  under  the  New  Empire. 


287 


threats.  Thanks  to  the  help  of  Osiris  and  of  other  soul- 
protecting  gods,  such  as  Anubis,  it  triumphed  in  the  end  over  all 
obstacles,  and,  as  the  sun  reappears  each  morning  upon  the 
eastern  horizon,  it  arrived  surely  at  last  at  those  celestial  dwellings 
where  it  became  incorporated  among  the  gods. 

The  Egyptian  imagination  spared  no  effort  to  represent  with 
the  greatest  possible  precision  those  mysterious  regions  where  the 
soul  had  to  undergo  its  appointed  tests.  Such  beliefs  afforded  a 
wide  scope  for  the  individual  influence  of  the  artist  and  the  poet, 
and  accordingly  we   find  that  they  were  modified   with   a   rapidity 


■AHrf 


Fig.    184. — The    weighing    of  actions.      (From    an    illustrated    Ritual   of  the   Dead  in    the 

British   Museum. ) 


which  is  unique  in  Egyptian  art.  But  the  Egyptians  were 
accustomed,  from  such  early  times,  to  give  a  concrete  form  to  all 
their  ideas,  that  they  were  sure  to  clothe  the  plastic  expression 
of  this  theme  in  a  richness  and  brilliancy  of  colour  which  we  do 
not  find  to  the  same  degree  in  any  other  people  of  antiquity.  On 
the  other  hand,  although  they  did  not  escape  the  operation  of  the 
eternal  law  of  change,  their  temperament  was  sufficiently  con- 
servative to  give  to  each  of  their  creations  a  peculiar  fixity  and 
consistency.  Their  Hades,  if  we  may  call  it  so,  took  on  a  very 
definite  form,   and    features    which    varied    but    little    throuorh    a 


288 


A  History  of  Art  in  Ancient  Egypt. 


J  '\ 


long  course  of  centuries  ;  and  this  form  is  practically  that  which 
we  find  in  the  sepulchres  of  the  great  Theban  kings  and  in 
some  belonging  to  private   individuals. 

It  was  through  long  and  gloomy  galleries,  like  those  of  the 
avpiy^,  that  the  perilous  voyage  of  the  soul  had  to  be  undertaken. 
A  boat  carried  it  over  the  subterranean  river,  for  in  a  country 
which  had  the  Nile  for  its  principal  highway,  every  journey, 
even  that  of  the  sun  through  space,  was  looked  upon  as  a 
navigation.  Spacious  saloons  were  imagined  to  exist  among 
those  galleries,  chambers  where  the  infernal  gods  and  their 
acolytes  sat  enthroned  in  all  the  majesty  of  their  office  ;  and  so  the 

passages  of  the  tomb  were  expanded 
here  and  there  into  oblong  or  square 
chambers,  their  roofs  supported  by  pillars 
left  in  the  living  rock.  On  either  side 
of  the  audience  chambers  the  imaeina- 
tion  placed  narrow  passes  and  defiles, 
in  which  the  walls  seemed  to  close  in 
upon  the  soul  and  bar  its  progress  ; 
tortuous  corridors  and  gloomy  gulfs 
were  fixed  in  these  defiles,  in  which 
the  terrible  ministers  of  divine  ven- 
geance held  themselves  in  ambush, 
prepared  to  harass  the  march  of  souls 
not  yet  absolved,  and  to  overwhelm 
with  frio^htful  tortures  those  ao-ainst 
whom  sentence  had  already  been  pro- 
nounced. The  tomb,  therefore,  had  its 
snares  and  narrow  passages,  its  gaping  depths  and  the  mazes 
of  its  intersecting  and  twisting  corridors.  To  complete  the 
resemblance  nothing  more  was  required  than  to  paint  and  chisel 
upon  the  walls  the  figures  of  those  gods,  genii,  and  monsters  who 
peopled  the  regions  below.  On  one  side  the  pious  king  may  be 
seen,  escorted  by  Amen-Ra  and  the  other  divinities  whom  he  had 
worshipped  during  life,  advancing  to  plead  his  cause  before  Osiris  ; 
on  the  other,  the  punishment  of  the  wicked  helps  to  give  i^c/al  to 
the  royal  apotheosis  by  the  contrasts  which  it  affords. 

Thus  the  tombs  of  the  Theban  period  embody  the  Egyptian 
solution  of  the  problem  which  has  always  exercised  mankind. 
Their  subterranean  corridors  were   a   reproduction  upon  a  small 


Fig.  1S5. — Anubis,  in  a  funerary 
pavilion  ;  from  a  bas-relief. 
{Description  de  VAgvpte,  i.,  pi. 
74.) 


The  Tomb  under  the  New  Empire.  289 

scale  of  the  leading  characteristics  of  the  under  world,  and 
we  should  commit  a  great  mistake  were  we  to  look  upon  the 
series  of  pictures  which  decorate  its  walls  as  mere  ornament 
resulting  from  a  desire  for  luxury  and  display.  Between  the  ideal 
models  of  these  pictures  and  the  pictures  themselves  the 
Egyptians  established  one  of  those  mutual  confusions  which  have 
always  been  readily  accepted  by  the  faithful.  Nothing  seemed 
more  natural  to  the  Egyptian,  or  to  the  Ethiopian  who  was  his 
pupil,  than  to  ascribe  the  power  of  speech  and  movement  to  the 
images  of  the  gods,  even  when  they  had  painted  or  carved  them 
with  their  own  hands.  This  INI.  Maspero  has  shown  by  an 
ingenious  collation  of  various  texts.^  The  chisel  which  created 
such  tangible  deities  gave  them  something  more  than  the 
appearance  of  life.  Each  god  exercised  his  own  proper  function 
in  that  tomb  which  was  a  reproduction  in  small  of  the  regions  of 
the  other  world.  His  gfestures  and  the  written  formulae  which 
appeared  beside  him  on  the  walls,  each  had  their  protective  or 
liberating  power.  To  represent  the  king  in  his  act  of  self- 
justification  before  Osiris  was  in  some  measure  to  anticipate 
that  justification.  The  reality  and  the  image  were  so  intimately 
commingled  in  the  mind  of  the  believer  that  he  was  unable 
to  separate   one   from   the  other. 

Did  the  roval  tombs  contain  statues  of  the  defunct  ?  None 
have  been  found  in  any  of  those  already  opened,  and  yet  there  is 
a  chamber  in  the  tomb  of  Rameses  IV.  which  appears  from  its 
inscriptions  to  have  been  called  the  Statue  chamber,  while  another 
apartment  in  its  neighbourhood  is  reserved  for  the  funerary 
statuettes.  The  tombs  of  private  individuals  contained  statues  ; 
why  then  should  none  have  been  put  in  those  of  the  sovereigns  ? 
The  commemorative  sanctuary,  the  external  funerary  temple,  was 
adorned  with  his  image  often  repeated,  which  in  order  that  it 
might  be  in  better  keeping  with  the  magnificence  of  its 
surroundings,  and  should  have  a  better  chance  of  duration,  was 
colossal  in  its  proportions.  In  the  inclosures  of  the  temples  of 
the  two  Rameses  and  upon  the  site  of  the  Amenophium,  the 
remains  of  these  huge  figures  are  to  be  counted  by  dozens,  most 
of  them  are  of  rose  granite  from  Syene  ;  the  smallest  are  from 
24  to  28  feet  high,  and  some,  with  their  pedestals,  are  as  much  as 
from  55  to  more  than  60  feet.     The  two  colossi  of  Amenophis  III., 

'  Recucil  lie  Travaux,  vol.  i.  pp.  155-159. 
VOL.    I.  r    P 


290  A  History  of  Art  in  Ancient  Egypt. 


the  Pharaoh  whom  the  Greeks  called  Memnon,  reached  the 
latter  height.  Flayed,  mutilated,  dishonoured  as  they  have 
been,  these  gigantic  statues  are  still  in  place.  They  should 
be  seen  in  autumn  and  from  a  little  distance  as  they  raise 
their  solitary  and  imposing  masses  above  the  inundated  plain, 
when  their  size  and  the  simplicity  of  their  lines  will  have  an 
effect  upon  the  traveller  which  he  will  never  forget  (Fig.  20, 
and   Plate   vi.). 

In  the  royal  tombs  at  Thebes,  as  in  those  at  Memphis,  the 
approach  to  the  mummy-chamber  is  not  by  a  well,  but  by  an 
inclined  plane.  The  only  wells  which  have  been  discovered  in 
the  tombs  of  the  Bab-el- Molouk  are,  if  we  may  use  the  term, 
false  wells,  ingeniously  contrived  to  throw  any  would-be  violator 
off  the  right  scent.  We  have  already  mentioned  one  of  these 
false  wells  as  existing  in  the  tomb  of  Seti.  In  the  pyramids 
the  corridor  \yhich  leads  to  the  mummy-chamber  is  sometimes  an 
ascending  plane,  but  in  the  Theban  tomb  it  is  always  descending. 
At  the  end  of  the  long  descent  the  mummy- chamber  is  reached 
with  its  sarcophagus,  generally  a  very  simple  one  of  red  granite, 
which  has  hitherto,  in  every  instance,  been  found  empty. 

It  is  doubtful  whether  the  sarcophagus-chamber  was  closed 
by  a  door  or  not.  It  is  known  that  tombs  were  sometimes  thus 
closed  ;  some  of  the  doors  have  been  found  in  place,"  and  in  a 
few  of  the  texts  mention  is  made  of  doors,-  but  not  the  slightest 
vestige  of  one  has  yet  been  discovered  in  the  royal  sepulchres 
at  Thebes.  "  All  the  doorways  have  sills  and  grooved  jambs,  as 
if  they  had  been  closed,  but  no  trace  of  hinges  or  of  the  leaves 
of  a  door  itself  have  been  found."  ^  It  is  possible  that  they 
were  never  put  in  place.  The  exact  and  accurate  spirit  which 
marks  all  the  work  of  Egyptian  artists  would  lead  them  to 
prepare  for  the  placing  of  a  door  at  the  entrance  to  each  chamber  ; 
but  at  the  same  time  it  is  obvious  that  a  few  panels  of  sycamore 
would  do  little  to  stop  the  progress  of  any  one  who  should  attempt 
to    violate    the    royal   sepulchre.     This   latter    consideration  may 

^  One  was  found  in  a  Theban  tomb  opened  by  Rhind  (Thebes,  &:c.,  pp.  94  and  95). 
In  the  tomb  of  Ti  easily  recognized  traces  of  a  door  were  found  (B.t:DEKER,  Unter 
^gyptefi,  p.  405) ;  nothing  but  a  new  door  was  required  to  put  the  opening  in  its 
ancient  state. 

2  See  one  of  the  great  inscriptions  at  Beni-Hassan,  interpreted  by  M.  Maspero 
{Recueil,  etc.  vol.  i.  p.  168). 

3  Description  de  VEgypte,  {A?itiquitts,  vol.  iii.  p.  35). 


The  Tomb  under  the  New  Empire.  291 

have  caused  them   to  abstain  from   expending   time  and  trouble 
upon  a  futile  precaution. 

These  tombs  seem  to  have  varied  greatly  in  size  from 
reasons  similar  to  those  which  determined  the  dimensions  of  the 
pyramids,  namely,  the  length  of  reign  enjoyed  by  their  respective 
makers.  Cheops,  Chephren,  and  Mycerinus  continually  added 
to  the  height  and  mass  of  their  tombs  until  death  put  an  end  to 
the  work.  In  the  same  way,  Seti  and  Rameses  never  ceased 
while  they  lived  to  prolong  the  quarried  galleries  in  the  Bab- 
el-Molouk.  As  these  galleries  w^ere  meant  to  be  sealed  from  the 
sight  of  man,  this  prolongation  was  caused,  no  doubt,  by  the  desire 
to  develop  to  the  utmost  possible  extent  those  pictures  which 
were  to  be  so  powerful  for  good  over  the  fortunes  of  the  defunct 
in  the  under  world. 

Apart  from  the  question  of  duration,  reigns  which  w^ere  glorious 
would  give  us  larger  and  more  beautiful  tombs  than  those  which 
were  obscure  and  marked  by  w^eakness  in  the  sovereign.  The 
three  great  Theban  dynasties  included  several  of  those  monarchs 
who  have  been  called  the  Louis  the  Fourteenths  and  the 
Napoleons  of  Egypt, ^  and  it  was  but  natural  that  they  should 
employ  the  crowd  of  artificers  and  artists  which  their  enterprises 
gathered  about  them,  for  the  excavation  and  decoration  of  their 
own  tombs.  Either  for  this  reason  or  for  some  other,  there 
is  an  extraordinary  difference  between  king  and  king  in  the 
matter  of  their  tombs.  Even  when  we  admit  that  a  certain 
number  of  royal  sepulchres  have  so  far  escaped  discovery,  it  is 
difficult  to  find  place  for  all  the  sovereigns  of  the  nineteenth  and 
twentieth  dynasties  in  the  two  Valleys  of  the  Ki7igs.  Many 
things  lead  us  to  believe  that  several  of  those  princes  were 
content  with  very  simple  tombs  ;  some  of  them  may  have  been 
merely  buried  in  the  sand.  Thus  Mariette  discovered,  at 
DraJi-Aboii l-Neggah,  the  mummy  of  Queen  Aah-hotep,  of  the 
nineteenth  dynasty,  some  few  feet  beneath  the  surface.  The 
mummy  chamber  consisted  of  a  few  ill-adjusted  stone  slabs.  Like 
other  mummies  found  on  the  same  place,  it  seemed  never  to  have 
been  disturbed  since  it  had  been  placed  beneath  the  soil.  It  was 
gilt  all  over,  and  was  decorated  with  jewels  which  now  form  some 
of  the  most  priceless  treasures  of  the  Boulak  Museum. 

The  private  tombs  in  the  Theban  necropolis,  which  are   much 
^  A.  Rhone,  l Egypte  a petitcs  journecs^  p.  104. 


292 


A  History  of  Art  in  Ancient  Egypt. 


more  numerous  than  those  of  kings,  do  not,  Hke  the  latter,  belong 
to  a  single  period  in  the  national  history.  The  most  ancient 
among  them  date  back  to  the  eleventh  dynasty.  There  are 
some  also  of  the  Sait  period,  and  a  few  contemporary  with  the 
Ptolemies  and  the  Roman  emperors.  But  by  far  the  greater 
number  belong  to  that  epoch  which  saw^  Thebes  promoted  to  be 
the  capital  of  the  whole  country,  to  the  centuries,  namely, 
between  Amosis,  the  conqueror  of  the  Hyksos,  and  the  last 
of  the  Ramessides. 

These  tombs  are  distinguised  by  great  variety,  but,  before 
noticing  the  principal  types  to  which  they  may  be  referred,  the 
points  which  distinguish  them  from  the  royal  burying-places 
should  be  indicated.      The  private   sepulchre   is   never  subdivided 


Fig.  186. — Plan  and  section  of  a  royal  tomb.      [Description  de  F Egypte,  vol.  ii.,  pi.  79.) 

like  those  of  the  kings.  This  subdivision  is  to  be  explained  by 
the  exceptional  position  of  the  sovereign  mid-way  between  his 
subjects  and  the  national  deities.  A  funerary  chapel  cut  in  the 
sides  of  the  mountain  would  obviously  be  too  small  for  the 
purposes  to  which  the  commemorative  part  of  the  tomb  of  a 
Rameses  or  a  Seti  would  be  put.  On  the  other  hand,  no  private 
individual  could  hope  to  receive  royal  honours  nor  to  associate 
his  memory  with  the  worship  of  the  great  Egyptian  gods.  We 
find  that  for  him  the  chapel  always  remains  closely  connected 
with  the  mummy  chamber.  Sometimes  it  is  in  front  of  it, 
sometimes  above  it,  but  in  neither  case  does  it  ever  fail  to  torm 
an  integral  part  of  the  tomb,  so  that  the  latter  preserves  at  once 
its  traditional  divisions  and  its  indissoluble  unitv. 


The  ToxMB  under  the  New  Empire.  293 

We  may  say  the  same  of  the  well,  which  plays  the  same  part 
in  the  private  tombs  of  the  New  Empire  as  in  the  Mastaba  and 
the  Spcos  of  the  Ancient  and  Middle  Empires.  In  almost 
every  instance  the  mummy  chamber  is  reached  by  a  well, 
whether  the  tomb  be  constructed  in  the  plain  or  in  the  side 
of  the  mountains.  It  is  seldom  so  deep  as  those  of  Gizeh 
or  Sakkarah  ;  its  depth  hardly  exceeds  from  20  to  30  feet ; 
but  its  arrangement  is  similar  to  those  in  the  early  necropolis. 
The  mummy  chamber  opens  directly  upon  it.  Sometimes 
there  are  two  chambers  facing  each  other  at  the  foot  of  the  well, 
and  of  unequal  heights.^  After  the  introduction  of  the  corpse, 
which  was  facilitated  by  notches  cut  in  two  faces  of  the  well, 
the  door  of  the  mummy  chamber  was  built  up — the  well  filled 
in.  In  a  few  exceptional  instances,  the  tombs  of  private 
individuals  seem  to  have  had  no  well,  and  the  innermost 
chamber,  as  in  the  case  of  ro^-al  tombs,  received  the  mummy. "^ 
In  such  cases  it  is  very  necessary  to  make  sure  that  explorers 
have  not  been  deceived  by  appearances.  In  these  dusty  interiors 
the  carefully  sealed  opening  might  easily  escape  any  but  the 
most  careful  research  ;  and  as  for  a  sarcophagus,  when  one  is 
found  in  such  a  chamber,  it  may  have  been  placed  there  long 
after  the  making  of  the  tomb.  Such  usurpations  are  by  no  means 
unknown.^  In  the  time  of  the  Ptolemies,  influential  people,  such 
as  priests  and  military  functionaries,  made  them  without  scruple. 
The  venerable  mummies,  dating  from  the  time  of  Rameses,  were 
thrown  into  a  corner  ;  their  cases  were  made  use  of,  sometimes 
for  the  mummy  of  the  usurper,  sometimes  for  more  ignoble 
purposes.  In  more  than  one  of  these  usurpations  the  new 
comer  has  been  placed  in  a  chamber  constructed  for  some 
other  object. 

'  Passalacqua  describes  a  tomb  of  this  kind  in  detail  in  his  Cataloi^iie  raisoufie 
et  historiqiie  des  Antiquites  decouvertes  en  Egypte  (8vo,  1826).  This  tomb  had  been 
visited  and  pillaged  at  some  unknown  epoch.  One  of  the  two  chambers  had  been 
opened  and  stripped,  but  the  second,  which  opened  lower  down  the  well,  and  on 
the  other  side,  escaped  the  notice  of  the  violators  (pp.  1 18-120).  In  the  tomb 
opened  by  Rhind  {Thebes,  its  Tombs,  etc.  pi.  5,  v.),  the  well  gave  access  to  four 
chambers  of  different  sizes  arranged  round  it  like  the  arms  of  a  cross. 

"  Description  de  V Egypte,  {Antiquites,)  plates,  vol.  ii.  p.  78. 

^  Rhind  describes  one  of  the  most  curious  of  these  substitutions  in  his  chapter  IV. 
In  that  case  an  usurper  of  the  time  of  Ptolemy  established  himself  and  all  his 
family  in  the  mummy  chambers  at  the  foot  of  the  well,  after  relegating  the  statues 
and  mummies  of  the  rightful  owner  and  his  people  to  the  room  above. 


2  94 


A   History  of  Art  in  Ancient  Egypt. 


The  well  and  mummy  chamber  in  the  rock  are,  then, 
found  almost  universally,  but  the  form  of  the  rest  of  the  tomb 
varies  according  to  its  date  and  site.  Those  in  the  plain  are 
arranged  differently  to  those  in  the  hill  sides.  But  it  must  be 
understood  that  when  we  speak  of  the  plain  in  connection  with 
the  Theban  tombs,  we  do  not  mean  the  space  over  which  spread 
the  waters  of  the  TJile.  We  mean  the  gentle  sandy  slopes 
which  lie  between  the  foot  of  the  cliffs  and  the  cultivated  fields, 
a  narrow  band  which  widens  a  little  between  the  long  spurs  which 
the  mountains  throw  out  towards  the  river.  In  these  land-aulfs 
the  rock  crops  up  here  and  there,  and  nowhere  is  it  covered  by 
more  than  a  thin  layer  of  sand.  Such  a  soil,  being  above  the 
reach  of  the  annual  inundation,  was  marvellously  well-fitted  both 


Figs.  187,  188. — Theban  tombs  from  the  bas-ieliets.     (From  Wilkinson,  ch.  xvi.) 

for   the    construction    of   the    tomb   and    for    the   preservation   of 
the  mummy. 

In  this,  which  may  be  called  the  level  part  of  the  necropolis, 
the  tombs  have  left  but  slight  and  ill-defined  traces.  Their 
superstructures  have  almost  entirely  disappeared,  and  yet  some 
which  have  now  completely  vanished  were  seen  by  travellers 
to  Thebes  in  the  first  half  of  the  present  century.  By  comparing 
what  they  tell  us  with  the  figured  representations  in  bas-relief 
and  manuscript,  we  may  form  some  idea  of  the  aspect  which 
this  part  of  the  cemetery  must  formerly  have  presented.  The 
tombs  which  it  contained  were  built  upon  the  same  principles 
as  those  of  Abydos  ;  a  square  or  rectangular  structure  with 
slighdy  sloping  walls  was  surmounted  by  a  small  pyramid.      There 


The  Tomb  under  the  New  Empire. 


295 


was,  however,  an  essential  difference  between  the  two.  At 
Abydos  the  nature  of  the  subsoil  compelled  the  architect  to 
contrive  the  mummy  chamber  in  the  interior  of  his  own  structure  ; 
at  Thebes,  on  the  other  hand,  there  was  nothing  to  prevent  him 
from  being  faithful  to  a  tradition  which  had  manifest  advantages, 
and  to  intrust  the  corpse  to  the  keeping  of  the  earth,  at  a  depth 
below  the  surface  which  would  ensure  it  greater  safety  both 
from  violence  and  from  natural  causes  of  decay.  At  Thebes  the 
rock  was  soft  enough  to  be  cut  with  sufficient  ease,  and  yet  firm 
enouo-h  to  be  free  from  all  dano-er  of  settlement  or  disinteera- 
tion.  The  soil  of  all  this  region  is 
honeycombed  with  mummy  pits,  which 
have  long  ago  been  pillaged  and  are 
now  filled  up  with  sand.  The  super- 
structure was  built  above  the  well  and 
inclosed  the  funerary  chapel.  Some- 
times it  was  surmounted  by  a  small 
pyramid  ;  sometimes  it  was  a  quad- 
rangular mass  standing  upon  a  surbase, 
with  a  pilaster  at  each  angle  and  a 
boldly  projecting  cornice  at  the  top.  The 
oldest  of  the  known  tombs  of  Apis  may 
be  taken  as  specimens  of  this  latter  class, 
which  must  also  have  been  represented 
at  Thebes.  These  are  contemporary 
wifh  the  eighteenth  dynasty,  and  were 
discovered  by  M.  Mariette  at  Sakkarah.^ 

These  little  monuments  have  either 
been  destroyed  since  185 1  or  covered 
by  the  sand  (see  Fig.  190). 

The  other  type  is  that  of  the  Speos. 


Fig.  189. — Theban  tomb  from  a 
bas-relief.  (From  Wilkinson, 
cli.  xvi.) 


We  have  here  seen  that 
it  dates  from  the  Ancient  Empire,  but  came  into  general  employ- 
ment and  obtained  its  full  development  under  the  First  and  Second 
Theban  Empires.  We  have  already  given  some  idea  of  the 
architectural  character  and  of  the  decoration  of  the  royal  sepulchres, 
we  must  now  indicate  the  peculiarities  which  in  these  respects 
distinguish  the  tombs  of  the  kings  from  those  of  their  subjects. 
One   of  the  points  of  difference  has  already  been  noticed  ;   the 

'  At/iencet/in  Francais,   1855,  p.  55   {Renseignemenis   si/r  les  soixante  qttotre  Apis 
troiivh  dans  le  Scrapeum  de  Memphis). 


296 


A  History  of  Art  in  Ancient  Egypt. 


employment  of  vertical  wells  instead  of  inclined  planes  as 
approaches  to  the  mummy  chamber.  The  most  extensive  of 
all  the  llieban  catacombs  is  that  of  a  private  individual,  the  priest 
Petamounoph  (Fig.  191).-^  In  this  the  galleries  have  not  less  than 
895  feet  of  total  length,  besides  which  there  are  a  large  number  of 
chambers,  the  whole  being  covered  with  painted  reliefs.  But  this 
tomb  is  quite  exceptional.  The  great  majority,  those  of  Rekhmara, 
for  instance,  and  others  excavated  in  the  hill  of  Sheikh-Abd-el- 
Gournak,  are  composed  of  two  or  three  chambers  at  most,  united 
by  corridors.  The  mummy  pit  opens  sometimes  upon  the  corridor 
between  two  of  the  chambers,  sometimes  upon  the  innermost 
chamber,  sometimes  upon  a  corrider  opening  out  of  the  latter. 
Rhind  tells  us  that  he  followed  one   of  these  corridors  for  about 


Fig.  190. — A  tomb  of  Apis.     From  Maiiette. 

300   feet  beyond  the  chamber   without   arriving  at   the    mummy 
pit,  the  air  then  became  too  bad  for  further  progress.- 

The  chamber  for  the  funerary  celebration  is  easily  recognized 
by  its  decorations.  It  is  sometimes  the  first,  but  more  often  the 
second  in  order,  in  which  case  the  first  acts  as  a  sort  of  vestibule. 
A  considerable  number  of  tombs  are  very  simple  in  arrangement. 
The  door  gives  access  to  a  rectangular  chamber,  from  6  to  10  feet 
high  and  10  to  12  wide,  which  extends,  in  a  direction  parallel  to 
the  wall,  for  from  12  to  24  feet.  This  chamber  is  the  funerary 
chapel.      From   its  posterior  wall  a  passage  opens  which  is  nearly 

^  This  view  is  obtained  by  a  series  of  liorizontal  and  vertical  sections  in  the 
rock  to  the  right  of  the  galleries.  By  this  operation  we  are  enabled  to  show  the 
subterranean  parts  of  the  tomb. 

2  Rhind,  Thebes,  etc.  p.  43. 


m 


c^ 


.2 

> 

5J 


C 


> 
"5 
ex 


rt 


o 
c 
a 
o 
S 
rt 

PL. 


<1J 


0^ 


VOL.    I. 


Q  Q 


The  Tomb  under  the  New.  Empire. 


299 


equal  in  height  and  width  to  the  chamber.  It  has  a  gentle  slope 
and  penetrates  into  the  rock  to  a  distance  of  some  25  to  35  feet, 
terminating,  in  some  cases,  in  the  mummy  chamber  itself,  but 
more  frequently  in  a  small  apartment  containing  the  opening  of 
a  mummy  pit.^ 

It  must  not  be  imagined  that  all  the  tombs  were  decorated  ; 
there  are  many  which  have  received  neither  painted  nor  carved 
ornament,  and  in  others  the  ornament  has  never  been  carried 
beyond  the  first  sketch.  But  even  in  those  which  are  quite  bare, 
the  walls  are,  in  nearly  every  instance,  covered  with  a  coat  of 
white  stucco. 

As  the  funerary  chapel  was  contained  in  the  tomb  itself,  no 
effort  could  be  made  to  mask  or  conceal  the  entrance,  which 
accordingly  was  taken  advantage  of  for  the   display  of  ornament. 


Fig.  192. — The  most  simple  form  of 
Theban  tomb  ;  from  Rhind. 


Fig.  193. — Tomb  as  represented  upon 
a  bas-relief  ;  from  Rhind. 


But  no  attemjjt  was  made  to  cut  architectural  facades  in  the  cliffs 
like  those  at  Beni- Hassan  ;  not  more  than  one  or  two  sepulchres 
have  yet  been  discovered  which  have  facades  made  up  of  those 
columns  which  have  been  called  protodoric.  The  makers  of 
these  tombs  were  usually  content  with  dressing  the  surface  of 
the  rock  above  and  around  the  entrance.  The  latter,  with  its 
sloping  lintel  above  a  cornice,  stands  in  the  centre  of  an  almost 
perpendicular  wall  which  acts  as  its  frame  or  background.  In 
the  uninjured  state  of  the  sepulchre  this  wall  was  more  or  less 
concealed  by  a  construction  similar  to  those  which  we  have 
described  in  speaking  of  the  tombs  in  the  plain.  According  to 
all  appearances,  one  of  these  little  buildings,  a  cube  of  masonry 
crowned    by   a   pyramidion,   was    placed    before    the    doorway   of 

'  Rhixd,  Thebes,  etc.  pp.  56,  57. 


300  A   History  of  Art  in  Ancient  Egypt. 

every  tomb.  It  is  difficult  to  say  whether  it  was  of  sufficient  size 
to  contain  a  funerary  chamber  or  not.  It  may  have  been  no  more 
than  a  sohd  erection  of  small  size,  meant  only  to  mask  the 
entrance  and  to  indicate  its  situation  to  those  concerned.  The 
wealthy,  indeed,  may  have  been  only  too  pleased  to  thus  call 
public  attention  to  the  position  of  their  gorgeously  decorated 
selpulchres. 

The  little  pyramids  of  crude  brick  which  we  find  upon  the 
irregular  rocky  slopes  of  the  Koiirnet-el-Mourrayi,  above  the  little 
window-shaped  openings  with  which  the  rock  is  honeycombed, 
probably  answered  a  similar  purpose.  Of  these  some  are  still 
standing,  and  others  have  left  unmistakable  traces  upon,  the 
slope.  They  seem  to  have  existed  in  great  numbers  in  this  part 
of  the  necropolis,  which  seems  to  have  been  set  apart,  about  the 
time  of  the  eighteenth  dynasty,  for  the  priests. 

Although  they  hardly  varied  from  the  two  or  three  types 
consecrated  by  custom,  these  little  buildings  could  easily  have 
been  made  to  present  slight  differences  one  from  another.  When 
they  existed  in  their  entirety,  they  must  have  given  a  very 
different  aspect  to  the  cemetery  from  that  which  it  presents  with 
its  rocky  slopes  burnt  by  the  sun  into  one  harsh  and  monotonous 
tint,  varied  only  by  the  black  and  gaping  mouths  of  the  countless 
tombs.  The  sides  which  they  turned  to  the  city  and  the  river 
were  adorned  with  those  brilliant  colours  of  which  the  Egyptian 
architects  were  so  fond,  and,  spaced  irregularly  but  never  very 
far  apart,  they  were  sprinkled  over  the  ground  from  the  edge 
of  the  plain  to  the  topmost  ridges  of  the  hills.  Nearly  all  of 
them  ended  in  a  pyramid,  but  the  varying  dimensions  of  their 
bases  and  their  different  levels  above  the  plain,  gave  diversity  to 
the  prospect,  while  here  and  there  the  slender  apex  of  an  obelisk 
rose  above  the  private  tombs  and  signalized  the  sleeping-place 
of  a  king.  It  has  been  very  justly  remarked,  that  the  best  idea 
of  an  Egyptian  cemetery  in  its  best  time  is  to  be  gained  by  a 
visit  to  one  of  those  Italian  Campo-Santos,  that  of  Naples,  for 
example,  where  the  tombs  of  many  generations  lie  closely 
together  under  a  blazing  sun,^  There,  too,  many  sepulchral 
facades  rise  one  above  another  upon  the  abrupt  slope  of  a  hill 
into  which  the  graves  are  sunk.  A  comparison  with  the  cemetery 
of  Pere-Lachaise,  or  with   that  at   Constantinople,  would  not   be 

*  Rhind,  Thebes,  etc.  p.  55. 


The  Tomb  under  the  New  Empire.  301 

just  because  no  trees  could  flourish  in  the  Theban  rocks,  at 
least  in  the  higher  part  of  the  necropolis.  In  those  districts  which 
border  closely  upon  the  irrigation  channels,  the  tombs  seem  to 
have  had  their  gardens  and  fountains.  Palms  and  sycamores 
appear  to  have  been  planted  about  them,  and  here  and  there, 
perhaps,  the  care  of  survivors  succeeded  in  rearing  flowers 
which  would  shed  their  perfumes  for  the  consolation  of  the  dead.^ 

Were  there  statues  in  the  courtyards  by  which  many  of  these 
tombs  were  surrounded  ?  There  is  no  doubt  that  such  statues 
were  placed  in  the  rock-cut  sepulchres ;  all  the  museums  of 
Europe  have  specimens  which  come  from  the  Theban  tombs. 
The  latter  were  opened  and  despoiled,  however,  at  such  an 
early  period  that  very  few  of  these  figures  have  been  found  in 
place  by  those  who  have  visited  the  ruins  of  Egypt  for  legitimate 
motives.  We  have,  however,  the  evidence  of  explorers  who 
have  penetrated  into  tombs  which  were  practically  intact.  They 
tell  us  that  the  statue  of  the  deceased,  accompanied  often  by  that 
of  his  wife  and  children,  was  placed  against  the  further  wall  of 
the  innermost  chamber.'^  In  some  tombs,  a  niche  is  cut  in  the 
wall  for  this  purpose,^  in  others  a  dais  is  raised  three  or  four 
steps  above  the  floor  of  the  chamber.^  Here,  too,  is  found  the  sar- 
cophagus, in  basalt  when  the  defunct  was  able  to  afford  such  a 
luxury,  and  the  canopic  vases,  which  were  sometimes  of  stone, 
especially  alabaster,  sometimes  of  terra  cotta,  and  now  and  then 
of  wood,  and  were  used  to  hold  the  viscera  of  the  deceased. 
These  vases  were  four  in  number,  protected  respectively  by  the 
goddesses   Isis,  Nephtys,  Neith,  and  Selk  (Fig.  196). 

During  the  period  of  which  we  have  just  been  treating,  the  taste 
for  these  huee  rock-cut  tombs   was  not  confined  to  Thebes  and 

'  Masveko,  Rcaiei!  de  Travaiix,  \o\.\\.  Y>-  105.  The  formula  which  is  generally 
found  upon  the  funerary  steles  of  the  eighteenth  and  nineteenth  dynasties  hints  at 
this  :  "  That  I  may  walk  daily  upon  the  border  of  my  fountain ;  that  my  soul  may 
rest  upon  the  branches  of  the  funerary  garden  which  has  been  made  for  me,  that  each 
day  I  may  be  out  under  my  sycamore  !  "  These  desires  may  be  taken  literally,  as 
is  proved  by  two  steles  in  the  museums  of  Turin  and  Boulak,  which  bear  representa- 
tions of  tombs  upon  their  lower  portions.  The  latter,  which  we  reproduce,  comes 
from  the  Theban  necropolis. 

^  Most  of  these  statues  were  of  calcareous  stone,  but  in  the  Descriptiofi  de 
r Egypte  {Af2tiquites,  vol.  iii.  p.  34)  two  granite  ones  are  mentioned. 

^  In  the  tomb  of  Amenemheb,  for  instance,  discovered  by  Professor  Ebers.  See 
also  Description  de  l' Egypte,  vol.  iii.  p.  41. 

*  Description  de  r Egvpte  {Antiquites,  vol.  iii.  p.  34). 


502 


A   History  of  Art  in  Ancient  Egypt. 


its  immediate  vicinity  ;  we  find  obvious  traces  of  them  in  the  city 
which  then  held  the  second  place  in  Egypt,  namely,  in  Memphis, 
where  a  son  of  the  sovereign  resided  as  viceroy.  It  was  in  the 
reign  of  Rameses   II,,  that  the  fourth  of  his  hundred  and  seventy 


Fig,  194. — Stele  in  the  Boulak  Museum,  showing  tombs  with  gardens  about  them.    From  Maspero. 


children    began  what    is    now  called    the  little   Serapeum,  in   the 
neighbourhood  of  the    Great   Pyramids,^      Until    then   each  Apis 

'  It  is  no  part  of  our  plan  to  describe  this  discovery,  which  did  so  much  honour 
both  to  the  perspicacity  and  the  energy  of  Mariette,  We  refer  all  those  who  are 
interested  in  the  matter  to  the  article  contributed  by  M,  E.  Desjardins  to  the 
Revue  des  Deux-Mondcs  of   March    15,    1874,  under   the   title:  Les  Dkouvertes  de 


> 

3 
O 
-1 


a 


43 


a 
o 


U3 

3 

bo 

ri 

O 
o 


ON 


C5 


The  Tomb  under  the  New  Empire. 


305 


bull  had  had  a  tomb  apart,  a  tomb   in   which  everything   was  of 
small  dimensions.     This  royal  prince  was  especially  vowed  to  the 


Fig.  196. — Canopic  vase  of  alabaster.     Louvre. 

worship  of  Ptah  and  Apis,  for  whom  he  inaugurated  new  rites. 
He  began  the  excavation  of  a  grand  gallery,  and  lined  it  on  each 

r Egyptologie  francaise^  les  Missions  et  les  Travaux  de  M.  Alariette.  Many  precious 
details  will  also  be  found,  some  of  them  almost  dictated  by  Mariette,  in  the  L Egypte 
a  petites  Joiirnees  of  M.  Arthur  Rhone  (pp.  212-263).  This  work  includes  two 
plans,  a  general  plan  and  a  detailed  plan  of  the  subterranean  galleries,  which  were 
supplied  by  the  illustrious  author  of  the  excavations  himself;  views  of  the  galleries 
are  also  given,  and  reproductions  of  various  objects  found  in  the  course  of  the 
exploration.  We  may  also  mention  the  Choix  des  Monuments  du  Scraphnn,  a  collec- 
tion of  ten  engraved  plates  published  by  Mariette,  and  the  great  work,  unfortunately 
incomplete,  which  he  commenced  under  the  title  :  Le  Scrapaun  de  Memphis 
(folio,  Paris,  Gide,  1858).  In  the  second  volume  of  Foiiilles  et  Decouvertes  (Didier, 
8vo,  1873,  2  vols.)  Beule  has  given  a  very  good  description  of  the  bold  but  fortunate 

VOT,.    T.  .  R    R 


3o6 


A   History  of  Art  in  Ancient  Egypt, 


side  with  small  chambers  which  were  increased  in  number  as  each 
successive  Apis  died  and  required  a  sepulchre.  This  gallery  and 
its  chambers  served  for  700  years  (see  Figs.  197  and  198). 

The  funerary  architecture  of  the  Sait  epoch  seems  to  have  had 
an  originality  of  its  own,  but  we  are  unable  to  form  an  opinion 
from  any  existing  remains.  Not  a  trace  is  extant  of  those  tombs 
in  which  the  princes  of  the  twenty-sixth  dynasty  were,  according 
to  Herodotus,  placed  one  after  another.  Here  are  the  words 
of    the     Greek    historian  :    ol    8e    (the    Egyptians)   fiiv    (A pries) 

arrtTTVL^ai',  /cat  tirebra  e6ay\rav  ev  rrjcn  irarpwrjat  racfirjai, — at  Se  etcrt 
IV  T(p  ipcp  Trjs  KOrjvairjs,  a'^)(OTarw  rov  /jieydpov  ecnovrt,  apLareprj^ 
%epo9 — tda\p-av  St  ^atrat  iravTas  rovs  tK  vo/xov  rovrou  <yevo/jb6vov9 
/SaatXeay   eaco    ei>    tc3  Ip^.   kol    •jap   to    tov   ^AfiacrLos    arjjxa   eKaarepco   fieu 


Fig.  197. — View  of  the  grand  gallery  in  the  Apis  Mausoleum  ;  from  Marietta. 


f^ajL  TOV  pL^'yapov  7;  to  tov  ATrpteco  Kat  tcov  toltov  npoTTaToprov'  eart 
fievTOt,  Kai  TovTu  iv  ti,  avXjj  tov  ipov,  iraaTas  Xidivr}  /u.eyaXr},  koI  rjafcrjjjb'tvr] 
aTvXoLol  Te  (f^oiviKas  Ta  BtvSpea  fj,ejj,(.fX7]fj,6V0tai,  Kat  tt)  aWrj  SaTravr/. 
eaco  Be  ev  tt;  TracTTaSi,  Bi^a  Ovpcu/jLaTa  ecTTijKe'  ev  8e  toIctl  dvpco/xaat 
7j  OrjKr]  tan. 

campaign  which,  begun  in  the  month  of  October,  1850,  brought  fame  to  a  young 
man  who  had,  until  then,  both  open  enmity  and  secret  intrigue  to  contend  against. 

'  Herodotus,  ii.  169.  "The  Egyptians  strangled  Apries,  but,  having  done  so, 
they  buried  him  in  the  sepulchre  of  his  fathers.  This  tomb  is  in  the  temple  of 
Athene  {N'eith),  very  near  the  sanctuary,  on  the  left  hand  as  one  enters.  The  natives 
of  Sais  buried  all  the  kings  which  belonged  to  their  nome  within  this  temple,  and, 
in  fact,  it  also  contains  the  tomb  of  Amasis,  as  well  as  that  of  Apries  and  his  family, 
but  the  former  is  not  so  close  to  the  sanctuary  as  the  former,  but  still  it  is  within  the 
buildings  of  the  temple,  in  a  large  chamber  constructed  of  stone,  with  columns  in 
the  shape  of  the  trunks  of  palm-trees,  and  richly  decorated  besides,  which  incloses 


The  Tomb  under  the  New  Empire.        i 


o 


07 


Preceding  centuries  afford  no  example  of  a  tomb  placed  within 
a  temple  like  this.^ 

First  of  all  the  royal  mummy  was  entombed  in  the  bowels  of  an 
artificial  mountain,  secondly,  under  the  Theban  dynasties,  in  those 
of  a  real  one  ;  but  at  Sais,  it  rests  above  the  soil,  in  the  precincts 
of  a  temple,  where  curious  visitors  come  and  go  at  their  will,  and 
nothing  but  a  pair  of  wooden  doors  protects  it  from  disturbance. 
Such  an  arrangement  seems  inconsistent  with  all  that  we  know  of 
the  passionate  desire  of  the  Egyptians  to  give  an  eternal  duration 
to  their  mummies.  We  have  every  reason  to  believe  that  this 
desire  had  shown  no  diminution  at  the  time  of  the  twenty-sixth 
dynasty,  and  we  can  hardly  admit  that  Psemethek  and  his 
successors  were  less  impelled  by  it  than  the  meanest  of  their 
subjects. 

The  explanation  of  the  apparent  anomaly  is  to  be  found,  we 
believe,  in  the  peculiar  nature  of  the  soil  of  Lower  Egypt.  The 
Sait  princes  were  determined  to  leave  their  mummies  in  the  city 
which  they  had  filled  with  magnificent  buildings  and  had  turned 
into  the  capital  of  all  Egypt.  Both  speos  and  mummy  pit,  how- 
ever, were  out  of  the  question.  Sais  was  built  in  the  Delta  ;  upon 
an  alluvial  soil  which  was  wetted  through  and  through,  as  each 
autumn  came  round,  by  the  water  of  the  Nile.  Neither  hill  nor 
rock  existed  for  many  miles  in  every  direction.      It  was,  therefore, 


a  kind  of  niche  or  shrine  with  folding  doors,  in  which  the  mummy  is  placed." 
This  is  one  of  the  most  difficult  passages  in  Herodotus,  and  has  given  much 
trouble  to  translators  and  commentators.  See  Larcher's  note  (ii.  565),  and  the 
j)assage  in  Stobjeus  (serm.  xli.  p.  251),  which  he  cites  in  justification  for  the  sense 
which  is  here  given  to  the  word  dypw/jLara.  Strabo  is  content  with  but  a  line  on 
this  subject :  "  Sais,"  he  says,  "  especially  worships  Athene  (Neith).  The  tomb 
of  Psammitichos  is  in  the  very  temple  of  that  goddess"  (xvii.  18). 

'  Herodotus  affirms  (ii.  129-132)  that  Mycerinus  caused  the  body  of  his  daughter 
to  be  inclosed  in  the  flank  of  a  wooden  cow,  richly  gilt,  and  he  says  that  "  the  cow  in 
question  was  never  placed  in  the  earth."  In  his  time  it  was  exposed  to  the  view  of 
all  comers  in  a  magnificently  decorated  saloon  of  the  royal  palace  of  Sais.  We  may 
be  allowed  to  suggest  that  Herodotus  was  mistaken  in  the  name  of  the  prince  ; 
Mycerinus  is  not  likely  to  have  so  far  abandoned  all  the  funerary  traditions  of  his 
time,  or  to  have  entombed  the  body  of  his  daughter  in  a  spot  so  distant  from  his 
own  pyramid  at  Gizeh.  There  is  one  hypothesis,  however,  which  may  save  us  from 
the  necessity  of  once  again  accusing  the  Greek  historian  of  misunderstanding  what 
was  said  to  him  ;  in  their  desire  to  weld  together  the  present  with  the  past,  and  to 
collect  into  their  capital  such  national  monuments  as  might  appeal  to  the  imagina- 
tions of  their  subjects,  the  Sait  jirinces  may  have  transported  such  a  curiously  shaped 
sarcophagus  either  from  the  pyramid  of  Mycerinus  or  from  some  small  pyramid  in 
its  neighbourhood. 


3o8 


A  History  of  Art  in  Ancient  Egypt. 


absolutely  necessary  that  the  tomb  should  be  a  constructed  one 
upon  the  surface  of  this  soil.  It  would  seem  that  the  pyramid 
would  have  been  the  best  form  of  tomb  to  ensure  the  continued 
existence  of  the  mummy,  but,  to  say  nothing  of  the  difficulty  of 
finding  a  satisfactory  foundation  for  such  a  structure  upon  a  soft 
and  yielding  soil,  the  pyramid  had,  for  many  ages,  been  completely 
out  of  fashion.  Egyptian  art  was  entirely  occupied  with  richer 
and  more  varied  forms,  forms  which  admitted  of  the  play  of  light 
and  shade,  and  of  al|  the  splendour  of  carved  and  painted  decora- 
tion.    The  pyramid  being  rejected,  no  type  remained  but  that  of 


Fjg.  198. — Sepulchral  chamber  of  an  Apis  bull  ;  from  Mariette. 


a  building  which  should  inclose  both  mummy  chamber  and 
funerary  chapel  under  one  roof,  or,  at  least,  within  one  bounding 
wall.  There  was  also,  it  is  true,  the  Abydos  type  of  sepulchre, 
with  its  mummy  chamber  hidden  in  the  thickness  of  its  base ; 
but  it  was  too  heavy  and  too  plain,  it  was  too  nearly  related  to  the 
pyramid,  and  it  did  not  lend  itself  readily  to  those  brilliant 
compositions  which  distinguish  the  last  renascence  of  Egyptian 
art.  But  the  hypostyle  hall,  the  fairest  creation  of  the  national 
genius,  was  thoroughly  fitted  to  be  the  medium  of  such  picturesque 
conceptions   as  were   then   required,  and   it    was   adopted  as  the 


The  Tomb  under  the  New  Empire.  309 


nucleus  of  the  tombs  at  Sais.  A  hall  divided,  perhaps,  into  three 
aisles  by  tall  shafts  covered  with  figures  and  inscriptions,  afforded 
a  meeting-place  and  a  place  of  worship  for  the  living.  The 
mummy  chamber  was  replaced  by  a  niche,  placed,  doubtless, 
in  the  wall  which  faced  the  entrance,  and  the  well,  the  one 
essential  constituent  of  an  early  Egyptian  tomb,  was  suppressed. 
Such  arrangements  as  these  afforded  much  less  security  to  the 
mummy  than  those  of  Memphis  or  Thebes,  and  to  compensate 
in  some  measure  for  their  manifest  disadvantages,  the  tomb  was 
placed  within  the  precincts  of  the  most  venerable  temple  in  the 
city,  and  the  security  of  the  corpse  was  made  to  depend  upon  the 
awe  inspired  by  the  sanctuary  of  Neith.  As  the  event  proved, 
this  was  but  a  slight  protection  against  the  fury  of  a  victorious 
enemy.  Less  than  a  year  after  the  death  of  Amasis,  Cambyses 
tore  his  body  from  its  resting-place,  and  burnt  it  to  ashes  after 
outrao^inof  it  in  a  childish  fashion.^ 

The  tombs  of  these  Sait  kings,  consisting  of  so  many  com- 
paratively small  buildings  in  one  sacred  inclosure,  remind  us 
of  what  are  called,  in  the  modern  East,  hirbehs,  those  sepulchres 
of  Mohammedan  saints  or  priests  which  are  found  in  the  immediate 
neighbourhood  of  the  mosques.  Vast  differences  exist,  of  course, 
between  the  Saracenic  and  Byzantine  styles  and  that  of  Ancient 
Egypt,  but  yet  the  principle  is  the  same.  At  Sais,  as  in  modern 
Cairo  or  Constantinople,  iron  or  wooden  gratings  must  have 
barred  the  entrance  to  the  persons  while  they  admitted  the  glances 
of  visitors  ;  rich  stuffs  were  hung  before  the  niche,  as  the  finest 
shawls  from  India  and  Persia  veil  the  coffins  which  lie  beneath  the 
domes  of  the  modern  burial-places.  Perhaps,  too,  sycamores  and 
palm-trees  cast  their  shadows  over  the  external  walls.^  The  most 
hasty  visitor  to  the  Bosphorus  and  the  Golden  Horn  can  hardly 
fail  to  remember  the  suburb  of  Eyoub,  where  the  turbehs  of  the 
Ottoman  princes  stand  half  hidden  among  the  cypresses  and 
plane  trees. 

The  material  condition  which  compelled  the  Sait  princes  to 
break  with   the  customs  of  their  ancestors,  affected  the  tombs  of 

'  Herodotus,  iii.  16.  Upon  this  subject  see  an  interesting  article  by  M.  Eugkne 
Revillout,  entitled  :  Le  Roi  Amasis  et  les  inercenaires  Grecs,  selon  les  Donncs 
d' Herodote  et  les  Renseig7iements  de  la  Chronique  Demotique  de  Paris.  {Revue 
Egyptologique,  first  year  ;  p.  50  ef  seq.) 

2  There  are  two  passages  in  Herodotus  (ii.  91,  and  138)  from  which  we  may 
infer  that  the  Egyptians  were  fond  of  planting  trees  about  their  tcmitles. 


J 


lo  A   History  of  Art  in  Ancient  Egypt. 


private  individuals  also.  Throughout  the  existence  of  the 
Egyptian  monarchy  the  inhabitants  of  the  Delta  were  obliged 
to  set  about  the  preservation  of  their  dead  in  a  different  fashion 
to  that  followed  by  their  neighbours  in  Upper  Egypt  ;  their 
mummies  had  to  be  kept  out  of  reach  of  the  inundation.  Isolated 
monuments,  like  those  of  Abydos,  would  soon  have  filled  all  the 
available  space  upon  artificial  mounds,  such  as  those  upon  which 
the  cities  of  the  Delta  were  built.  The  problem  to  be  solved 
was,  however,  a  simple  one.  Since  there  could  be  no  question  of 
a  lateral  development,  like  that  of  the  Theban  tombs,  or  of  a  down- 
ward one,  like  that  of  the  Memphite  mummy  pits,  it  was  obvious 
that  the  development  must  be  upwards.  A  beginning  was  made 
by  constructing,  at  some  distance  from  a  town,  a  platform  of  crude 
brick,  upon  which,  after  its  surface  had  been  raised  above  the 
level  of  the  highest  floods,  the  mummies  were  placed  in  small 
chambers  closely  packed  one  against  another.  As  soon  as  the 
whole  platform  was  occupied,  another  layer  of  chambers  was 
commenced  above  it.  Champollion  discovered  the  remains  of 
two  such  cemeteries  in  the  immediate  neiofhbourhood  of  Sais. 
The  larger  of  the  two  was  not  less  than  1,400  feet  long,  500  feet 
wide,  and  80  feet  high  ;  an  enormous  mass  "  which  resembled,"  he 
said,  "a  huge  rock  torn  by  lightning  or  earthquake."^  No  doubt 
was  possible  as  to  the  character  of  the  mass  ;  Champollion  found 
among  the  debris  both  canopic  vases  and  funerary  statuettes. 
Within  a  few  years  of  his  death  Mariette  undertook  some  fresh 
excavations  in  the  same  neighbourhood ;  they  led  to  no  very 
important  results,  but  they  confirmed  the  justice  of  the  views 
enunciated  by  Champollion.  Most  of  the  objects  recovered  were 
in  a  very  bad  state  of  preservation  ;  the  materials  had  been  too 
soft,  and  in  time  the  dampness,  which  had  impregnated  the  base 
of  the  whole  structure,  had  crept  upwards  through  the  porous 
brick,  and  turned  the  whole  mass  into  a  gigantic  sponge. 

These  tombs  resemble  those  of  the  kinofs  in  havinsf  no  well ; 
and  as  for  the  funerary  chapel  we  do  not  as  yet  know  whether  it 
existed  at  all,  how  it  was  arranged,  or  what  took  its  place. 
Perhaps  each  of  the  more  carefully  constructed  tombs  was 
divided  into  two  parts,  a  chamber  more  or  less  decorated  and  a 
niche  contrived  in  the  masonry,  like  the  rock-cut  ovens  of  the 
Phoenician  catacombs.  As  soon  as  the  mummy  was  introduced, 
'  Lettrcs  Ecrites  d' Ei^vf'/e  ct  de  K'ubic,  2nd  edition,  186S,  p.  41. 


The  Tomb  under  the  New  Empire.  311 

the  niche  was  walled  up,  while  the  chamber  would  remain  open 
for  the  funerary  celebrations.  In  order  that  the  tombs  situated 
at  some  heicrht  above  the  level  of  the  soil,  and  in  the  middle  of 
the  block  of  buildings,  should  be  reached,  a  complicated  system 
of  staircases  and  inclined  planes  was  necessary.  In  the  course 
of  centuries  the  tombs  of  the  first  layer  and  especially  those  in  the 
centre  of  the  mass,  were  overwhelmed  and  buried  from  sight  and 
access  by  the  continual  aggregation  above  and  around  them. 
The  families  to  which  they  belonged,  perhaps,  became  extinct, 
and  no  one  was  left  to  watch  over  their  preservation.  Had  it 
not  been  for  the  infiltration  of  the  Nile  water,  these  lower  strata 
of  tombs  would  no  doubt  have  furnished  many  interesting  objects 
to  explorers.  In  any  case  it  would  seem  likely  that,  if  deep 
trenches  were  driven  through  the  heart  of  these  vast  agglomera- 
tions of  unbaked  brick,  many  valuable  discoveries  would  be  made.^ 
Such  a  system  left  slight  scope  to  individual  caprice ;  space  must 
have  been  carefully  parcelled  out  to  each  claimant,  and  the 
architect  had  much  less  elbow  room  than  when  he  was  cutting 
into  the  sides  of  a  mountain  or  building  upon  the  dry  soil 
of  the  desert.  In  the  royal  tombs  alone,  if  time  had  left  any 
for  our  inspection,  could  we  have  found  materials  for  judging 
of  the  funerary  architecture  of  Sais,  but,  as  the  matter  stands, 
we  are  obiieed  to  be  content  with  what  we  can  sfather  from 
Theban  and  Memphite  remains  as  to  the  prevailing  taste  of 
the  epoch. 

Upon  the  plateau  of  Gizeh,  to  the  south  of  the  Great  Pyramid, 
Colonel  Vyse  discovered  and  cleared,  in  1837,  an  important 
tomb  to  which  he  gave  the  name  of  Colonel  Campbell,  then 
British  Consul-General  in  Egypt.  The  external  part  of  the 
tomb  had  entirely  disappeared,  but  we  may  conclude  that  it  was 
in  keeping  with  the  subterranean  portion.  The  maker  of  the 
tomb  had  taken  the  trouble  to  define  its  extent  by  a  trench  cut 
around  it  in  the  rock.  The  external  measurements  of  this  trench 
are  89  feet  by  74.  A  passage  had  been  contrived  from  one  of  its 
faces  to  the  well,  which  had  been  covered  in  all  probability  by  an 
external  structure.  The  well  opens  upon  a  point  nearer  to  the  north 
than  the  south,  and  its  dimensions  are  quite  exceptional.  It  is  54 
feet  4   inches  deep,  and   31    feet  by  26  feet  8  inches  in  horizontal 

'  Similar  structures  exist  in  lower  Chaldnea,  and  have  furnished  many  inscriptions 
of  great  interest  and  value  to  assyriologists. 


312 


A  History  of  Art  in  Ancient  Egypt, 


section  ;   it  terminates   in  a  chamber  which  is  covered  by  a  vault 
1 1  feet  2  inches   thick.      It  was  not  however  in   this  chamber,  but 

in  small  lateral  grottos  that 
several  sarcophagi  in  granite, 
basalt,  white  quartz,  &c.,  were 
found.  The  remains  of  two 
other  wells  were  traced.  This 
tomb  dates  from  the  time  of 
Psemethek   I. 

In  the  necropolis  of  Thebes 
there  is  a  whole  district,  that  of 
the  hill  Bl  Assassif,  where  most 
of  the  tombs  belong  to  the 
twenty-sixth  dynasty.  Their 
external  aspect  is  very  different 
from  that  of  the  Theban  sepulchres.  The  entrance  to  the 
subterranean    galleries    is    preceded    by    a    spacious    rectangular 


Fig.  199.  — Sectiin  in  perspective  of 
"  Campbell's  tomb,"  from  the  plans 
and  elevations  of  Perring. 


t 


y///y'^ZiSZl 


<^';'i-- ;"i '  iii;\\v ' 


Amvyyo\//X:'A,. 


Y\o.  200. — Vertical  section  in  perspective  of  the  sarcophagus  chamber  of  the  above  tomb  ; 

compiled  from  Perring. 


courtyard,    excavated   in    the   rock  to   a  depth   of  10  or    12    feet. 
This  court  was  from  80  to    100  feet  long  and  from  /\o  to  80  feet 


The  Tomb  under  the  New  Empire. 


313 


wide;  it  was  surrounded  by  a  stone  or  brick  wall,  and  reached 
by  a  flight  of  steps.  A  pylon-shaped  doorway  gave  access  to  the 
courtyard  from  the  side  next  the  rock,  another  door  of  similar 
shape  opened  upon  the  plain  ;  but  some  tombs  are  entirely  closed 
(see  Fig.  201)  except  towards  the  mountain,  from  which  side  they 
may  be  entered  by  one  or  two  openings. 

The  subterranean  part  of  these  tombs  varies  in  size.  In  some 
of  them  a  gallery  of  medium  length  leads  to  a  single  chamber. 
In  others,  and  these  form  the  majority,  there  is  a  suite  of  rooms 
connected  by  a  continuous  gallery.     To  this  latter  group  belongs 


Fig.  201. — A  Tomb  on  El-Assafif  (drawn  in  perspective  from  the  plans  and  elevations  of  Prisse). 


the  largest  of  all  the  subterranean  Theban  tombs,  that  of  Peta- 
mounoph  (Fig.  191).  We  have  already  noticed  the  extraordinary 
dimensions  of  its  galleries  ;  there  are  also  two  wells  which  lead  to 
lower  sets  of  chambers.  All  the  walls  of  this  tomb  are  covered 
with  sculptured  reliefs.  In  the  first  chambers  these  are  in  very 
bad  condition,  but  they  improve  as  we  advance,  and  in  the 
farthest  rooms  are  remarkable  for  their  finish  and  good  preserva- 
tion. The  exterior  of  this  sepulchre  is  worthy  of  the  interior. 
The  open  court,  which  acts  as  vestibule,  is  100  feet  long  by 
80  wide.  An  entrance,  looking  towards  the  plain,  rises  between 
two  massive  walls  of  crude  brick,  and,  to  all  appearance,  was 
once  crowned  by  an  arcade  ;  within  it  a  flight  of  steps  leads 
VOL.    I.  s   s 


314  A  History  of  Art  in  Ancient  Egypt. 

down  into  the  court.  Another  door,  pierced  through  the  Hme- 
stone  rock,  leads  to  a  second  and  smaher  court  which  is  surrounded 
by  a  portico.  From  this  peristyle  a  sculptured  portal  leads  to 
the  first  subterranean  chamber,  which  is  53  feet  by  23,  and 
once  had  its  roof  supported  by  a  double  range  of  columns.  The 
next  chamber  is  ;^t,  feet  square.  With  a  double  vestibule 
and  these  two  great  saloons  there  was  no  lack  of  space  for 
gatherings  of  the  friends  and  relations  of  the  deceased. 

Neither  at  Memphis  nor  at  Thebes  do  the  tombs  of  this  late 
period  contain  any  novel  elements,  but  they  are  distinguished 
by  their  size  and  the  luxury  of  their  decoration.  In  some,  the 
wells  are  much  wider  than  usual  ;  in  others  it  is  upon  the  external 
courts  and  upon  those  double  gateways  which  play  a  part  similar 
to  that  of  the  successive  pylons  before  a  Theban  temple,  that 
extra  care  is  bestowed.  Vaults  are  frequently  employed  and 
help  to  give  variety  of  effect.  Private  tombs  become  as  large 
as  those  of  sovereigns,  and  similar  tendencies  are  to  be  found 
in  the  sculpture.  The  Egyptian  genius  was  becoming  exhausted, 
and  it  endeavoured  to  compensate  for  its  want  of  invention  and 
creative   imagination  by   an    increase   in  richness   and   elegance, 

A  chronological  classification  is  only  possible  in  the  cases  of 
those  tombs  which  bear  inscriptions  and  figures  upon  their  walls. 
At  Memphis,  as  at  Thebes,  the  remains  of  thousands  of  tombs 
are  to  be  found  which  o-ive  no  indication  of  their  date.  Some- 
times  they  are  deep  mummy  pits,  slightly  expanding  at  the 
bottom  ;  sometimes,  as  at  Thebes,  the  rock  is  honeycombed  with 
graves  between  the  border  of 'the  cultivated  land  and  the  foot 
of  the  Libyan  chain.  In  the  mountains  themselves  there  are 
hundreds  of  small  chambers,  with  bare  walls  and  often  extremely 
minute  in  size,  cut  in  the  sides  of  the  cliffs.  Finally,  there  are 
the  vast  catacombs,  in  each  chamber  of  which  the  mummies  of 
labourers  and  artisans  were  crowded,  often  with  the  instruments 
of  their  trade  by  their  sides. ^  Pits  full  of  mummified  animals 
are  also  found  among  the  human  graves.  Rhind  saw  some 
hundreds  of  the  mummies  of  hawks  and  ibises  taken  from  a 
tomb  at  the  foot  of  the  Drah-Abou'1-Neggah.  They  were  each 
enveloped  in  bandages  of  mummy  cloth,  and  beside  them 
numerous   small   boxes,   each   with   a  carefully    embalmed   mouse 

1   Rhind,  Thebes,  etc.  p.  51.      Belzoni,  A^irrative  of  the  Operations,  etc.  p.  167. 


The  Tomb  under  the  New  Empire.  315 

inside  it,  were  found.      The  lids  of  these  boxes  had  each  a  wooden 
mouse  upon  it,  sometimes  gilt.^ 

We  have  endeavoured  to  notice  all  that  is  of  importance  in  the 
funerary  architecture  of  Egypt,  because  the  Egyptian  civilization, 
as  we  know  it  through  the  still  existing  monuments,  carries  us 
much  farther  back  than  any  other  towards  the  first  awakening 
of  individual  thought  and  consciousness  in  mankind.  The 
primitive  conceptions  of  those  early  periods  were,  of  course, 
different  enough  from  those  to  which  mankind  was  brought  by 
later  reflection,  but  nevertheless  they  were  the  premises,  they 
contained  the  germ  of  all  the  development  that  has  followed  ;  and 
to  thoroughly  understand  the  origin  and  constitution  of  this 
development  it  was  necessary  to  follow  it  up  to  its  source,  to  the 
clearness  and  transparency  of  its  springs. 

The  art  of  Egypt  is  the  oldest  of  all  the  national  arts,  and  the 
oldest  monuments  of  Egypt  are  its  tombs.  By  these  alone  is 
that  earliest  epoch  in  its  history  which  we  call  the  Ancient 
Empire  known  to  us. 

In  later  ages  the  country  was  covered  with  magnificent  temples 
and  sumptuous  palaces,  but,  even  then,  the  tomb  did  not  lose  its 
pre-eminent  importance.  The  chief  care  of  the  Egyptian  in  all 
ages  was  his  place  of  rest  after  death.  Rich  or  poor,  as  soon  as 
he  arrived  at  full  age  he  directed  all  his  spare  resources  towards 
the  construction  and  decoration  of  his  tomb,  his  happy,  his 
eternal  "  dwelling,"  with  which  his  thoughts  were  far  more  pre- 
occupied than  with  that  home  in  the  light,  upon  which,  whether 
it  were  a  miserable  mud  hut  or  a  vast  edifice  of  brick  or  wood, 
he  looked  with  comparative  indifference,  regarding  It  as  an  encamp- 
ment for  a  day  or  a  mere  hotel  for  a  passing  traveller.  The  tombs, 
when  not  hollowed  from  the  living  rock,  were  built  with  such 
solidity  and  care  that  they  have  survived  in  thousands,  while  the 
palaces  of  great  sovereigns  have  perished  and  left  no  trace,  and 
the  temples  which  have  been  preserved  are  very  few  in  number. 

Being  mostly  subterranean  and  hidden  from  the  eye  of  man, 
sepulchres  preserved  the  deposits  entrusted  to  them  much  better 
than   buildings  upon  the  surface.     Those  among  the  latter  which 

1  Rhixd,  p.  52.  Among  the  mummified  animals  found  at  Thebes,  Wilkinson 
also  mentions  monkeys,  sheep,  cows,  cats,  crocodiles,  etc.  See  Belzoni, 
Narrative,  p.  187. 


3i6  A   History  of  Art  i.v  Ancient  Egypt. 

were  not  completely  destroyed,  were,  in  very  remote  times, 
damaged,  pillaged,  stripped,  and  mutilated  in.  a  thousand  ways. 
All  that  subsists  of  their  decoration — shattered  colossi  and  bas- 
reliefs  often  broken  and  disfigured — tells  us  nothing  beyond  the 
pomps  and  triumphs  of  official  history.  The  tombs  have  suffered 
much  less  severely.  The  statues,  bas-reliefs,  and  paintings  which 
have  been  found  in  them,  seem,  in  many  instances,  to  have  been 
the  work  of  the  very  men  whose  footprints  were  found  in  the 
sand  which  covered  their  floors  when  they  were  opened.^  The 
pictures  offered  to  our  eyes  by  the  walls  of  the  private  tombs  are 
very  different  from  those  which  we  find  in  the  temples.  All 
classes  of  the  people  appear  in  them  in  their  every-day  occupa- 
tions and  customary  attitudes.  The  whole  national  life  is 
displayed  before  us  in  a  long  series  of  scenes  which  comment 
upon  and  explain  each  other.  Almost  all  that  we  know  concerning 
the  industrial  arts  of  Egypt  has  been  derived  from  a  study  of 
her  tomb-houses.  In  every  hundred  of  such  objects  which  our 
museums  contain  at  least  ninety-nine  come  from  those  safe 
depositories. 

In  order  to  give  a  true  idea  of  the  national  character  of  the 
Egyptians  and  to  enable  the  originality  of  their  civilization  to 
be  thoroughly  understood,  it  was  necessary  to  show  the  place 
occupied  in  their  thoughts  by  the  anticipation  of  death  ;  it  was 
necessary  to  explain  what  the  tomb  meant  to  them,  to  what 
sentiments  and  beliefs  its  general  arrangements  and  its  principal 
details  responded  ;  it  was  necessary  to  follow  out  the  various 
modifications  which  were  brought  about  by  the  development  of 
religious  conceptions,  from  the  time  of  the  first  six  dynasties  to 
that  of  the  Theban  Empire. 

The  brilliant  architectural  revival  which  distinguished  the  first 
and  second  Theban  Empires  was  mainly  due  to  this  develop- 
ment of  religious  thought.  Almost  all  the  peculiarities  of  the 
Memphite  tomb  are  to  be  explained  by  the  hypotheses  with  which 
primitive   man  is"  content.      But  when  mature  reflection   evolved 

1  When  jSIariette  discovered  the  tomb  of  the  Apis  which  had  died  in  the  twenty- 
sixth  year  of  the  reign  of  Rameses  II.,  the  fingers  of  the  Egyptian  mason  who 
laid  the  last  stone  of  the  wall  built  across  the  entrance  to  the  tomb  were  found 
marked  upon  the  cement,  and  "  when  I  entered  the  sarcophagus-chamber  I  found 
upon  the  thin  layer  of  dust  which  covered  the  floor  the  marks  made  by  the  naked 
feet  of  the  workmen  who  had  placed  the  god  in  his  last  resting  place  3,200  years 
before."     (Quoted  by  RHOXfc;  in  L Egypte a  Petites  Journees,  p.  239.) 


The  Tomb  under  the  New  Emtire. 


17 


higher  types  for  the  national  gods,  when  polytheism  came  to  be 
superimposed  upon  fetishism,  the  hour  arrived  for  the  temple 
to  take  its  proper  place  in  the  national  life,  for  majestic  colonnades 
and  massive  pylons  to  be  erected  on  the  banks  of  the  life-giving 
river.  The  temple  was  later  than  the  tomb,  but  it  followed 
closely  upon  its  footsteps,  and  the  two  were,  in  a  fashion,  united 
by  those  erections  on  the  left  bank  of  the  Nile,  under  the  Theban 
necropolis,  which  partook  of  the  character  of  both.  The  temple 
is  the  highest  outcome  of  the  native  genius  during  those  centuries 
which  saw  Egypt  supreme  over  all  the  races  of  the  East,  supreme 
partly  by  force  of  arms,  but  mainly  by  the  superiority  of  her 
civilization. 


CHAPTER  IV. 

THE  SACRED  ARCHITECTURE  OF  EGYPT. 

§   I. —  TJie   Temple  under  tJie  Ancient  Empire. 

No  statue  of  a  god  is  known  which  can  be  confidently  referred 
to  the  first  six  dynasties.  Hence  it  has  sometimes  been  asserted 
that  at  that  early  period  the  Egyptian  gods  were  not  born,  if  we 
may  use  the  expression,  that  the  notions  of  the  people  had  not 
yet  been  condensed  into  any  definite  conception  upon  the  point. 
Some  writers  incline  to  believe  that  Egyptian  thought  had  not 
yet  reached  the  point  where  the  polytheistic  idea  springs  up,  that 
they  were  still  content  with  those  fetishes  which  retained  no 
slight  hold  upon  their  imaginations  until  a  much  later  period. 
Others  affirm  that  the  absence  of  gods  is  due  to  the  fact  that 
the  Egyptian  people  were  so  near  to  the  first  creation  of  mankind 
that  they  had  not  yet  forgotten  those  religious  truths  Avhich  were 
revealed  to  the  fathers  of  our  race.  They  believe  that  Egypt 
began  with  monotheism,  and  that  its  polytheistic  system  was 
due  to  the  gradual  degradation  of  pure  doctrine  which  took  place 
among  all  but  the  chosen  people. 

We  shall  not  attempt  to  discuss  the  latter  hypothesis  in  these 
pages.  It  is  a  matter  of  faith  and  not  of  scientific  demonstration. 
But  to  the  first  hypothesis  we  shall  oppose  certain  undoubted 
facts  which  prove  it  to  be,  at  least,  an  exaggeration,  and  that 
Egypt  was  even  in  those  early  days  much  farther  advanced, 
more  capable  of  analysis  and  reflection,  than  is  generally  imagined. 
M.  Maspero,  in  his  desire  for  enlightenment  upon  this  point, 
searched  the  epitaphs  of  the  ancient  empire,  and  found  in  their 
nomenclature  most  of  the  sacred  names  which,  in  later  phases 
of  the  national  civilization,  designate  the  principal  deities   of  the 


The  Temple  under  the  Ancient  Empire.  319 

Egyptian  pantheon.^  The  composite  proper  names  often  seem 
to  express  .individual  devotion  to  some  particular  deity,  and 
to  indicate  some  connection  between  the  latter  and  the  mortal 
who  bore  his  name  and  lived  under  his  protection.  These 
divinities  must,  then,  have  already  been  in  existence  in  the 
minds  of  the  Egyptians.  The  most  that  can  be  said  is  that 
they  had  not  yet  arrived  at  complete  definition  ;  art,  perhaps, 
had  not  vet  eiven  them  those  unchanorino-  external  features  and 
characteristics  which  they  retained  to  the  last  days  of  paganism. 
It  is  quite  possible  that  they  were,  more  often  than  not,  represented 
by  those  animals  which,  in  more  enlightened  times,  served  them 
for  symbols. 

If  the  inscription  and  the  figured  representations  still  existing 
upon  a  certain  stele  which  was  found  a  short  distance  eastward 
of  the  pyramid  of  Cheops-  are  to  be  taken  literally,  we  must 
believe  that  that  monarch  restored  the  principal  statues  of  the 
Egyptian  gods  and  made  them  pretty  much  what  we  see  them 
in  monuments  belonofino-  to  times  much  more  recent  than  his. 
The  upper  part  of  the  stele  in  question  shows  the  god  ot 
generation,  Horus,  Thoth,  Isis  in  several  different  forms,  Nephthys, 
Selk,  Horus  as  the  avenger  of  his  father,  Harpocrates,  Ptah, 
Setekh,  Osiris,  and  Apis.  These  statues  would  seem  to  have 
been  in  gold,  silver,  bronze  and  wood.  Mariette,  however,  is 
inclined  to  think  that  this  stele  does  not  date  from  the  time 
of  Cheops,  that  it  is  a  restoration  made  during  the  middle,  or, 
perhaps,  even  during  the  New  Empire.  On  attempting  to 
restore    so    venerable    a    lelic    of    the    author    of    the    greatest 

1  We  may  take  a  few  of  those  in  the  Boulak  Museum  at  random  :  Ra-Hotep 
(No.  590),  Hathor-En-Khe'ou  (588),  Ra-Xefer  (23),  Ra-Our  (25),  Sokar-Kha-Ca-u 
(993),  Noum-Hotep  (26),  Hathor-Xefer  (41),  Ptah-Asses  (500),  Ptah-Hotep,  iv:c. 
The  names  of  several  deities  are  to  be  found  in  the  inscription  upon  the  wooden 
coffin  or  mummy-case  of  Mycerinus,  now  in  the  British  Museum.  (Maspero, 
Histoire  Ancienne,  p.  75).  A  priest  of  Apis  is  mentioned  upon  a  tomb  of  the 
fourth  dynasty;  Osiris  is  invoked  in  the  steles  of  the  sixth  d} nasty.  (Boulak 
Catalogue,  Xo.  41.) 

Amen,  or  Ammon,  is  never  mentioned  on  the  monuments  of  the  Ancient  Empire  ; 
his  first  appearance  is  contemporary  with  the  twelfth  dynasty.  (Grebault,  Hymue 
a  Ammofi-Ra,  introduction,  part  iii.  p.  i3"6.)  This  is  natural  enough.  Amen  was 
a  Theban  god,  and  Thebes  does  not  seem  to  have  existed  in  the  time  of  the 
Ancient  Empire. 

2  Notice  des  principaux  ]\lonumei:ts  exposes  dans  les  Galeries  provisoires  dii  Mmee 
d'An/iquites   Eg},ptieniies  a  Boulak.     (Edition  of  1876,  Xo.  582.) 


320  A  History  of  Art  in  Ancient  Egypt. 

architectural  work  in  their  country,  the  scribes  may  have  allowed 
themselves  to  add  figures  treated  in  the  style  of  their  own  day 
to  the  ancient  text.  It  is  equally  doubtful,  moreover,  whether 
the  text  itself  dates  back  to  the  earlier  period,  and  we  need, 
perhaps,  accept  as  fact  only  this  :  that  Cheops  restored  an  already 
existing  temple,  assigned  to  it  certain  sacred  offerings  as 
revenues,  and  restored  the  statues  of  gold,  silver,  bronze,  and 
wood,  which  adorned  the  sanctuary. 

It  may  be  that  the  divine  effigies  were  abundant  even  in  those 
early  days,  but  that  they  have  failed  to  survive  to  our  day.  The 
portraits  of  so  many  private  individuals  have  been  preserved 
because,  in  their  desire  to  afford  a  proper  support  to  their  double, 
they  multiplied  their  own  images  to  as  great  an  extent  as  their 
means  would  allow.  Between  the  third  and  the  sixth  dynasties 
the  multiplication  of  these  portrait  statues  went  on  at  a  prodigious 
rate,  and  their  number  may  be  judged  from  the  fact  that  twenty 
were  taken  from  the  scrdab  alone  of  the  tomb  of  Ti.  The  greater 
their  number,  the  greater  was  the  chance  that  one  of  them  would 
escape  destruction. 

The  ingenuity  of  man  combined  with  the  process  of  nature 
to  preserve  these  figures  to  generations  in  the  remote  future, 
to  a  time  when  they  could  excite  interest  enough  to  save  them 
from,  destruction  and  to  ensure  them  a  chance  of  eternal  existence. 
To  the  thick  walls  of  the  mastabas,  to  the  well-concealed  serdabs, 
and,  more  than  all,  to  the  constantly  increasing  mask  of  sand 
laid  upon  the  cemeteries  of  Gizeh  and  Sakkarah  by  the  winds 
from  the  desert,  did  they  owe  their  preservation  through  the 
troublous  times  which  were  in  store  for  Egypt. 

In  their  more  exposed  situations  in  the  temples  and  private 
houses,  the  ima2fes  of  the  crods  ran  far  crreater  risks  than  the 
private  statues.  The  material  of  those  which  were  of  gold,  silver, 
or  bronze,  would  excite  dangerous  cupidity,  while  the  wooden  ones 
were  pretty  sure,  sooner  or  later,  to  be  destroyed  or  damaged  by 
fire.  The  stone  statues  mieht  be  overthrown  and  broken  and 
replaced  by  others  of  a  later  fashion,  besides  which  a  vast  number 
of  them  have,  perished  in  the  lime-kilns. 

We  see,  then,  that  supposing  Cheops  and  Chephren  to  have 
paid  their  devotions  before  statues  of  I  sis  and  Osiris,  Ptah 
and  Hathor,  there  is  nothing  very  astonishing  in  the  total 
disappearance    of   those    figures.      To    argue  from    their    absence 


The  Temple  under  the  Ancient  Empire,  321 

that  in  those  early  days  the  gods  of  the  Egyptians  were  not  yet 
created,  and  consequently  that  there  were  no  temples  erected  in 
their  honour,  Is  to  hazard  a  gratuitous  assertion  which  may  at  any 
time  be  disproved  by  some  happy  discovery,  such  as  that  which 
gave  us,  twenty  years  ago,  the  statue  of  Chephren  now  in  the 
Boulak  Museum.  Before  that  statue  was  found  it  might  similarly 
have  been  contended  that  the  series  of  royal  effigies  only 
commenced  with  the  first  Theban  Empire. 

We  possess,  moreover,  at  least  one  divine  effigy  which,  in  the 
opinion  of  all  contemporary  archaeologists,  dates  from  the  time  of 
the  ancient  monarchy,  namely,  the  great  Sphinx  at  Gizeh  (Fig. 
157)-  We  learn  from  epigraphic  writings  that  this  gigantic  idol, 
combining  the  body  of  a  crouching  lion  with  the  head  of  a  man, 
represents  Hor-em-Khou,  or  '  Horus  in  the  shining  Sun,' 
corresponding  to  the  Harmachis,  or  rising  sun,  of  the  Greeks. 
According  to  the  stele  above  quoted,  it  was  carved,  long  before 
the  time  of  Cheops,  out  of  a  natural  rock  which  reared  its  head 
above  the  sand  in  this  part  of  the  necropolis  ;  here  and  there  the 
desired  form  was  made  out  by  additions  in  masonry.^ 

As  primitive  Egypt  had  gods  she  must  have  had  temples. 
Few  traces  of  them  are  to  be  found,  however,  and  their  almost 
total  disappearance  is  mainly  owing  to  causes  which  merit 
careful   notice. 

Before  they  began  to  erect  stone  buildings,  the  early  Egyptians 

^  The  total  height  of  the  Sphinx  is  66  feet ;  the  ear  is  6  feet  4  inches  high  ;  the 
nose  is  6  feet,  the  mouth  7  feet  9  inches,  wide.  The  greatest  width  of  the  face  across 
the  cheeks  is  14  feet  2  inches.     If  cleared  entirely  of  sand  the  Sphinx  would  thus  be 
higher  than  a  five-storied  house.   For  the  history  of  the  Sphinx,  the  different  restora- 
tions which    it  has  undergone,  and  the  aspect  which  it  has  presented  at  different 
epochs,  see  Mariette,  Questions  relatives  mix  iwnvelles  Fouilles.     Our  plan  (Fig. 
204)  shows  the  wide  flight  of  steps  which  was  constructed  in  the  time  of  Trajan 
to  give  access  to  a  landing    constructed  immediately  in  front  of  the  fore-paws. 
Between  these  paws  a  little  temple  was  contrived,  where  the  steles  consecrated  by 
several  of  the  Theban  kings  in  honour  of  the  Sphinx  were  arranged.     Caviglia  was 
the  first  to  bring  all  these  matters  to  light,  in  1817,  but  the  ensemble,  as  it  now  exists, 
only  dates  back  to  the  Roman  epoch.     It  is  curious  that  neither  Herodotus,  nor 
Diodorus,  nor  Strabo,  mention  the  Sphinx.     Pliny  speaks  of  it  (N.  H.  xxxvi.  17); 
some  of  the  information  which  he  obtained  was  valuable  and   authentic,  but  it  was 
mixed  with  errors  ;  it  was  said  to  be,  he  tells  us,  tiie  tomb  of  the  king  Armais,  but  he 
knows  that  the  whole  figure  was  painted  red.  The  Denkmaler  of  Lepsius  (vol.  i.  pi.  30) 
gives  three  sections  and  a  plan  of  the  little  temple   between  the  paws.     The  same 
work  (vol.  V.  pi.  68)  contains  a  reproduction  of  the  great  stele  of  Thothmes  relative 
to  the  restoration  of  the  Sphinx. 

VOL.    I.  T    I- 


32  2  A   History  of  Art  in  Ancient  Egypt. 

made  constant  use  of  wood  for  many  ages.  Various  bas-reliefs 
and  paintings  prove  that  this  latter  material  was  never  entirely 
abandoned,  but  after  stone  and  brick  came  into  general  use  it 
was  reserved  for  special  purposes ;  it  was  usually  employed  in 
those  lighter  and  more  ephemeral  edifices  in  which  rapidity  of 
construction  was  the  chief  point  required.  When,  in  the  remains 
of  the  early  dynasties,  we  see  the  characteristics  of  wooden 
constructions  so  closely  imitated  in  stone,  we  are  constrained  to 
believe  that  wood  then  played  a  much  more  important  part  than 
under  the  Theban  princes.  Either  brick  or  stone  was  absolutely 
necessary  for  a  tomb,  because  they  alone  had  sufficient  durability, 
but  it  is  quite  possible  that  most  of  the  temples  were  of  wood. 
With  the  help  of  colour  and  metal,  wood  could  be  easily  made 
to  fulfil  all  the  conditions  of  a  temple.  The  shrine  which  enclosed 
either  a  statue  or  some  symbolic  object,  the  portico  which 
surrounded  the  inner  court,  the  furniture,  the  doors,  the  high 
palisade  which  enclosed  the  sacred  precinct,  might  all  have  been 
of  wood.  When  destroyed  by  accident  or  damaged  by  time,  such 
a  structure  could  be  quickly  and  easily  restored. 

We  may  admit,  however,  that  from  the  epoch  of  the  Pyramids 
onwards,  such  cities  as  Thinis,  Abydos,  Memphis,  and  others, 
constructed  their  temples  of  stone,  which,  in  the  then  state  of 
architectural  skill,  they  could  have  clone  without  any  serious 
difficulty.  The  chief  cause  of  the  disappearance  of  the  early  temples 
was  the  construction  of  those  that  came  after  them.  The  national 
taste  chano-ed  with  the  centuries,  and  the  time  came  when  the 
comparative  simplicity  of  the  primitive  erections  was  unable  to 
satisfy  the  longing  of  the  people  for  magnificence  and  splendour. 
New  temples,  more  vast  and  sumptuous  than  the  old,  were 
constructed,  and  the  substance  of  their  predecessors  was,  as  a 
matter  of  course,  employed  in  their  construction.  Sometimes  a 
fragmentary  inscription  or  a  piece  of  sculpture  betrays  the 
restoration  ;  and,  here  and  there,  inscriptions  on  the  later  building 
go  so  far  as  to  preserve  the  name  of  the  architect  of  the  first.  An 
instance  of  this  occurs  at  Denderah.  Champollion  discovered 
that  the  Ptolemaic  temples  almost  always  replaced  structures 
dating  from  the  great  Theban  or  Sait  dynasties.  The  island  of 
Philee,  however,  affords  an  exception  to  this  rule.^     These  temples 

^  Champollion,  Lettres  d'Agypte  et  de  Nubie,  pp.  125,  143,  and   166.     Under 
both  the  temples  at  Ombos,  ChampolHon  discovered  remains  of  a  building  of  the 


The  Temple  under  the  Ancient  Empire.  323 

he  calls  "  second  editions."  But  in  some  cases  they  were  third 
or  even  fourth  editions. 

But  in  spite  of  all  these  rearrangements  and  restorations,  a  few 
sacred  buildings  of  the  early  period  were  still  in  existence  during 
the  Roman  occupation  of  the  country,  and  were  then  shown  as 
curiosities.  This  we  may  gather  from  a  passage  in  Strabo. 
After  having  described,  with  much  precision,  the  disposition  of 
certain  buildings  which  are  easily  recognized  as  temples  built 
under  the  princes  of  the  New  Empire,  he  adds  :  "  At  Heliopolis, 
however,  there  is  a  certain  buildino^  with  several  ranees  of 
columns,  which  recalls,  by  its  arrangement,  the  barbarous  style  ; 
because,  apart  from  the  great  size  of  the  columns,  their  number 
and  their  position  in  several  long  rows,  there  is  nothing  graceful 
in  the  building,  nothing  that  shows  any  power  of  artistic  design  ; 
effort,  and  impotent  effort,  is  its  most  striking  characteristic."  ^ 
Lucian,  too,  was  thinking  of  the  same  building  in  his  treatise  upon 
the  Syrian  goddess,  when  he  said  that  the  Egyptians  had,  in 
ancient  times,  temples  without  sculptured  decorations.- 

One  of  these  '  barbarous '  temples,  as  Strabo  calls  them,  is 
supposed  to  have  been  discovered  in  the  small  building  disinterred 
by  Mariette  in  1853,  at  about  50  yards  distance  from  the  right 
foot  of  the  Sphinx  in  a  south-easterly  direction.  Mariette  cleared 
the  whole  of  the  interior,  and  by  means  of  a  flight  of  steps  well 
protected  from  the  sand,  he  provided  easy  access  to  it.  But  he  left 
the  external  walls  buried  as  he  found  them,  and  so  they  still 
remain. 

The  entrance  is  by  a  passage  about  66  feet  long  and  7  wide, 
which  runs  almost  in  an  easterly  direction  through  the  massive 
masonry  which  constitutes  the  external  wall.  About  midway 
along  this  passage  two  small  galleries  branch  off;  that  on  the 
right  leads  to  a  small  chamber,  that  on  the  left  to  a  staircase 
giving  access  to  the  terrace  above.  At  the  end  of  the  passage  we 
find  ourselves  at  one  of  the  angles  of  a  hall,  running  north  and 
south,  and  about  83  feet  long  by  23  wide.  The  roof  is  sup- 
ported   by    six    quadrangular     piers.       These    are    monoliths    16 

time  of  Thothmes  III.  The  same  thing  occurred  at  Edfou  and  at  Esneh.  We 
except  Philae,  because  there  is  good  reason  to  believe  that  in  the  time  of  the  Ancient 
Empire  that  island  did  not  exist,  and  that  the  cataract  was  then  at  Silsilis. 

^  Strabo,  xvii.  128  :  OvSev  4'xet  x°^P'-^^  "^^^  ypacfiLKov,  etc. 

2  Lucian,  §  3  :  'A^odvot  vtjol,  etc. 


324 


A  History  of  Art  in  Ancient  Egypt. 


feet  6  inches  high  and  3  feet  4  inches  by  4  feet  8  inches  in  section. 
Several  of  their  architraves  are  still  in  place.  These  are  stones 
about  JO  feet  in  length. ^  From  the  eastern  side  of  this  hall 
another  opens  at  right  angles.  This  second  hall  is  about  57  feet 
long  and  30  wide,  and  its  roof  was  supported  by  ten  columns 
similar  to    those  we  have  already  mentioned. 

From  the  south-west  anofle  of  the  first  hall  there  is  a  short 
corridor  which  leads  to  six  deep  niches  in  the  masonry,  arranged 
in  pairs  one  above  the  other,  and  apparently  intended  for  the 
reception  of  mummies. 


^^  -^  >-;;ti±££t^. 


Fig.    202. — The   Temple  of  the   Sphiax  (from   an  unpubhshed  plan  by  Mariette) 


In  the  middle  of  the  eastern  wall  of  this  same  large  chamber 
there  is  a  short  and  wide  passage  which  leads  to  a  third  and  last 
hall,  parallel  to  the  one  with  six  columns  ;  it  has  no  supporting 
pillars,  but  there  is,  in  the  centre  of  the  floor,  a  deep  well  which 
Mariette  cleared  from  the  sand  with  which  it  was  filled.     There 

1  The  piers  are  not  quite  equidistant ;  their  spacing  varies  by  some  centimetres. 
Exact  symmetry  has  been  sacrificed  in  consequence  of  the  different  lengths  of  the 
stones  which  formed  the  architrave. 


The  Temple  under  the  Ancient  Empire. 


325 


had  been  water  in  it,  because  it  was  sunk  below  the  level  of  the 
Nile.  At  the  bottom  nine  broken  statues  of  Chephren  were 
found  ;  they  were  not  copies,  one  from  the  other,  but  represented 
the  king  at  different  periods  of  his  life.  Several  stone 
cynocephali  were  also  found. 

At  each  end  of  this  hall  there  is  a  small  chamber  communicating 
with  it  by  short  corridors.  One  of  these,  that  in  the  northern 
angle  of  the  temple,  seems  to  have  communicated  with  the 
outward  air  by  an  irregular  opening  in  the  masonry. 


i »» 1  ^  _. 


Fig.  203. — Interior  of  the  Temple  of  the  Sphinx  (from  a  sketch  by  M.   Ernest  Desjardins)- 


The  materials  employed  in  the  interior  of  this  building  are 
rose  granite  and  alabaster.  The  supporting  piers  are  of  granite, 
the  lining  slabs  of  the  walls  and  the  ceiling,  alabaster.  Both 
these  materials  are  dressed  and  fixed  with  care  and  knowledge, 
but  in  no  part  of  the  temple  is  the  slightest  hint  at  a  moulding 
or  at  any  other  sort  of  ornament  to  be  found.  The  pillars  are 
plain  rectangular  monoliths ;  the  walls  are  without  either  bas- 
reliefs  or  paintings,  and  there  is  not  a  trace  of  any  inscription  on 
any  part  of  the  building.  The  external  walls  are  constructed 
of  the   largest   limestone  blocks   which  are  to  be  found  in   Egypt. 


326  A  History  of  Art  in  Ancient  Egypt. 


In  these  days  none  of  their  outward  faces  are  visible,  but 
according  to  Mariette,  who,  doubtless,  had  inspected  them  by 
means  of  temporary  excavation,  nothing  is  to  be  seen  but 
"  smoothlv  polished  surfaces,  decorated  with  long  vertical  and 
horizontal  grooves  skilfully  interlaced  ;  in  one  corner  there  is  a 
door,  the  only  one,  and  that  very  small."  ^ 

For  the  last  thirty  years  there  has  been  much  controversy  as 
to  the  true  character  of  this  curious  monument.  Mariette  himself 
allows  us  to  see  that  he  could  not  convince  himself  of  its  real 
meaning :  "  It  cannot  be  doubted  that  this  buildine  dates  from 
the  time  of  the  pyramids  ;  but  is  it  a  temple  or  a  tomb  ?  Its  ex- 
ternal appearance  is,  it  must  be  confessed,  more  that  of  a  tomb  than 
of  a  temple.  From  a  distance  it  must  have  looked  not  unlike  a 
mastaba  from  Sakkarah  or  Abousir,  which  it  but  slightly  excelled 
in  size.  The  six  deep  niches  which  exist  in  the  interior  recall 
the  internal  arrangements  of  the  pyramid  of  Mycerinus  and  the 
Mastabat-el-Faraoun,  and  the  general  plan  resembles  that  of 
several  other  tombs  in  the  neighbourhood.  It  appears,  therefore, 
that  the  hypothesis  which  would  make  it  a  sepulchre  might  be 
upheld  without  violating  the  rules  which  should  guide  the 
archaeologist.  .  .  .  On  the  other  hand  it  may,  very  naturally, 
be  asserted  that,  as  the  Sphinx  is  a  god,  it  must  be  the  Temple 
of  the  Sphinx."  ^ 

This  latter  hypothesis  seems  to  have  found  most  favour  with 
Mariette.  The  rectangular  niches,  which  at  first  seemed  to  him 
to  be  intended  for  funerary  purposes,  were  accounted  for  in 
another  way.  "  May  they  not  be  here,"  he  asks,  "  what  the 
crypt  is  at  the  temple  of  Denderah  ?  "  And  he  does  not  hesitate 
to  employ  the  terms  Temple  of  the  SpJiinx,  and  Temple  of 
Hannachis.  He  does  not  give  his  reasons,  but  to  some  extent 
we  can  supply  them.  Every  mastaba  of  any  importance  has 
funerary  representations  upon  it,  and  inscriptions  containing  both 
the  name  of  the  deceased  and  those  magical  formula;  which  we  have 
already  explained  ;  the  v>^alls  display  his  portrait  and  the  whole 
course  of  his  posthumous  life.  The  humblest  of  these  tombs 
shows  at  least  a  stele  upon  which  the  name  of  the  defunct  is  inscrib- 
ed together  with  the  prayer  which  is  to  insure  him  the  benefit  of 

1  Mariette,    Questions    relatives    aux   nouvelles     Fouilles    a   fain    en  Egypte. 
{Acadeinie  des  hiscripiions,  Comptes  Rendus des  Seances  de  FAnvce,  1877,  pp.  427-473.) 
"  Jti}ieraire  des  Invites  du  Vice-roi^  p.  99. 


The  Temple  under  the  Ancient  Empire.  327 

the  funerary  offerings  mentioned  upon  it.  The  tomb  is  thus 
consecrated  to  the  use  of  some  particular  person,  of  an  individual 
whose  name  is  placed  upon  it,  and  who  is  exclusive  owner  of  it 
and  its  contents  to  all  eternity.  In  this  temple  there  is  no  sign 
of  such  individual  appropriation.  Its  total  size  is  rather  in 
excess  of  that  of  the  largest  mastaba  yet  discovered  ;  its  materials 
are  finer  and  its  construction  more  careful.  The  bareness 
of  the  walls,  therefore,  can  hardly  be  attributed  to  want  of 
means  on  the  part  of  the  proprietor. 

It  is  true  that  in  many  tombs  the  decorative  works  have  never 
advanced  beyond  the  sketch  stage  ;  but  here,  although  the 
building  is  in  a  good  state  of  preservation,  not  the  slightest 
sign  is  to  be  discovered  that  any  funerary  ornamentation  had 
ever  been  attempted.  It  is  difficult  to  see  how  such  an  anomaly 
is  to  be  accounted  for  except  by  the  supposition  that  this  is  not 
a  tomb,  and  was  never  intended  to  be  one. 

An  examination  of  the  well  leads  to  the  same  conclusion.  In 
the  mastaba  the  well  is  simply  a  vertical  corridor  of  approach  to 
the  mummy  chamber.  Here  there  is  neither  sarcophagus  nor 
any  place  to  put  one  ;  no  enlargement  of  the  well  of  any  kind. 
But  of  the  three  parts  into  which  the  typical  Egyptian  tomb 
may  be  divided,  the  most  important  is  the  mummy  chamber.  It 
is  the  only  one  of  three  which  is  absolutely  indispensable.  It 
could,  in  itself,  furnish  all  the  necessary  elements  of  a  place  of 
sepulture,  because  it  could  ensure  the  safety  and  repose  of  the 
corpse  entrusted  to  it.  Where  there  is  no  mummy  chamber  there 
can  hardly  be  a  tomb,  strictly  speaking. 

The  anomalous  character  of  these  arrangements,  supposing  the 
building  to  be  a  tomb,  disappears  when  it  is  looked  upon  as  a 
temple.  Its  bareness  and  simplicity  agree  entirely  with  the 
descriptions  given  by  Plutarch  and  the  pseudo-Lucian  of  those 
early  Egyptian  temples  which  the  one  saw  with  his  own  eyes  and 
the  other  knew  by  tradition.  A  well  for  providing  the  water 
required  by  the  Egyptian  ritual  and  by  the  ablutions  of  the  priests 
would  be  in  its  proper  place  in  such  an  edifice,  while  the  similarity 
between  its  general  arrangements  and  those  of  the  mastabas  may 
easily  be  accounted  for  by  the  inexperience  of  the  early  architect. 
The  forms  at  his  command  were  too  few  and  too  rieid  to  enable 
him  to  mark,  with  any  certainty,  the  different  purposes  of  the 
buildings  which  he  erected.      The  architect  of  this  temple  seems, 


328  A  History  of  Art  in  Ancient  Egypt. 

however,  to  have  done  his  best  to  express  the  distinction.  In 
none  of  the  Memphite  mastabas  do  we  find  such  spacious  chambers 
or  so  many  large  and  well-wrought  monolithic  columns. 

Many  hypotheses  have  been  put  forward  in  the  attempt  to 
reconcile  these  two  explanations  of  the  "  Temple  of  the  Sphinx." 
but  we  cannot  discuss  them  here.  "  Why,"  asks  Mariette,  in  his 
recently  published  memoir,  "  should  not  the  temple  of  the  Sphinx 
be  the  tomb  of  the  king  who  made  the  Sphinx  itself  ?  "  This 
question  we  may  answer  by  two  more  :  Why  did  not  that  king 
decorate  the  walls  of  his  tomb  ?  and  why  did  he  have  neither 
sarcophagus  nor  sarcophagus  chamber  ?  Others  have  seen  in  it 
the  chapel  in  which  the  funerary  rites  of  Chephren  were 
performed ;  ^  a  theory  which  was  of  course  suggested  by  the 
discovery  of  that  king's  statues  in  the  well.  These  statues, 
we  are  told,  must  formerly  have  been  arranged  in  one  of  the 
chambers,  and,  in  some  moment  of  political  tumult,  they  must 
have  been  cast  into  the  well  either  by  foreign  enemies  or  by  the 
irritated  populace. 

In  all  probability  we  shall  never  learn  the  true  cause  of  this 
insult  to  the  memory  of  Chephren,  and  it  seems  to  us  to  be 
hazarding  too  much  to  affirm  that,  because  the  statues  of  that  king 
were  found  in  it,  the  building  we  are  discussing  must  have  been 
his  funerary  chapel.  It  is  very  near  the  Sphinx,  and  it  is  a 
considerable  distance  from  the  second  pyramid,^  which,  moreover, 
had  a  temple  of  its  own.  According  to  all  analogy,  the  funerary 
chapel  would  be  in  the  immediate  neighbourhood  of  the  mummy 
for  whose  benefit  it  was  erected. 

In  the  absence  of  any  decisive  evidence  either  one  way  or  the 
other,  the  most  reasonable  course  is  to  look  upon  this  building  as 
the  temple  in  which  the  worship  of  the  neighbouring  Colossus  was 
carried  on  :  as  the  temple  of  Harmachis,  in  a  word.  This  solution 
derives  confirmation  from  the  following  facts  mentioned  by 
Mariette:  "The  granite  stele,  erected  by  Thothmes  IV.  to 
commemorate  the  works  of  restoration  undertaken  by  him,  was 
placed  against  the  right  shoulder  of  the  Sphinx,  that  is  to  say,  at 
the  point  nearest  to  the  building  which  we  are  discussing.  In 
later  years  this  stele  and  some  others  representing  scenes  of 
adoration  which  were  added  by  Rameses  II.,  were  combined  into 

'  B^DEKER,  Guide  to  Lo7cer  Egypt,  p.  350. 
-  The  actual  distance  is  about  670  yards. 


The  Temple  under  the  Ancient  Empire.  329 


a  sort  of  small  building,  which  almost  directly  faced  any  one  comino- 
out  of  the  temple."  ^  One  of  Mariette's  favourite  projects  was  to 
clear  the  sphinx  down  to  its  base,  to  clear  all  the  space  between  it 
and  the  temple  (see  Fig.  204),  and  finally  to  build  a  wall  round 
the  whole  group  of  sufficient  height  to  keep  it  free  from  sand  in 
the  future.  In  Mariette's  opinion  such  an  operation  could  hardly 
fail  to  bring  to  light  more  than  one  monument  of  great  antiquity, 
of  an  antiquity  greater,  perhaps,  than  that  of  the  pyramids.  In 
any  case  it  would  lay  open  the  material  connection  between 
the  great  idol  and  its  temple,  and  would  help  us  to  reconstitute 
the  most  ancient  group  of  religious  buildings  in  existence. 

Those  structures  which  are  generally  called  the  temples  of  the 
Pyramids  belong  to  the  same  class  of  architecture  (Fig.  127).  We 
have  already  mentioned  them,  and  explained  how  they  are  to  the 
pyramids  what  the  funerary  chamber  is  to  the  mastaba.  We  must 
return  to  them  for  a  moment  in  their  capacity  as  temples.  The 
deceased  kings  in  whose  honour  they  were  erected  were 
worshipped  within  their  w^alls  even  down  to  the  time  of  the 
Ptolemies.  They  are  in  a  much  worse  state  of  preservation  than 
the  Temple  of  the  Sphinx.  Unlike  the  latter  they  were  not 
protected  by  the  sand,  and  their  materials  could  readily  be  carried 
away  for  the  construction  of  other  buildings.  Nothing  remains 
but  the  lower  courses  of  the  walls  and  their  footings,  so  that  no 
exact  agreement  has  yet  been  come  to  even  as  to  their  ground 
plan.  We  shall  quote,  however,  the  description  given  by  Jomard 
of  the  temple  belonging  to  the  third  pyramid.  The  French 
savants,  whose  visit  to  Egypt  took  place  nearly  a  century  ago, 
saw  many  things  which  have  disappeared  since  their  time. 

"  The  building  situated  to  the  east  of  the  third  pyramid  is 
remarkable  for  its  arrangement,  its  extent,  and  the  enormous  size 
of  the  blocks  of  which  it  is  composed.  In  plan  it  is  almost 
square,  being  177  feet  by  186.  On  its  eastern  side  there 
is,  however,  a  vestibule  or  annexe  103  feet  long  and  46 
wide.  .  .  .  Outside  the  vestibule  there  is  a  vast  courtyard  with 
two  lateral  openings  or  posterns ;  beyond  this  there  are  several 
spacious  saloons,  five  of  which  are  still  in  existence  ;  the  farthest  of 
these  is  the  same  size  as  the  vestibule,  and  is  exactly  opposite  to 
the  centre  of  the  pyramid,  from  which  it  is  but   43   feet  distant. 

^  Mariette,  Questions  relatives  aux  nouvelles  Fotiiltes,  etc. 
VOL.   I.  U    U 


330  A  History  of  Art  in  Ancient  Egypt. 

But   I    could  see  no  opening  in  that  part  of  the  wall  which  faced 
the  pyramid. 

"  The  general  symmetry  of  the  arrangement,  however,  suffices 
to  prove  the  connection  between  the  two  buildings. 

"  After  havine  studied  the  construction  and  the  materials  of  the 
Theban  edifices,  I  was  astonished  by  the  size  of  the  stones  here 
made  use  of,  and  the  care  with  which  they  were  fixed.  The  walls 
are  6  feet  9  inches  thick  ;  a  thickness  which  is  determined  by  that 
of  the  stones  employed.  Their  length  varies  from  12  to  23  feet. 
At  first  I  took  these  blocks  for  the  face  of  the  rock  itself, 
elaborately  worked  and  dressed,  and  I  might  not  have  discovered 
my  mistake  but  for  the  cemented  joints  between  the  courses. 

"  The  eastward  prolongation  or  annexe  is  formed  of  two  huge 
walls,  which  are  not  less  than  13  feet  4  inches  thick.  It 
may  well  be  asked  why  such  walls  should  have  been  constructed, 
seeing  that  had  they  been  of  only  half  the  thickness  they  would 
have  been  quite  as  durable  and  solid. 

"  This  building  forms,  as  it  were,  the  continuation  of  an  enclined 
plane  or  causeway  laid  out  at  right  angles  to  the  base  line  of 
the  third  pyramid,  and  leading  up  to  it."  ^ 

Jomard  appears  to  have  found  no  traces  of  pillars  in  any  part 
of  the  edifice  ;  but  Belzoni,  whose  description  is,  however,  both 
short  and  confused,  seems  to  have  found  them  in  the  temple  of  the 
second  pyramid.  He  speaks  oi  "d.  portico,  and  he  adds  that  some 
of  its  blocks  were  24  feet  high,"  or  about  the  same  height  as  the 
monoliths  in  the  Temple  of  the  Sphinx.  Such  blocks  would, 
of  course,  be  the  first  to  be  carried  off  and  used  elsewhere. 

In  spite  of  this  difference  many  of  the  peculiar  arrangements  of 
the  sphinx  temple  are  repeated  in  these  buildings.  There  is  the 
same  squareness  of  plan,  the  same  multiplicity  of  internal 
chambers,  the  same  employment  of  huge  masses  of  stone  and  the 
same  care  and  skill  in  dressing  and  fixing  them.  It  is  now 
impossible  to  say  whether  these  buildings,  when  complete,  were 
decorated  or  not ;  it  is  certain  that  at  the  present  day  no  sign  of 
any  ornamentation,  either  carved  or  painted,  is  to  be  found  upon 
them. 

We  see,  then,  that  the  religious  architecture  of  the  early  empire 
is  represented  by  a  very  small  number  of  monuments,  ot  which 

'  Description  de  T Egypte,  Ant.,  vol.  v.  p.  654. 

-  Belzoni,  Narrative  of  tJie  Operafio7is,  ect.  pp.  261-2. 


o 
p 

4) 


■  3 
O 

M 

c 


c 


a. 

CO 


C/3 


J3 


o 

a 


The  Temple  under  the  Middle  Empire.  333 


only  one  is  in  a  good  state  of  preservation.  Wlien  we  recall  the 
texts  which  we  have  quoted,  when  we  compare  the  temple  of  the 
Sphinx  with  tombs  like  the  pyramids  or  the  sepulchre  of  Ti,  we 
must  acknowledge  that  the  energies  of  the  Egyptians  during  the 
early  dynasties  were  mainly  directed  to  their  resting-places  after 
death,  that  the  worship  of  the  dead  held  the  largest  place  in  their 
religious  life.  Their  temples  were  small  in  size,  insignificant  in 
height,  and  severe  in  their  absence  of  ornament.  They  give 
slight  earnest  of  the  magnificent  edifices  which  the  country  was  to 
rear  some  ten  or  fifteen  centuries  afterwards  at  the  command  of 
the  great  Theban  pharaohs.  The  monolithic  pillars,  however,  of 
which  we  have  spoken,  give  some  slight  foretaste  of  a  feature 
which  was  to  reach  unrivalled  majesty  in  the  hypostyle  halls  of 
Karnak  and  Luxor. 


§  2.    The  Temple  tmder  the  Middle  Einph'e. 

No  temples  constructed  under  the  first  Theban  empire  are  now 
in  existence  ;  and  yet  the  Egyptians  had  then  generally  adopted 
the  worship  of  all  those  deities  whose  characters  and  attributes 
have  been  made  known  to  us  through  the  monuments  of  the  New- 
Empire.  The  Theban  triad  received  the  homage  of  the 
Ousourtesens  and  Amenemhats  ;  its  principal  personage,  Amen, 
or  Ammon,  identified  with  Ra,  already  showed  a  tendency  to 
become  a  supreme  deity  for  the  nation  as  a  whole.  To  him 
successful  sovereigns  attributed  their  successes  both  of  peace  and 
war.  As  the  god  of  the  king  and  of  the  capital.  Amen  acquired 
an  uncontested  superiority  throughout  the  whole  valley  of  the 
Nile,  which  affected,  however,  neither  the  worship  of  the  local 
deities,  nor  the  homage  paid  by  every  man  and  woman  in  Egypt 
to  Osiris,  the  god  to  whom  they  looked  for  happiness  beyond  the 
grave. 

Art,  at  this  period,  had  advanced  so  far  that  there  was  no  longer 
any  difficulty  in  marking  the  distinction  between  the  temple  and 
the  tomb.  In  the  sepulchres  at  Beni- Hassan  which  date  from  the 
twelfth  dynasty,  we  find  two  very  different  kinds  of  support,  and 
there  was  nothing  to  prevent  the  forms  employed  in  these  rock-cut 
chambers  from  being  made  use  of  in  constructed  buildings,  seeing 
especially  how  skilful  the  Egyptians  had   shown  themselves  to  be 


334  A  History  of  Art  in  Ancient  EcvrT. 

in  working  the  excellent  materials  provided  for  them  by  nature. 
The  architect  could,  if  he  had  chosen,  have  multiplied  to  infinity 
those  stone  supports  which  his  distant  predecessors  had  employed, 
apparently  with  some  inkling  of  their  future  possibilities.  The 
obelisk  set  up  by  Ousourtesen  at  Heliopolis,  proves  that  the  cutting 
and  polishing  of  those  monoliths  was  understood  in  his  time,  'and 
as  the  obelisk  seems  always  to  have  been  closely  combined  with 
the  pylon,  it  is  not  improbable  that  the  religious  edifices  of  the 
time  of  Ousourtesen  were  prefaced  by  those  huge  pyramidoid 
masses.  The  hypostyle  halls,  the  pylons,  and  the  obelisks  of 
the  New  Empire  differed  from  those  of  the  Middle  Empire  rather 
in  their  extent  and  in  the  mao-nificence  of  their  decoration,  than 
in  their  o-eneral  arranofement. 

Of  all  the  temples  then  constructed,  the  only  one  which  has  left 
any  apparent  traces  is  that  which  was  erected  at  Thebes  by  the 
princes  of  the  twelfth  dynasty  to  the  honour  of  Amen,  It  forms 
the  central  nucleus  around  which  the  later  buildings  of  Karnak 
have  been  erected.  The  name  of  Ousourtesen  is  to  be  read  upon 
the  remains  of  the  polygonal  columns  which  mark,  it  is  believed, 
the  site  of  the  sanctuary  properly  speaking,  between  the  granite 
chambers  and  the  buildings  of  Thothmes  III.  ;  these  columns,  like 
those  at  Beni-Hassan,  are  hexagonal  in  section.^ 

Of  many  other  buildings  erected  at  this  period,  nothing  is  left  to 
us  beyond  tradition  and  the  mere  mention  of  them  in  various  texts. 
This,  however,  is  sufficient  to  prove  their  existence.  We  shall 
choose  examples  of  them  from  the  two  extremities  of  Egypt. 
Nothing  has  been  found  of  that  great  tem[)le  at  Heliopolis  which 
all  the  Greek  travellers  visited  and  described,  but  we  know  that  a 
part,  at  least,  of  its  buildings  dated  from  the  time  of  the  first 
Theban  empire,  because  a  MS.  at  Berlin,  published  by  Herr 
Ludwig  Stern  in  1873,  narrates  the  dedication  of  a  chapel  by 
Ousourtesen.  It  is  probable  that  the  obelisk  was  in  the  portion 
then  built  and  consecrated  to  the  god  Ra. 

At  Semneh,  in  Nubia,  the  fortress  on  the  left  bank  of  the  river 
contains  a  temple  of  Thothmes  HI.,  which,  according  to  the 
pictures  and  inscriptions  which  cover  its  walls,  is  no  more  than  a 

^  The  little  that  now  remains  of  the  columns  and  foundations  of  the  ancient 
temple  is  marked  in  the  plan  which  forms  plate  6  of  Mariette's  Karnak,  Fig.  a. 
In  plate  8  the  remains  of  all  statues  and  inscriptions  which  date  from  the  same 
period  are  figured.     See  also  pages  ^(y,  37,  and  41-45  of  the  text. 


The  Temple  under  the  New  Emtire.  335 

restoration  of  one  built,  in  the  first  instance,  in  honour  of 
Ousourtesen  III.  This  latter  prince  was  deified  at  Semneh  after 
his  death,  and  his  worship  continued  for  more  than  ten  centuries. 
His  temple,  which  had  fallen  into  ruin  during  the  first  reigns  of  the 
eighteenth  dynasty,  was  reconstructed  by  Thothmes,  and  that 
prince  is  represented  doing  homage  to  the  local  deities,  among 
whom  Ousourtesen  may  be  discovered  presenting  his  pious 
successor  to  the  other  gods. 

Many  more  instances  might  be  given,  but  the  monuments  of 
the  second  Theban  empire  demand  our  attention.  A  Thothmes 
or  an  Amenophis,  a  Seti  or  a  Rameses,  could  dispose  of  all  the 
resources  of  a  rich  country  and  of  an  aged  civilization  for  the 
construction  of  their  edifices,  edifices  so  great  and  splendid  that 
they  ran  no  risk  of  being  destroyed  in  later  times  for  the  sake  of 
constructing  others  still  more  sumptuous  ;  besides  which  they  were 
built  at  the  zenith  of  the  national  greatness,  at  the  moment  when, 
in  the  Egyptian  character,  all  the  energy  of  an  unconquered 
people  was  combined  with  the  knowledge  and  experience 
resulting  from  an  old  and  complex  social  system.  In  the  later 
ages  of  the  monarchy  a  few  unimportant  additions  were  made,  an 
obelisk  or  a  pylon  here,  there  a  court,  a  colonnade,  or  a  few 
chambers  ;  but  the  great  temples  of  the  New  Empire  have  come 
down  to  us  with  few  modifications  beyond  those  caused  by  the 
three  thousand  years  through  which  they  have  existed,  and  we 
have  little  difficulty  in  restoring  them,  on  paper,  to  the  condition  In 
which  they  were  left  by  the  great  monarchs  of  the  eighteenth, 
nineteenth,  and  twentieth  dynasties.  The  later  additions,  although 
they  render  the  ground-plans  more  complicated,  fail  to  hide  or 
materially  affect  the  general  characteristics  of  the  buildings,  and  in 
no  way  prevent  us  from  recognizing  and  defining  the  spirit  and 
originality  of  their  conception. 


§  3.    The   Temple  tinder  the  New  Empire. 

Before  we  cross  the  threshold  of  the  great  Theban  temples  and 
attempt  to  evolve  order  out  of  their  complexity  of  courts,  halls, 
porticos  and  colonnades.  It  may  be  convenient  to  describe  their 
approaches.  Each  temple  had  its  external  and  accessory  parts 
which   had  their  share  in  the  religious  ceremonies  of  which  It  was 


A   History  of  Art  in  Ancient  Egypt, 


the  theatre,  and  it  would  be  difficult  to  make  its  economy  under- 
stood unless  we  began  by  noticing  them  in  detail. 

One  of  the  first  signs  which  denoted  to  visitors  the  proximity  of 
an  Egyptian  temple  was  what  the  Greek  travellers  called  a  Bpofios, 
that  is  to  say  a  paved  causeway  bordered  on  each  side  with  rams 
or  sphinxes,  their  heads  being  turned  inwards  to  the  road.  These 
avenues  vary  in  width,  that  at  Karnak  is  76  feet  between  the 
inner  faces  of  the   pedestals  ;  ^  within  the  precincts  of  the  sacred 


Fig.  205. — Ram,  or  Kriosphinx,  from    Karnak. 


edifice,  between  the  first  and  second  pylon,  this  width  underwent  a 
considerable  increase.  The  space  between  one  sphinx  and  another 
on  the  same  side  of  the  causeway  was  about  13  feet.  The  dromos 
which  led  from  Luxor  to  Karnak  was  about  2,200  yards  long; 
there  must,  therefore,  have  been  five  hundred  sphinxes  on  each 
side  of  it.  At  the  Serapeum  of  Memphis  the  sphinxes  which 
Mariette   found  by    digging   70  feet    downwards    into    the   sand 

^  Mariette,  Karnak,  p.  4. 


The  Temple  under  the  New  Empire.  ^j-r 


were  still   nearer  to  one  another  ;  ^    the  dromos  which  they  lined 
was  found  to  be  50  feet  wide  and  about  1,650  yards  long. 

Following  our  modern  notions  we  should,  perhaps,  expect  to  find 
these  causeways  laid  out  upon  an  exactly  rectilinear  plan.  Thev 
are  not  so,  however.  It  has  sometimes  been  said  that  one  of  the 
characteristic  features  of  Egy-ptian  architecture  is  its  dislike,  or 
rather  hatred,  of  a  rigorous  symmetn.-.  Traces  of  this  hatred  are 
to  be  found  in  these  avenues.  The  ver}-  short  ones,  such  as 
those  which  extend  between  one  pylon  and  another,  are  straight, 
but  those  which  are  prolonged  for  some  distance  outside  the  build- 
ings of  the  temple  almost  always  make  some  abrupt  turns.  The 
Serapeum  dromos  undergoes  several  slight  changes  of  direction, 
in  order,  no  doubt,  to  avoid  the  tombs  between  which  its  course  lay. 
We  find  the  same  thing  at  Karnak,  where  the  architect  must  have 
had  different  motives  for  his  abandonment  of  a  straiofht  line.  At 
the  point  where  the  man-headed  sphinxes  of  Horus  succeed  to  those 
sphinxes  without  inscriptions  the  date  of  which  ]\Iariette  found  it 
impossible  to  determine,  the  axis  of  the  avenue  inclines  gently  to 
the  left. 

These  avenues  of  sphinxes  are  always  outside  the  actual  walls 
ot  the  temple,  from  which  it  has  been  inferred  that  they  were 
merely  ornamental,  and  without  religious  signification. - 

Some  of  the  great  temples  have  several  of  these  avenues  leading 
up  to  their  different  gates.  It  is  within  these  gates  only  that  the 
sacred  inclosure  called  by  the  Greeks  the  -efi€vo9  commences. 
The  religious  ceremonies  were  all  performed  within  this  space, 
which  was  inclosed  by  an  encircling  wall  built  at  sufficient  distance 
from  the  actual  temple  to  allow  of  the  marshalling  of  processions 
and  other  acts  of  ritual. 

These  outer  walls  are  of  crude  brick.  At  Karnak  thev  are 
about  2)3  feet  thick,  but  as  their  upper  parts  have  disappeared 
through  the  perishable  nature  of  the  material,  it  is  impossible  to 
say   with   certainty  what    their  original    height   may  hav^e    been.-^ 

^  We  ma)'  infer  from  what  Mariette  says  that  they  were  separated  from  one 
another  by  a  distance  of  1 2  feet  4  inches. 

-  Mariette,  Karnak,  p.  5.  We  find,  however,  that  sphinxes  were  sometimes 
placed  in  the  interior  of  a  temple.  The  two  fine  sphinxes  in  rose  granite  which 
form  the  chief  ornaments  of  the  principal  court  of  the  Boulak  museum,  were 
found  in  one  of  the  inner  halls  of  the  temple  at  Karnak.  They  date,  probably, 
from  the  time  of  Thothmes  III.,  to  whom  this  part  of  the  building  owes  its  existence. 

2  Description,  etc.  ;  Description  ginirale  Je  T/icfics,  section  viii.  §   i. 

VOL.    L  XX 


T,T,S  A  History  of  Art  in  Ancient  Egypt. 

Their  summits,  with  their  crenellated  parapets,  must  have  afforded 
a  continuous  platform  connected  with  the  flat  tops  of  the  pylons 
by  flights  of  steps. 

"  These  inclosing  walls  served  more  than  one  purpose.  They 
marked  the  external  limits  of  the  temple.  They  protected  it 
against  injury  from  without.  When  their  height  was  considerable, 
as  at  Denderah,  Sais,  and  other  places,  they  acted  as  an  impenetrable 
curtain  between  the  profane  curiosity  of  the  external  crowd  and  the 
mysteries  performed  within  ;  and  when  they  had  to  serve  their  last 
named  purpose  they  were  constructed  in  such  a  fashion  that  those 
without  could  neither  hear  nor  see  anything  that  passed. 

"It  is  probable  that  the  walls  of  Karnak  served  all  three 
purposes.  There  are  four  of  them,  connected  one  with  another  by 
avenues  of  sphinxes,  and  all  the  sacred  parts  of  the  building, 
except  a  few  chapels,  are  In  one  of  the  four  inclosures.  .  .  .  Their 
height  was  at  least  sufficient  to  prevent  any  part  of  the  inside  from 
being  overlooked  from  any  quarter  of  the  city,  so  that  the  cere- 
monies In  the  halls,  under  the  colonnades,  or  upon  the  lakes  could 
be  proceeded  with  in  strict  isolation  from  the  outer  world. ^  We 
may  therefore  perceive  that,  on  certain  occasions,  these  inclosures 
would  afford  a  sanctuary  which  could  not  easily  be  violated,  while 
they  would  keep  all  those  who  had  not  been  completely  initiated 
at  a  respectful  distance  from  the  holy  places  within."^ 

These  walls  were  pierced  In  places  by  stone  doorways,  embedded 
in  the  masses  of  crude  brick,  whose  highest  parts  always  rose 
more  or  less  above  the  battlements  of  the  wall  (Fig.  206).  At 
those  points  where  the  sphinx  avenues  terminated,  generally  at  the 
principal  entrance  of  the  temple  but  sometimes  at  secondary 
gateways,  these  portals  expanded  Into  those  towering  masses  which 
by  their  form  as  well  as  their  size,  so  greatly  impress  the  traveller 
who  visits  the  ruins  of  ancient   Egypt.     These  masses  have   by 

^  The  wall  of  the  principal  inclosure  at  Denderah,  that  on  the  north,  is  not  less 
than  33  feet  high,  and  between  30  and  40  thick  at  the  base.  Its  surface  is 
perfectly  smooth  and  naked,  without  ornament  of  any  kind,  or  even  rough-cast. 
(Mariette,  Denderah,  p.  27.)  At  Karnak  the  bounding  walls  are  in  a  much  worse 
state  of  preservation  ;  they  are  ten  or  twelve  centuries  older  than  those  of  Denderah, 
and  those  centuries  have  had  their  effect  upon  the  masses  of  crude  brick.  Our  only 
means  of  estimating  their  original  height  is  by  comparing,  in  the  representations 
furnished  to  us  by  certain  bas-reliefs,  the  height  of  walls  with  that  of  the  pylons 
on  which  they  abut. 

2  Mariette,  Karnak,  pp.  5,  6. 


Fig.  206. — Gateway  and  boundary  wall  of  a  temple  ;  restored  by  Ch.  Chipiez. 


The  Temple  under  the  New  Empire.  341 

common  consent  been  na.med  py/o7is.  They  seem  to  have  been  in 
great  favour  with  the  architects  of  Egypt,  who  succeeded  by  their 
means  in  rendering  their  buildings  still  more  original  than  they 
would  have  been  without  them.^ 

The  pylon  is  composed  of  three  parts  intimately  allied  one  with 
another  ;  a  tall  rectangular  doorway  is  flanked  on  either  hand  by  a 
pyramidal  mass  rising  high  above  its  crown.  Both  portal  and 
towers  terminate  above  in  that  hollow  gorge  which  forms  the 
cornice  of  nearly  all  Egyptian  buildings.  Each  angle  of  the 
towers  is  accentuated  by  a  cylindrical  moulding,  which  adds  to  the 
firmness  of  its  outlines.  This  moulding  bounds  all  the  flat 
surfaces  of  the  pylon,  which  are,  moreover,  covered  with  bas-reliefs 
and  paintings.  It  serves  as  a  frame  for  all  this  decoration,  which  it 
cuts  off  from  the  cornice  and  from  the  uneven  line  which  marks  the 
junction  of  the  sloping  walls  with  the  sandy  soil.  From  the  base 
of  the  pylon  spring  those  vertical  masts  from  whose  summits  many 
coloured  streamers  flutter  in  the  sun.-  In  consequence  of  the  inclina- 
tion of  the  walls,  these  masts,  being  themselves  perpendicular,  were 
some  distance  from  the  face  of  the  pylon  at  its  upper  part.  Brackets 
of  wood  were  therefore  contrived,  through  which  the  masts  passed 
and  by  which  their  upright  position  was  preserved  ;  without  some 
such  support  they  would  either  have  been  liable  to  be  blown  down 
in  a  high  wind,  or  would  have  had  to  follow  the  inclination  of  the 
wall   to  which    they   were  attached,   which    would    have   been  an 

'  The  word  ttuAwv  strictly  means  the  place  before  the  door  (like  Ovpm),  or  rather 
great  door  (upon  the  augmentative  force  of  the  suffix  wv,  wros,  see  Ad.  Regnier, 
Traite  de  la  Forinatioji  des  Mots  dans  la  Laiigue  Grecque,  §  184).  Several  passages 
in  PoLYBius  {Thesaurus^  s.  v.)  show  that  in  the  military  language  of  his  time  the 
term  was  employed  to  signify  a  fortified  doorway  with  its  flanking  towers  and  other 
defences.  We  may  therefore  understand  why  Diodorus  (i.  47)  made  use  of 
it  in  his  description  of  the  so-called  tomb  of  Osymandias.  Strabo  (xvii.  i, 
28)  preferred  to  use  the  word  irpoirvXixiv.  Modern  usage  has  restricted  the  word 
propylajim  to  Greek  buildings,  and/j7f«  to  the  great  doorways  which  form  one  of 
the  most  striking  features  of  Egyptian  architecture. 

'■^  We  learn  the  part  played  by  these  masts  and  banners  in  Egyptian  decoration 
entirely  from  the  representations  in  the  bas-reliefs.  The  facade  of  the  temple  of 
Khons  is  illustrated  in  one  of  the  bas-reliefs  upon  the  same  building.  That  relief 
was  reproduced  in  the  Description  de  PEgypte  (vol.  iii.  pi.  57,  Fig.  9),  and  is  so  well 
known  that  we  refrained  from  giving  it  in  these  pages.  It  shows  the  masts  and 
banners  in  all  their  details.  Another  representation  of  the  same  kind  will  be  found 
in  Cailliaud,j  Voyage  a  Meroe,  plates,  vol.  ii.  pi.  64,  Fig.  i.  See  in  the  text, 
vol.  iii.  p.  298.  It  is  taken  from  a  rock-cut  tomb  between  Dayr-el-Medinet  and 
Medinet-Abou. 


342  A  History  of  Art  in  Ancient  Egypt. 

unsightly  arrangement.  The  interiors  of  the  pylons  were  partly 
hollow ;  they  inclosed  small  chambers  to  which  access  was 
obtained  by  narrow  staircases  winding  round  a  central  square 
newel.  The  object  of  these  chambers  seems  to  have  been  merely 
to  facilitate  the  manoeuvring  of  the  masts  and  their  floating  banners, 
because  when  the  latter  were  in  place,  the  small  openings  which 
gave  light  to  the  chambers  were  entirely  obscured. 

If  the  pylons  had  been  intended  for  defensive  purposes,  the 
doors  in  their  centres  would  have  been  kept  in  rear  of  the  flanking 
towers,  as  in  more  modern  fortifications.  But  instead  of  that 
being  the  case  they  are  slightly  salient,  which  proves  conclusively 
that  their  object  was  purely  decorative. 

The  pylon  which  we  have  taken  as  a  type  of  such  erections, 
is  one  of  those  which  inclose  a  doorway  opening  in  the  centre  of 
one  of  the  sides  of  the  brick  inclosure,  it  may  be  called  an  external 
pylon,  or  a  pi^o-pylon,  to  make  use  of  the  word  proposed  by  M. 
Ampere,  but  in  all  temples  of  any  importance  several  pylons 
have  to  be  passed  before  the  sanctuary  is  reached.  At  Karnak, 
for  instance,  in  approaching  the  great  temple  from  the  temple 
of  Mouth,  the  visitor  passes  under  four  pylons,  only  one  of  which, 
the  most  southern,  is  connected  with  the  inclosing  wall.  So, 
too,  on  the  west.  After  passing  the  pylon  in  the  outer  wall, 
another  has  to  be  passed  before  the  hypostyle  hall  is  reached,  and 
a  third  immediately  afterwards.  Then,  behind  the  narrow  court 
which  seems  to  cut  the  great  mass  of  buildings  into  two  almost 
equal  parts,  there  are  three  more  at  very  slight  intervals.  Thus 
M.  Mariette  counts  six  pylons,  progressively  diminishing  in  size, 
which  lie  in  the  way  of  the  visitor  entering  Karnak  by  the  west 
and  passing  to  the  east.     At  Luxor  there  are  three. 

A  glance  at  our  general  view  of  the  buildings  of  Karnak  will 
give  a  good  idea  of  the  various  uses  to  which  the  Egyptian  archi- 
tect put  the  pylon. ^  There  is  the  pro-pylon  ;  there  are  those 
pylons  which,  when  connected  with  curtain  walls,  separate  one 
courtyard  from  another ;  there  are  those  again,  which,  placed 
immediately  in  front  of  the  hypostyle  halls,  form  the  fa9ades 
of  the    temples  properly    speaking.     The  temple  is  always   con- 

'  This  plate  (iv.)  is  not  a  picturesque  restoration ;  it  is  merely  a  map  in  relief. 
Only  those  buildings  are  marked  upon  it  which  have  left  easily  traceable  remains. 
No  attempt  has  been  made  to  reconstruct  by  conjecture  any  of  those  edifices  which 
are  at  present  nothing  but  confused  heaps  of  debris. 


The  Temple  under  the  New  Empire.  34 


0Hv3 


cealed  behind  a  pylon,  whose  summit  rises  above  it  while  its 
two  wings  stretch  beyond  it  laterally  until  they  meet  the  rect- 
angular wall  which  incloses  the  sanctuary. 

The  dimensions  of  pylons  vary  with  those  of  the  temples  to 
which  they  belong.  The  largest  still  existing  is  the  outer  pylon 
of  the  great  temple  of  Karnak.  It  was  constructed  in  Ptolemaic 
times.  Its  two  chief  masses  are  146  feet  high,  or  about  equal 
to  the  Vendome  column  in  Paris.  This  pylon  is  376  feet 
wide  at  the  widest  part  and  50  feet  thick.  The  first  pylon 
at  Luxor,  which  was  built  by  Rameses  II.,  is  less  gigantic  in  its 
proportions  than  this  ;  it  is,  however,  76  feet  high,  each  of 
its  two  great  masses  is  100  feet  wide,  and  the  portal  in  the 
middle  is  56  feet  high  (see  Fig.  207). 

In  those  temples  which  were  really  complete,  obelisks  were 
erected  a  few  feet  in  front  of  the  pylons,  and  immediately  behind 
the  obelisks,  in  contact  with  the  pylons  themselves,  were  placed 
those  colossal  statues  by  which  every  Egyptian  monarch  commemo- 
rated his  connection  with  the  structures  which  were  reared  in 
his  time.  The  obelisks  are  generally  two  in  number,  the  colossi 
vary  from  four  to  six  for  each  pylon,  according  to  the  magnificence 
of  the  temple.  The  obelisks  range  in  height  from  about  60  to 
100  feet,  and  the  statues  from  20  to  45  feet.^  Obelisks  and 
colossal  statues  seem  to  have  been  peculiarly  necessary  outside 
the  first,  or  outer,  pylon  of  a  temple.  This  produced  an 
effect  upon  the  visitor  at  the  earliest  moment,  before  he  had 
entered  the  sacred  inclosure  itself.  But  they  are  also  to  be 
found  before  the  inner  pylons,  a  repetition  which  is  explained 
by  the  fact  that  such  temples  as  those  of  Karnak  and  Luxor 
were  not  the  result  of  a  single  effort  of  construction.  Each 
of  the  successive  pylons  which  met  the  visitor  during  the  last 
centuries  of  Egyptian  civilization  had  been  at  one  time  the  front 
of  the  whole  edifice. 

To  complete  our  description  of  the  external  parts  of  the  temple 

'  The  obelisk  of  Ousourtesen  at  Heliopolis  is  20*27  metres,  or  67  feet  6  inches, 
high;  the  Luxor  obehsk  at  Paris,  22-80  metres,  or  76  feet;  that  in  the  piazza 
before  St.  Peter's  in  Rome,  83  feet  9  inches ;  that  of  San  Giovanni  Laterano, 
the  tallest  in  Europe,  is  107  feet  2  inches;  and  that  of  Queen  Hatasu,  still  standing 
amid  the  ruins  at  Karnak,  32-20  metres,  or  107  feet  4  inches.  This  is  the 
highest  obelisk  known.  [The  Cleopatra's  Needle  on  the  Victoria  Embankment 
is  only  68  feet  2  inches  high. — Ed.] 


344  A  History  of  Art  in  Ancient  Egypt. 

we  have  yet  to  mention  those  small  lakes  or  basins  which  have 
been  found  within  the  precincts  of  all  the  greater  temples.  Their 
position  within  the  inclosing  walls  suggests  that  they  were  used 
for  other  purposes  beyond  such  ablutions  as  those  which  are 
prescribed  for  all  good  Mohammedans.  If  nothing  but  washing 
was  in  view  they  might  have  been  outside  the  inclosure,  so  that 
intending  worshippers  could  discharge  that  part  of  their  duty 
before  crossing  the  sacred  threshold  ;  but  their  situation  behind 
the  impenetrable  veil  of  such  walls  as  those  we  have  described, 
suggests  that  they  had  to  play  a  part  in  those  religious  mysteries 
which  could  not  be  performed  within  sight  of  the  profane.  Upon 
certain  festivals  richly  decorated  boats,  bearing  the  images  or 
emblems  of  the  gods,  were  set  afloat  upon  these  lakes.  As 
the  diurnal  and  nocturnal  journeys  of  the  sun  were  looked  upon 
as  voyages  by  navigation  across  the  spaces  of  heaven  and 
through  the  shadows  of  the  regions  below,  it  may  easily  be 
understood  how  a  miniature  voyage  by  water  came  to  have  a 
place  in  the  worship  of  deities  who  were  more  or  less  solar  in 
their  character. 

We  have  now  arrived  upon  the  threshold  of  the  temple  itself, 
and  we  must  attempt  to  describe  and  define  that  edifice,  distin- 
guishing from  each  other  its  essential  and  accessory  parts. 

When  we  cast  our  eyes  for  the  first  time  either  upon  the  con- 
fused but  imposing  ruins  of  Karnak  themselves,  or  upon  one  of 
the  plans  which  represent  them,  it  seems  a  hopeless  task  to  evolve 
order  from  such  a  chaos  of  pylons,  columns,  colossal  statues  and 
obelisks,  from  such  a  tangled  mass  of  halls  and  porticos,  corridors 
and  narrow  chambers.  If  we  begin,  however,  by  studying  some  of 
the  less  complex  structures  we  soon  find  that  many  of  these 
numerous  chambers,  in  spite  of  their  curious  differences,  were 
repetitions  of  one  another  so  far  as  their  significance  in  the 
general  plan  is  concerned.  When  a  temple  was  complete  in 
all  its  parts  any  monarch  who  desired  that  his  name  too  should  be 
connected  with  it  in  the  eyes  of  posterity,  had  no  resource  but  to 
add  some  new  building  to  it,  which,  under  the  circumstances 
supposed,  could  be  nothing  but  a  mere  replica  of  some  part  already 
in   existence.^     They  took  some  element  of  the  general  plan,  such 

'  At  Thebes,  still  existing  inscriptions  prove  this  to  be  the  case,  and  at  Memphis 
the  same  custom  obtained,  as  we  know  from  the  statements  of  the  Greek  travellers. 
The  temple  of  Ptah — the  site  of  which   seems  to  be   determined  by  the  colossal 


"a, 


o 

X 


c 

a 


C3 


O 
M 

O 


VOL.    I. 


V  ^' 


The  Temple  under  the  New  Empire.  347 

as  the  hypostyle  hall  at  Karnak,  and  added  to  it  over  and  over 
again,  giving  rise  to  interesting  changes  in  the  proportion, 
arranofement  and  decoration. 

One  of  the  most  intelligent  of  the  ancient  travellers,  namely, 
Strabo,  attempted  the  work  of  discrimination  which  it  is  now 
our  duty  to  undertake.  He  wrote  for  people  accustomed  to 
the  clear  and  simple  arrangements  of  the  Greek  temple,  and 
he  attempted  to  give  them  some  idea  of  the  Egyptian  temple, 
such  as  he  found  it  in  that  Heliopolis  whose  buildings  made  such 
an  impression  upon  all  the  Greeks  who  saw  them.^ 

His  description  is,  perhaps,  rather  superficial.  It  says  nothing 
of  some  accessory  parts  which  were  by  no  means  without  their 
importance,  and  those  details  which  most  strongly  attracted  the 
author's  attention  are  not  mentioned  in  their  natural  order,  w^hich 
would  seem  to  be  that  in  which  the  visitor  from  without  would 
meet  them  in  his  course  from  the  main  door  to  the  sanctuary. 
But  Strabo  had  one  great  advantage  over  a  modern  writer.  He 
saw  all  these  great  buildings  in  their  entirety,  and  could  follow 
their  arrangement  with  an  easy  certainty  which  is  impossible  in 
our  day,  when  so  many  of  them  present  nothing  but  a  confused 
mass  of  ruins,  and  some  indeed,  such  as  the  temple  at  Luxor,  are 
partly  hidden  by  modern  ruins.  We  shall,  then,  take  Strabo  for 
our  guide,  but  we  shall  endeavour  to  give  our  descriptions  in 
better  sequence  than  his,  and  to  fill  up  some  of  the  gaps  in  his 
account  by  the  study  of  those  remains  which  are  in  the  best  state 
of  preservation.  In  our  descriptions  we  shall  advance  from 
simple  buildings  to  those  which  are  more  complex.  We  should 
soon  lose  the  thread  of  our  argument  if  we  were  to  begin  by 
attacking  temples  which  are  at  once  so  complicated  and  so  muti- 
lated as  those  of  Karnak  and  Luxor.      The  character  of  each  of 

statue  of  Rameses  which  still  lies  there  upon  its  face — must  have  rivalled  Karnak 
in  extent  and  in  the  number  of  its  successive  additions.  According  to  Diodorus 
(i.  50)  it  was  Moeris  (Amenemhat  III.)  who  built  the  southern  propylons  of  this 
temple,  which,  according  to  the  same  authority,  surpassed  all  their  rivals  in  mag- 
nificence. At  a  much  later  period,  Sesostris  (a  Rameses)  erected  several  colossal 
monoliths,  from  20  to  30  cubits  high,  in  front  of  the  same  temjjle  (Diodorus, 
cap.  Ivii.  ;  Herodotus,  ii.  140)  ;  at  the  same  time  he  must  have  raised  obelisks  and 
constructed  courts  and  pylons.  Herodotus  attributes  to  two  other  kings,  whom  he 
names  Rhampsinite  and  Asychis,  the  construction  of  two  more  pylons  on  the  eastern 
and  western  sides  of  the  temple  (ii.  121  and  136).  Finally  Psemethek  I.  built  the 
southern  propylons  and  the  pavilion  where  the  Apis  was  nursed  after  his  first  discovery. 
Herodotus,  ii.  153.)     I.  Strabo,  xvii,  i,  28. 


34^  A  History  of  Art  in  Ancient  Egypt. 

the  elements  of  an  Egyptian  temple  of  this  period  will  be  readily 
perceived  if  we  begin  our  researches  with  one  which  is  at  once 
well  preserved,  simple  in  its  arrangements,  and  without  those 
successive  additions  which  do  so  much  to  complicate  a  plan. 

Of  all  the  ruins  at  Thebes  the  Temple  of  K/ions,  which  stands 
to  the  south-west  of  the  great  temple  at  Karnak,  is  that  which 
most  completely  fulfils  these  conditions.^  Time  has  not  treated  it 
very  badly,  and,  although  the  painted  decoration  may  be  the  work 
of  several  successive  princes,  we  are  inclined  to  believe  from  the 
simplicity  of  the  plan  that  most  of  the  architectural  part  of  the 
work  was  begun  and  completed  by  Rameses  III. 

The  advanced  pylon,  or  propylon,  which  stands  some  forty 
metres  in  front  of  the  whole  building  and  was  erected  by  Ptolemy 
Euergetes,  may  be  omitted  from  our  examination.  The  really 
ancient  part  of  the  structure  begins  with  the  rows  of  sphinxes 
which  border  the  road  behind  the  propylon.  They  lead  up  to  a 
pylon  of  much  more  modest  dimensions  than  that  of  Ptolemy. 
In  front  of  this  pylon  there  is  no  trace  of  either  obelisks  or  colossal 
figures.  As  the  whole  temple  is  no  more  than  about  233  feet 
long  and  67  feet  wide,  it  may  not  have  been  thought  worthy  of 
such  ornaments,  or  perhaps  their  small  size  may  have  led  to  their 
removal.  In  any  case,  Strabo  appears  to  have  seen  religious 
edifices  in  front  of  which  there  were  neither  obelisks  nor  the 
statues  of  royal  founders. 

Immediately  behind  this  pylon  lay  a  rectangular  court  sur- 
rounded by  a  portico  of  two  rows  of  columns  standing  in  front  of 
a  solid  wall.  In  this  wall  and  in  the  columns  in  front  of  it  we 
recognise  the  wings  of  which  Strabo  speaks  ;  the  two  walls  of  the 
smne  height  as  those  of  the  temple,  zvhich  are  prolonged  in  fi'ont  of  the 
pronaos.  There  is  but  one  difficulty.  Strabo  says  that  the  space 
between  these  walls  diminishes  as  they  approach  the  sanctuary.^ 
His  court  must  therefore  have  been  a  trapezium  with  its  smallest 
side  opposite  to  the  pylon,  rather  than  a  rectangle.  We  have 
searched  in  vain  for  such  a  form  among  the  plans  of  those  pharaonic 

1  This  is  the  temple  which  the  members  of  the  Egyptian  institute  call  the  Great 
Southern  Temple.  In  the  background  of  our  illustration  (Fig.  208)  the  hypostyle 
hall  and  the  southern  pylons  of  the  Great  Temple  are  seen. 

"  Toi)  Se  Trpoi'dov  Trap'  eKarepov  TrpoKeirai  to,  Xeyofxeva  Trrepa-  ecm  oe  ravra  Irrovij/r]  tw 
vacp  Tei^T]  8vo,  Kar  dp^^a?  jxkv  a(f>€crTWTa  dir'  dXXr^Xwv  [xiKpov  irXeov,  7]  to  7r\uT0<;  ecrrt  t^s 
Kpr^TTtSos  Tov  v€(s),  €7r€tT'  €ts  TO  TTpocrOev  TTpoLovTi  KOT  eTTtvcuouo'as  ypa/A/Atts  fJ-^XP'-  '^''IX'^^ 
irevTTjKovTa  -q  k^Kovra. — SxRADO,  xvii,  i,  28. 


a 


a 

bO 

i) 

fcJO 

c 

■? 

o 

M 
tn 


o 


c 

CI 


s 
o 

'u 
O 


O 


4J 

B 


00 

o 


.ivilWi"V'\VA\,il-.i^....iUi'w 


The  Temple  under  the  New  Empire.  351 

temples  which  have  been  measured.  In  every  instance  the  sides 
of  the  peristylar  court  form  a  rectangular  parallelogram.  It  must, 
apparently,  have  been  in  a  Ptolemaic  temple  that  Strabo  noticed 
these  converging  sides,  and  even  then  he  was  mistaken  in  sup- 
posing such  an  arrangement  to  be  customary.  The  Ptolemaic 
temples  which  we  know,  those  of  Denderah,  Edfou,  Esneh,  have  all 
a  court  as  preface  to  the  sanctuary,  but  in  every  case  those  courts 
are  rectangular.  In  the  great  temple  of  Philse  alone  do  we  find 
the  absence  of  parallelism  of  which  Strabo  speaks,^  the  peristylar 
court  which  follows  the  second  pylon  is  rather  narrower  at  its 
further  extremity  than  immediately  behind  the  pylon.  In 
presence  of  this  example  of  the  trapezium  form  we  may  allow- 
that  it  is  quite  possible  that  in  the  temples  of  Lower  and  Middle 
Egypt,  which  have  perished,  the  form  in  question  was  more 
frequently  employed  than  in  those  of  Upper  Egypt,  where, 
among  the  remains  of  so  many  buildings,  we  find  it  but  once. 

To  return  to  the  Temple  of  Khons.  From  the  courtyard  of 
which  we  have  been  speaking,  a  high  portal  opens  into  a  hall  of 
little  depth  but  of  a  width  equal  to  that  of  the  whole  temple. 
The  roof  of  this  hall  is  supported  by  eight  columns,  the  central 
four  being  rather  higher  than  the  others.^  It  is  to  this  room  that 
the  name  of  hypostyle  hall  has  been  given.  We  can  easily  under- 
stand how  Strabo  saw  in  it  the  equivalent  to  the  pronaos  o^  the 
Greek  temples.  We  know  how  in  the  great  peripteral  buildings 
of  Greece  and  Italy,  t]\& pronaos  prefaced  the  entrance  to  the  cella 
with  a  double  and  sometimes  a  triple  row  of  columns.  Except 
that  it  is  entirely  inclosed  by  its  walls,  the  Egyptian  hypostyle 
had  much  the  same  appearance  as  the  Greek  proimos.  Its 
name  in  those  texts  which  treat  of  its  construction  is  the  large 
Jiall ;  but  it  is  also  called  the  Hall  of  Assembly  and  the  Hall  of  the 
Appearance,  terms  which  explain  themselves.  Only  the  kings  and 
priests  were  allowed  to  penetrate  into  the  sanctuary  for  the 
purpose  of  bringing  forth  the  emblem  or  statue  of  the  god  from 
the  tabernacle  or  other  receptacle  in  which  it  was  kept.  This 
emblem  or  figure  was  placed  either  in  a  sacred  boat  or  in  one  of 
those  portable  wooden  tabernacles  in  which  it  was  carried  round 
the  sacred  inclosure  to  various  resting  places  or  altars.  The 
crowd  of  priests  and  others  who  had  been  initiated  but  were  of 

^   Descriptio7i  de  F Egypte^  Aniiquifcs.  vol.  i.  pi.  5. 
-  Description  de  /' Egypte,  vol.  iii.  55. 


352 


A   History  of  Art  in  Ancient  Egypt. 


inferior  rank  awaited  the  appearance  of  the  deity  in  the  hypostyle 
hall,  in  which  the  cortege  was  marshalled  before  emerging  into 
the  courts. 

The  second  division  of  the  temple,  for  Strabo,  was  the  sanc- 
tuary, or  (TriKos.  In  this  Temple  of  Khons  it  was  a  rectangular 
chamber,  separated  by  a  wide  corridor  running  round  its  four  sides 
from  two  smaller  chambers,  which  filled  the  spaces  between  the 
corridor  and   the   external   walls.       In   this   hall    fragments    of    a 


Fig.  209 — The  ban,  ov  sacred  boat ;  from  the  temple  of  Elephantine. 


granite  pedestal  have  been  discovered,  upon  which  either  the  dan 
or  sacred  boat,  which  is  so  often  figured  upon  the  bas-reliefs 
(Fig.  209),  or  some  other  receptacle  containing  the  peculiar 
emblem  of  the  local  divinity,  must  have  been  placed.  Strabo 
was  no  doubt  correct  in  saying  that  the  ai-jKos  differed  froni  the  cella 
of  the  Greek  temple  in  that  it  contained  no  statue  of  the  divinity, 
but  nevertheless  it  must  have  had  something  to  distinguish  it 
from  the  less  sacred  parts  of  the  building.  This  something  was  a 
kind   of  little   chapel,    tabernacle,    or   shrine,   closed    by    a  folding 


The  Temple  under  the  New  Empire.  353 

door,  and  containing  either  an  emblem  or  a  statue  of  the  divinity, 
before  which  prayers  w^ere  recited  and  religious  ceremonies  per- 
formed on  certain  stated  days.  Sometimes  this  shrine  was  no 
more  than  an  inclosed  niche  in  the  wall,  sometimes  it  was  a  little 
edifice  set  up  in  the  middle  of  the  sanctuary.  In  those  cases  in 
which  it  was  a  structure  of  painted  and  gilded  wood,  like  the  ark 
of  the  Hebrews,  it  has  generally  disappeared  and  left  no  trace 
behind.  The  tabernacle  in  the  Turin  Museum  (Fig  210)  is  one 
of  the  few  objects  of  the  kind  which  have  escaped  complete 
destruction.  In  temples  of  any  importance  the  shrine  was 
hollowed  out  of  a  block  of  granite  or  basalt.  A  monolithic  chapel 
of  this  kind  is  still  in  place  in  the  Ptolemaic  temple  of  Edfou  ; 
it  bears  the  royal  oval  of  Nectanebo  I.^  Examples  are  to  be 
found  in  all  the  important  European  museums.  One  of  the  finest 
belongs  to  the  Louvre  and  bears  the  name  of  Amasis  ;  it  is  of 
red  granite  and  is  entirely  covered  with  inscriptions  and  sculpture 
(Fig.  211).^  It  must  resemble,  on  a  smaller  scale,  the  tabernacle 
prepared  in  the  Elephantine  workshops,  under  Amasis,  for  the 
temple  of  Neith,  at  Sais,  which  so  greatly  excited  the  admiration 
of  Herodotus.'^ 

The  doors  of  the  shrine  were  kept  shut  and  even  sealed  up. 
The  king  and  the  chief  priest  alone  had  the  right  to  open  them 
and  to  pay  their  devotions  before  the  image  or  symbol  which 
they  inclosed.  This  seems  clearly  proved  by  the  following 
passage  from  the  famous  stele  discovered  by  Mariette  at  Gebel- 
Barkal,  upon  which  the  Ethiopian  conqueror  Piankhi-Mer-Amen 
celebrates  his  victories  and  the  occupation  of  Egypt  from  south  to 
north.  After  noticing  the  capture  of  Memphis  he  tells  us  that  he 
stopped  at  Heliopolis  in  order  that  he  might  sacrifice  to  the  gods 
in  the  royal  fashion  :   "  He   mounted  the   steps  which   led  to   the 

1  According  to  Gau,  there  was,  in  181 7,  a  well  preserved  tabernacle  in  the 
sanctuary  of  the  temple  at  Debout,  in  Nubia.  {Antiquites  de  la  Niibie,  1821,  pi.  v. 
Figs.  A  and  B.) 

^  De  Rouge,  Notice  des  Monuments,  etc.  (Upon  the  ground  floor  and  the  stair- 
case.) Monuments  Divers,  No.  29.  The  term  noos  has  generally  been  applied  to 
these  monuments,  but  it  seems  to  us  to  lack  precision.  The  Greeks  used  the  word 
mo?  or  reojs  to  signify  the  temple  as  a  whole.  Abd-el-Latif  describes  with  great 
admiration  a  monolithic  tabernacle  which  existed  in  his  time  among  the  ruins  of 
Memphis,  and  was  called  by  the  Egyptians  the  Green  Chamber.  Makrizi  tells  us 
that  it  was  broken  up  in  1349.      {Description  de  VEgypte,  Ant.,  vol.  v.  pp.  572,  573.) 

3  Herodotus,  ii.  175. 

VOL.    I.  Z   Z 


:54 


A  History  of  Art  in  Ancient  Egypt. 


frreat  sanctuary  in  order  that  he  might  see  the  god  who  resides 
in  Ha-benben,  face  to  face.  Standing  alone,  he  ch'ew  the  bolt, 
and  swung  open  the  folding  doors  ;  he  looked  upon  the  face  of  his 
father  Ra  in  Ha-benben,  upon  the  boat  Mad,  of  Ra,  and  the 
boat  Seket,  of  Shou  ;  then  he  closed  the  doors,  he  set  sealing 
clay  upon   them  and   impressed    it   with   the   royal   signet."  ^ 

From    the    description    of    Strabo    we    should    guess    that    the 
Egyptian  temple  ended  with   the  sanctuary.     Such   was  not   the 


Fig.  2IO. — Portable  tabernacle  of  painted  wood,  19th  dynasty.      In  the  Turin  Museum. 


case  however.  Like  most  of  the  Greek  temples,  the  Egyptian 
temple  had  its  further  chambers  which  served  nearly  the  same 
purposes  as  the  oTriaOcFjo/jiot  of  the  Greeks.  Thus  in  the  Temple  of 
Khons,  the  sanctuary  opens,  at  the  rear,  into  a  second  hypostyle 
hall   which   is   smaller  than   the  first  and   has   its   roof   supported 

^  Translated  by  Maspero,  Histoire  Ancienue,  p.  385.  The  wliole  inscription  has 
been  tfcinslated  into  English  by  the  Rev.  T.  C.  Cook,  and  published  in  vol.  ii.  of 
Jiecords  of  the  Past. — Ed. 


The  Temple  under  the  New  Empire. 


155 


by  only  four  columns  instead  of  eight.  Upon  this  hall  open 
four  small  and  separate  chambers  which  fill  up  the  whole 
space  between   it  and   the   main    walls. 

Similar  general  arrangements  to  those  of  the  Temple  of  Khons 
are  to  be  found  in  even  the  largest  temples.  The  second  hypostyle 
hall  is  however  much  larger  and  the  chambers  to  which  it  gives 
access  much  more  numerous.  It  is  not  easy  to  determine  the 
object  of  each  of  these  small  apartments  ;  in  the  Pharaonic  temples 
they  are  usually  in  very  bad  condition,  but  in  some  of  the 
Ptolemaic  buildings,  such  as  the  temples  of  Edfou  and  Denderah, 
they  are  comparatively  well  preserved.  In  the  last  named  the 
question    is    complicated    by    the    existence    of    numerous    blind 


Fig.  211. — Granite  tabernacle  :  in  the  Louvre. 

passages  contrived  in  the  thickness  of  the  walls.  The  stone 
which  stopped  the  opening  into  these  passages  seems  to  have 
been  manipulated  by  some  secret  mechanism.^  Some  of  the 
sacred  images  and  such  emblems  as  were  made  of  precious 
materials  were  kept  in  these  hiding  places.  Their  absolute 
darkness    and     the    coolness    which    accompanied     it,    were    both 

1  As  M.  Maspero  has  remarked  (Aii/n/aire  dc  F Association  dis  Etudes  Grecqiies, 
1877,  p.  135),  these  secret  passages  remind  usof  ihe  movable  stone  -which,  according 
to  Herodotus  (ii.  121),  the  architect  of  Rhampsinit  contrived  in  the  wall  of  the 
royal  treasure-house  which  he  was  commissioned  to  build.  Herodotus's  story  was 
at  least  founded  upon  fact,  as  the  arrangement  in  question  was  a  flwourite  one  with 
Egyptian  constructors. 


JD^ 


A   History  of  Art  in  Ancient  Egypt. 


conducive  to  the  preservation  of  delicately  ornamented  objects 
in  such  a  climate  as  that  of  Egypt. 

It  was  this  part  of  the  temple,  then,  that  the  Greeks  called 
the  treasure-house.  It  inclosed  the  material  objects  of  worship. 
Some  of  its  chambers,  however,  were  consecrated  to  particular 
divinities  and  seem  to  have  had  somewhat  of  the  same  character 
as  the  apsidal  chapels  of  a  Roman  Catholic  Church.  They  are 
material  witnesses  to  the  piety  of  the  princes  who  built  them 
and  who  wished  to  associate  the  divinities  in  whose  honour 
they  were  raised  with  the  worship  of  the  god  to  whom  the  temple 
as  a  whole  had  been  dedicated.  Whether  store-rooms  or  chapels, 
these  apartments  might  be  multiplied  to  any  extent  and  might 
present  great  varieties  of  aspect.  At  Karnak,  therefore,  where 
they  communicate  with  long  and  wide  galleries,  they  are  very 
numerous.  One  of  them  was  that  small  chamber  which  was 
dismantled  thirty  years  ago  by  Prisse  d' Avennes  and  transported  to 
Paris.  It  is  known  as  the  Hall  of  Ancestors.  In  it  Thothmes  III. 
is,  in  fact,  represented  in  the  act  of  worshipping  sixty  kings 
chosen  from  among  his  predecessors  on  the  Egyptian  throne. 

The  last  feature  noticed  by  Strabo  in  the  small  temple  taken 
by  him  as  a  type,  was  the  sculpture  with  which  its  walls  were 
lavishly  covered.  These  works  reminded  him  of  Etruscan  sculp- 
ture and  of  Greek  productions  of  the  archaic  period,  but  we  can 
divine  from  the  expressions  ^  of  which  he  makes  use,  that  he 
perceived  the  principles  which  governed  the  Egyptian  sculptor 
to  be  different  from  those  of  the  Greeks.  The  Greek  architect 
reserved  certain  strictly  circumscribed  places  for  sculpture,  such 
as  the  friezes  and  pediments  of  the  temples,  while  in  Egypt  it 
spreads  itself  indiscriminately  over  every  surface.  In  the  temple 
of  Khons,  as  in  every  other  building  of  the  same  kind  at  Thebes, 
we  find  this  uninterrupted  decoration.  Mariette  has  shown  the 
interesting  nature  of  these  representations  and  their  value  to 
the  historian. 

We  have  still  to  notice,  always  keeping  the  same  edifice 
in  view,  two  original  points  in  the  characteristic  physiognomy 
of  the  Egyptian  temple  which  seem  to  have  escaped  the  attention 
of  the  Greek  traveller. 

In  the  Greek  temple  there  is  no  space  inclosed  by  a  solid 
wall   but   that   of   the   cella,    which,    by    its    purpose,    answers    to 

^    ' Avay\v(f)a<;  8'  i^ovcTLV  ol  rorj^oi  ovtol  fx^yaXuiv  elbu)\<t)v  (StRABO,  XVll,  I,   28). 


The  Temple  under  the  New  Empire.  357 

the  (jrjKos  of  the  Egyptian  buildings.  Both  the  peristyle  and 
the  pronaos  are  open  to  the  air  and  to  the  view  of  all  comers  ; 
the  statues  of  the  pediments,  the  reliefs  of  the  friezes  are  all 
visible  from  outside,  and  the  eye  rejoices  freely  both  in  master- 
pieces of  sculpture  and  in  the  long  files  of  columns,  which 
vary  in  effect  as  they  are  looked  at  from  different  points  of  view. 
The  appearance  of  the  Egyptian  temple  is  altogether  dissimilar. 
The  peristylar  court,  the  hypostyle  hall,  the  sanctuary  and  its 
adjuncts,  in  a  word  the  whole  combination  of  chambers  and 
courts  which  form  the  temple  proper,  is  surrounded  by  a  curtain 
wall  which  is  at  least  as  hio^h  as  the  buildinos  which  it  incloses. 
Before  any  idea  of  the  richness  and  architectural  magnificence 
of  the  temple  itself  can  be  formed,  this  wall  must  be  passed. 
From  the  outside  nothing  is  to  be  seen  but  a  great  retangular 
mass  of  building,  the  inclined  faces  of  which  seem  to  be  endea- 
vouring to  meet  at  the  top  so  as  to  give  the  greatest  possible 
amount  of  privacy  and  security  to  the  proceedings  which  take 
place  within.  The  Egyptian  temple  may,  in  a  word,  be  compared 
to  a  box  (Fig,  61),  and  in  such  buildings  as  that  dedicated  to 
Khons,  the  box  is  a  simple  rectangular  one.  The  partitions 
which  separate  its  various  halls  and  chambers  are  kept  within 
the  main  wall.  But  in  larger  buildings  the  box  is,  partially  at 
least,  a  double  one.  When  we  examine  a  plan  of  the  great 
temple  at  Karnak,  we  see  that  all  the  back  part  of  the  vast 
pile,  all  that  lies  to  the  west  of  the  open  passage  and  the 
fourth  pylon,  is  inclosed  by  a  double  wall.  A  sort  of  wide 
corridor,  open  to  the  sky,  lies  between  the  outer  wall  and  that 
which  immediately  surrounds  the  various  chambers.  This  outer 
wall  is  absent  only  on  the  side  closed  by  the  inner  pylon.  In 
some  temples,  especially  in  those  of  the  Ptolemaic  period,  the 
hypostyle  hall  is  withdrawn  some  distance  behind  the  courtyard, 
and  the  sanctuary  behind  the  hypostyle  hall.  This  arrangement 
is  repeated  in  the  position  of  the  two  walls.  The  inner  one 
embraces  the  chambers  of  the  temple  and  follows  their  irregulari- 
ties ;  the  other  describes  three  sides  of  a  rectancjle  leavinQf  a 
wider  space  at  the  back  of  the  temple  than  at  the  sides.  The 
pylon,  as  we  have  said,  supplies  the  fourth  side.  This  outer 
wall  has  no  opening  of  any  kind.  It  is  true  that  at  Karnak 
lateral  openings  exist  in  the  hypostyle  hall  and  in  the  courtyard, 
but  those  parts  were  less  sacred  in  their  character  than  the  inner 


58 


A  History  of  Art  in  AnciExXT  Egypt. 


Fig.  212. — General  plan  of  the  Great  Temple  at  Karnak. 


chambers  to  which  they 
gave  access.  From  the 
point  where  the  wall  be- 
comes double,  that  is  from 
the  posterior  wall  of  the 
hypostyle  hall,  there  are 
no  more  external  openings 
of  any  kind.  To  reach 
the  presence  of  the  deity 
the  doors  of  the  fourth 
and  fifth  pylons  had  to  be 
passed.  The  high  and 
thick  wall,  without  opening 
of  any  kind,  which  in- 
closed the  sanctuary  and 
its  dependencies  like  a 
cuirass,  was  no  doubt  in- 
tended to  avert  the  possi- 
bility of  clandestine  visits 
to  the  holy  place. 

The  evident  desire  of 
the  architect  to  hide  his 
porticos  and  saloons  be- 
hind an  inpenetrable  cur- 
tain of  limestone  or  sand- 
stone suffices  to  prove  that 
shadow  rather  than  sun- 
shine was  wanted  in  the 
inner  parts  of  the  ternple. 
When  the  slabs  which 
formed  the  roofs  of  the 
temple  of  Khons  were 
all  in  place — they  are  now 
mostly  on  the  ground — it 
must  have  been  very  dark 
indeed.  The  hypostyle 
hall  communicated  directly 
and  by  an  ample  doorway 
with  the  open  courtyard, 
wdiich  was  bathed   in   the 


The  Temple  under  the  New  Empire.  ^59 


v5; 


constant  sunlight  of  Egypt  ;  besides  which  there  were  openings  just 
under  the  cornice  and  above  the  capitals  of  the  columns.  When  the 
door  was  open,  therefore,  there  would  be  no  want  of  light,  although 
it  would  be  softened  to  a  certain  extent.  The  sanctuary  was 
much  darker.  The  light  which  came  through  the  door  was 
borrowed  from  the  hypostyle  hall.  The  hall  with  the  four 
supporting  columns  and  the  chambers  which  surrounded  it  were 
still  worse  provided  than  the  sanctuary  ;  the  first  named  was 
feebly  illuminated  by  small  openings  in  the  stone  roof,  the  latter 
were  in  almost  complete  darkness.  The  only  one  which  could 
have  enjoyed  a  little  light  was  that  which  lay  on  the  central  axis 
of  the  building.  A  few  feeble  rays  may  have  found  their  way 
to  this  chamber  when  the  doors  of  the  temple  were  open,  but, 
as  a  rule,  they  seem  to  have  been  closed.  Marks  of  hinges 
have  been  found  in  the  Egyptian  temples,  and  it  is  certain  that 
the  sanctuary  was  permanently  closed  in  some  fashion  against 
the  unbidden  visits  of  the  curious,^ 

We  shall  return  elsewhere  to  the  illumination  of  the  Egyptian 
temples,  and  shall  discuss  the  various  methods  made  use  of 
to  ensure  sufficient  light  for  the  enjoyment  of  the  sumptuous 
decorations  lavished  upon  them  ;  here,  however,  it  will  be 
sufficient  if  we  indicate  their  general  character,  which  is  the  same 
in  all  the  religious  edifices  in  the  country. 

The  largest  and  best  lighted  chambers  are  those  nearest  to  the 
entrance.  As  we  leave  the  last  pylon  behind  and  penetrate  deeply 
into  the  temple,  the  light  gradually  becomes  less  and  the  chambers 
diminish  in  size,  until  the  building  comes  to  an  end  in  a  number  of 
small  apartments  in  which  the  darkness  is  unbroken.  There  are 
even  some  temples  which  become  gradually  narrower  and  lower 
from  front  to  back  ;  this  is  especially  the  case  with  those  which 
have  a  double  wall  round  their  more  sacred  parts. 

This  progressive  diminution  is  even  more  clearly  marked  in  a 
vertical  section  than  in  one  taken  horizontally.  The  pylon  is 
much   higher    than  any  other  point  in   the   building.     After   the 

'  Description  de  TEgypte,  Afifiguites,  vol.  i.  p.  219.  The  authors  of  the  Descrip- 
tion gcnlrale  de  Tliebcs  noticed  recesses  sunk  in  the  external  face  of  one  of  the 
pylons  at  Karnak,  which  they  believed  to  be  intended  to  receive  the  leaves  of  the 
great  door  when  it  was  open  (p.  234);  they  also  noticed  traces  of  bronze  pivots 
upon  which  tlie  doors  swung  (p.  248),  and  they  actually  found  a  pivot  of  sycamore 
wood. 


360  A   History  of  Art  in  Ancient   Egypt. 

pylon,  in  the  temple  of  Khons,  comes  the  portico  which  surrounds 
the  courtyard.  Next  come  in  their  order  the  columns  of  the 
hypostyle  hall,  the  roof  of  the  sanctuary,  the  roof  of  the  chamber 
with  four  columns,  and  the  roof  of  the  last  small  apartment  which 
rests  upon  the  inclosing  wall.  Between  the  large  hypostyle  hall 
and  the  smaller  one  there  is  a  difference  in  height  amountinor  to  a 
quarter  of  the  whole  height  of  the  former. 

In  the  most  important  temples,  such  as  those  of  Karnak, 
Luxor,  and  the  Ramesseum,  the  same  law  of  constajit  diminution 
in  height  from  front  to  rear  holds  good,  with  the  exception  that  in 
their  cases  it  is  the  hypostyle  hall  which  is  the  highest  point  in  the 
building  after  the  pylons.  In  this  hall  their  architects  have 
raised  the  loftiest  columns,  and  it  is  after  these  that  the  pro- 
gressive diminution  begins.  The  longitudinal  section  of  the 
temple  of  Luxor  (Fig.  213)  and  the  general  view  of  Karnak 
(plate  iv.)  illustrate  this  statement. 

As  the  roofs  of  the  temple  chambers  are  gradually  lowered, 
their  carefully  paved  floors  are  raised,  but  not  to  an  equal  degree. 
In  the  temple  Khons  four  steps  lead  up  from  the  court  to  the 
hypostyle  hall,  and  one  step  from  the  hall  to  the  sanctuary. 
Similar  arrangements  are  found  elsewhere.  At  Karnak  a  con- 
siderable flight  is  interposed  between  the  courtyard  and  the 
vestibule  of  the  hypostyle  hall.  At  Luxor  the  level  of  the  second 
court  is  higher  than  that  of  the  first.  In  the  Ramesseum  there  are 
three  flights  of  steps  between  the  first  and  second  hypostyle  hall. 

All  these  buildings  are  provided  with  staircases  by  which  their 
flat  roofs  may  be  reached.  These  roofs  seem  to  have  been  freely 
opened  to  the  people.  The  interiors  of  the  temples  were  only  to 
be  visited  by  the  priests,  except  on  a  few  stated  days  and  in  a 
fashion  prescribed  by  the  Egyptian  ritual  ;  but  the  general  public 
were  allowed  to  mount  to  the  roofs,  just  as  with  us  they  are 
allowed  to  ascend  domes  and  belfries  for  the  sake  of  the  view  over 
the  surrounding  buildings  and  country.  The  numerous  ^'r*^;^//, 
some  in  the  hieroglyphic,  others  in  the  demotic  character,  which 
are  still  to  be  seen  upon  the  roof  of  the  temple  of  Khons,  attest 
this  fact. 

We  thus  find  the  characteristic  features  of  Egyptian  architecture 
united  in  a  single  building  in  this  temple  of  Khons  ;  but,  even 
at  Thebes,  no  such  similarity  between  one  building  and  another 
is  to  be  found  as  in  the  great  temples  of  Greece.     In  passing  from 


,0    *2l 


GENERAL  VIEW  OF  THE  EXISTING   BUILDINGS  AT  KARNAK 
RESTORED  BY'CHAPLES   CH1PIK7. 


n.  T  X. 


A-^^r^  ■*•■*-  A-»T*^TT-»XT'-T-»  l-i     ^^^■%TT\r 


IS  to  De 


lOLina  as  in  me  great  tempnis  of  Greece.     In  passing  from 


The  Temple  under  the  New  Empire. 


361 


the  Parthenon  to  the  temple  of  Theseus 
or  to  that  of  Jupiter  Olympius,  from  a 
Doric  to  an  Ionic,  and  from  an  Ionic  to  a 
Corinthian  building,  certain  well  marked 
variations,  certain  changes  of  style,  pro- 
portion and  decoration  are  seen.  But  the 
differences  are  never  sufficient  to  embarrass 
the  student  of  those  buildings.  The  object 
of  each  part  remains  sufficiently  well 
defined  and  immutable  to  be  easily  re- 
cognised by  one  who  has  mastered  a  single 
example.  In  Egypt  the  variations  are 
much  greater  even  amonof  buildings  erected 
during  a  single  dynasty  and  by  a  single 
architect.  After  the  attentive  study  of 
some  simple  and  well  marked  building, 
like  the  temple  of  Khons,  the  visitor  pro- 
ceeds to  inspect  the  ruins  of  Karnak, 
Luxor,  the  Ramesseum,  Medinet-Abou  or 
Gournah,  and  attempts  to  restore  something 
like  order  in  his  mind  while  walking-  amone 
their  ruins.  But  in  vain  are  the  rules 
remembered  which  were  thought  to  apply 
to  all  such  buildings  ;  they  are  of  little 
help  in  unravelling  the  mazes  of  Karnak 
or  Luxor,  and  at  each  new  ruin  explored 
the  visitor's  perplexities  begin  anew. 

The  variations  are,  in  fact,  very  great, 
but  they  are  not  so  great  as  they  seem 
at  the  first  glance.  They  are  generally  to 
be  explained  by  those  developments  and 
repetitions  of  which  Egyptian  architects 
were  so  fond.  We  shall  endeavour  to 
demonstrate  this  by  glancing  rapidly  at 
each  of  the  more  celebrated  Theban  build- 
ings in  turn.  Our  purpose  does  not 
require  that  we  should  describe  any  of 
them  in  detail,  as  we  have  already  done 
in  the  case  of  the  temple  of  Khons,  and 
we    shall    be    content    with    noticincr   their 


..^M 


'Y^ 


I 

I 
I 


1 
I 


N 

<U 

U 
>> 

o 


o 


O 
O 


rt 

.s 

•4—' 

'So 
c 
o 

h-1 


VOL.    I. 


3    A 


362  A  History  of  Art  in  Ancient  Egypt. 

variations  upon  the  type  established  by  our  study  of  the  minor 
monument. 

Let  us  take  Karnak  first.  This,  the  most  colossal  assemblage 
of  ruins  which  the  world  has  to  show,  comprises  no  less  than 
eleven  separate  temples  within  its  four  inclosing  walls  of  crude 
brick.  The  longest  axis  of  this  collection  of  ruins  is  that  from 
north  to  south  ;  it  measures  about  1,560  yards  ;  its  transverse  axis 
is  620  yards  long.  The  whole  circuit  of  the  walls  is  nearly  two 
EnMish  miles  and  a  half.^ 

The  first  thing  that  strikes  us  in  looking  at  a  general  map  of 
Karnak  is  that  Egyptian  temples  were  not  oriented.^  The  Great 
Temple  is  turned  to  the  west,  that  of  Khons  to  the  south,  that  of 
Mouth  to  the  north.  There  is  some  doubt  as  to  the  name  which 
should  be  given  to  several  of  these  buildings.  Two  of  the  most 
important  are  consecrated  to  those  deities  who,  with  Amen,  form 
the  Theban  triad.  The  highest  and  largest  of  them  all,  that 
which  is  called  the  Gj'eat  Temple,  is  dedicated  to  Amen-Ra. 

We  are  here  concerned  with  the  latter  building  only.  We 
reproduce,  on  a  much  larger  scale  and  in  two  parts  (Fig.  214  on 
on  page  363,  and  Fig.  215  on  page  367),  the  plan  given  on  page 
358  (Fig.  212).  A  few  figures  will  suffice  to  give  an  idea  of  the 
several  dimensions.  From  the  external  doorway  of  the  first 
western  pylon  to  the  eastern  extremity  of  the  building,  the  length, 
over  all,  is  1,215  f^^t-  ^^s  greatest  width  is  that  of  the  first 
pylon,  namely  376  feet.  The  total  circumference  of  the 
bounding  wall  is  about  3,165  feet.  The  outside  curtain  wall  of 
brick  is  from  2,500  to  2,700  yards  in  length,  which  corresponds 
closely  to  the  13  stadia  said  by  Diodorus  to  be  the  circumference 
of  the  oldest  of  the  four  great  Theban  temples,^ 

After  passing  the  first  pylon  (No.  i  on  the  plan)  we  find  our- 
selves in  a  peristylar  court  answering  to  that  in  the  temple  of 
Khons.  On  our  right  and  left  respectively  we  leave  two  smaller 
temples,  one  of  which  (C  on  plan)  cuts  through  the  outer  wall  and 

'  These  measurements  are  taken  from  Martettf,  Voyage  dans  la  Ilaide-Egypte. 
vol.  ii.  p.   7. 

2  We  have  not  given  a  general  map.  In  order  to  do  so  we  should  either  have 
had  to  overpass  the  limits  of  our  page,  or  we  should  have  had  to  give  it  upon  too 
small  a  scale.  Our  fourth  plate  will  give  a  sufficiently  accurate  idea  of  its  arrange- 
ment. The  plan  in  Lepsius's  Denk)iuder  (part  i.  pi.  74-76)  occupies  three 
entire  pages. 

^  Diodorus,  i.  46. 


w 


fWV/////////// 7/. "/////// '.   :■  '' 


MmX 


il  (S  #  #  (0  ®  # 

mmmm  m  m 


n    n    n     n.  \ 


#  #  #  #  #  # 


;^!5» 


i 


WM 


m 


O 
O) 


n 


_rL 


n-.-n-_JV.^i    iU^ 


A. 


n     n     n     n 


■ 


m 


rs 


or 
C 

ByiBHnSn  or 


'■■■■■  ■  i   ' i-'M 


0            J           IC          J5          so          °^          30  J\^ 
1 I I : 1 > . 


Fig.   214.— Plan  of  the  anterior  portion  of  the  Great  Temple  at   Karnak.      From  the  plan  of 

M.  Bnine. 


The  Temple  under  the  New  Empire.  365 

was  built  by  Rameses  III.  ;  Seti  II.  was  the  author  of  the  other  (D). 
Those  two  buildings  are  older  than  the  court  and  its  colonnades. 
When  the  princes  of  the  tv/enty-second  dynasty  added  this 
peristyle  to  the  already  constructed  parts  of  the  great  temple, 
they  refrained  from  destroying  those  monuments  to  the  piety  of 
their  ancestors.  We  also  may  regard  these  temples  as  mere 
accidents  in  the  general  arrangement.  We  may  follow  the  path 
marked  out  down  the  centre  of  the  court  by  the  remains  of  an 
avenue  of  columns  which  dates  from  the  times  of  the  Ethiopian 
conquerors  and  of  the  Bubastide  kings  (E).  After  the  second 
pylon  (2)  comes  the  hypostyle  hall,  the  wonder  of  Karnak,  and 
the  largest  room  constructed  by  the  Egyptians  (F).  It  is  340 
feet  long  by  i  70  wide.-"-  One  hundred  and  thirty-four  colossal 
columns  support,  or  rather  did  once  support,  the  roof,  which,  in 
the  central  portion,  was  not  less  than  76  feet  above  the  floor ; 
in  this  central  portion,  twelve  pillars  of  larger  proportions  than 
the  others  form  an  avenue  ;  these  columns  are  1 1  feet  10  inches 
in  diameter  and  more  than  ;^2>  ^^^^  '^^^  circumference,  so  that,  in 
bulk,  they  are  equal  to  the  column  of  Trajan.  They  are,  without 
a  doubt,  the  most  massive  pillars  ever  employed  within  a  building. 
From  the  ground  to  the  summit  of  the  cube  which  supports  the 
architrave,  they  are  yo  feet  high.  Right  and  left  of  the 
central  avenue  the  remaining  122  columns  form  a  forest  of 
pillars  supporting  a  flat  roof,  which  is  lower  than  that  of  the 
central    part    by    33    feet.      The  cathedral    of    Notre    Dame,  at 


'  These  are  the  figures  given  by  Mariette  {Itineraire  de  la  Hmcfe-Egypte,  p.  135). 
Other  authorities  give  340  feet  by  177.  Diodorus  ascribed  to  the  temple  of  which 
he  spoke  a  height  of  45  cubits  (or  69  feet, 3  inches).  This  is  slightly  below  the  true 
height.  We  may  here  quote  the  terms  in  which  ChampoUion  describes  the  impression 
which  a  first  sight  of  these  ruins  made  upon  him  :  "  Finally  I  went  to  the  palace,  or 
rather  to  the  town  of  palaces,  at  Karnak.  There  all  the  magnificence  of  the  Pharaohs 
is  collected  ;  there  the  greatest  artistic  conceptions  formed  and  realised  by  mankind 
are  to  be  seen.  All  that  I  had  seen  at  Thebes,  all  that  I  had  enthusiastically  admired 
on  the  left  bank  of  the  river,  sunk  into  insignificance  before  the  gigantic  structures 
among  which  I  found  myself.  I  shall  not  attempt  to  describe  what  I  saw.  If  my 
expressions 'were  to  convey  but  a  thousandth  part  of  what  I  felt,  a  thousandth  part 
of  all  that  might  with  truth  be  said  of  such  objects,  if  I  succeeded  in  tracing  but  a 
faint  sketch,  in  the  dimmest  colours,  of  the  marvels  of  Karnak,  I  should  be  taken, 
at  least  for  an  enthusiast,  perhaps  for  a  madman.  I  shall  content  myself  with 
saying  that  no  people,  either  ancient  or  modern,  have  had  a  national  architecture 
at  once  so  sublime  in  scale,  so  grand  in  expression,  and  so  free  from  littleness  as 
that  of  the  ancient  F.gyptians."     {Lettres  iV Egypte^  pp.  79,  So.) 


366  A   History  of  Art  in  Ancient  Egypt. 


Paris,  would  stand   easily  upon  the  surface  covered  by  this  hall 
(see  Plate  V). 

Its  proportions  are  very  different  from  those  of  the  correspond- 
ing chamber  in  the  little  temple  of  Khons,  but  yet  it  fills 
the  same  office  in  the  general  conception,  it  is  constructed  on  the 
same  principle  and  lighted  in  the  same  fashion.  To  use  the 
expression  of  Strabo,  we  have  here  a  real  pronaos  or  ante-temple, 
because  a  passage,  open  to  the  sky,  intervenes  between  it  and  that 
part  of  the  building  which  contains  the  sanctuary.  The  four 
doorways  ^  with  which  this  vast  hall  is  provided  seem  to  indicate 
that  it  was  more  accessible  than  the  parts  beyond  the  passage 
just  mentioned. 

We  cannot  pretend  to  determine  the  uses  of  all  those  chambers 
which  encumber  with  their  ruins  the  further  parts  of  the  great 
building.  It  is  certain  however  that  between  them  they  constitute 
the  naos,  or  temple  properly  speaking.  They  are  surrounded  by 
a  double  wall  and  there  is  but  one  door  by  which  they  can  be 
reached — -precautions  which  suffice  to  prove  the  peculiarly  sacred 
character  of  this  part  of  the  whole  rectangle.  In  which  of  these 
chambers  are  we  to  find  the  amo^  ^  Was  it,  as  the  early  observers 
thought,  in  those  granite  apartments  which  are  marked  H  on  the 
plan  ^  This  locality  was  suggested  by  the  extra  solicitude  as 
to  the  strength  and  beauty  of  those  chambers  betrayed  by  the 
use  of  a  more  beautiful  and  costly  material  upon  them  than  upon 
the  rest  of  the  temple.  Moreover,  the  chamber  (H)  which  is 
situated  upon  the  major  axis  of  the  temple  bears  a  strong 
resemblance  in  shape,  as  well  as  position,  to  the  sanctuary  of 
the  temple  of  Khons,  in  the  case  of  which  no  doubt  was  possible. 
Or  must  we  follow  Mariette  when  he  places  the  sanctuary  in  the 
middle  of  the  eastern  court  (I  in  plan)  ?  All  traces  of  it  have 
now  almost  vanished,  but  Mariette  based  his  opinion  upon  the 
fact  that  in  the  ruins  of  this  court  alone  are  to  be  found  any 
traces  of  the  old  temple  dating  back  to  the  clays  of  the  Amen- 
emhats  and  Ousourtesens  of  the  twelfth  dynasty.  Pie  does 
not  attempt  to  account,  however,  for  those  carefully  built  granite 
apartments  which  seem  to  most  visitors  to  be  the  real  sanctuary, 
or,  at  least,  the  sanctuary  of  the  temple  as  reconstructed  and 
enlarged  by  the  princes  of  the  second  Theban  Empire. 

^  Including  a  postern  of  comparatively  small  dimensions,  there  are  five  doorways 
to  the  hypostyle  hall. — Ed. 


a  A  a_st  m  K  Si  a 


Fig.  215. — The  Great  Temple  at  Karnak  ;  inner  portion  ;  from  the  plan  of  M.  Brunc. 


INTERIOR    OF    THE    HYPOSTYLE    HALL  AT   KARNAK 


bcs  Cii  Chtriswi 


The  Temple  under  the  New  Empire. 


169 


In  the  actual  state  of  the  ruins  the  doubts  on  this  point  are, 
perhaps,  irremovable.  But  the  final  determination  of  the  question 
would  be  of  no  particular  moment  to  our  argument.  For  our 
purposes  it  is  sufticient  to  note  that  in  the  Great  Temple,  as  in 
the  Temple  of  Khons,  the  sanctuary  was  surrounded  and  followed 
by  a  considerable  number  of  small  apartments.  In  the  Great 
Temple  these  chambers  are  very  numerous  and  some  of  them 
are  large  enough  to  require  central  supports  for  their  ceilings  in 
the  form  of  one  or  more  columns.  In  other  respects  they  are 
similar  to  those  in  the  Temple  of  Khons. 


Fig.  216. — Karnak  as  it  is  at  present.     The  ruins  of  a  pylon  and  of  the  hypostyle  hall. 


The  resemblance  between  the  two  temples  is  completed  by 
the  existence  in  both  of  a  minor  hypostyle  hall  behind  the 
sanctuary.  The  hall  of  four  columns  of  the  smaller  building 
corresponds  to  the  large  saloon  called  the  Hall  of  Thothmes, 
in  the  Great  Temple  (J).  The  roof  of  this  saloon  is  supported 
by  twenty  columns  disposed  in  two  rows  and  by  square  piers 
standing  free  of  the  walls.  It  is  146  feet  wide,  and  from  53  to 
57   feet  deep.      Immediately  before   the   granite  apartments,  and 


VOL.     L 


B 


0/ 


o  A   History  of  Art  in  Ancient   Egypt, 


between  the  fifth  and  sixth  pylons,  another  hall,  also  with 
two  ranges  of  columns  but  not  so  deep  as  the  last,  is  introduced. 
Its  position  shows  it  to  be  meant  for  a  vestibule  to  the  naos 
properly  speaking.  The  fine  Court  of  the  Caryatides  with  its 
Osiride  pillars,  the  first  chamber  entered  by  the  visitors  who 
penetrate  into  the  temple  proper,  seems  to  have  been  designed 
for  a  similar  purpose  (G). 

If  we  wish,  then,  to  evolve  some  order  out  of  the  seeming  chaos 
ot  Karnak ;  if  we  wish  to  find  among  its  ruins  the  essential 
characteristics,  the  vital  organs,  if  we  may  put  it  so,  of  the 
Egyptian  temple,  we  have  only  to  apply  the  method  of  analysis 
and  reduction  suggested  by  examination  of  simpler  monuments, 
and  to  take  account  of  the  long  series  of  additions  which  resulted 
in  the  finally  stupendous  dimensions  of  the  whole  mass.^  These 
additions  may  be  distinguished  from  one  another  by  their  scale 
of  proportions  and  by  their  methods  of  construction.  When 
rightly  examined  the  gigantic  ruins  of  the  great  temple  of  Amen 
betray  those  simple  lines  and  arrangements,  which  form,  as  we 
have  shown,  the  original  type. 

The  same  remark  may  be  applied  to  the  other  great  building 
on  the  left  bank  of  the  river,  the  Temple  of  Luxor.  There, 
too,  the  architecture  is,  to  use  the  words  of  Champollion,  the 
"  architecture  of  giants."  From  the  first  pylon  to  the  innermost 
recesses  of  the  sanctuary  the  building  measures  about  850  feet. 
No  traveller  can  avoid  being  deeply  impressed  by  the  first  sight 
of  its  lofty  colonnades,  by  its  tall  and  finely  proportioned  pillars 
rearing  their  majestic  capitals  among  the  palms  and  above  the 
huts  of  the  modern  village.  These  columns  belong  to  the  first 
hypostyle  hall,  and,  were  they  not  buried  for  two-thirds  of  their 
height,  they  would  be,  from  the  ground  up  to  the  base  of  their 
capitals,  rather  more  than  50  feet  high ;  the  capitals  and  the 
cubes  above  them  measure  about  18  feet  more. 

The  plan  of  Luxor  is  more  simple  than  that  of  Karnak  ;  it 
was  built  in  two  "  heats  "  only,  to  borrow  an  expression  from  the 

^  A  plan  of  the  successive  accretions  is  given  in  plates  6  and  7  of  Mariettes' 
work.  The  different  periods  and  their  work  are  shown  by  changes  of  tint.  The 
same  information  is  given  in  another  form  in  pages  36  and  37  of  the  text.  The 
complete  title  of  the  work  is  as  follows  :  Karnak^  Etude  topogmphique  et  archeo- 
logiqite.  avec  u/i  Appendice  comprenant  les  principaux  Textes  hieroglyphiques.  Plates 
in  folio  ;  text  in  a  4to  of  88  pages  (1875). 


The  Temple  under  the  New  Empire. 


ZT^ 


athletes,  under  Amenophis  III.  and 
Rameses  II.  In  later  periods  it 
underwent  some  insignificant  re- 
touches, and  that  is  all.  It  is 
narrower  than  its  great  neighbour, 
and  covers  a  very  much  less  space 
of  ground,  neither  has  it  so  many 
chambers,  and  yet  we  are  in  some 
respects  more  at  a  loss  in  attempt- 
ing to  assign  their  proper  uses  to 
its  apartments  and  in  finding  some 
equivalent  for  them  in  the  elementary 
type  from  which  we  started,  than 
we  were  in  the  larger  temple. 

It  is  true  that  the  proper  character 
of  the  naos  is  better  marked  at  Luxor 
than  elsewhere.  The  sanctuary  may 
be  determined  at  a  glance.  It 
consists  of  a  rectangular  chamber 
standing  in  the  middle  of  a  large 
square  hall ;  it  is  the  only  chamber 
in  the  whole  building  for  which 
granite  has  been  used  ;  it  has  two 
doors,  one  in  each  end,  exactly  upon 
the  major  axis  of  the  building.  The 
hall  in  which  it  is  placed  is  preceded 
by  a  vestibule,  and  surrounded  by 
those  small  chambers  which  are 
always  found  in  this  part  of  a 
temple.  So  far,  then,  there  is 
nothing  to  embarrass  us ;  every- 
thing is  in  conformity  with  the 
principles  which  have  been  laid 
down. 

The  real  difficulty  begins  when 
we  look  round  us  for  the  p7'o- 
naos,  and  examine  the  hypostyle 
halls.  Here,  as  elsewhere, 
there  is  a  hall  of  modest  dimen- 
sions   beyond    the    sanctuary.       It 


to         43        20 


Fig.  217. — Plan  of  the  Temple  of  Luxor. 


1^2  A   History  of  Art  in  Ancient  Egypt. 

is  supported  by  twelve  columns.     There  is  another,  much  wider 
and  deeper,  in  front  of  the  naos  ;  it  has  thirty-two  of  those  lofty 
columns    of   which    we    have    already    spoken.       By    its    design, 
situation,  and   the  spacing  of  its   columns,   it  reminds    us   of  the 
hypostyle    hall    of    Karnak.      It    differs    from    it    in    being    open, 
without  any  external  wall   towards  the   court ;  so  that   it  may  be 
called   a   portico  with   four   ranges  of  columns.      Moreover,  again 
unlike   the    Karnak   hall,    it   is   by   no   means  the   most   imposing 
feature   of   the   whole    edifice.     The   greatest    elevation    and    the 
most  imposing  proportions,  so  far  as  the  interior  of  the  building 
is   concerned,  are   to   be   found   in   the  great   gallery   which   leads 
from  the  first  to  the  second  court,  from  the  second  to  the  third 
pylon.      This    gallery  is    in  effect  a  hypostyle  hall,   but  it  differs 
profoundly  from   the    superb    edifice   which    bears    that    name    at 
Karnak.      It    is    lon^    and    narrow    and   looks    more  like  a  mere 
covered  corridor   than  an  ample  hall  in   which    the   eager   crowd 
could  find  elbow-room. 

The  place  occupied  by  this  hall  in  the  whole  composition  is 
equally  singular.  It  has  been  ascertained  that  the  first  pylon 
and  the  peristylar  courtyard  behind  it  date  from  the  time  of 
Rameses  II.,  while  all  the  rest  of  the  building,  from  what  is  at 
present  the  second  pylon  inwards,  was  built  by  Amenophis  III. 
The  doorway  in  the  second  pylon  leads  immediately  into  the 
grand  gallery,  some  1 76  feet  long,  of  which  we  have  been 
speaking. 

We  can  hardly  tell,  therefore,  where  to  look  for  the  true 
pronaos  at  Luxor.  In  that  part  of  the  ground  plan  where  it 
is  generally  found  there  is  nothing  but  an  open  portico,  which  is 
considerably  lower  than  the  highest  parts  of  the  building.  The 
great  colonnade,  again,  is  separated  from  the  naos  by  an  open 
court,  so  that  it  ought,  perhaps,  to  be  classified  as  what  the 
Greeks  called  a  propylseum  ;  but  yet  it  is  a  hall,  inclosed  and 
covered,  of  great  size  and  height,  and  richly  decorated,  like  the 
hypostyle  halls  which  we  have  already  described  and  others  which 
we  have  yet  to  notice.^ 

'  In  presence  of  this  double  range  of  superb  columns  one  is  tempted  to  look 
upon  them  as  the  beginning  of  a  hypostyle  hall  which  was  never  finished,  to  suppose 
that  a  great  central  nave  was  constructed,  and  that,  by  force  of  circumstances 
unknown,  the  aisles  were  never  begun,  and  that  the  builders  contented  themselves 
by  inclosing  and  preserving  their  work  as  far  as  it  had  gone. 


•    fill 


It     ^iji  # 

■'V    V'« 


o 

h-1 


13 


C 


The  Temple  under  the  New  Empire.  375 

Another  peculiarity  of  Luxor  is  its  change  of  axis.  The  first 
pylon,  that  of  Rameses,  is  not  parallel  with  the  two  built  by 
Amenophis  ;  the  angle  at  which  they  stand  Is  a  very  perceptible 
one.  Neither  is  the  doorway  of  this  pylon  in  alignment  with  the 
other  doorways  on  the  major  axis  of  the  building.  No  justification 
or  even  explanation  of  this  irregularity,  which  is  unique  among  the 
Theban  temples,  has  been  discovered. 

If  we  cross  the  Nile  and  land  upon  the  plain  which  stretches 
between  the  river  and  the  Libvan  hills,  we  find  ourselves  in  the 
presence  of  those  temples,  the  Ramesseum,  Medinet-Abou,  and 
Gournah,  whose  funerary  destination  we  have  already  noticed. 
These  are  royal  chapels  erected  in  connection  with  the  royal 
tombs  in  their  neighbourhood,  they  are  cenotaphs  filled  with  the 
memories  of  the  great  Theban  princes,  and  with  representations 
of  their  exploits.  Consequently  we  do  not  find  In  them  those 
complications  which,  in  the  great  temples  of  the  right  bank,  mark 
the  successive  dynasties  to  which  their  final  form  was  due.  But 
yet  the  difference  in  general  appearance  is  not  great ;  there  is 
however,  one  distinction  which,  as  it  goes  far  to  prove  the 
peculiar  character  of  these  buildings,  should  be  carefully  noticed. 
In  no  one  of  them,  if  we  may  judge  from  plans  which  have  been 
made,  has  any  chamber  or  structure  been  found  which  corresponds 
to  the  sanctuary  or  a^Kosf  of  the  temples  of  Amen  or  Khons.  The 
absence  of  such  a  chamber  might  easily  be  explained  by  our 
supposition  that  these  buildings  were  funerary  chapels  ;  as  such 
they  would  require  no  depository  for  those  mysterious  symbols 
of  this  or  that  deity  which  the  temples  proper  contained  :  they 
were  the  lineal  descendants  of  the  upper  chambers  In  the 
mastabas,  in  which  no  rudiment  of  such  a  thing  Is  to  be  found. 
On  the  other  hand,  we  have  reason  to  believe  that  the  great 
Theban  divinities  were  associated  In  the  worship  paid  to  deceased 
kings.  If  that  were  so  these  funerary  temples  might  well  have 
been  arranged  like  those  of  the  right  bank.  The  inner  portions 
of  the  Ramesseum  and  of  Medinet-Abou  are  so  ruinous  that  the 
question  cannot  be  settled  by  the  examination  of  their  remains. 

The  Ramesseum  certainly  appears  to  have  been  the  monument 
described  by  Diodorus  as  the  Tomb  of  Osyniandias,  a  name  which 
has  never  been  satisfactorily  explained.^  It  Is  also  called  by  the 
Instittit  d Egyptc,    the   Palace  of  Memnon  and    the  Memnonlum, 

'  Diodorus,  i.  47-49. 


376  A   History  of  Art  ix  Ancient  Egypt. 

upon  the  faith  of  Strabo's  identification  of  Ismandes  and  Memnon.^ 
It  is  to  Champollion  that  this  building  owes  the  restoration  of  its 
true  title,  under  which  it  is  now  generally  known. 

Without  being  so  colossal  as  Karnak,  the  size  of  the  Ramesseum 
would  astonish  us  anywhere  but  in  Egypt.  When  it  was  com- 
plete, it  must  have  been  as  large  as  Luxor  before  the  additions 
of  Rameses  II.  were  made,  if  not  larger.  The  first  pylon 
was  226  feet  wide;  the  whole  of  its  upper  part  is  destroyed. ^ 
Immediately  behind  this  pylon  comes  a  vast  peristylar  court, 
almost  square  on  plan  (186  feet  by  173).  On  the  left  the 
remains  of  a  double  colonnade  exist,  which  must  at  one  time 
have  extended  along  at  least  two  sides  of  the  quadrangle.  At  the 
further  end  of  this  court  and  directly  facing  the  back  of  the  pylon, 
was  a  colossal  statue  of  Rameses.  Although  seated,  this  statue 
was  more  than  56  feet  high  ;  its  fragments  now  cover  a  con- 
siderable amount  of  the  courtyard.  A  grand  doorway,  pierced 
through  the  centre  of  the  wall  upon  which  the  defeat  of  the 
Khetas  is  painted,  leads  to  a  second  court,  a  little  less  extensive 
than  the  first.  Right  and  left  there  are  porticos,  each  with  a 
double  range  of  columns.  On  the  side  of  the  entrance  and  on 
that  opposite  to  it  there  are  single  ranges  of  Osiride  figures. 
Many  of  these  figures  are  still  standing  ;  they  are  3  i  feet  high. 

Three  flights  of  steps  lead  up  from  this  court  into  a  vestibule 
ornamented  with  two  colossal  busts  of  Rameses  and  with  a  row 
of  columns.  From  this  vestibule  the  hypostyle  hall  is  reached 
by  three  doorways  of  black  granite.  It  measures  136  feet  wide 
and  103  deep.  Its  roof  is  supported  by  forty-eight  columns,  in 
eiofht  ranofes  of  six  each,  counting  from  front  to  rear.  Five 
of  these  eight  ranges  are  still  standing  and  still  afford  support 
to  a  part  of  the  ceiling.      This  latter  is  painted  with  golden  stars 

^  Strabo,  xvii.  i.  42.  In  another  passage  (xvii.  i.  46)  he  seems  to  place  the 
Memnonium  close  to  the  two  famous  colossi.  He  would,  therefore,  seem  rather  to 
have  had  in  view  an  "  Amenophium,"  the  remains  of  which  have  been  discovered  in 
the  immediate  neighbourhood  of  the  two  colossi.  The  French  savants  suspected  this 
to  be  the  case,  but  they  often  defer  to  the  opinions  of  their  immediate  predecessors 
among  Egyptian  travellers.     (^Description  generale  de  Thebes,  section  iii.) 

2  This  pylon  stands  in  the  foreground  of  our  view  (Fig.  220).  The  face  which 
is  here  shown  was  formerly  covered — as  we  may  judge  from  the  parts  which  remain 
— with  pictures  of  battles ;  and  that  we  might  not  have  to  actually  invent  scenes  of 
combat  for  our  restoration,  we  have  borrowed  the  ornamentation  of  the  first  pylon 
of  tlie  Temple  of  Khons.  The  scale  of  our  cut  is  too  small,  however,  to  show  any 
details. 


The  Temple  under  the  New  Empire. 


Z11 


upon   a  blue  ground,  In   imitation  of  the  vault  of  heaven.      The 
side  walls  have  entirely  disappeared.^ 

This  hall  resembles  that  at   Karnak,  both  in  its  plan  and  in  its 
general    appearance.      The   mode   of  lighting   is  the   same  ;    the 


si 


-.  ( 


:3^' 


I  < 


^^^ 


■•vy/"\*''"l\^^*'v^ 


\\  \ 


mwi 


^  \^#§.^ 


=*.  i  =i^ 


Sftf 


^%:'jnhi 


S  K^ 


%^iiii#f) 


^^^liiliilHtefe 


tf-.  r^:r^.. 


*■  '^'^ 


m^;^f0f 


^-^i^  r  ,^ "  I  W//^/v  //-<iM^ 


il" 


iM  'Mil   r.W^ 


•.%^ 


"  •    :rg"'      19"   8,1)    1^^^ 


e  • 


::  D 


^.■;;;:ffnip 


^#?-' 


.1. 


1)       iO      20      31)      45      SO 


Fig. 219. — Plan  of  the  Kamesseum  (from  Lepsius.) 

arrangement  is  the  same  ;  there  is  in  both  a  wide  passage  down 
the  centre,  supported  by  columns  thicker  and  higher  than  the 
rest,  from  which  they  are  also  distinguished  by  the  nobility  of 
their   bell-shaped   capitals.      At   Karnak   the   hall  was   begun    by 

^  Lepsius,  Denkmceler,  part  i.  plates  88  and  89.  The  engineers  of  tlie  Instihit 
d'Egypte  fell  into  an  error  in  speaking  of  this  hall.  They  failed  to  notice  that  it 
was  smaller  than  the  second  court,  and  they  accordingly  gave  it  sixty  columns. 
[Description  gl'iieralc  de  T/icbes,  \o\.  i.  p.  132.) 

VOL.    I.  3    <' 


Tf^S  A    History  of  Art  in  Ancient  Egypt. 

Rameses  I.  and  Seti  ;  Rameses  II.  did  no  more  than  carry  on 
the  work  of  his  predecessors.  He  heard  the  chorus  of  admiration 
with  which  the  completion  of  such  a  superb  building  must  have 
been  hailed,  and  we  can  easily  understand  that  he  was  thereby 
incited  to  reproduce  its  happy  arrangement  and  majestic  propor- 
tions in  the  great  temple  which  he  was  erecting  in  his  own  honour 
on  the  left  bank  of  the  river. 

Ambitious  though  he  was,  Rameses  II.  could  not  attempt  to 
give  the  colossal  dimensions  of  the  great  temple  of  Amen  to 
what  was,  after  all,  no  more  than  the  chapel  of  his  own  tomb. 
The  great  hall  at  Karnak  required  three  reigns,  two  of  them  very 
long  ones,  for  its  completion.  In  the  Ramesseum  an  attempt  was 
made  to  compensate  for  inferior  size  by  extra  care  in  the  details 
and  by  the  beauty  of  the  workmanship.  The  tall  columns 
of  the  central  nave  were  no  more  than  thirty-six  feet  high, 
including  base  and  capital,  the  others  were  only  twenty-five  feet ; 
but  they  surpassed  the  pillars  at  Karnak  by  the  elegance  of 
their  proportions. 

The  admiration  excited  in  us  by  the  ruins  of  Karnak  is  mingled 
with  astonishment,  almost  with  stupefaction,  but  at  the  Rames- 
seum we  are  more  charmed  although  we  are  less  surprised.  We 
see  that,  when  complete,  it  must  have  had  a  larger  share  than  its 
rival  of  that  beauty  into  which  merely  colossal  dimensions  do 
not  enter.^ 

Beyond  the  hall  there  are  wide  chambers,  situated  upon  the 
major  axis  of  the  building,  and  each  with  its  roof  supported  by 
eight  columns.  Beyond  them  again  there  is  a  fourth  and  smaller 
chamber  which  has  only  four  columns.  Round  these  rooms  a 
number  of  smaller  ones  are  gathered  ;  they  are  all  in  a  very  frag- 
mentary state,  and  among  them  no  vestige  of  anything  like  a 
secos  has  been  found.  On  the  other  hand,  the  bas-reliefs  in  one 
of  the  larger  rooms  seem  to  confirm  the  assertion  of  Diodorus, 
in  his  description  of  the  Tomb  of  OsymandiaSy  that  the  library 
was  placed  in  this  part  of  the  building.^ 

The  Ramesseum  was  formerly  surrounded  by  brick  structures 
of  a  peculiar  character,  some  of  which  are  yet  to  be  found  in  good 
preservation  at  about  50  metres  from  the  north  face  of  the 
building.       They    consist    of   a   double    range   of    vaults    closely 

1  See  Ebers,  ^'Egypten,  vol.  ii.  pp.  309  et  seq. 
-  Ibid.,  p.  312. 


"  X"  "^-^  ^T'. 

Mm 


i^CAirj. 


"\  "•\m^%\ 


Fig.   220. — The   Ramesseum.     Bird's-eye   view   of  the   general   arrangement,    restored    by 

M.   Ch.  Chipiez. 


The  Temple  under  the  New  Empire. 


381 


abutting  on  each  other,  numbering  from  ten  to  twelve  in  each 
range,  and  surmounted  by  a  platform.  If  it  be  true  that  a  library 
was  included  in  the  building,  these  curious  structures,  which  are 
situated  within  the  outer  bounding  wall  of  the  temple,  may  have 
contained  rooms  for  lodging  and  instructing  students,  as  well  as 
chambers  for  the  priests.  In  that  case  Rameses  would  deserve 
the  credit  of  having  founded,  like  the  Mussulman  sovereigns,  a 
n^t'dress^,  or  sort  of  university,  by  the 
side  of  his  tu7'dch  and  mosque.  Additional 
probability  is  given  to  this  conjecture 
both  by  certain  discoveries  which  have 
been  made  in  tombs  near  the  Ramesseum 
and  by  the  evidence  of  several  papyri.^ 
But  for  these  texts  we  should  be  inclined 
to  believe  that  these  remains  are  the 
ruins  of  storehouses. 

About  a  thousand  yards  south-west  of 
the  Ramesseum  rises  the  group  of  build- 
ings which  is  known  by  the  name  of 
the  modern  villaafe  of  Medinet-Abou, 
It  was  not  until  the  second  half  of  the 
present  century  had  commenced  that 
they  were  cleared  from  the  debris  and 
modern  huts  which  concealed  many  of 
their  parts.  The  group  is  composed  of 
three  distinct  building's  in  one  enclosure. 
The  oldest  is  a  temple  built  by 
Thothmes  II.  and  Thothmes  III,  and 
afterwards  enlarged  by  the  Ptolemies 
and  the  Roman  Emperors  (A  on 
plan).  The  other  two  date  from  the 
time  of  Rameses  III.,  the  founder  of  the  twentieth  dynasty. 
They  both  lie  upon  the  same  axis,  they  are  connected  by  a  sphinx 
avenue,  and  they  must  certainly  be  considered  as  two  parts  of  one 
whole.  The  first  of  the  three  which  we  encounter  in  approaching 
the  group  from  the  river  is  known  as  the  Royal  Pavilion  or 
Pavilion  of  Rameses  III.  (B).  Ninety  yards  farther  to  the  north 
we  come  upon  the  great  temple,  the  funerary  character  of  which  we 
have  already  explained  (C) .    It  is  a  second  Ramessewn,  and  to  avoid 

'  Ebers,  yEg}'pfen,  vol.  ii.  p.  312. 


Fig.    221. — General    plan   of   the 
buildings  at  Medinet-Abou. 


;82 


A   History  of  Art  in  Ancient  Egypt. 


confusion  it  is  generally  known  as  the  Great  Temple  of  Medinct-Abou. 
We  shall  return  to  the  Royal  Pavilion  presently,  and,  as  for  the 
Temple  of  Thothmes,  which  was  consecrated  to  Amen,  its  really 
ancient  portion  is  of  too  little  importance  to  detain  us  long.  It 
consists  merely  of  an  isolated  secos  surrounded  on  three  sides  by 
an  open  gallery  upheld  by  square  piers  and,  upon  the  fourth,  by 
a  block  containing  six  small  chambers  (Fig.  222). 

The  great  temple,  however,  whose  picturesque  ruins  attract  every 

visitor  to  Thebes,  deserves  to  be 
carefully  considered  even  in  our 
summary  review.^  It  bears  a  strik- 
inor  resemblance  to  the  Rames- 
seum.  Their  dimensions  are 
nearly  the  same.  The  first  pylon 
at  Medinet-Abou  is  210  feet 
wide.  The  two  courts  which  follow 
and  isolate  the  second  pylon  are 
severally  113  feet  by  140,  and  126 
feet  by  136.  The  plan  of  Me- 
dinet-Abou does  not  differ  (223)  in 
any  very  important  points  from 
n  [7  that   of   the    Ramesseum.     Upon 

idZ^Zini    ^^     ^     ^  two  of  its  sides  only,  those  which 

are  at  right  angles  to  the  face  of 
the  pylon,  the  first  quadrangle 
has  colonnades.  One  of  these 
colonnades,  that  on  the  right 
of  a  visitor  entering  the  temple, 
consists  of  a  row  of  pillars  faced 
with  caryatides  of  Osiris.  These 
Osiride  piers  are  repeated  in 
the  second  court,  where  a  double 
colonnade,  five  steps  above  the  pavement,  leads  to  the  pronaos. 
The  latter  seems  too  small  for  the  two  peristyles.  It  has  only 
twenty-four  supporting  columns,  in  four  rows  of  six  each,  counting 

1  The  plan  in  the  Description  de  P Egvpte  {Aniiqiiiles,  vol.  ii.  pi.  4)  does  not  go 
beyond  the  back  wall  of  the  second  court.  That  of  Lepsius  goes  to  the  back  of 
the  hypostyle  hall.  {Deiiii/jicBkr,  part.  i.  pi.  92.)  Ours  is  much  more  comprehensive 
— it  goes  three  stages  farther  back ;  it  was  communicated  to  us  by  M.  Brune,  who 
measured  the  building  in  1866. 


^;Mix  :?.u  ■ 


D     I      '-:^?;ii!wy>v'&^ 


n 


h — fr 


^■^ 


=0  0= 


Fig.  222. — Plan  of  the  Temple  of  Thothmes 
(Champollion,  Notices  descripiives,  p.  314.) 


The  Temple  under  the  New  Empire. 


3^3 


from  front  to  back 
of  the  building. 
These  columns  are 
smaller  in  section 
than  those  of  the 
peristyles,  and  the 
eight  which  con- 
stitute the  central 
nave  do  not  differ 
from  their  com- 
panions.^ This  hy- 
postyle  hall  lacks, 
therefore,  some  of 
the  distinoruishinof 
characteristics  of 
its  rivals  elsewhere. 
Its  unambitious  ap- 
pearance is  all  the 
more  surprising  af- 
ter the  noble  pro- 
portions and  rich 
decorations  of  the 
two  external  courts. 
The  effect  of  the 
hall  is  still  farther 
lessened  by  the  fact 
that  it  does  not 
occupy  the  whole 
width  of  the  build- 

'  Here  M.  Perrot  is 
in  error,  as  may  be  seen 
by  reference  to  his  own 
plan.  The  columns  of 
the  central  passage  of 
the  hypostyle  hall  are 
similar  in  section  to 
those  of  the  two  peri- 
styles, except  that  their 
bases  are  flattened  late- 
rally in  a  somewhat  un- 
usual fashion. — Ed. 


Fig.  223.— -Plan  of  the  great  Temple  at  Medinet-Abou. 
(Communicated  by  M.  Brune. ) 


384  A  History  of  Art  in  Ancient  Egypt. 

ing.       Ranges  of  apartments   are   introduced  between  it  and   the 
external  walls  of  the  temple. 

Was  there  a  sanctuary  behind  this  hypostyle.hall  ?  It  would 
seem  rather,  according  to  the  recent  investigations  of  Mariette, 
that  upon  the  major  axis  of  the  temple  there  were  two  small  halls, 
each  supported  by  eight  columns,  like  those  in  the  Ramesseum  ; 
around  these  many  small  chambers  would  be  grouped  in  the 
fashion  which  is  almost  universal  in  this  part  of  an  Egyptian 
religfious  buildincr.  The  little  that  can  be  discovered  as  to  this 
point  has  its  importance  in  establishing  a  comparison  between  the 
temple  of  Rameses  II.  and  that  of  Rameses  III.,  because  it 
might  prove  that  the  similarity,  which  we  have  mentioned  as 
existing  between  the  more  public  parts  of  the  two  edifices, 
extended  to  the  sanctuary  and  its  dependencies  in  the  rear. 
The  last  of  the  great  Theban  Pharaohs  certainly  drew  much  of 
his  inspiration  from  the  work  of  his  illustrious  predecessors.  In 
their  present  state  of  mutilation  it  is  impossible  to  decide  which 
was  the  finer  of  the  two  in  their  complete  state.  To  the  fine 
hypostyle  hall  of  the  Ramesseum,  Medinet-Abou  could  oppose  the 
Royal  Pavilion  which  rose  in  front  of  the  temple  and  grouped 
itself  so  happily  with  the  first  pylon,  affording  one  of  the  most 
effective  compositions  in  the  whole  range  of  Egyptian  architecture. 

The  rest  of  the  temples  in  this  neighbourhood  and  within  the 
enclosures  at  Karnak  are  all  more  or  less  intimately  allied  to  the 
type  we  have  established,  and  need  not  be  noticed  in  detail.^ 

We  have  good  reason  to  believe  that  the  type  of  temple  which 
we  have  described  was  a  common  one  in  other  parts  of  Egypt 
than  Thebes.  The  temples  of  Memphis,  of  Heliopolis  and  of 
the  Delta  cities,  have  perished  and,  practically,  left  no  trace 
behind  ;  but  the  great  buildings  constructed  by  the  Theban  con- 
querors outside  the  limits  of  Egypt  proper,  in  Nubia,  are  in 
comparatively  good  preservation.  One  of  these,  the  Temple 
of  Soleb,  built  by  Thothmes  III.  and  reconstructed  by  Amen- 
ophis  III.,  must  have  borne  a  strong  resemblance  to  the 
Ramesseum,  so  far  as  can  be  judged  through  the  discrepan- 
cies in  the  available  plans  of  the  first-nanied  building.  Cailliaud 
only   allows  it  one  peristylar  court,   while  Hoskins   and   Lepsius 

1  A  few  of  these  buildings — that,  for  instance,  on  the  right  of  the  great  lake — 
seem  to  have  been  very  peculiar  in  arrangement,  but  their  remains  are  in  such  a 
state  of  confusion  that  it  is  at  present  impossible  to  describe  their  plans. 


The  TE^rpLE  under  the  New  Empire.  385 

give  it  two.  According  to  Cailliaud,  its  hypostyle  hall,  which  must 
have  been  a  very  beautiful  one,  contained  forty-eight  columns. 
After  it  came  another  hall,  with  a  roof  supported  by  twelve 
columns.  This  was  surrounded  by  small  chambers,  the  remains 
of  which  are  very  confused.  In  the  plan  given  by  Lepsius  there  are 
two  hypostyle  halls  with  a  wall  between  them,  an  arrangement  which 
is  also  found  at  Abydos.  The  outer  one  must  have  had  twenty- 
four  columns,  the  largest  in  the  building,  and  the  second  forty,  of 
rather  less  diameter  ;  the  remainder  of  the  temple  has  disappeared.^ 

We  find  analogous  arrangements  in  the  great  temple  of  Napata 
{Gebel-Bai'kal).  Built  by  Amenophis  III,  when  Napata  was  the 
seat  of  an  Egyptian  pro-consul,  and  repaired  by  Tahraka  when 
Ethiopia  became  supreme  over  Egypt,  this  temple  resembles  the 
Theban  buildings  in  its  plan.  From  a  peristylar  court  enclosed 
between  two  pylons,  we  pass  into  a  hypostylar  hall  containing 
forty-six  columns  ;  behind  this  hall  comes  the  sanctuary,  in  its 
usual  position,  with  its  entourage  of  small  chambers.  We  may 
call  this  the  classic  type  of  Egypt. 

The  tem.ples  which  we  have  hitherto  examined  are  chiefly 
remarkable  for  the  simplicity  of  their  plan.  A  single  sanctuary 
forms  the  centre  and,  so  to  speak,  the  heart  of  the  whole 
composition.  Pylons,  peristylar  courts  and  hypostylar  halls,  are 
but  anterooms  and  vestibules  to  this  all  important  chamber ; 
while  the  small  apartments  which  surround  it  afford  the  necessary 
accommodation  for  the  material  adjuncts  of  Egyptian  worship.  In 
the  great  temple  at  Karnak,  the  anterior  and  posterior 
dependencies  are  developed  to  an  extraordinary  extent,  but 
this  development  is  always  in  the  direction  of  the  length,  or  to 
speak  more  accurately,  of  the  depth  of  the  building.  The  smaller 
faces   of  the  whole  rectangle  are  continually  carried  farther  from 

*  Cailliaud,  Voyage  a  Miroe,  plates,  vol.  ii.  pi.  9-14.  Lepsius,  Dejikmceler, 
part  i.  pi.  T16,  117.  HosKiNS,  Travels  in  Ethiopia,  plates  40,  41,  and  42.  The 
plan  given  by  Hoskins  agrees  more  with  that  of  Lepsius  than  with  Cailliaud,  but  it 
only  shows  the  beginning  of  the  first  hypostyle  hall  and  nothing  of  the  second. 
These  divergences  are  easily  understood  when  it  is  remembered  that  nothing  but 
some  ten  columns  of  two  different  types  remain  ///  situ,  and  that  the  mounds  of 
debris  are  high  and  wide.  In  order  to  obtain  a  really  trustworthy  plan,  this  accumu- 
lation would  have  to  be  cleared  away  over  the  whole  area  of  the  temple.  All  the 
plans  show  a  kind  of  gallery,  formed  of  six  columns,  in  front  of  the  first  pylon  ;  it 
reminds  us  in  some  degree  of  the  great  corridor  at  Luxor  ;  by  its  general  form, 
however,  rather  than  its  situation. 

VOL.    I.  3D 


386  A  History  of  Art  in  Ancient  Egypt. 

each  other  by  the  additions  of  fresh  chambers  and  architectural 
features,  which  are  distributed,  with  more  or  less  regular 
alternation,  on  the  right  and  left  of  the  major  axis  which  always 
passes  through  the  centre  of  the  secos.  The  building,  therefore, 
in  spite  of  many  successive  additions  always  contrives  to  preserve 
the  unity  of  its  organic  constitution. 

But  all  the  great  buildings  In  Egypt  which  were  constructed  for 
the  service  of  religion  were  not  so  simply  designed.  A  good 
instance  of  a  more  complex  arrangement  Is  to  be  found  In  the 
great  temple  at  Abydos  (Fig.  224).  It  was  begun  by  Setl  I.  and 
finished  by  Rameses  II.  Marlette  freed  It  from  the  debris  and 
modern  hovels  which  encumbered  It,  and,  thanks  to  his  efforts, 
there  are  now  few  monuments  in  Egypt  whose  inner  arrangements 
can  be  more  clearly  and  certainly  perceived. 

Its  general  shape  is  singular.  The  courts  and  the  pronaos 
compose  a  narrow  and  elongated  rectangle,  with  which  the  parts 
corresponding  to  the  sanctuary  and  Its  dependent  chambers 
form  a  right  angle  (see  Fig.  224).  This  salient  wing  has  no 
corresponding  excrescence  on  the  other  side.  We  might  consider 
the  building  unfinished,  but  that  there  is  no  sign  whatever  that 
the  architect  meant  to  complete  it  with  another  wing  at  the 
opposite  angle.  The  Egyptians  were  never  greatly  enamoured 
of  that  exact  symmetry  which  has  become  one  of  the  first  artistic 
necessities  of  our  time. 

Still  more  surprising  than  the  eccentricity  of  Its  plan,  are  the 
peculiar  arrangements  which  are  to  be  found  In  the  interior  of 
this  temple.  As  at  Medinet-Abou  and  the  Ramesseum,  there 
are  two  courts,  each  preceded  by  a  pylon.  After  these  comes 
the  pronaos.  The  courts  differ  from  those  at  Thebes  in  having 
no  peristyles  or  colonnades.  The  only  thing  of  the  kind  is  a  row 
of  square  pillars  standing  before  the  inner  wall  of  the  second  court 
(see  plan).  This  is  a  poor  equivalent  for  the  majestic  colonnades 
and  files  of  caryatides  which  we  have  hitherto  encountered. 

The  suppression  of  the  portico  has  a  great  effect  upon  the 
appearance  of  these  two  courts.  It  deprives  them  of  the  rich 
shadows  cast  by  the  long  colonnades  and  their  roofs  of  the 
Theban  temples,  and  the  long  walls  must  have  seemed  rather  cold 
and  monotonous  in  spite  of  the  bas-reliefs  and  paintings  which 
covered  them.  Their  absence,  however,  is  not  allowed  to  affect 
the  general  lines  of  the  plan. 


;5 


Ifa^F^  ooooooooooo 

OOOOOOOOOOG 

ooooooooooo 

□  □  □  □  a  f 

OOGOOOOOQG  GO 
OO0OOOOO 


We  have  given  neither  an  elevation 
nor  a  section  of  the  temple  at  Abydos, 
because  neither  the  one  nor  the  other 
was  to  be  had.  The  building  was 
hardly  known  until  Mariette  freed  it 
from  the  debris  with  which  it  was  en- 
gulphed.  He,  too,  studied  rather  as 
an  egyptologist  than  as  an  architect, 
and  was  content  with  making  known  its 
internal  arrangements  by  a  plan.  This 
plan  does  not  appear  to  be  minutely 
exact.  A  little  farther  on  we  shall  have 
to  speak  of  a  peculiarity  which  exists 
at  Abydos,  but  which  is  not  hinted  at 
in  the  adjoining  plan ;  some  of  the 
columns  are  coupled  in  the  first  hypo- 
style  hall.  We  take  this  fact  from  the 
Description^  where  the  measurements 
are  given  in  a  fashion  which  forbids  all 
doubt  of  their  fidelity. 


* 


QDaDQO    naaeBQ 


:  , ir~~7^ 


<//■■ 

m. 


WA 


I 


i 


« 


^4 

i 


m 


J   -L 


Fig.  224, — Plan  of  the  Temple  at  Abydos  (from  Mariette.) 


The  Temple  under  the  New  Empire.  389 

It  is  when  we  arrive  at  the  pronaos  that  we  fail  to  recognize 
the  disposition  to  which  we  have  grown  accustomed.  There  is 
no  central  nave,  with  its  columns  of  extra  size  and  more  careful 
desio-n,  leading  to  the  closed  door  of  the  sanctuary.  There  are 
two  hypostyle  halls,  the  first  supported  by  twenty-four,  the  second 
by  thirty-six  columns.  They  are  separated  by  a  wall  pierced  with 
seven  doorways,  each  doorw^ay  corresponding  to  one  of  the  aisles 
between  the  columns.  In  the  farther  wall  of  the  second  of  these 
halls,  there  are  seven  more  doorways,  corresponding  to  the  last 
named,  and  opening  upon  seven  oblong  vaulted  saloons,  all  of  one 
size  and  completely  isolated  one  from  another. 

By  their  situation  on  the  plan,  by  their  form,  and  by  the 
decoration  of  their  walls,  these  vaulted  chambers  declare 
themselves  to  be  so  many  sanctuaries.  Each  one  of  them  is 
dedicated  to  some  particular  deity,  whose  name  and  image  appear 
in  the  decorations  of  the  chamber  itself  and  also  upon  the  lintel 
of  the  door  outside.  These  names  and  images  are  again  repeated 
upon  all  the  surfaces  presented  by  the  aisle  which  leads  up  to 
the   door. 

The  seven  deities  thus  honoured,  beginning  at  the  right,  are 
Horus,  Isis,  Osiris,  Amen,  Harmachis,  Ptah,  and  Seti  himself, 
whom  we  thus  find  assimilated  with  the  greatest  of  the  Egyptian 
gods.  Each  chamber  contains  a  collection  of  thirty-six  pictures, 
which  are  repeated  from  one  to  another  with  no  changes  beyond 
those  rendered  necessary  by  the  substitution  of  one  god  for 
another.  These  pictures  deal  with  the  rites  which  would  be 
celebrated  by  the  king  in  each  of  the  seven  sanctuaries. 

Behind  this  septuple  sanctuary  there  is  a  secondary  hypostyle 
hall,  just  as  we  find  it  behind  the  single  sccos  of  the  ordinary 
temple.  Its  roof  was  supported  by  ten  columns,  and  access  to  it 
was  obtained  through  the  third  sanctuary,  that  of  Osiris.  This 
part  of  the  temple  is  in  a  very  fragmentary  condition.  Very  little 
is  left  of  the  boundinof  walls,  but  it  has  been  ascertained  that 
several  of  these  chambers  were  dedicated  to  one  or  other  of  the 
deities  between  whom  the  naos  was  apportioned.  Thus  one  of  the 
chambers  referred  to  was  placed  under  the  protection  of  Osiris, 
another  under  that  of  Horus,  and  a  third  under  that  of  Isis. 

The  decoration  of  the  southern  wing  of  the  temple  seems 
never  to  have  been  completed.  It  contains  a  long  corridor, 
a    rectangular  court   with    an    unfinished    peristyle,   several  small 


390 


A   History  of  Art  in  Ancient  Egypt. 


chambers  with  columns,  and  a  flight  of  steps  leading  up  on  to 
the  flat  roof.  A  dark  apartment  or  crypt,  divided  into  two  stories 
by  a  floor  of  large  stone  slabs,  may  have  been  used  as  a 
storehouse. 

These  farthest  apartments  seem  to  have  been  arranged  in  no 
sort  of  order.  We  shall  not  here  enter  into  such  matters  as 
the  construction  of  the  seven  parallel  vaults  in  the  naos  ;  for  that 
a  future  opportunity  will  be   found  ;  ^   at  present  our  business  is  to 


Fig.  225. — Seti,  willi    the    attributes  of  Osiri>,    between  Amen,   to  whom  he   is   paying  homage, 

and  Chnoum. 


make  the  differences  between  the  temple  at  Abydos  and  that  of 
Khons  and  its  congeners,  clearly  understood.  The  distinction  lies 
in  the  seven  lono^itudinal  subdivisions,  beofinninof  with  the  seven 
doors  in  the  facade  of  the  hypostyle  hall,  and  ending  in  the 
vaulted  chambers  which  form  the  same  number  of  sanctuaries. 
Seen  from  outside,  the  temple  would  not  betray  its  want  of  unity  ; 


^  Full  particulars  of  the  more  obscure  i)arts  of  the  temple  at  Abydos  will  be 
found  in  Mariette's  first  volume. 


The  Temple  under  the  New  Empire.  391 

it  was  surrounded  by  a  single  wall,  the  complex  naos  was  prefaced 
by  courts  and  pylons  in  the  same  fashion  as  in  the  temples  of 
Thebes  which  we  have  already  noticed,  and  it  would  not  be  until 
the  building  was  entered  and  explored  that  the  fact  would  become 
evident  that  it  was  seven  shrines  in  one,  seven  independent 
temples  under  one  roof^ 

At  Thebes  also  we  find  a  temple  w^hich,  by  its  internal 
arrangements,  resembles  that  of  Abydos.  It  is  called  sometimes 
the  Palace  and  sometimes  the  Temple  of  Gournah ;  in  the 
inscriptions  it  is  called  the  House  of  Sell.  Two  propylons,  one 
about  fifty  yards  in  front  of  the  other,  form  an  outwork  to  the 
main  building,  with  which  they  are  connected  by  an  avenue  of 
sphinxes.  It  is  probable  that  they  were  originally  the  doorways 
through  brick  walls,  now  demolished,  which  formed  successive 
enclosures  round  the  temple.  The  dromos  led  up  to  the  pronaos, 
which  was  reached  by  a  few  steps.  The  front  of  the  naos  is  a 
portico  of  simple  design,  consisting  of  ten  columns  between  two 
square  pilasters,  the  whole  being  166  feet  long  by  10  feet 
deep.  Eight  of  these  fine  columns  are  still  erect.  The  wall  at 
the  back  of  the  portico  is  pierced  by  three  doorways,  to  which 
three  distinct  compartments  or  divisions  of  the  interior  correspond 
(see  plan,  Fig.  226). 

The  only  feature  in  which  these  compartments  resemble  one 
another  is  their  independence.  They  are  isolated  from  one 
another  by  walls  which  run  from  front  to  back  of  the  naos.  The 
most  important  and  elaborate  of  the  three  compartments  is  the 
middle  one.  Its  entrance  doorway  opens  directly  upon  a  hall 
which  is  the  largest  in  the  whole  temple.  It  is  eighteen  metres 
long,  its  roof  is  supported  by  six  columns  similar  to  those  of  the 
portico'  already  mentioned,  and  ranged  around  it  are  nine  small 
chambers,  the  pictures  in  which  illustrate  the  apotheosis  of 
Seti,  who,  often  indued  with  the  attributes  of  Osiris,  is  sometimes 
shown  doing  homage  to  the  Theban  triad  of  gods,  and  more 
especially  to  Amen-Ra,  sometimes  as  himself  the  object  of 
worship.  The  central  one  of  these  chambers  opens  upon  a 
hall  where  the  roof  is  supported  by  four  square  pillars,  and  upon 
this  hall  again  four  small  apartments  open.  These  can  hardly  be 
mere  storehouses,  but  they  have  suffered  so  greatly  that  no  certain 

'  Upon  the  funerary  character  of  the  great  temple  at  Abydos,  see  Ebers, 
^.gypten,  vol.  ii.  pp.  234,  235. 


39: 


A   History  of  Art  in  Ancient  Egypt. 


Tig.  226.— Plan  of  the  Ttmple  of  Gournah. 


opinion  can  be  formed  as 
to  their  real  purposes. 

The  riQ;ht-hand  com- 
partment  is  in  a  very  bad 
state,  but  enough  of  it 
remains  to  show  that  its 
arrangements  were  quite 
different  from  those  of  its 
neighbour  and  much  less 
complex.  So  far  as  we 
can  judge,  the  larger  part 
of  it  was  taken  up  with  a 
peristylar  court  or  hall 
seventy-six  feet  long  and 
forty-six  wide.  Behind 
this  the  site  of  three  rect- 
angular chambers  may  be 
distinguished.  Every  wall 
which  is  still  standing 
bears  representations  of 
Rameses  II.  paying  his 
devotions  to  the  Theban 
eods. 

The  left  compartment 
is  in  better  preservation 
than  the  right,  and  its 
arrano-ements  are  more 
like  those  of  the  central 
part  of  the  naos.  It  is 
not  so  large,  however, 
and  it  contains  no  hypo- 
style  hall.  It  has  six 
chambers  placed  in  two 
sets  of  three,  the  one  set 
behind  the  other.  Here 
we  find  Rameses  I.,  the 
founder  of  the  dynasty, 
honoured  by  his  son 
Seti  I.  and  his  grandson 
Rameses  II. 


^^\  I   s 


^^feaj-ScS 


rrM- 


^ 


V~    i 


n 


-M 


--    -^^ 


-->' 


.f^F 


f=?_==a 


a  ^jna±Kcei-i^-i?-i. 


ik; 


I     I 


■u- 


a. 


o 


^1 


-I 


■^ 


X 

r1 

:3 

o 

O 

*— . 

o 

"1 

1) 

J- 

<f 

C3 

aj 

"1 

H 

0) 

•J 

<3 


i3 


0) 

a 
ci 


r  f^i.f'^""^ 


iT 


(Tl 


"M 


6 


CO 


o 

g 

o 


o 


O 

o 


o 


,J3 
rt 
C 

S 
o 

O 

<a-< 
O 


o 


t/3 

c 
o 

-I 


CO 

d 


VOL.    I. 


3    E 


The  Temple  under  the  New  Empire.  395 

The  great  temples  of  Abydos  and  Gournah  were  built  by  the 
same  sovereigns,  Seti  I.  and  Rameses  II.  Perhaps,  too,  their 
plans  were  traced  by  the  same  architect.  The  resemblance 
between  them  is  so  great  that  they  may  be  looked  upon  as 
variants  of  one  type,  of  a  type  which  is  distinguished  by  the 
juxtaposition  of  similar  parts  grouped  laterally  one  by  the  side 
of  the  other.  Each  of  the  chapels  which  we  have  described  was 
self  contained,  the  subsidiary  chambers  which  were  required  for 
the  routine  of  worship  were  grouped  round  it,  either  on  one  side, 
as  at  Abydos,  or  in  the  angles  of  the  sanctuary  itself,  as  at 
Gournah.  With  such  slight  differences  of  detail  as  this,  the  two 
buildings  were  built  upon  the  same  principle.  At  Gournah  the 
division  is  tripartite,  and  the  three  compartments  vary  in  their 
arrangements  ;  at  Abydos  they  are  seven  in  number,  and  exactly 
similar  in  design.  A  temple  thus  cut  into  three  parts,  or  seven, 
reminds  us  of  the  seed-pods  of  certain  plants,  in  which  the 
fertilizing  grain  is  divided  between  several  cells.  But  whether 
these  are  numerous  or  few,  the  naos  never  has  any  great  depth. 
It  seems  as  if  the  absence  of  a  true  organic  centre  arrested  the 
development  of  the  building ;  we  find  no  signs  of  an  edifice 
which,  like  the  temple  of  Amen  at  Karnak,  might  be  developed 
almost  to  infinity  without  losing  its  unity. 

On  the  other  hand,  there  were  a  few  temples  in  which  a  severe 
and  extreme  unity  was  the  distinguishing  characteristic.  In  Upper 
Egypt  and  Nubia  a  few  examples  of  the  class  are  still  to  be  seen. 
As  a  rule  they  date  from  the  eighteenth  dynasty,  but  there  were  a 
few  temples  of  the  same  kind  erected  under  the  Ptolemies.^  It 
seems  probable,  therefore,  that  they  were  common  to  all  the 
periods  of  Egyptian  history,  and  to  the  conquered  provinces,  as 
well  as  to  Egypt  proper.  They  were  erected  within,  and  in  the 
neighbourhood  of,  those  cities  whose  importance  was  not  sufficient 
to  demand  such  great  monumental  works  as  the  temples  of  Thebes 
or  Abydos,  of  Memphis  or  Sais.  We  might  call  them  chapels, 
raised  either  to  the  honour  of  the  local  deities,  or  for  the  purpose 
of  commemorating    the  passage  of  some  conquering  prince  and 


'  We  may  cite  as  a  peripteral  temple  of  the  Ptolemaic  epoch  the  building  at  Edfou, 
railed,  in  the  Description,  the  Little  7"cwplc  {Atitiqtiiics,  vol.  i.  plates  62-65).  It 
differs  from  the  Pharaonic  temples  of  the  same  class  in  having  square  piers  only  at 
the  angles,  the  rest  of  the  portico  being  supported  by  columns. 


o 


96 


A  History  of  Art  in  Ancient  Egypt. 


the   homage   paid   by  him   to  the   deity  to  whom    he   looked   for 
protection  and  victory. 

In  these  chapels  there  are  neither  internal  peristyles  nor  hypo- 
styles  ;  there  are  none  of  those  subsidiary  chambers  among  which 
it  is  sometimes  so  easy  to  lose  our  way.  There  is,  in  fact,  nothing 
but  a  rectangular  chamber  and  a  portico  about  it,  and,  in  most 
cases,  it  would  appear  that  a  short  dromos,  consisting  of  a  few 
pairs  of  sphinxes,  lent  dignity  to  the  approach. 

The  best  proportioned  and  perhaps  the  most  interesting  building 
of  this  class  is  the  little  sandstone  temple  built  by  Amenophis  III. 
at  Elephantine,  upon  the  southern  frontier  of  Egypt.  It  was  dis- 
covered at  the  end  of  the  last  century  by  the  draughtsmen  of  the 
French  Expedition,  and  named  by  them  the  Temple  of  the  Sotith} 

This  little  building  no  longer  exists.  It  was 
destroyed  in  1822  by  the  Turkish  Governor  of 
Assouan,  who  had  a  mania  for  building.  Happily 
the  plans  and  drawings,  which  we  reproduce, 
seem  to  have  been  made  with  great  care. 

The  total  area  of  the  temple,  at  the  floor 
level  of  the  cella,  was  40  feet  by  31.  It  was 
raised  upon  a  well-built  rectangular  base  of 
almost  the  same  lateral  dimensions,-  and  7  feet 
6  inches  high  to  the  pavement  of  the  portico. 
From  the  earth  level  to  the  top  of  the  cornice 
the  temple  was  21  feet  6  inches  in  height.  A 
flight  of  steps,  enclosed  between  two  walls  of  the 
same  height  as  the  stylobate,  led  up  to  the  portico.  The  portico 
itself  was  composed  of  square  piers  and  round  columns.  Two  of 
the  latter  were  introduced  in  the  centre  of  each  of  the  smaller  faces 
of  the  building,  while  the  side  galleries  were  enclosed  by  seven 
square  piers,  inclusive  of  those  at  the  angles.  A  dwarf  wall  about 
three  feet  in  height  bounded  the  gallery  on  the  outside,  and 
afforded  a  base  for  the  piers  ;  the  circular  columns  on  each  side 
of  the  entrance  alone  stood  direcdy  upon  the  pavement  of  the 
gallery,  and  were  thus  higher  by  about  three  feet  than  either  the 
piers  or  the  columns  in  the  corresponding  facade  at  the  rear.  The 
oblong  chamber  enclosed  by  this  portico  had  two  entrances,  one  at 

1  Description  de  fEgypte,  Antiqiiites,  vol.  i.  plates  34-38- 

2  This  base  contained  a  crypt,  no  doubt  for  the  sake  of  economising  the  material. 
There  seems  to  have  been  no  means  of  access  to  it,  either  from  without  or  within. 


Fig.  229. — Plan  of  the 
Templeof  Elephantine. 
{DcscnptioH  deVEgypte, 
i-  35-) 


The  Temple  under  the  New  Empire. 


397 


the  top  of  the  steps,  the  other  at  the  back.^  The  first  named  was 
indicated  as  the  true  entrance  to  the  building-  by  the  slight  salience 
of  its  jambs  and  lintel,  by  the  increased  size  of  the  columns  in  front 
of  it,  and  by  its  position  with  regard  to  the  steps. 

One  more  peculiarity  must  be  noticed.  Neither  in  piers  nor  in 
walls  do  we  find  that  inward  slope  which  is  almost  universal  in 
Egyptian  exteriors.  The  lines  are  vertical  and  horizontal.  This 
is  not  the  effect  of  caprice  ;  the  architect  had  a  good  reason  for 
neglecting  the  traditions  of  his  profession.  By  avoiding  the  usual 
inclination  towards  the  centre,  he  gave  to  his  small  creation  a 
dignity  which  it  would  otherwise  have  missed,  and,  in  some 
degree,  concealed  its  diminutive  size. 


Fig.    230. — View   in    perspective   of   the    Temple  of  Elephantine   (from   the   Description  de 

VEg^'pte,  i.  35). 


In    spite   of   Its   modest    dimensions,    this    temple  was   without 
neither  beauty  nor  grandeur.      Its  stylobate  raised  it  w^ell  above 

1  Our  plan.  etc.  shows  the  temple  as  it  must  have  left  the  hands  of  the  architect, 
according  to  the  authors  of  the  Description  de  V Egyple.  Jomard  (pi.  35,  Fig.  i) 
has  imported  a  small  chamber  into  his  plan,  placing  it  behind  the  large  hall  as  a  sort 
of  opisihodomos  ;  but  he  bids  us  remark  that  it  was  constructed  of  different  materials, 
and  in  a  diftlerent  bo7id,  from  the  rest  of  the  temple.  It  showed  no  trace  of  the 
sculptured  decoration  which  covered  all  the  rest  of  the  temple.  This  chamber  was 
therefore  a  later  addition,  and  one  only  obtained  at  the  expense  of  the  continuous 
portico,  the  back  part  of  which  was  enclosed  with  a  wall  in  which  the  columns 
became  engaged.  According  to  Jomard,  this  alteration  dates  from  the  Roman 
period,  but  however  that  maybe,  in  our  examination  of  the  temple  we  may  disregard 
an  addition  which  appears  to  have  been  so  awkwardly  managed. 


398 


A  History  of  Art  in  Ancient  Egypt. 


the  plain,  while  the  steps  in  front  gave  meaning  and  accent  to  its 
elevation.  The  wide  spacing  of  the  columns  in  front  allowed  the 
richly  decorated  doorway  to  be  seen  in  effective  grouping  with  the 
long  perspectives  of  the  side  galleries.  The  piers  on  the  flanks 
were  more  closely  spaced  than  the  columns  of  the  facade,  and  the 
contrast  was  heightened  by  the  simplicity  of  their  form.  The 
dignity  of  the  entablature  and  the  bold  projection  of  the  cornice 
added  to  the  effect  of  the  whole,  and  emphasized  the  well-balanced 
nature  of  the  composition.  The  Egyptian  architects  never  pro- 
duced a  building  better  calculated  to  please  modern  tastes.  Its 
symmetry  and  just  proportion  appeal  directly  to  those  whose 
artistic  ideas  are  founded  upon  the  creations  of  the  Greeks 
and  Romans. 


ID  H  '7  'J 

1  ic.  231. — Longitudinal  section  of  the  Temple  of  Elephantine  (from  the  Description,  i.  35). 

This  sympathy  was  conspicuously  felt  by  those  who  discovered 
the  little  monument.  "  The  arrangement,"  says  Jomard,  "  is  a 
model  of  simplicity  and  purity.  .  .  .  The  Temple  of  Ele- 
phantine is  pleasing  as  a  whole,  and  commands  our  attention." 
But  the  purity  and  harmony  of  its  lines  are  not  its  only  claims  to 
our  admiration.  The  pleasure  which  it  causes  us  to  feel  is  partly 
the  result  of  its  resemblance  to  a  well-known  and  much  admired 
type,  that  of  the  Greek  temple.  In  all  essentials  the  arrangements 
are  the  same,  a  cella  raised  upon  an  important  base  and  surrounded 
by  a  colonnade. 

The  general  arrangement  of  the  Elephantine  structure  has  even 
its  name  in  the  technical  language  of  the  Greek  architects,  they 
would  call  it  a  pei'ipteral  temple,  because  the  colonnade  goes 
completely  round  it.  Nowhere  else  do  we  find  such  a  striking 
resemblance  between  Greece  and   Egypt.      But  for  the  mouldings, 


The  Temple  under  the  New  Empire.  399 

the  sculptured  decorations,  and  the  inscribed  texts,  we  should  be 
tempted  to  see  in  it  a  building  of  the  Ptolemaic  period,  Greek  in 
conception  and  plan,  but  decorated  in  the  Egyptian  taste.  Such 
a  mistake  would,  however,  be  impossible  in  these  days,  and  even 
at  the  end  of  the  last  century.  The  French  savants  knew  enough 
to  prevent  them  falling  into  such  an  error.  They  were  unable  to 
read  the  hieroglyphics,  but  the  general  physiognomy  of  the 
building  told  them  that  it  could  boast  of  a  venerable  antiquity. 
In  coming  to  this  conclusion  they  were  right,  but  they  should 
have  stopped  there  instead  of  attempting  to  establish  a  direct 
connection,  as  cause  and  effect,  between  the  Egyptian  building  and 
the  temples  of  Greece.  We  shall  not  here  discuss  the  delicate 
question  of  the  indebtedness  of  Greek  artists  to  those  of  Egypt, 
but  we  may  allow  ourselves  to  make  two  observations.  In  the 
first  place,  the  temples  built  upon  this  plan  were  very  small,  and 
must  have  attracted  very  little  notice  indeed  from  strangers 
dazzled  by  the  wonders  of  Sais,  Memphis,  and  Thebes  ;  and  the 
buildings  in  those  great  cities  did  not  offer  the  peculiar  charac- 
teristics which,  we  are  asked  to  believe,  inspired  the  early  Greek 
architects.  In  the  second  place,  if  there  had  been  any  direct 
imitation  of  an  Egyptian  model,  we  should  have  found  in  the 
copy  at  least  some  passing  trace  of  those  square  piers  which  were 
so  continually  and  successfully  used  by  the  Egyptian  architects  ; 
but  in  the  Greek  peripteral  temples  the  external  colonnades 
are  always  made  up  exclusively  of  circular  columns.  The  Greek 
architect  hardly  ever  made  use  of  the  square  pier,  except 
in  the  form  of  a  pilaster,  to  give  strength  to  the  extremities 
of  a  wall. 

Would  it  not  be  much  simpler  to  admit  that  we  have  here  one 
of  those  coincidences  which  are  so  frequent  in  the  history  of  the 
arts  ?  Human  nature  is  pretty  much  the  same  all  over  the  world. 
When  human  skill  has  been  employed  at  different  times  and  in 
different  countries,  in  supplying  similar  wants  and  solving  almost 
identical  problems,  it  has  been  led  to  results  which  vary  only  in 
the  minor  details.  These  variations  are  more  or  less  marked 
according  to  race  characteristics  or  material  surroundings.  When 
examined  closely  the  circumstances  of  mankind  are  never  found 
unchanged  from  one  period  or  one  race  to  another,  but  a  superficial 
resemblance  is  enoucrh  to  ensure  that  their  artistic  creations  shall 
have    many   important   points   in  common.       In    no  pursuit    does 


400  A  History  of  Art  in  Ancient  Egypt. 

the  human  mind  turn  in  a  narrower  circle  than  in  architecture.  The 
purpose  of  the  building  on  the  one  hand,  and  the  qualities  of  the 
material  on  the  other,  exercise  a  great  influence  upon  form.  But 
the  purposes  for  which  important  buildings  are  erected  are  very 
few,  neither  are  the  materials  at  the  command  of  the  architect 
very  many.  The  possible  combinations  are  therefore  far  from 
numerous.  Take  two  races  placed  in  conditions  of  climate  and 
civilization  which  may  fairly  be  called  analogous  ;  put  the  same 
materials  in  the  hands  of  their  architects  and  jjive  them  the  same 
programme  to  carry  out  ;  is  it  not  almost  certain  that  they  would 
produce  works  with  many  features  in  common,  and  that  without 
any  knowledge  of  each  other's  work  ?  From  this  point  of  view 
only,  as  it  seems  to  us,  should  the  type  of  building  just  described 
be  regarded.  If  the  temple  at  Elephantine  had  possessed  no  other 
interest  but  that  belonging  to  it  as  an  example  of  Egyptian 
temple  building,  we  might  have  omitted  all  mention  of  it,  or  at 
least  devoted  but  a  few  words  to  it.  And  yet  such  types  are 
scarce.  The  French  explorers  found  a  second  temple  of  the  same 
class  not  far  from  the  first  ;  now,  however,  it  exists  only  in 
their  drawings.^  A  third  has  been  discovered  In  Nubia,  which 
must  resemble  the  two  at  Elephantine  very  strongly  ;  we  mean 
the  temple  constructed  by  Thothmes  III.  on  the  left  bank  of  the 
river,  at  Semneh.  Although  It  has  suffered  greatly,  traces  of  a 
portico  are  to  be  found  about  the  cella,  and  it  has  been  ascertained 
that  this  portico  consisted  both  of  square  piers  and  columns.^ 
Finally,  at  El-kab  (Elllthya),  In  Upper  Egypt,  there  is  a  temple 
constructed  upon  the  same  plan  ;  it  differs  from  that  at  Elephantine 
in  having  only  two  circular  columns,  those  upon  the  facade  ;  all  the 
rest  of  the  peristyle  consists  of  square  piers. '^  The  oldest  part  of 
the  temple  built  by  Thothmes  II.  and  Thothmes  III.  at  Medinet- 
Abou  presents  an  analogous  arrangement.  The  sanctuary  is  there 
surrounded  on  three  sides  by  a  portico  of  square  piers  (Fig.  222). 

There  is  nothing  to  forbid  the  supposition  that  these  temples 
were  once  much  more  numerous  in  the  valley  of  the  Nile,  but  it 

^  la  tlie  Description  de  I'Egypie  it  is  called  T/ie  NortJicrn  Temple  (see  vol.  i.  pi.  38, 
Figs.  2  and  3).  The  only  difference  noted  by  Jomard  was  in  the  ornamentation 
of  the  capitals. 

^  Lepsius  Denkmceler^  part  i.  pi.  113. 

■^  Description,  Antiqiiites,  vol.  i.  pi.  71,  Figs,  i,  2,  3,  4 ;  letterpress,  vol.  i.  ch.  vi. 
This  temple  is  50  feet  long,  31  wide,  and  15  feet  8  inches  high. 


The  Temple  under  the  New  Empire. 


401 


appears  certain  that  they  were  always  of  small  dimensions.  If  like 
those  of  Sais  and  Memphis,  the  temples  of  Thebes  had  vanished 
and  left  no  trace  behind,  we  might  have  been  led  to  believe  that 
some  of  the  great  religious  buildings  of  the  Egyptians  had  been 
in  this  form  ;  but  we  have  Luxor  and  Karnak,  Medinet-Abou  and 
the  Ramesseum,  Gournah  and  Abydos  ;  we  have  several  important 
temples  built  in  Ethiopia  by  Egyptian  conquerors,  and  others  erected 
by  the  Ethiopian  sovereigns  in  imitation  of  Egyptian  architecture.^ 
When  we  compare  these  remains  with  one  another  and  call  to 
mind  the  words  of  Strabo  and  of  other  ancient  travellers  as  to 
the  monuments  which  have  been  destroyed,  we  are  forced  to 
this  general  conclusion,  that  it  was 
within  the  high  external  walls  of  their 
buildings,  around  courts  open  to  the 
sky  or  as  supports  for  wide  and  lofty 
halls,  that  the  Egyptians  loved  to 
group  their  mighty  piers  and  columns. 
When  the  portico  was  outside  it  was 
so  placed  because  there  was  no  room 
for  it  within.  When  the  temple  was 
reduced  to  a  single  narrow  chamber, 
so  small  that  there  was  no  room 
for  columns  and  that  the  walls  could 
support  the  roof  without  help,  the 
colonnade  was  relegfated  to  the  ex- 
terior,  where  it  served  to  give  im- 
portance to  the  cella,  and  to  clothe 
and  beautify  it. 

The  peripteral  arrangement,  which  is  a  constant  principle  in 
Greek  architecture,  is  no  more  than  a  rare  accident  in  that  of 
Egypt.  But  in  spite  of  this  difference  the  similarity,  which  might 
be  called  a  chance  likeness,  if  the  word  chance  had  any  place  in 
history,  is  full  of  interest  for  the  historian  of  art. 

The  following  facts  are  sufficient  to  prove  that  it  was  the  small 
size  of  these  peripteral  temples  that  first  suggested  the  external 
situation  of  their  colonnades.  As  lone  as  the  cella  was  laro-e 
enough  to  admit  supports  of  the  ordinary  diameter  without  encum- 
bering the  space  or  destroying  its  proportions,  we  find  the  columns 


Fig.  232. — Temple  of  Amenophis  III. 
at  Eilithyia  ;  from  Lepsius. 


^  See  Lepsius  for  plans  of  these  buildings;  Denkma^ler.  part  i.  plates  125,  127, 
and  128. 


VOL.    I. 


3    F 


402  A   History  of  Art  in  Ancient  Egypt. 


inside.  Of  this  the  temple  of  Amenophis  III.  at  Eilithyla,  a 
plan  and  section  of  which  we  take  from  Lepsius  (Figs.  232  and 
233),^  is  an  instance.  It  is  prefaced  by  a  chamber,  very  ruinous, 
and  wider  than  it  is  deep.  It  is  now  difficult  to  say  whether  this 
was  an  uncovered  court  or  a  hypostyle  hall.^  Immediately  abut- 
ting upon  it  comes  the  naos,  a  rectangular  chamber  measuring 
internally  28  feet  by  22  feet  6  inches.  The  roof  might  very  pos- 
sibly have  been  supported  by  the  four  columns,  as  their  bases 
were  4  feet  in  diameter.  A  niche  contrived  in  the  further  wall 
of  the  naos  acted  the  part  of  a  secos. 

Here  too  we  find  a  very  simple  form  of  temple,  but  the  naos 
being  large  enough  to  admit,  and  even  to  demand,  the  use  of 
internal  columns,  it  never  entered  the  architect's  head  to  surround 
it  with  a  portico  externally.  Thus  arranged,  the  chapel,  as  we 
have  called  these  buildings,  was  nothing  more  than  an  epitome  of 
the  temple,  and  there  is  no  need  for  insistance  upon  the  variations 
which  it  presents  upon  a  single  theme,  upon  a  first  principle  which 
sometimes  was  developed  into  a  colossal  structure  like  that  at 
Karnak,  sometimes  reduced  until  it  resulted  in  buildings  where  a 
few  paces  carry  the  visitor  from  one  extremity  to  the  other. 

We  may  say  the  same  of  those  subterranean  temples  which 
are  called  speos  or  hemi-speos,  grotto,  or  half -grot  to,  according  to 
whether  they  are  entirely  rock  cut,  or  prefaced  by  architectural 
constructions.  They  are  chiefly  found  in  Lower  Nubia,  a  fact 
which  has  sometimes  been  explained  by  the  natural  configuration 
of  the  soil.  Tn  that  portion  of  the  Nile  Valley  the  river  is 
embraced  so  closely  by  the  rocks  between  which  it  flows  that  it 
would,  we  are  told,  have  been  difficult  to  find  a  site  for  a  con- 
structed temple.  In  this,  however,  there  is  some  exaggeration. 
If  we  examine  a  map  of  Nubia  we  shall  find  many  places  where 
either  one  or  the  other  of  the  two  chains  of  hills  fall  back  from 
the  river  far  enouQrh  to  allow  a  considerable  intervening  fringe  of 
level  ground.  This  is  cropped  and  tilled  by  little  groups  of 
natives,  who  live,  as  a  rule,  at  the  mouth  of  those  wadis,  or  dry 
torrent    beds,   which    intersect    the    mountains.      These   strips   of 

^  Denkmcele?;  part  i.  pi.  100. 

^  The  internal  measurements  of  this  chamber  were  26  feet  by  33.  Lepsius 
gives  it  four  columns,  but  at  present  there  are  only  the  remains  of  one  to  be  found. 
Almost  the  same  arrangements  are  to  be  found  in  the  Temple  oi Sedei/?ga.  (Lepsius, 
De?ik!!ja'Ier,  part  i.  pi.  115.) 


The  Temple  under  the  New  Emph-ie. 


40: 


arable  land  are  always  either  level  or  of  a  very  gentle  slope.  It 
would,  therefore,  not  be  very  difficult  to  obtain  a  site  for  such  little 
oratories  as  were  required  for  the  scanty  population,  for  the  soldiers 
in  the  nearest  military  post,  for  the  engineers  and  workmen  in  some 
neighbouring  quarry.  Even  supposing  that  it  pleased  the  king  to 
choose  some  deserted  site  in  a  conquered  province  for  the  erection 
of  some  durable  memorial  of  his  prowess,  no  very  large  building 
would  be  required.  Great  temples  were  reserved  for  populous 
cities,  in  which  the  king,  the  military  commanders,  and  the 
priests  resided,  in  which  the  popular  ceremonies  of  religion  were 
performed. 

The  Egyptian  architect  did  not  hesitate  to  cut  away  part  of  the 
side  of  a  mountain  when  it  was  the  only  means  open  to  him  of 
obtaining  a  level  site  for  building.  In  this  fashion  Seti  obtained 
a  site  for  his  great  temple  at  Abydos.  The  same  thing  might 
have  been  done,  at  much  less  cost,  for  these  little  Nubian  temples. 


'X. 


Fig.  233. — Temple  of  Ameaophis  IH.  at  Eilithyia  ;  longitudinal  section,  fiom  Lepsius. 

It  would  always  have  been  easy  with  pick  and  chisel  to  adapt 
some  ridge  or  cornice  of  the  cliffs  for  their  reception,  or  to  cut  a 
sort  of  courtyard  in  the  slope  of  the  hill,  in  which  a  small  temple 
miofht  have  been  erected.  We  must  not  seek,  then,  for  a  reason 
for  the  multiplication  of  these  rock  temples  in  the  Nubian  section 
of  the  Nile  Valley  either  in  natural  conditions  or  in  the  want 
of  architectural  resource.  Even  in  Egypt  proper  there  are 
chapels  cut  in  the  flanks  of  the  hills  ;  near  Beni- Hassan  there  is 
the  Speos  Ai^teinidos,  and  near  Assouan,  close  to  the  quarries  of 
Gebel  Silsilis,^  there  is  another.  Below  the  first  cataract,  how- 
ever, these  grottos  are  as  rare  as  they  are  numerous  on  the 
other  side  of  the  frontier,  where,  indeed,  they  sometimes  rise  to 
a  magnificence  of  which  nothing  else  in  Egypt,  unless  it  be  the 
finest  of  the  sepulchral  excavations  at  Thebes,  can  give  an  idea. 
How  are  we  to  account  for  this  difference,  or  rather  contrast  ? 

^  See,  for  Gebel  Silsilis,  Lepsius,  Denkmcckr,  part  i.  pi.  102. 


404  A  History  of  Art  in  Ancient  Egypt. 

This  question  is  more  easily  asked  than  answered.  The 
following  explanation  seems  to  us,  however,  the  most  probable. 

Ethiopia  was  not  Egypt.  Although  they  were  closely  connected 
as  early  as  the  sixth  dynasty,  the  former  never  lost  its  character 
of  a  conquered  province.  In  Ethiopia  men  did  not  feel  so  sure 
of  the  morrow  as  in  Egypt  proper.  Between  the  sixth  and  the 
eleventh  dynasty  the  hold  of  Egypt  upon  Ethiopia  had  been  lost 
at  least  once.  Reconquered  by  the  kings  of  the  first  Theban 
period,  it  regained  its  independence  during  the  domination  of 
the  Hyksos  ;  the  eighteenth  dynasty  had,  therefore,  to  begin  the 
work  of  subjugation .  all  over  again,  and  it  did  its  work  more 
thoroughly  than  any  of  its  predecessors.  Then,  when  the 
Egyptian  sceptre  ruled  as  far  south  as  Napata  and  the  great 
bend  of  the  Nile,  the  governors  of  the  southern  provinces  must 
have  been  continually  employed  in  repelling  the  incursions  of  the 
negroes  from  Upper  Ethiopia,  and  in  suppressing  the  warlike 
tribes  who  lived  within  the  conquered  frontier.  At  such  times 
the  king  himself  must  often  have  been  compelled  to  take  the  field 
and  lead  his  armies  in  person.  A  constructed  temple,  especially 
when  of  small  size,  would  be  in  great  risk  of  destruction  in  a 
country  exposed  to  the  repeated  incursions  of  savage  tribes  ; 
columns  and  piers  would  soon  be  overturned  by  their  ruthless 
arms.  But  chambers  cut  in  the  living  rock  would  offer  a  much 
stouter  resistance ;  the  decorations  might  be  scraped  down  or 
daubed  over,  but  the  time  and  patience  required  for  any  serious 
attack  upon  the  limestone  or  granite  sides  and  piers  would  not  be 
forthcominof.  Such  damaw  as  could  be  done  in  a  short  time  and 
by  the  weapons  of  the  invaders  could  readily  be  repaired  when 
the  raid  was  over. 

We  think  it  probable,  therefore,  that  subterranean  architecture 
was  preferred  throughout  this  region  because  the  political  con- 
dition of  the  province  was  always  more  or  less  precarious,  rather 
than  because  the  configuration  of  the  country  required  it.  Where 
security  was  assured  by  the  presence  of  a  strong  and  permanent 
garrison,  as  at  Semneh  and  Kumneh,  we  find  constructed  temples 
just  as  we  do  in  Egypt.  They  are  found,  too,  in  those  localities 
— Soleb  and  Napata  for  instance — where  there  was  a  large  urban 
population,  and  therefore  fortifications  and  troops  for  their  defence. 
Everywhere  else  it  was  found  more  convenient  to  confide  the  temple 
to  the  guardianship  of  its  own  materials,  the  living  rock,  and  to  bury 


The  Temple  under  the  New  Empire.  405 


it  in  faces  of  the  cliffs.  This  kind  of  work,  moreover,  was  per- 
fectly easy  to  Egyptian  workmen.  For  many  centuries  they  had 
been  accustomed,  as  we  have  seen,  to  hollow  out  the  flanks  of 
their  mountains,  and  to  decorate  the  chambers  thus  obtained,  for 
the  last  resting-places  of  their  dead.  In  the  execution  of  such 
works  they  must  have  arrived  at  a  degree  of  practised  skill  which 
made  it  as  easy  for  them  to  cut  a  speos  like  the  great  temple  at 
Ipsamboul,  as  to  build  one  of  the  same  size.  This  fact  probably 
had  its  weight  in  leading  the  conquerors  of  Nubia  to  fill  it  with 
underground  temples.  Such  a  method  of  construction  was  at 
once  expeditious  and  durable,  a  double  advantage,  which  would 
be  greatly  appreciated  in  the  early  years  of  the  occupation  of 
the  province.  When  security  was  established,  the  same  process 
continued  to  be  used  from  love  for  the  art  itself.  When 
Rameses  II.  cut  those  two  caves  in  the  rock  at  Ipsamboul,  whose 
facades,  with  their  gigantic  figures,  have  such  an  effect  upon  the 
travellers  of  to-day,  it  was  neither  because  he  was  pressed  for 
time,  nor  because  he  was  doubtful  of  the  tenure  of  his  power. 
The  military  supremacy  of  Egypt  and  the  security  of  her 
conquests  seemed  to  be  assured.  The  Egyptian  monarch  carved 
the  cliffs  of  Ipsamboul  into  gigantic  images  of  himself  because  he 
wished  to  astonish  his  contemporaries  and  their  posterity  with  the 
boldness  and  novelty  of  the  enterprise.  At  Thebes  he  had  built, 
on  the  right  hand  of  the  river,  the  hall  of  Karnak  and  the  pylons 
of  Luxor  ;  on  the  left  bank,  the  Temple  of  Seti  and  the  Ramesseum. 
For  these  he  could  have  imagined  no  pendant  more  original  or 
more  imposing  than  the  great  temple  carved  from  a  natural  hill, 
in  front  of  which  statues  of  the  sovereign,  higher  than  any  of 
those  which  adorned  the  courtyards  at  Thebes,  would  see  countless 
generations  of  Egyptians  pass  before  their  feet  in  their  journeys 
up  and  down  the  Nile.  The  hypostyle  hall  at  Karnak  was  a 
marvel  of  constructed  architecture,  the  great  temple  at  Ipsamboul 
was  the  masterpiece  of  that  art  which  had  been  so  popular  with 
the  Egyptians  from  the  earliest  periods  of  their  civilization,  the  art 
which  imitated  the  forms  of  a  stone  building  by  excavations  in  the 
living  rock. 

Subterranean  architecture  had,  of  course,  to  go  through  a 
regular  course  of  development  before  it  was  capable  of  such 
works  as  the  tomb  of  Seti,  at  Thebes,  and  the  .temples  of  Ipsam- 
boul.     In  the   necropolis  of   Memphis,   and   in   that  of   the    First 


4o6 


A  History  of  Art  in  Ancient  Egypt. 


Theban  Empire,  its  ambition  was  more  easily  satisfied.  So,  too, 
the  first  rock-cut  temples  were  of  very  modest  dimensions,  They 
date  from  the  eighteenth  dynasty.  Two  of  them  are  to  be  found 
in  the  neighbourhood  of  Ipsamboul  but  on  the  other  side  of  the 
river,  one  near  the  castle  of  Addeh,  the  other  at  Feraig.  The 
latter  was  cut  by  the  king  Harmhabi  (or  Armais).  It  is  composed 
— as  also  is  that  of  Addeh — ^of  a  hall  supported  by  four  columns, 
two  lateral  chambers,  and  a  sanctuary.  There  is  an  equally  small 
speos  in  Egypt  which  dates  from  the  same  period  ;  it  is  the  grotto 
at  Beni-Hassan,  which,  ever  since  antique  times,  has  been  known 
as  the  Speos  Artei7iidos.  The  goddess  Sekhet,  to  which  it  was 
consecrated,  had  been  identified  with  the  Greek  Artemis.  It  was 
begun  by  Thothmes  III.,  carried  on  by  Seti  I.,  and  seems  never 
to  have  been  finished.     The  temple  proper  is  prefaced  by  a  kind 


Fig.  234. — The  speos  at  Addeh  ;         Fig.  235. — The  speis  at  Addeh.     Longitudmal 
plan  from  Horeau.  section  ;  from  Horeau. 


of  portico  of  square  pillars  cut,  with  the  roof  which  they  support, 
from  the  limestone  rock.  A  narrow  passage  about  nine  feet  deep 
leads  to  the  naos,  which  is  a  quadrangular  chamber  about  thirteen 
feet  square,  with  a  niche  in  the  further  wall  in  which  an  image  of 
the  lion-headed  goddess  probably  stood. ^  The  most  important  of 
the  rock-cut  chapels  of  Silsilis  was  also  inaugurated  by  Harmhabi 
and  restored  and  embellished  by  Rameses  II.-  The  hemispeos  at 
Redesieh,  in  the  same  district,  is  a  work  of  Seti  I.'^ 

Only  one  subterranean  temple  later  than  the  nineteenth  dynasty 

^  Description  de  I' Egypte,  Antiquites,  vol.  iv.  pi.  65,  Fig,  i.  The  French  draughts- 
men thought  this  building  was  a  disused  quarry,  and  give  nothing  but  a  picturesque 
view  of  the  fa9ade. 

2  Lepsius,  Deiikmaler,  part  i.  pi.  102  ;  Rosellim  (vol.  iii.  pi.  32,  Fig.  3)  gives  a 
view  of  the  interior  of  t4ie  Silsilis. chapel. 

2  Lepsius,  Deiihnceler,  part  i.  pi.  loi. 


The  Temple  under  the  New  Empire. 


40; 


wm^ 
M 


^M 


p 


S 


<^pSi5^0|| 


is  known  to  us,  namely,  that  which  is  cut  in  the  flanks  of  the 
Gebel-Barkal  at  Napata.^  It  is  called  the  Typhonium,  on  account 
of  the  erimacinor  fio-ures  which  stand  before 
the  piers.  It  dates  from  the  time  of  Tahrak  , 
and  was  one  of  the  works  with  which  the 
famous  Ethiopian  decorated  his  capital  in  the 
hope  that  it  might  become  a  formidable  rival 
to  those  great  Egyptian  cities  which  he 
had  taken  and  occupied.-  All  the  other  rock- 
cut  temples  were  the  work  of  Rameses  II.  ; 
they  are,  as  we  ascend  the  Nile,  Beit-el-Wali, 
near  Kalabcheh  (Figs.  236  and  237);  Gherf- 
Hossein,  or  Gircheh,  Wadi-Seboua,  Dayr, 
and  Ipsamboul, 

We  may  give   Gherf-H ossein  as   a    good 
example  of   the   hemlspeos  (Figs.    238   and 
239).      It    was   approached   from    the    river   by   a  broad  flight  of 
steps,    decorated    with    statues    and     sphinxes,    of    which    but    a 


m 


fm 


^~^^^^s^ 


Fig.  235. — Plan  of  speos  at 
Beit-el-Wali  ;  from  Prisse. 


-I — I — i- 


-I 1 1 1 \- 


FiG.  237. — Longituciinal  section  of  the  speos  at  Beit-el-Wali  ;  from  Prisse. 

few  fragments  now  remain.  A  pylon  gave  access  to  a  rect- 
angular court,  on  the  right  and  left  sides  of  which  stood  five 
piers  faced  with  colossal  statues  of  Rameses   II.     These  statues 


^  Lepsius,  Denkviceler,  part  i.  pi.  127. 

^  There  are  also  ahemispeos  or  two  of  the  Ptolemaic  period.  That,  for  instance, 
of  which  the  plans  are  given  in  plate  loi  of  Lepsius's  first  part,  was  begun  by 
Ptolemy  Eiiergetes  II, 


4o8 


A   History  of  Art  in  Ancient  Egypt, 


^^^^^S^^^^!!^|^ON(^-^^j^^W-> 


uOn  * 


were  about  twenty-six  feet  high.      Next,  and  at  a  slightly  higher 
level,   came  a  hypostyle  hall ;    its  roof  was  supported   by  twelve 

square  piers,  those  forming  the 
central  avenue  being  of  caryatid 
form  and  higher  than  the  others. 
The  subterranean  part  of  the  temple 
begins  with  a  passage  cut  in  the 
rock  on  the  further  side  of  this  hall. 
This  passage  leads  to  a  long  trans- 
verse vestibule,  from  which  open 
two  lateral  chambers,  and  three 
from  its  further  side.  The  furthest 
chamber  on  the  major  axis  of  the 
whole  building  was  the  sanctuary. 
This  is  proved  by  its  position,  its 
shape,  and  the  niche  which  is  cut 
in  its  further  wall.  Four  deities 
are  sculptured  in  this  niche,  and  in 
spite  of  the  ill-usage  to  which  they 
have  been  subjected,  one  of  them 
can  still  be  identified  as  Ptah,  the 
chief  god  of  the  temple.^ 

We  find  almost  the  same 
arrangements  in  the  hemispeos  of 
Wadi-Asseboua.'^  That  of  Derri 
(Figs.  240  and  241)  is  more 
simple.  There  are  neither  dromos 
nor  pylon,  properly  speaking,  and 
only  four  caryatid  pillars  ;  but 
there  is  an  open  court  with  a 
hypostyle  hall  and  a  sanctuary  cut  in  the  rock.     At  the  back  of 

^  This  description  has  been  mainly  taken  from  the  plate  given  by  Prisse  {Histoire 
de  VArt  Egyptie/i,  vol.  i.).  There  are  discrepancies,  however,  between  it  and  both 
the  inscription  of  Isambert  and  the  plan  of  Horeau  {Panorama  d'Egypie  et Nubie), 
discrepancies  which  may  probably  be  referred  to  the  bad  condition  of  the  structural 
part  of  the  building.  According  to  Prisse's  measurements  the  dromos,  from 
its  commencement  to  the  foot  of  the  first  pylon,  was  about  fifty-five  yards  long, 
and  the  rest  of  the  temple,  to  the  back  of  the  niche,  was  about  as  much  again. 
The  rock-cut  part  was  only  about  ten  yards  deep. 

-  The  resemblance  between  Prisse's  plan  of  Gherf-Hossein  and  Horeau's  plan  of 
Wadi-Asseboua  is  so  great  as  to  suggest  that  one  of  the  two  writers  may  have 
made  a  mistake. 


Fig.  238. — Plan  of  the  hemisptos  of 
Gherf-Hcssein  ;  from  Prisse. 


The  Temple  under  the  New  Empire. 


409 


the    sanctuary  there    is   a   stone   bench  upon  which  three  statues 
were  seated. 

The  two  temples  of  Ipsamboul  are  so  well  known  and  have 
been  so  often  illustrated  and  described,  that  they  need  not  detain 
us  long.  The  chief  thing  to  be  noticed  here  is  that  they  are 
without   any  external  and  constructed    part,  and    that  from  their 


Ai 


'    ''-j;!fct,.i?^i.Ky^||^j|j|j|i^ 


A 


'0  ^  10  15  20  — 

Fig.  239. — Gherf-Hossein,  longitudinal  section  ;  from  Prisse. 


position,  high  above  the  river  and  close  to  it,  it  was  impossible 
that  they  could  have  any  dromos  ;  and  yet  between  the  doorway 
of  the  speos  and  the  river  bank  there  were  steps  which  are  now 
either  worn  away  by  the  action  of  the  floods  or  hidden  by  the 
ddbris  from  the  cliffs.  The  facades  of  these  temples  were,  however, 
as  richly  decorated  and  as  monumental  in  their  way  as  those  of 
the  most  sumptuous  buildings  in  Thebes. 


ElG.  240. — Plan  of  the  hemispeos 
of  Derri ;  from  Horeau. 


Fig.    241. — Longitudinal   section,    Derri; 
from  Horeau. 


The  prototype  of  these  facades  is  the  Theban  pylon.  They 
have  the  same  trapeziform  surfaces  covered  with  figures  and 
inscriptions,  circumscribed  by  a  moulding  and  crowned  by  a 
cornice  in  bold  relief;  they  arc  inclined  from  the  perpendicular, 
and    they    afford     a    background     to     the    statues     of    the    king 


VOL.    I. 


4IO  A   History  of  Art  in  Ancient  Egypt. 


who  caused  them  to  be  made.  The  chief  difference  is  in  the 
situation  of  these  statues.  In  the  case  of  a  built  temple  they  are 
monoliths,  brought  from  a  distance  and  erected  in  front  of  the 
pylon.  But  space  was  wanting  for  such  an  arrangement  at 
Ipsamboul  ;  besides  which  it  was  better,  for  many  reasons,  that 
the  whole  edifice  should  be  homogeneous,  and  that  the  statues 
should  be  carved  in  the  rock  from  which  its  chambers  were  to 
be  cut.  The  way  to  do  this  was  obvious.  The  colossi  had 
but  to  recede  a  pace  or  two  so  as  to  be  incorporated  in  the 
substance  of  the  pylon  itself. 

At  Ipsamboul  there  are,  as  we  have  seen,  two  temples  close  to 
one  another.  Their  fa9ades,  though  conceived  in  the  same  spirit, 
executed  by  the  same  processes,  and  having  a  good  deal  in  common 
in  their  design,  are  yet  by  no  means  similar.  That  of  the  temple  of 
Hathor,  generally  called  the  Smaller  Temple,  is  on  a  smaller  scale 
than  the  Gi^eat  Temple,  but  perhaps  its  design  is  the  happier  and 
more  skilful  of  the  two.  The  front  is  90  feet  wide  and  nearly 
40  high.  It  is  ornamented  by  six  colossal  upright  statues,  four  of 
them  Rameses  himself,  the  other  two  his  wife  Nefert-Ari.  These 
statues,  which  are  about  34  feet  high,  are  separated  one  from 
another  by  eight  buttresses,  two  of  them  acting  as  jambs  for  the 
door,  above  which  they  unite  and  become  a  wide  band  of  flat 
carving  marking  the  centre  of  the  facade.  The  gentle  salience  of 
these  buttresses  forms  a  framework  for  the  statues  (see  Fig.  242), 
which  are  chiselled  with  great  care  and  skill  in  the  fine  yellow 
sandstone  of  which  the  mountain  consists. 

The  fa9ade  of  the  Great  Temple  is  much  larger.  It  is  about 
130  feet  wide  by  92  high.  It  is  not  divided  by  buttresses  like  the 
other,  but  it  has  a  bold  cornice  made  up  of  twenty- two  cynocephalic 
figures  seated  with  their  hands  upon  their  knees.  Each  of  these 
animals  is  sculptured  in  the  round,  and  is  only  connected  with  the 
face  of  the  rock  by  a  small  part  of  its  posterior  surface.  They 
are  not  less  than  seven  feet  high.  A  frieze,  consisting  of  a 
dedicatory  inscription  carved  in  deep  and  firmly  drawn  hiero- 
glyphs runs  below  the  cornice.  Above  the  doorway  a  colossal 
figure  of  Ra  is  carved  in  the  rock,  and  on  each  side  of  him 
Rameses  is  depicted  in  low  relief,  in  the  act  of  adoration.  This 
group  occupies  the  middle  of  the  facade.  But  the  most  striking 
feature  of  the  building  is  supplied  by  the  four  colossi  of 
Rameses   placed  two  and  two  on   either  side  of  the  door.     They 


3 
O 


'a, 

a 


s 

(A 


I      ~ 


7 


The  Temple  under  the  New  Emph^e. 


413 


are  the  largest  in  Egypt.  From  the  sole  of  the  feet  to  the  apex  of 
the  pschent  which  the  king  bears  on  his  head,  they  are  about  sixty- 
five  feet  in  height.  Rameses  is  seated,  his  hands  upon  his  thighs, 
in  the  pose  ordinarily  made  use  of  for  the  royal  statues  at  the 
entrances  of  the  temples.      In  spite  of  these  enormous  dimensions 


10  )5  40  ' 


Fig.  243. — Plan  of  the  smaller 
temple. 


Fig.   244. — Perspective  of  the  principal  chamber  in  the 
smaller  temple  ;  from   Horeau. 


the  workmanship  is  very  fine.  The  countenance,  especially,  is 
remarkable  for  its  combination  of  force  and  sweetness,  an 
expression  which  has  been  noticed  by  all  the  travellers  who 
have   written  upon   Ipsamboul. 

The  interiors  of  the  two  temples  are  still  more  different  than  the 


-SiN^iT^;^^■^SS^■^^T:^->^"^^'^^Nf^,  X 


Fig.  245. — Longitudinal  section  of  the  smaller  temple  ; 
from  Horeau. 


!  *    IV   ; — 

j   i>:     It    ' 


1 


^1 


Fig.   246. — Plan  of  the 
Great  Temple. 


exteriors,  and,  in  this  instance,  the  variations  are  entirely  in  favour 
of  the  greater  monument.  The  total  depth  of  the  smaller  edifice 
is  about  ninety  feet.  A  single  hall,  supported  by  six  square 
Hathor-headed  pillars,  precedes  the  sanctuary.  The  latter  is 
nothing   but   a   narrow   gallery,    in   the    middle  of   which   a   small 


414 


A  History  of  Art  in   Ancient  Egypt. 


chamber  or  niche  is  cut,  in  which  the  rock-carved  cow  of  Hathor 
may^  be  seen  with  a  statue  between  its  legs.  The  other  temple 
is  a  orreat  deal  laro-er.  Its  total  length  is  about  i8o  feet.  The 
first  hall  is  60  feet  long  and  53  wide  ;  the  roof  is  supported  by 
eight  pillars,  against  each  of  which  a  colossal  figure  33  feet  high 
is  placed.  A  doorway  in  the  middle  of  the  further  side  leads  to  a 
second  chamber  not  quite  so  large  as  the  first,  and  supported  by 
four  thick  square  pillars.      Three  openings  in  its  furthest  side  lead 


Fig.  247. — Perspective  of  the  principal  hall  in  the  Great  Temple  ;  from  Horeau. 

to  a  third  chamber,  as  wide  as  the  second,  but  only  lo  feet  deep. 
Through  this  the  innermost  parts  of  the  speos  are  reached  ;  they 
consist  of  three  small  chambers,  those  on  the  left  and  right  being 
very  small  indeed,  while  that  in  the  centre,  the  adytum,  is  about 
13  feet  by  23.  In  the  middle  of  this  chamber  was  an  altar,  or 
table  for  offerings;  at  the  back  of  it  a  bench  with  four  seated 
statues.  The  walls  of  both  temples  are  covered  with  pictures  like 
those   of  Luxor,   Karnak,   and  the  Ramesseum.      They  represent 


The  Temple  under  the  New  Empire. 


417 


the  battles  and  triumphs  of  Rameses,  and  the  king  seated  upon 
the  laps  of  goddesses,  who  act  as  the  tenderest  of  nurses. 

Besides  the  halls  which  form  the  main  body  of  the  temple,  the 
plan  shows  eight  lateral  chambers,  some  perpendicular  to  the 
major  axis  of  the  building,  others  falling  upon  it  obliquely.  Several 
of  these  do  not  seem  to  have  been  finished.  There  are  indi- 
cations that  they  were  utilized  as  depositories  for  the  objects 
worshipped  in  the  temple. 

We  have  now  briefly  noticed  the  principal  rock-cut  temples  in 
Egypt  and  Nubia.  Neither  in  plan  nor  in  decoration  do  they 
materially  differ  from  the  temples  of  wrought  masonry.  The 
elements  of  the  building  are  the  same,  and  they  are  arranged  in 
the  same  order — an  avenue  of  sphinxes  when  there  is  room  for 
it,  colossi  before  the  entrance,  a  colonnaded  court,  a  hypostyle  hall 


Fig.  249. — Longitudinal  section  of  the  Great  Temple  ;  from  Iloreau. 


acting  as  a  pronaos,  a  naos  with  its  secos,  or  sanctuary  ;  but  some- 
times one,  sometimes  many  of  these  divisions  are  excavated  in  the 
living  rock.  Sometimes  only  the  sanctuary  is  subterranean,  some- 
times the  hypostyle  hall  is  included,  and  at  Ipsamboul  the  whole 
temple  is  in  the  mountain,  from  the  secos  to  those  colossal  statues 
which  generally  form  the  preface  to  the  pylon  of  the  constructed 
temple. 

Except  in  the  case  of  the  peristylar  court,  the  interior  of  the 
rock-cut  temple  did  not  differ  so  much  in  appearance  from  that  of 
the  constructed  edifice  as  might  at  first  be  imagined.  We  have 
already  explained  how  scantily  lighted  was  the  interior  of  the 
Egyptian  temple  ;  its  innermost  chambers  were  plunged  in  almost 
complete  darkness,  so  that  the  absolute  night  which  was  involved 
in  their  being  excavated  in  the  heart  of  a  mountain  was  no  very 


VOL.    L 


II 


Ai8  A  History  of  Art  in  Ancient  Egypt. 


great  change  from  the  obscurity  caused  by  the  thick  walls  and 
heavy  roofs  of  the  edifices  in  the  plain.  In  the  case  of  a  hemi- 
speos  the  internal  effect  must  have  been  almost  identical  with  that 
of  any  other  religious  building.  In  the  great  temple  of  Ipsam- 
boul  the  daylight  does  not  penetrate  beyond  the  second  hall  ;  from 
that  point  onwards  artificial  light  is  necessary  to  distinguish 
objects,  but  the  Egyptians  were  so  thoroughly  accustomed  to  a 
mysterious  solemnity  of  shadow,  to  a  "  dim  religious  light,"  in  their 
temples,  that  the  darkness  of  the  speos  would  seem  no  drawback 
in  their  eyes. 

The  column  occurs  very  seldom  in  these  subterranean  temples.^ 
Even  those  chambers  which  correspond  to  the  hypostyle  hall  by 
their  places  in  the  excavation  and  the  general  characteristics  of  their 
form,  are  hardly  ever  supported  by  anything  but  the  rectangular 
piers  in  use  in  the  early  ages  of  the  monarchy  ;  but  these  piers  are 
often  clothed  with  an  elaborate  decoration  which  is  unknown  in 
the  works  of  the  primitive  architects.  This  preference  for  the  pier 
is  easily  to  be  explained  by  the  necessity  for  having  supports  of 
sufficient  strength  and  solidity  to  bear  the  weight  of  the  super- 
incumbent mountain. 

Another  and  more  constant  peculiarity  of  the  underground 
temples,  is  the  existence  in  them  of  one  or  more  seated  statues 
carved  from  masses  of  rock  expressly  left  in  the  furthest  recesses 
of  the  excavation.  These  statues,  which  represent  the  presiding 
deity  of  the  place  and  his  acolytes,  do  not  occur  in  the  constructed 
temples.  In  the  latter  the  tabernacle  which  stood  in  the  secos 
was  too  small  to  hold  anything  larger  than  a  statuette  or  emblem. 
We  think  that  the  cause  of  this  difference  may  be  guessed.  At 
the  time  these  rock  temples  were  cut,  the  Pharaohs  to  whom  they 
owed  their  existence  no  doubt  assigned  a  priest  or  priests  to  each. 
But  their  position,  sometimes  in  desert  solitudes,  as  in  the  case 
of  the  Speos  Artemidos,  sometimes  in  places  only  inhabited  for  an 
intermittent  period,  in  the  quarries  at  Silsilis  for  instance,  or  in 
provinces  which  had  been  conquered  by  Egypt  and  might  be 
lost  to  her  again,  rendered  it  impossible  that  they  could  be  served 
and  guarded  in  the  ample  fashion  which  was  easy  enough  in 
the  temples  of  Memphis,  Abydos  and  Thebes.  All  these 
considerations     suggested    that,    instead    of    a    shrine    containing 

^  There  are  two  polygonal  columns  resembling  those  at  Beni-Hassan  in  the  small 
speos  at  Beit-el-Wali  (Fig.  237). 


S'PBO.f  PF.  I.OuBST 


/ 


0  J    to        so         Jo        -fo 

I I 1 I I l_ 


so' 


Fig.  250. — Dayr-el-Bahari  ;  according  to  M.  liiuii'. 


The  Temple  under  the  New   Empire.  421 

some  small  figure  or  emblem,  statues  of  a  considerable  size,  from 
six  to  eight  or  ten  feet  high,  should  be  employed,  and  that  they 
should  be  actually  chiselled  in  the  living  rock  itself  and  left 
attached  to  it  by  the  whole  of  their  posterior  surfaces.  By  their 
size  and  by  their  incorporation  with  the  rock  out  of  which  both 
they  and  their  surroundings  were  cut,  such  statues  would  defend 
themselves  efficiently  against  all  attempts  on  the  part  of  enemies. 
In  spite  of  their  age  several  of  these  statues  came  down  to  us  in 
a  sufficiently  good  state  of  preservation  to  allow  Champollion  and 
his  predecessors  to  recognize  with  certainty  the  divine  personages 
whom  they  represented.  During  the  last  fifty  years  they  have 
suffered  as  much  at  the  hands  of  ignorant  and  stupid  tourists  as 
they  did  in  the  whole  of  the  many  centuries  during  which  they 
were  exposed  to  all  the  vicissitudes  of  Egyptian  history.^ 

Our  study  of  the  Egyptian  temple  would  not  be  complete 
without  a  few  words  upon  the  buildings  called  Dayr-el-BahariP' 
By  their  extent,  their  picturesqueness,  and  the  peculiar  nature  of 
their  situation,  these  ruins  have  always  had  a  great  effect  upon 
foreign  visitors.  Those  who  know  Thebes  will,  perhaps,  be 
surprised  at  our  having  said  so  little  about  them  hitherto, 
especially  as  they  are  older  than  most  of  the  buildings  over 
which  we  have  been  occupied.  We  have  not  yet  described  them 
because  they  do  not  belong  to  any  of  the  categories  which  we 
have  been  treating  ;  they  form  a  class  by  themselves  ;  their  general 
arrangement  has  no  parallel  in  Egypt,  and  therefore  we  have 
reserved  them   to  the  last. 

The  building  in  question  is  situated  at  the  foot  of  the  Libyan 
chain,  in  a  deep  amphitheatre  hollowed  out  by  nature  in  the 
yellow  limestone  rocks  which  rise  on  the  north-west  of  the 
necropolis.  On  two  sides,  on  the  right  and  at  the  back,  it  rests 
against  perpendicular  walls  of  rock  cut  by  the  pickaxe  and 
dominating  over  the  built  part  of  the  temple.  On  the  left  this 
natural  wall  is  absent  and  is  replaced  by  an  inclosure  of  bricks 
(Figs.  250  and  251). 

^  For  Beit-el-Wali  and  Girc/ie/i,  see  plates  13,  30  and  31  in  Gau,  Antiquites  de  la 
Ntibie.  It  seems  that  the  statues,  when  they  were  drawn  by  him,  were  in  a  fairly 
good  state. 

2  These  words  mean  Convent  of  the  North.  The  name  is  derived  from  an 
abandoned  Coptic  convent  which  existed  among  the  ruins  of  the  ancient  building. 


42  2  A   History  of  Art  in  Ancient  Egyi'T. 

Under  such  conditions  we  need  feel  no  surprise  at  finding  part 
of  the  temple  subterranean.  In  backing"  his  work  against  the 
mountains  in  this  fashion  the  architect  must  haYe  been  partly 
impelled  by  a  desire  to  make  use  of  the  facilities  which  it  afforded. 
The  mausoleum  of  Hatasu,  unlike  the  other  funerary  chapels  at 
Thebes,  is,  then,  a  triple  hemispeos.  At  a  point  immediately 
opposite  to  the  door  in  the  external  pylon,  but  at  the  other 
extremity  of  the  building,  a  chamber  about  sixty-five  feet  deep 
was  excavated  in  the  rock.  This  must  have  acted  the  part  of  a 
sanctuary.  Right  and  left  of  it,  and  at  a  shorter  distance  from  the 
entrance,  there  are  two  more  groups  of  rock-cut  apartments.  The 
whole  arrangement  may "  be  compared  to  the  system  of  three 
apsidal  chapels  which  is  so  common  at  the  east  end  of  European 
cathedrals. 

In  approaching  this  temple  from  the  river  bank,  a  dromos  of 
sphinxes  had  to  be  traversed  of  which  very  scanty  traces  are  now 
to  be  found,  but  in  the  time  of  the  Institut  d' Egypt e  there  were 
still  two  hundred  of  them  to  be  distinguished,  a  few  of  the  last 
being  shown  in  the  restoration  figured  upon  the  opposite  page 
(Fig.  251).  At  the  end  of  the  dromos,  upon  the  spot  where  a 
few  traces  of  the  bounding  walls  still  remain,  we  have  placed  a 
pylon  with  a  couple  of  obelisks  in  front  of  it.  We  have  done  so 
not  only  because  nearly  all  the  important  temples  had  such  a 
preface,  but  also  because  Sir  Gardner  Wilkinson  says  that  he 
saw  the  foundations  of  two  obelisks  and  of  a  doorway.  After 
passing  the  pylon,  a  first  courtyard  was  entered,  which 
communicated  with  a  second  by  an  inclined  plane  stretching 
almost  across  its  width. ^  Here  the  arrangements  which  con- 
stituted the  real  originality  of  Dayr-el-Bahari  began.  The  whole 
interior  of  the  temple,  betw^een  the  pylon  and  the  commencement 
of  the  speos,  consisted  of  four  courtyards,  rising  in  terraces  one 
above  another  like  the  steps  of  a  gigantic  staircase.  The  walls 
upon  which  these  inclined  planes  and  terraces  were  constructed 
are  still  to  be  traced  in  places.  In  order  to  furnish  the  vast 
courts,  we  have  supposed  them  to  contain  seated  statues  at  regular 
intervals  alonsf  the  inner   faces  of  their  walls  ;  in  such  matters  of 

^  This  wide  inclined  plane  agrees  better,  as  it  seems  to  us,  with  the  indications 
in  M.  Brune's  plan  of  the  actual  remains  at  Dayr-el-Bahari,  than  the  narrow  flight 
of  steps  given  in  his  restoration ;  the  effect,  too,  is  better,  more  ample  and 
majestic. 


1       i       :i      i  .lit.  ''  /    W^^^B^^-^^:^  .:-^^ 


Fig.  251.— RcFtoratinn  in  persrective  of  Dayr-^l-Bahai  i,  by  Ch.  Chipioz. 


The  Temple  under  the  New  Empire.  425 


decorative  detail  a  little  conjecture  may  perhaps  be  allowed.^ 
As  for  the  portico  which  ornamented  the  further  side  of  the 
second  court,  its  remains  were  visible  even  before  the  excavations 
of  Mariette.- 

Those  excavations  have  since  1858  led  to  the  discovery  of 
the  porticos  of  the  third  court.  There  seems  to  have  been 
only  a  plain  wall  on  the  left  of  this  court,  while  on  the  right  there 
was  a  long  colonnade  which  masked  a  number  of  chambers  cut 
in  the  rock  which  rose  immediately  behind  it.  Facing  the 
entrance  to  the  court  there  was  also  a  colonnade  which  was 
cut  in  two  by  the  steps  leading  to  the  fourth  and  highest  terrace. 
In  the  middle  of  this  terrace  a  fine  doorway  leading  to  the 
principal  speos  was  raised.  While  all  the  rest  of  the  temple  was 
of  limestone,  this  doorway  was  built  of  fine  red  granite,  a  distinc- 
tion which  is  to  be  explained  by  its  central  situation,  facing  the 
gateway  in  the  pylon  though  far  above  it,  and  forming  the 
culminating  point  of  the  long  succession  of  terraces  and  inclined 
planes.  The  attention  of  the  visitor  to  the  temple  would  be 
instantly  seized  by  the  beauty  and  commanding  position  of  this 
doorway,  which,  moreover,  by  its  broad  and  mysterious  shadows, 
suggested  the  secos  hidden  in  the  flanks  of  the  mountains,  to 
which  all  the  courts  were  but  the  prelude. 

These  terraced  courts  have  surprised  all  visitors  to  the  cenotaph 
of  Hatasu.  "  No  one  will  deny,"  says  Mariette,  "  that  the  temple 
of  Dayr-el-Bahari  is  a  strange  construction,  and  that  it  resembles 
an  Egyptian  temple  as  little  as  possible !"  ^     Some  have  thought 

^  The  same  idea  caused  M.  Brune  to  place  sphinxes  upon  the  steps  between  the 
courts  ;  he  thought  that  some  small  heaps  of  debris  at  the  ends  of  the  steps  indicated 
their  situation ;  but  M.  Maspero,  who  recently  investigated  the  matter,  informs  us 
that  he  found  no  trace  of  any  such  sphinxes. 

-  We  must  refer  those  who  wish  to  study  the  remains  of  this  temple  in  detail  to 
the  \vork  devoted  to  it  by  M.  Maiiette.  The  plan  which  forms  plate  i  in  the  said 
w^ork  was  draw^n,  in  1866,  by  an  architect,  M.  Brune,  who  is  now  a  professor  at  the 
Ecole  des  Beaux  Arts.  M.  Brune  succeeded,  by  intelligent  and  conscientious 
examination  of  all  the  remains,  in  obtaining  the  materials  for  a  restoration  which 
gave  us  for  the  first  time  some  idea  of  what  this  interesting  monument  must  have 
been  in  the  great  days  of  Egypt.  Plate  2  contains  a  restored  plan  ;  plate  3  a 
view  in  perspective  of  the  three  highest  terraces  and  of  the  hill  which  forms  their 
support.  We  have  attempted  to  give  an  idea  of  the  building  as  a  whole.  Our  view- 
is  taken  from  a  more  distant  point  than  that  of  M.  Brune,  but  except  in  some  of 
the  less  important  details,  it  does  not  greatly  differ  from  his. 

^  Mariette,  Dayr-el-Bahari^  letterpress,  p.  10. 

VOL.    L  31 


426  A   History  of  Art  in  Ancient  Egypt. 

foreiofn  influence  was  to  be  traced  in  its  arranorements.  "  Are 
we  to  consider  it  an  accident,  asks  Ebers,  that  the  stepped 
building  at  Dayr-el-Bahari  was  built  shortly  after  an  Egyptian 
army  had,  under  Thothmes,  trodden  the  soil  of  Mesopotamia 
for  the  first  time,  and  found  monumental  buildings  constructed 
in  terraces  in  its  great  cities  ?  Why  did  the  Egyptians,  who 
as  a  rule  were  so  fond  of  repeating  themselves  that  they 
became  almost  incapable  of  inventing  new  forms,  never  imitate 
the  arrangements  of  this  imposing  building  elsewhere,  unless  it 
was  because  its  forms  reminded  them  of  their  foreign  enemies 
and  therefore  seemed  to  be  worthy  of  condemnation  ?  "  ^ 

We  are  content  with  asking  the  question  and  with  calling 
attention  to  its  interest.  The  materials  are  wanting-  for  a  definite 
answer  but  the  suggestion  of  Professor  Ebers  is  probable  enough. 
Twelve  or  thirteen  centuries  later  the  Persians,  after  their 
conquest  of  Egypt,  carried  back  with  them  the  notion  of  those 
hypostyle  halls  which  gave  to  the  buildings  of  Persepolis  so 
different  an  aspect  from  those  of  Assyria,  although  the  decorative 
details  were  all  borrowed  from  the  latter  country.  So  too  the 
Egyptians,  in  spite  of  the  pride  which  they  felt  in  their  ancient 
civilization,  may  have  been  unable  to  control  their  admiration 
when  they  found  themselves,  in  the  wide  plains  of  Persia,  before 
those  lofty  towers  with  their  successive  terraces,  to  which  access 
was  obtained  by  majestic  flights  of  steps.  It  seems  by  no  means 
unlikely  that  one  of  their  architects  should  have  attempted  to 
acclimatize  an  artistic  conception  which  was  so  well  calculated 
to  impress  the  imaginations  of  the  people ;  and  none  of  the 
sovereigns  of  Egypt  was  better  fitted  to  preside  over  such  an 
attempt  than  the  high  spirited  and  enterprising  Hatasu,  the 
queen  who  reared  two  obelisks  in  the  temple  of  Karnak,  one  of 
them  beinsf  the  highest  that  has  remained  erect ;  who  made  the 
first  recorded  attempt  at  acclimatization  ;  -  and  who  was  the  first 
to  launch  a  fleet  upon  the  waters  of  the  Red  Sea. 

^  Ebers,  ^gypten,  p.  285. 

^  Maspkro,  Histoire  A/icienne,  pp.  202,  203.  The  bas-reliefs  at  Dayr-el-Bahari 
represent  the  booty  brought  back  by  Hatasu  from  the  expedition  into  Fount.  Among 
this  booty  thirty-two  perfume  shrubs,  in  baskets,  may  be  distinguished  ;  these  shrubs 
were  planted  by  the  orders  of  Hatasu  in  the  gardens  of  Thebes.  On  the  subject 
of  Hatasu  and  her  expedition,  see  Maspero's  paper  entitled :  De  quelques 
Navigations  des  Egyptietis  sur  les  Cotes  de  la  Mer  Erythree  (in  the  Reinie  Historique, 
1878). 


The  Temple  under  the  New  Emph-ie.  427 

Whether  Hatasu's  architect  was  Inspired  by  those  artistic  crea- 
tions of  the  Chaldees  which,  as  time  went  on,  were  miiltiphed  over 
the  whole  basin  of  the  Euphrates  and  even  spread  as  far  as  northern 
Syria,  or  whether  he  drew  his  ideas  entirely  from  his  own  brain, 
his  work  was,  in  either  case,  deserving  of  high  praise.  In  most 
parts  of  the  Nile  Valley  sites  are  to  be  found  which  lend  them- 
selves readily  to  such  a  building.  The  soil  has  a  gentle  slope, 
upon  which  the  erection  of  successive  terraces  would  involve  no 
architectural  difficulties,  and  there  is  no  lack  of  rocky  walls  against 
which  porticoes  could  be  erected,  and  in  which  subterranean 
chambers  could  be  excavated.  Upon  a  series  of  wide  platforms 
and  easy  gradients  like  these,  the  pompous  processions,  which 
played  such  an  important  part  in  the  Egyptian  ritual,  could  defile 
with  great  effect,  while  under  every  portico  and  upon  every 
landing  place  they  could  find  resting  places  and  the  necessary 
shelter  from  the  sun.  Why  did  such  a  model  find  no  imitators  ? 
Must  we  seek  for  the  reason  in  the  apparent  reaction  against  her 
memory  which  followed  the  death  of  Hatasu  ?  "  The  Egyptian 
people  chose  to  look  upon  her  as  an  usurper  ;  they  defaced  the 
inscriptions  which  celebrated  her  campaigns  ;  they  effaced  her 
cartouches  and  replaced  her  titles  with  those  of  her  brothers."  ^ 

It  is  certain  that  nowhere  In  Egypt  has  any  building  of  con- 
siderable dimensions  been  discovered  in  which  the  peculiar 
arrano;ements  of  Dayr-el-Bahari  are  repeated.  At  most  it  may 
be  said  that  somethinof  of  the  same  kind  Is  to  be  found  in  those 
rock-cut  temples  of  Nubia  which  are  connected  with  the  river 
bank  by  a  dromos  and  flights  of  steps.  When  the  princes  of  the 
nineteenth  dynasty  wished  to  raise  funerary  temples  to  their 
memory  in  their  own  capital,  it  would  have  been  easy,  had  they 
chosen,  to  find  sites  upon  the  slopes  of  the  western  chain  similar 
to  that  which  Hatasu  had  employed  with  such  happy  results  ;  but 
they  preferred  a  different  combination.  They  erected  their  ceno- 
taphs in  the  plain,  at  som.e  distance  from  the  hills,  and  they  chose 
a  form  which  did  not  essentially  differ  from  that  of  the  great 
temples  on  the  opposite  bank' of  the  Nile. 

The  religious  architecture  of  Egypt,  in  all  Its  richness  and 
variety,  is  known  to  us  only  through  the  monuments  of  the 
second  Theban  Empire,  through  the  great  works  of  the  kings 
belonging   to   the  eighteenth  and  nineteenth  dynasties.      We  are 

^  Maspero,  Histoire  Ancicnue,  p.  203. 


428  A  History  of  Art  in  Ancient  Egypt. 


tempted,  however,  to  believe  that  the  architects -of  the  Sait  period 
must  have  introduced  fresh  beauties  into  the  plans,  proportions, 
and  decorations  of  those  temples  which  the  princes  of  the  twenty- 
sixth  dynasty,  in  their  desire  that  their  capital  and  the  other  cities 
of  the  Delta  should  rival  or  excel  the  magnificence  of  Memphis 
and  Thebes,  confided  to  their  skill.  Both  the  statues  and  the 
royal  tombs  of  the  Sait  period  have  characteristics  which  dis- 
tinguish them  from  those  of  earlier  epochs.  In  all  that  we 
possess  from  this  last  period  of  artistic  activity  in  Egypt,  there 
is  a  new  desire  for  elegance,  for  grace,  carried  sometimes  to  an 
extreme  which  is  not  free  from  weakness  and  affectation.  It  is 
probable  that  the  same  qualities  existed  in  the  religious  architecture 
of  Sais. 

Unhappily  all  the  buildings  constructed  in  Memphis  and  Lower 
Egypt  during  the  Sait  supremacy  have  disappeared  leaving 
hardly  a  trace  behind,  and  the  Greek  writers  have  left  us  nothing 
but  vague  accounts  to  supply  their  place.  Herodotus  goes  into 
ecstasies  over  the  propylsea,  that  is,  the  pylons  and  outer  courts, 
which  Amasis  added  to  the  temple  of  Neith  at  Sais,  and  over  the 
enormous  size  of  the  stones  employed.  He  describes  in  great 
detail  a  chapel  carved  out  of  a  single  block  of  Syene  granite,, 
which  Amasis  transported  from  the  quarries  at  great  cost  in  order 
that  it  might  be  erected  in  the  sanctuary  of  the  said  temple  ; 
unhappily  it  was  so  much  injured' on  the  journey  that  his  intention 
had  to  be  abandoned.^ 

All  that  we  learn  from  the  historian  is  that  the  Sait  princes 
made  use  of  colossal  stones  in  theij*  buildings  without  much 
regard  to  their  appropriateness,  but  simply  to  impress  their  con- 
temporaries with  an  exaggerated  idea  of  their  wealth  and  power. 
The  contractors  of  an  earlier  agfe  were  also  in  the  habit  of 
employing  blocks  which  seem  astonishing  to  us  from  their  length 
and  size,  but  they  were  never  used  except  when  they  were 
required,  to  cover  a  void  or  some  other  purpose  ;  the  earlier 
architects  never  made  the  mistake  of  seeking  for  difiiculties 
merely  to  show  how  cleverly  they  could  overcome  them. 

It  is  to  be  reofretted  that  we  know  so  little  of  the  monument 
attributed  by  Herodotus  to  Psemethek,  and  described  by  him  in 
the   followincr  terms  : — "  Having-  become   master  of  the  whole  of 

^  Herodotus,  ii,  175. 


The  Temple  under  the  New  E:\ipire.  429 

Egypt,  Psammitichos  constructed  those  propylaea  of  the  temple  of 
Hephaistos  which  h'e  to  the  south  of  that  building.  In  front  of 
these  propylaea  he  also  caused  to  be  constructed  an  edifice  in 
which  Apis  was  nourished  as  soon  as  he  had  manifested  himself. 
It  was  a  peristyle  ornamented  with  figures.  Colossal  statues, 
twelve  cubits  high,  were  employed  as  supports,  instead  of  columns."^ 
We  may  assume  that  these  colossi  were,  as  in  other  Egyptian 
buildings,  placed  immediately  in  front  of  the  real  supports,  and  did 
not  themselves  uphold  an  entablature.  Herodotus  was  not  an 
architect,  and,  in  taking  account  merely  of  the  general  effect, 
he  doubtless  used  an  expression  which  is  not  quite  accurate. 

The  most  important  point  to  be  noticed  in  this  short  extract 
from  the  Greek  historian  is  the  hint  it  contains  of  the  attempts 
at  originality  made  by  the  later  generations  of  Egyptians,  by 
"  men  born  too  late  in  too  old  a  century,"  and  of  the  means  by 
which  they  hoped  to  rival  their  predecessors.  The  architect  of 
Psemethek  borrowed  a  motive  which  had  long  been  disused, 
of  which,  however,  there  are  many  examples  at  Thebes,  and 
employed  it  under  novel  conditions. 

The  caryatid  form  of  pier  is  generally  found,  in  temples,  in  the 
peristyles  of  the  fore-courts  or  the  hypostyles  of  the  pronaos. 
Psemethek  made  use  of  it  for  the  decoration  of  what  was  no 
more  than  a  cattle  stable.^  The  stable  in  question  had,  it  must 
be  confessed,  a  god  for  its  inhabitant,  and  so  far  it  might  be  called 
a  "temple;  but  it  was  a  temple  of  a  very  peculiar  kind,  in  which 
the  arrangements  must  have  been  very  different  from  those 
required  in  the  abode  of  an  inanimate  deity.  In  it  the  god 
was  present  in  flesh  and  blood,  and  special  arrangements  were 
necessary  in  order  to  provide  for  his  wants,  and  to  exhibit  him  to 
the  crowd  or  conceal  him,  as  the  ritual  demanded.  The  problem 
was  solved,  apparently,  in  a  method  satisfactory  to  the  Egyptians, 
as  the  guide  who  attended  Herodotus  called  his  attention  to 
the  building  with  an  insistance  which  led  the  historian  to  pay  it 
special  attention. 

Herodotus  does  not  tell  us  what  form  the  caryatides  took  in  this 
instance.  It  is  unlikely  that  they  were  Osiride  figures  of  the  king, 
as   in   the  Theban   temples,    but   as  Apis  was   the   incarnation   of 

^  Herodotus,  ii.  153. 

-  Herodotus  uses  the  word   ai'Av/,  of  which  stable   or  cattle-shed  was  one   of  the 
primitive  meanings. 


430  A  History  of  Art  in  Ancient  Egypt. 


Ptah,  the  great  deity  of  Memphis,  they  may  very  possibly  have 
been  carved  in  the  imacre  of  that  a^od. 

Between  the  days  of  Cambyses  and  those  of  Alexander,  Egypt 
temporarily  recovered  her  independence  more  than  once.  The 
art  of  that  period — during  which  numerous  works  were  carried 
out  and  many  others  restored — was  a  prolongation  of  the  art  of 
the  Sait  princes.  Its  aims,  methods,  and  taste  were  entirely 
similar.  We  may,  therefore,  in  spite  of  the  limits  which  we  have 
imposed  upon  ourselves,  mention  a  work  carried  out  no  more  than 
fifty  years  before  the  Greek  conquest,  in  the  reign  of  Nectanebo  I. 
We  mean  the  small  buildinij  which  is  sometimes  called  the  southern 
temple,  in  the  island  of  Philai.  It  is  the  oldest  building  upon  the 
island,  all  the  rest  being  Ptolemaic  or  Roman. 

Its  arrangements  are  different  to  anything  we  have  hitherto 
encountered  in  relicfious  architecture.  There  are  no  internal  sub- 
divisions  of  any  kind,  nothing  which  resembles  a  secos.  According 
to  all  the  plans  which  have  been  published.  It  contained  only  one 
hall,  or  rather  rectangular  court,  Inclosed  by  fourteen  graceful 
columns  and  a  low,  richly-decorated  wall,  which  forms  a  kind  of 
screen  between  the  lower  part  of  the  columns.  This  screen  does 
not  extend  quite  half-way  up  the  columns  ;  these  latter  support  an 
entablature,  but  there  has  never  been  a  roof  of  any  kind.  There 
can  be  no  doubt  that  the  building  was  consecrated  to  I  sis,  whose 
Image  is  carved  all  over  it ;  but  could  an  edifice  thus  open  to  the 
outward  air  and  to  every  prying  eye  be  a  temple  ?  Ebers  is 
disposed  to  look  upon  it  as  a  waiting-room.^  Close  to  it  the 
remains  of  a  wide  staircase  are  to  be  traced,  against  which  boats 
were  moored,  and  upon  which  they  discharged  their  loads.  Thus 
the  faithful  who  came  to  be  present  at  the  rites  of  I  sis  would 
assemble  in  the  waiting-hall,  whence  they  would  be  conducted  by 
the  priests  to  that  sanctuary  which  became  the  object  of  so 
many  pilgrimages  In  the  later  years  of  the  Egyptian  monarchy. 

Certain  peculiarities  in  the  management  of  the  column,  which 
grew  Into  frequent  use  in  the  Ptolemaic  epoch,  are  here  encoun- 
tered for  the  first  time.  This  is  not  the  place  for  Its  detailed 
consideration,  but  one  must  point  it  out  as  a  second  result  of  the 
desire  shown  by  the  architects  of  the  period  to  achieve  new 
developments  without  breaking  the  continuity  of  the  national 
traditions.       Here,  as  in  the  monumental   cattle-shed  at  Memphis, 

^   Egypic.  etc.  ]i.  406. 


X 

3 
O 


c 
o 


a, 


g 
"3 

u 
<u 


-N 


The  Temple  under  the  New  Empire.  433 

there  is  no  invention  of  new  forms  ;  all  the  architectural  elements 
introduced  are  to  be  found  in  earlier  buildings.  It  is  the  general 
aspect  and  physiognomy  of  the  building  that  is  new.  Whatever 
we  may  call  it,  the  edifice  erected  by  Nectanebo  at  the  southern 
point  of  the  island  is  certainly  novel  in  form  ;  we  have  found 
nothing  like  it  either  in  Egypt  or  in  Nubia,  but  the  repetition  of  its 
forms  in  a  much  later  generation  proves  that  it  answered  to  a 
real  change  in  the  national  taste  and  to  new  aspirations  in  the 
national  genius.  Painting,  engraving,  and  photography  have 
given  us  countless  reproductions  of  the  picturesque  building 
which  rises  on  the  eastern  shore  of  the  island,  amid  a  bouquet 
of  palm-trees.  It  has  been  variously  called  the  bed  of  Pharaoh, 
the  eastern  temple,  the  great  hypcsthra,  the  stimmer-ho2tse  of 
Tiber ms,  &c.  It  is  nothing  more  than  a  replica  of  Nectanebo's 
creation  ;  it  is  larger  and  its  proportions  are  more  lofty,  but  its 
plan  is  quite  similar.^  In  the  sketch  lent  to  us  by  M.  Hector 
Leroux,  the  eastern  temple  is  seen  on  the  right,  while  the  left 
of  the  drawing  is  filled  up  with  the  pylons  of  the  great  temple 
of  Isis  (Fig.  252). 

If  we  knew  it  better,  we  should  probably  find  that  the  archi- 
tecture of  the  Sait  period  formed  the  transition  between  that  of 
the  second  Theban  empire  and  that  of  the  Ptolemies.  We  should 
find  in  it  at  least  hints  and  foreshadowings  of  those  original 
features  of  which  we  shall  have  to  speak  when  we  arrive  at  the 
Grseco- Egyptian  temples.  Unhappily,  as  none  of  the  temples 
built  by  Psemethek,  Amasis,  and  their  successors  have  been 
recovered  from  the  sands  of  Egypt,  we  shall  be  reduced  to 
conjecture  on  this  point.  But  must  all  hope  of  recovering  some- 
thinof  from  the  ruins  of  Sais  be  abandoned  ?  Mariette  himself 
made  some  excavations  upon  its  site,  and  confessed  that  he  was 
discouraged  by  their  result,  or  rather  by  their  want  of  result. 
Perhaps,  however,  deeper  and  more  prolonged  excavations  might 
bring  to  light  sufficient  indications  of  the  ordonnance  and  plans 
of  the  more  important  buildings  to  permit  of  some  attempt  at 
restoration  beine  made.^ 

'  The  temple  of  Ke?-dasc/i  or  Gartasse  in  Nubia  resembles  the  Eastern  Temple  at 
Philoe  in  plan  ;  its  date  appears  to  be  unknown. 

^  We  have  omitted  to  speak  of  those  little  temples  known  since  the  time  of 
Champollion  as  luammisi  or  places  for  accouchement,  because  the  existing  examples 
all  belong  to  the  Ptolemaic  period.  The  best  preserved  is  that  of  T^enderah.  It  is 
VOL.    L  3    K 


434  ^  History  of  Art  in  Ancient  Egypt. 


§  4.      General  Chai'acteristics  of  the  Egyptian  Temple. 

We  have  now  conducted  our  history  of  the  Egyptian  temple 
from  the  most  ancient  monument  to  which  that  title  can  be  given 
to  the  period  when  Greek  art,  introduced  into  the  country  by  the 
Macedonian  conquest,  began  to  have  an  influence  upon  many  of 
the  important  details,  if  not  upon  the  general  aspects  of  the 
national  architecture.  The  reader  will  not  be  surprised  to  find 
that  before  we  conclude  our  study  we  wish  to  give  a  res7t7nd  of  the 
leading  ideas  which  seem  to  be  embodied  in  the  temple,  and  to 
define  the  latter  as  we  see  it  in  its  finest  and  most  complete 
expression,  in  the  buildings  of  the  great  Theban  Pharaohs.  We 
cannot  do  better  for  our  purpose  than  borrow  the  words  of 
Mariette  upon  the  subject.  No  one  has  become  more  thoroughly 
acquainted  with  the  temples  of  the  Nile  valley.  He  visited  them 
all  at  his  leisure,  he  explored  their  ruins  and  sounded  most  of  them 
down  to  their  foundations,  and  he  published  circumstantial  descrip- 
tions of  Abydos,  Karnak,  Dayr-el-Bahari,  and  Denderah.  In 
these  monographs  and  in  the  ItinSrau'-e  de  la  Haitte-Egypte,  he 
returned  to  his  definition  again  and  again,  in  a  continual  attempt 
to  improve  it,  to  make  it  clear  and  precise.  We  shall  freely 
extract  from  his  pages  all  those  expressions  which  seem  to  us  to 
give  the  best  rendering  of  their  author's  ideas,  and  to  bring  out 
most  clearly  the  originality  which  belongs  to  the  monuments 
of   Avhich  he  treats.^ 

"  The    Egyptian    temple    must   not   be    confused    with  that    of 
Greece,   with    the    Christian    church,   or   with    the    Mohammedan 

probable,  however,  that  the  custom  of  building  these  little  edifices  by  the  side  of 
those  great  temples  where  a  triad  of  gods  was  worshipped  dated  back  as  far  as  the 
Pharaonic  period.  The  mammisi  symbolised  the  celestial  dwelling  in  which  the 
goddess  gave  birth  to  the  third  person  of  the  triad.  The  authors  of  the  Description 
called  them  Typ/wnia,  from  the  effigy  of  a  grimacing  deity  which  figures  in  their 
decoration.  This  deity  has,  however,  nothing  in  common  with  Set-Typhon,  the 
enemy  of  Osiris.  We  now  know  that  his  name  was  Bes,  that  he  was  imported  into 
Egypt  from  the  country  of  the  Aromati,  and  that  he  presided  over  the  toilette  of 
women.     (Ebers,  LEgypte^  etc.,  p.  255.) 

^  Mariette,  Itincraire,  pp.  13-16,  157-159;  Karnak,  p.  19;  Voyage  dans  la 
Haute-Egypte^  vol.  i.  pp.  15,  16. 


General  Characteristics  of  the  Egyptian  Temple.    435 

mosque.  It  was  not  a  place  for  the  meeting  of  the  faithful,  for 
the  recital  of  common  prayers ;  no  public  ritual  was  celebrated 
within  it ;  no  one  was  admitted  to  it  except  the  priests  and  the 
king.  The  temple  was  a  kind  of  royal  oratory,  a  monument 
reared  by  the  king  in  token  of  his  own  piety,  and  in  order  to 
purchase  the  favour  of  the  gods. 

"  The  elaborate  decoration  with  which  all  the  walls  of  the 
temples  are  covered  is  only  to  be  explained  by  admitting  this 
point  of  departure.  The  essential  element  of  this  decoration  is 
the  picture  ;  many  pictures  are  arranged  symmetrically  side  by 
side,  and  tiers  above  tiers  of  pictures  cover  the  walls  from  floor  to 
ceiling.  This  arrangement  never  varies,  and  the  same  may  be 
said  of  the  general  significance  of  the  pictures  :  on  the  one  hand, 
the  king,  on  the  other,  one  or  more  deities  ;  these  are  the  subjects 
of  all  the  compositions.  The  king  makes  an  offering  (meats,  fruits, 
flowers,  emblems)  to  the  god  and  asks  for  some  favour  at  his 
hands  ;  in  his  answer  the  deity  grants  the  favour  demanded.  The 
whole  decoration  of  a  temple  consisted  therefore  in  an  act  of 
adoration  on  the  part  of  the  monarch  repeated  in  various  forms. 
The  temple  was  therefore  the  exclusive  personal  monument  of  the 
prince  by  whom  it  was  founded  and  decorated.  This  fact  explains 
the  presence  of  those  precious  representations  of  battles  which 
adorn  the  external  w^alls  of  certain  temples.^  The  king  ascribed 
all  his  successes  in  the  field  to  the  immediate  protection  of  the 
gods.  In  combating  the  enemies  of  Egypt,  in  bringing  them  by 
thousands  to  the  capital,  in  employing  them  upon  the  construction 
of  their  temples,  he  was  performing  an  act  as  agreeable  to  the 
gods  as  when  offering  incense,  flowers,  and  the  limbs  of  the 
animals  sacrificed.  By  such  deeds  he  proved  his  piety  and 
merited  the  continuation  of  those  favours  for  which  the  erection 
of  a  temple  was  meant  to  be  an  acknowledgment." 

The  piety  and  gratitude  of  the  monarch  also  found  expression 
in  the  splendour  of  the  great  festivals  of  which  the  temple  was  the 
scene  several  times  in  the  course  of  the  year.  "  The  ceremonies 
consisted  for  the  most  part  of  great  processions,  issuing  from  the 
sanctuary  to  be  marshalled  in  the  hypostyle  hall,  and  afterwards 
traversing    the    open    courts  which  lay  between    the   buildings  of 

^  The  canal  figured  in  ficnt  of  the  Charict  of  Rameses,  in  Fig.  254  was,  according 
to  Ebers,  the  oldest  of  the  Suez  Canals,  the  one  dug  by  Seti  I.  This  canal  was 
defended  by  fortifications,  and  is  called  in  inscriptions  the  Cutting  (JO Egypte^  etc.). 


436 


A  History  of  Art  in  Ancient  Egypt. 


the  temple  and  the  great  wall  which  incloses  the  whole.  They 
perambulated  the  terraced  roofs,  they  launched  upon  the  lake  the 
sacred  barque  with  its  many-coloured  streamers.  Upon  a  few 
rare  occasions  the  priests,  with  the  sacred  images,  sallied  from  the 
inclosure  which  ordinarily  shielded  their  rites  from  profane  eyes, 
and,  at  the  head  of  a  brilliant  flotilla,  directed  their  course  to  some 
other  city,  either  by  the  Nile  or  by  the  waterway  which  they 
called  '  the  sacred  canal.'  "  ^ 

"  The  ensigns  of  the  gods,  the  coffers  in  which  their  effigies  or 
symbolic  representations  were  inclosed,  their  shrines  and  sacred 
barques  were  carried  in  these  processions,  of  which  the  kings  were 
the   reputed  conductors.      At  other  times  all  these  objects   were 


r     ^      Sc 


^^.2k ±  ^.A ^     ^  I  ->^       -k      ^      ^-       f^      ^r     -i^     -^     -^i  -A      -^      4c    i< 


-k      -K     ^      ~k       -A^~|  k       -^      ic     -k      -k     ie -k       A       >c      -^       y       -Jj     i< 


Fig.  253. — The  battle  against  the  Khetas,  Luxor,     (From  Champollion,  pi.  328.) 


deposited  in  the  naos.  Upon  the  occurrence  of  a  festival,  the 
priest  to  whom  the  duty  was  delegated  by  the  king  entered  the 
naos  and  brought  out  the  mysterious  emblem  which  was  hidden 
frorn  all  other  eyes  ;  he  covered  it  with  a  rich  veil,  and  it  was 
then  carried  under  a  canopy." 

A  ritual  to  which  so  much  "  pomp  and  circumstance "  was 
attached  required  material  appliances  on  a  great  scale.  The 
preservation  of  so  much  apparatus  required  extensive  store-rooms, 
which,  like  the  sanctuary  itself,  had  to  be  kept  in  almost  total 
darkness   in   order  to  preserve    the    sacred  vestments    and    other 

^  To  follow  these  processions  was  an  act  of  piety.  Upon  a  Theban  stele  we  find 
the  following  words  addressed  to  Amen-Ra  :  "  I  am  one  of  those  who  follow  thee 
when  thou  goest  abroad."  The  stele  of  Suti  and  Har,  architects  at  Thebes, 
translated  into  French  by  Paul  Pierret,  in  Reaieil  de  Travaiix,  p.  72. 


ON 

»— ( 

a 

n 


o 

a, 

u 

2 


c 


s 

H 

o 


o 
c 
1) 


o 


a 


X 


en 


S 

3 


B 


Pi 


General  Characteristics  of  the  Egyptian  Temple.    439 

objects  from  the  deteriorating  effects  of  sun,  dust,  heat,  and  the 
insects  which  they  engender.  There  is  nothing  in  the  texts  which 
seems  to  hint  at  the  celebration  of  any  rites  in  the  dark  parts  of 
the  temple  by  artificial  light,  and  no  trace  of  the  discoloration 
caused  by  smoke  has  been  found  upon  the  walls.  There  seems  to 
have  been  no  necessity  for  anything  beyond  the  most  subdued 
daylight  within  the  chambers  of  the  temple.  All  the  important 
part  of  the  ritual  was  performed  in  the  open  air,  and  the  few 
liturgical  acts  in  the  naos  were  short  and  took  place  before  a 
very  restricted  audience.  They  consisted  of  a  few  prayers  said  by 
the  king  or  by  the  chief  priest,  and  in  the  presentation  of  the 
traditional  offerings.  The  cares  of  maintenance  and  of  preparation 
for  the  periodical  festivals  had  also  to  go  on  in  that  part  of  the 
temple.  Such  duties,  however,  could  be  readily  discharged  by  the 
practised  and  disciplined  priests  in  the  half  light  of  the  sanctuary, 
and  even  the  almost  total  darkness  of  the  apartments  which  were 
ranged  behind  it. 

"  The  temples  show  no  trace  of  dwelling-places  for  the  priests, 
nor  of  places  for  initiation,  nor  of  any  contrivance  for  divination  or 
the  giving  of  oracles.     There  is  nothing  to  lead  us  to  suppose  that, 
except  the  king  and  the  priests,  any  of  the  public  were  admitted 
into   the  building,"   at   least  beyond  the   hypostyle  hall.      Certain 
privileged   individuals  or  classes  were  admitted  into  the   latter  on 
the  occasion  of  a  festival  ;  others,  less  fortunate,  were  compelled  to 
wait  in  the  courtyards.      It  was  their  right  to  be  the  first  to  see  the 
god  as  he  emerged  from  the  sanctuary  on  the  shoulders  of  the 
priests.      But  in  spite  of  their  vast  dimensions,  these  halls  would 
have  been  ill  fitted   for  the  uses  to  which  the  spacious  naves  of  a 
church  or  mosque  are  put.      The  huge  and  closely  spaced  columns 
would  embarras  the  movements  and  intercept  the  view  of  those 
who  crowded  about  their  bases.    It  was  only  in  the  central  aisle  that 
sufficient  space  was  left  for  the  easy  passage  of  a  procession.     The 
hypostyle  hall  was  lofty  and  wide  in  order  that  it  might  be  a  vesti- 
bule worthy  of  the  god  who  dwelt  in  the  sanctuary  beyond  it,  and 
in  order  that  it  might  bear  witness  by  its  magnificence  to  the  piety, 
wealth,  and  power  of  the   king   who   constructed  it.      It  offered  no 
place  in  which   the   faithful   could  assemble   to  listen   to  religious 
discourses,  to  unite  in  the  expression  of  their  faith  and  hope,  to 
sing  and  pray  in  common. 


440  A   History  of  Art  in  Ancient  Egypt. 

In  virtue  of  the  sanctuary  which  was  its  nucleus,  the  temple  was 
the  dwelling  of  the  god,  the  terrestrial  resting-place  to  which  the 
king,  his  son  and  the  nursling  of  the  goddesses,  came  to  offer  him 
thanks  and  to  do  homage  in  return  for  the  protection  and  support 
which  he  received.  The  temple  was  also,  in  virtue  of  those 
numerous  chambers  which  surrounded  the  sanctuary,  a  place  for 
the  preparation,  consecration,  and  preservation  of  holy  objects  ;  a 
huge  sacristy  to  which  access  was  forbidden  to  all  but  those  who 
were  specially  attached  to  the  service  of  the  god  and  charged  with 
the  custody  of  the  sacred  furniture. 

Such  being  the  origin  and  purpose  of  the  temple,  we  need  feel 
no  surprise  at  the  triple  fortification  behind  which  it  was  en- 
trenched. This  fortification  consisted,  in  the  first  place,  of  the  brick 
wall  which  formed  the  outermost  inclosure  ;  secondly,  of  the  wall 
of  masonry  which  embraced  the  temple  proper,  leaving  a  narrow 
passage  only  wide  enough  for  the  walk  of  a  sentry  ;  thirdly,  of 
the  wall  which  divided  the  really  secret  parts  of  the  building  from 
the  pronaos.  Now  that  the  line  of  the  external  wall  is  only  indi- 
cated by  a  gentle  swell  of  the  ground,  now  that  the  best  preserved 
of  the  inner  walls  are  broken  down  in  many  places,  and  now 
that  all  the  roofs  and  ceilinofs  have  fallen  and  encumbered  the 
floors,  it  is  difficult  enough  to  form  a  true  idea  of  the  former  ap- 
pearance of  the  Egyptian  temples.  Could  we  see  them  as  they 
left  their  architects'  hands,  we  should  be  struck  by  the  jealous 
severity  of  their  isolation,  by  the  austere  monotony  of  the 
screen  of  stone  which  was  interposed  between  the  eyes  of  the 
people  and  the  internal  splendours  of  the  building.  In  this  we 
should  find  the  chief  point  of  distinction  between  the  temples 
of  Egypt  and  those  great  religious  edifices  of  our  own  times 
with  which  we  half  involuntarily  compare  all  other  works  of  the 
kind. 

But  the  Greek  temple  was  no  more  a  church  than  its  Egyptian 
rival.  It  was  not  a  place  of  assembly  for  public  praise  or  religious 
teaching.  Its  cella  was  an  inclosed  chamber,  illuminated  only 
by  the  door  and  by  a  few  openings  contrived  in  the  roof,  and 
reserved  for  the  god  who  inhabited  it.  The  two  architects  in  fact, 
Egyptian  and  Greek,  had  the  same  points  of  departure ;  the 
problems  which  they  had  to  solve  strongly  resembled  each  other, 
and  yet   they  created    types    which    differed    very  greatly.     The 


General  Characteristics  of  the  Egyi'tian  Temple.    441 

Greek  temple  was  not  isolated  and  hidden  behind  a  stone  curtain  ; 
it  could  be  seen  from  all  sides  in  its  commanding  position  ;  its 
encircling  band  of  porticos  seemed  to  invite  all  comers  to  the 
shelter  of  their  galleries,  while  they  charmed  all  eyes  with  the 
play  of  light  and  shadow  afforded  by  their  alternate  voids  and 
solids.  The  colonnades  reserved  by  the  Egyptian  for  the 
decoration  of  courts  and  halls  were  placed  by  the  Greek  upon 
the  external    faces  of  his   temples,  and  although  the  task  of  both 


Fig.  255. — Tlie  godde;s  Anouke  suckling  Rameses  II.,  IJeit-Wali ;  from  Iloreaii. 


architects  was  to  fulfil  almost  identical  requirements,  this  trans- 
position of  the  elements  employed  was  sufficient  to  cause  a 
profound  difference  in  the  outward  expression,  in  the  physiognomy, 
of  their  several  works. 

Another   and   perhaps  still    more  characteristic  diff,i'ence  is   to 
be   found    in   the   fact   that   the    Greek    temple  is  not  susceptible, 


VOL.    T. 


442  A   History  of  Art  in  Ancient  Egypt. 

like  that  of  Egypt,  of  almost  infinite  extension.  Greece  never 
produced  anything  like  Karnak  or  Luxor  ;  even  in  the  centuries 
when  the  taste  for  the  colossal  eclipsed  the  love  for  the  great, 
she  never  dreamed  or  imao-ined  anvthinor  of  the  kind.  The 
Greek  temple  had  the  unity  of  a  living  organism.  Given  the 
main  dimensions,  and  the  elements  of  which  it  was  to  be  composed 
could  only  vary  within  very  narrow  limits.  In  accordance  with 
the  degree  of  luxury  desired,  the  cella  would  be  either 
surrounded  by  a  simple  wall  or  would  be  encircled  by  a  portico, 
but  this  portico  would  only  be  a  kind  of  adornment,  a  vesture 
which  would  be  more  or  less  rich  and  ample  according  to 
circumstances.  Behind  the  long  files  of  columns  on  either  side, 
behind  the  double  or  triple  rows  which  veiled  the  two  facades, 
the  body  of  the  temple  could  always  be  discerned,  just  as  the 
modelling  of  the  human  form  may  be  distinguished  under  the 
drapery  of  a  statue,  in  spite  of  the  folds  which  cover  it.  The  cella 
was  proportioned  to  the  sacred  figure  which  was  to  be  its  in- 
habitant, which,  again,  afforded  a  standard  by  which  the  proportions 
and  subjects  of  the  groups  which  filled  the  pediments,  and  of  the 
bas-reliefs  of  the  frieze,  as  well  as  the  height  of  the  columns  and 
the  projection  of  the  entablature,  were  determined.  Between 
all  these  parts  there  was  an  intimate  and  clearly  defined 
connection. 

When  a  plant  is  seen  bursting  from  the  seed,  we  are  able,  if 
we  know  the  species  to  which  it  belongs,  to  say  beforehand  what 
its  leaves,  its  flower,  and  its  fruit  will  be  like,  and  to  foretell  the 
limits  of  its  height.  It  is  the  same,  to  a  great  extent,  with  the 
Greek  temple.  The  trench  dug  to  receive  the  footing  stones 
of  the  cella  walls  is  the  hole  into  which  the  seed  is  thrown  from 
.  which  the  whole  temple  is  to  spring.  These  walls  rise  above 
the  level  of  the  ground,  the  building  progresses  to  completion, 
but  from  the  day  upon  wdiich  the  seed  was  sown,  from  the  day 
upon  which  the  foundation  was  laid,  the  temple  had  been  virtually 
complete.  Like  an  organic  body,  the  Greek  temple  inclosed 
within  itself  the  principle  of  its  own  growth,  the  law  which 
governed  its  development,  and  forbade  it  in  advance  to  excede 
certain  definite  limits. 

Such  was  not  the  case  with  the   Egyptian  temple.     In  those  of 
small   or  moderate   dimensions   this   unity  and   simplicity  of  plan 


General  Characteristics  of  the  Egyptian  TeiMple.    44, 


exists  to  a  certain  extent.  The  peripteral  temple  of  Elephantine 
and  even  the  temple  of  Khons  may  be  given  as  instances  of  this  ; 
in  them  there  is  much  with  which  the  most  exclusive  philo-Greek 
can  sympathize..  The  impression  received  from  the  ruins  of 
Abydos  or  Gournah,  still  more  from  those  of  Karnak  or  Luxor, 
is  very  different.  There  we  find  several  sanctuaries  closely 
wedged  together,  all  of  the  same  size  and  decorated  in  the  same 
fashion,  in  one  place  the  architect  has  built  seven  in  a  row,  and 
there  was  nothing  to  prevent  him  doubling  the  number  if  he  had 
chosen  to  do  so.  In  another  we  find  a  succession  of  courts,  of 
hypostyle  halls  and  chambers,  of  forests  of  columns.  Sometimes 
it  requires  considerable  search  to  pitch  upon  the  sanctuary,  which, 
again,  is  not  the  loftiest  part  of  the  building,  being  dominated  by 
hypostyle  hall  and  pylons. 

When  Egypt  had  arrived  at  the  summit  of  her  greatness  and 
wished  to  erect  temples  to  her  gods  which  should  be  worthy  both 
of  herself  and  of  them,  she  found  herself  obliged  either  to  sacrifice 
the  unity  of  the  temple  by  dividing  it  up  into  distinct  naves  and 
sanctuaries,  or  to  hide  the  main  parts  by  the  accessories  in  such 
a  fashion  that  the  sanctuary  seems  to  be  lost  among  the  annexes 
which  envelop  it  in  front  and  rear.  The  vestibule  and  other 
subsidiary  parts  mask  the  real  dwelling  of  the  god.  We  are 
sometimes  at  a  loss  to  decide  the  uses  of  all  the  chambers  of 
so  vast  and  complex  a  structure,  because  our  knowledge  of  the 
circumstances  of  ancient  Egyptian  worship  is  still  far  from 
complete.  It  is  significant  that  even  among  such  an  imposing 
pile  of  buildings  as  those  of  Karnak,  egyptologists  have  found 
it  impossible  to  agree  as  to  the  situation  of  the  heart  and  organic 
centre  of  the  whole.  That  centre  exists  ;  it  existed  before 
all  those  sumptuous  additions  of  which  it  was  the  cause.  But  it 
would  seem  that  its  influence  failed  to  make  itself  felt  bevond  a 
certain  distance.  The  temple  was  enlarged  by  additions  made 
at  its  two  extremities,  in  the  manner  of  an  inorganic  body,  so  that 
no  limit  could  be  logically  assigned  to  its  development.  Karnak, 
as  it  was  left  by  the  Pharaohs  and  their  successors,  is  the  most 
colossal  work  of  architecture  which  has  come  down  to  us  from 
antiquity,  and  yet  oiu'  imagination  can  give  to  it  even  greater 
dimensions  than  it  actually  possessed  without  injury  to  its 
artistic    expression.      If  the   worship   of  which   it   was    the    scene 


444 


A   History  of  Art  ix  Ancient  Egypt, 


had  endured  a  few  centuries  longer,  it  would  have  been  easy 
to  add  new  pylons,  new  courts,  and  new  hypostyle  halls  to 
those  already  existing  ;  but  had  the  worship  of  Athene  endured 
through  as  many  ages  as  that  of  Ptah  or  Amen,  it  would  have 
been  impossible  to  make  additions  to  the  Parthenon  as  it  left  the 
hands  of   Ictinus  and  Phidias. 


EN=&^OF    VOL.    I. 


LONDON:    R.    CLAY,    SONS,    AND    TAYLOR,    PRINTERS. 


^^    .' 


please  do  not  remove 
cArds  or  slips  from  this  pocket 


UNIVERSITY  OF  TORONTO  LIBRARY 


"  V 


5556 


Perrot,  Georges 

A  history  of  art  in 
ancient  Egypt 


v.l 


FOR  USE  IN 

LIBRARY 
ONLY 


Honaavam 


■UMdMMM 


■  ■"'  ■'I'Mww It— aiinninni«i>nMMMBM«ia«MMiamMj.»>^«iMiM»-^.-^._».— -—- »-