Skip to main content

Full text of "A History of Fine Art in Indian & Ceylon"

See other formats




Rei ^d by B. COL^ingTON 











Printed in Great Britain 


It was left to Vincent Smith to summarize the results of the first century of 
historical and archaeological research in India, and his Oxford History of 
India and Early History of India are still authoritative works of reference. 
His History of Indian Fine Art was the outcome of his realization of the im- 
portance of the evidence of archaeology in Indian studies, and of his per- 
ception that the aesthetic nationalism preached by Mr. Havell and Dr. 
Coomaraswamy in their pioneer works was not acceptable without deeper 
research and a more accurate appreciation of the facts. It may be admitted 
that the criticism of art was never his chief care, although he had acquired an 
unrivalled experience of the great sites of India during his long period of 
service. His strength lay in his determination to set down on paper what was 
known and what was conjectured, and in such order that it might be easily 
grasped. Since then new discoveries have been made and new facts been 
brought to light. Opinions have consequently changed. For one thing 
Indian sculpture and painting have taken their place among the familiar arts: 
mediocre Mughal paintings are at a premium in the sale room; Ajanta and 
Ellora are made much of in the advertising schemes of the Indian railways; 
the literature of Indian art grows apace. Yet in spite of changes and the 
passage of years Vincent Smith's chronology of Early India remains a de- 
fensible whole, and we are no nearer to an exact knowledge of the Graeco- 
Buddhist art of Gandhara. In fact, an examination of the reports of recent 
discoveries would suggest that we are suffering from an over-abundance of 
new material which we are unable to handle. 

Official research in India has many obstacles, the chief among which is, 
undoubtedly, conservation. Furthermore, until recently there has been no 
co-operation between the workers in the field and scholars in general. This 
period of isolation is passing. It was inevitable that the widening of research 
in the near east should raise questions concerning India, questions which 
must be answered. The problem is a hard one, but now that India has 
penetrated within the field of academic interest a more general and keener 
criticism will be brought to bear on Indian researches which must work for 
the good of all concerned. At the moment there is a distressing lack of atten- 
tion to matters of classification. No corpus of Indian pottery exists, and very 
little information about the pottery of the classic Indian sites has been made 
accessible to scholars out of India. Apart from the many dark periods of 
Indian chronology no agreed period classification is in use, leading to much 
confusion, especially when an attempt is made to translate dynastic periods 
into geography. Furthermore, there is urgent need for a technical analysis 


of Gupta and medieval architecture after the manner of M. Jouveau-Dubreuirs 
excellent Archeologie du Sud de VInde. 1 In fact there is so much to be done 
that it is essential that any account of Indian history or art should confess 
the fact. In the present state of Indian knowledge to attempt the encyclo- 
paedic is to mislead. At the moment a consecutive account, rather than a 
detailed account, is wanted. 

This, as has been said, was the great merit of the History of Indian Fine 
Art y and an attempt has been made to preserve it. The text of this revision 
is therefore not weighted down with recent references and theories, which, 
however, may be arrived at by means of the short bibliography provided. 
The greatest change is the unification of the dual accounts of the history of 
architecture and sculpture, which does away with a considerable amount of 
repetition. The plates and page illustrations have also been rearranged and 
a large number of new illustrations introduced. A great deal of aesthetic 
comment has, furthermore, been omitted, partly because it can quite well 
be left to the student himself, and partly because it was considered desirable 
to shorten the book as much as possible. Lastly, it must be stated that the 
matter dealing with Indian paintings is intended to be only general and intro- 
ductory. In this subject much research is still necessary before 'schools' can 
be accurately distinguished from 'periods'. 

The question of the period classification to be adopted was a difficult one. 
The dynastic periods in common use are seldom accurately datable, often 
of long duration, and always lead to geographical complications. 2 They are 
moreover archaeologically unreal. Three dynastic periods, however, may be 
preserved because they provide a sufficiently accurate chronology and because 
they happen to coincide with the evidence of the sculptures: these are the 
Mauryan, Kushan, and Gupta. It may be pointed out that our knowledge 
does not at present warrant our speaking of 'cultures' in India, if we use the 
word in its accredited archaeological sense. An accurate knowledge of Indian 
pottery would enable us to do so and would doubtless necessitate a com- 
pletely new classification. 

The present position with regard to the transliteration of place-names is 
complicated. Since the publication of the Index to the Archaeological 
Survey Reports there has been a tendency to Sanskritize place-names whole- 
sale. It is only necessary here to point out that the central authority on Indian 
subjects, the Imperial Gazetteer, adopts a system based 'upon the usage now 

1 A corpusof Indian Pottery and another of Indian 2 See Cousens, Chalukyan Architecture, a title 

Beads are being formed by the India Research which excludes the discussion of the Ellora 

Committee (Royal Anthropological Institute), Kailasa with its cognate building at Pattadkal. 

52 Upper Bedford Place, W. A short analysis Also Coomaraswamy, Hist, of Ind. Art, use of 

of medieval pillars, mouldings, and motives is 'Gupta' to include the Badami caves which are 

included in the editor's Medieval India (Ernest Chalukyan. 
Benn, Ltd.) which is now in active preparation. 


generally adopted', that is to say the system of the district gazetteers. These 
will always be the source-books of Indian studies, and the position is consider- 
ably confused by departing from their usage. As a whole, the place-names of 
India have acquired their present rendering at the hands of the early 
administrators and the compilers of route-books, not to speak of railway 
time-tables. In this edition the classes of consonants are not distinguished. 
In the first edition long vowels were marked 'where necessary as a guide to 
pronunciation \ They are here consistently omitted, and students in doubt 
are referred to the Gazetteer or the volumes given in the short bibliography. 

I have to thank the Directors of the various Museums indicated in the list 
of illustrations, Mr. Ajit Ghose, and also the following, for permission to 
reproduce photographs : 

The Archeological Survey of India (Mysore, Madras, Western India, and 
Ceylon) for Plates 2, 3 b and c, 7 b, 15 b, 16, 17 B and c, 19 a-d, 20 B, 24 B, 
31 a and B, 38 c, 41 d, 43, 44, 46 b and c, 58, 59, 60, 64, 66, 67, 68, 70 a, 
71 a, 75 b, 76 a, 78, 79, 80, 81 c, 83, 88, 89, 90, 94, 96, 97, 98, 99, 100, 101, 
102 B, 103, 104, 124 B, 136 c, 137, and 138; the India Office for Plates 4, 5, 
6, 11, 12 A, 13 a, 22, 23, 39 c, 50, 51, 52 b and c, 65, 82, 84, 87, 91, and 93; 
the Council of the Royal Asiatic Society for Plates 57 B, and 77 B and c; the 
Council of the Royal Institute of British Architects for Plate 69 b ; Professor 
A. A. Macdonell for Plates 10, 52 A, 85 a, 123 A, and 132; Dr. Fiihrer for 
Plates 15 a and 21 ; Messrs. C. Whittingham and Griggs for Plates 27, 28, 29, 
30 b, 31 c, 32, 33, 34 A and b, 35 a and b, 37, 38 d, 39 B, and 40 c; Dr. A. 
Nell for Plates 106, 107 b, 108 A and B, and 109 a; Dr. A. K. Coomaraswamy 
for Plates 102 A, 105, no a and b, 112 b, 113, and 165; Messrs. Klein and 
Peyerl, Madras, for Plate 69 a; Council of the Asiatic Society of Bengal 
for Plate 70 b; and Professor Griinwedel for Plate 120 c. 

I have especially to thank Mr. Laurence Binyon and Dr. Barnett for their 
kind assistance. I should also like to take this opportunity of stating my 
gratitude and offering my best wishes to my colleagues of the University 
of Cincinnati where the revision of this book was done, especially to the 
Dean of the Graduate School and the Dean of the College of Engineering 
and Science, to W. Semple, R. Robinson, B. A. G. Fuller, R. Casey, and 
H. Feis. I must acknowledge special gratitude to Miss Abbot, who was 
then in charge of the Art Library and is now on the staff of the Municipal 
Library. Her co-operation was of the greatest assistance to me. 





Part I. Architecture 
Part II. Sculpture 


Part I. The Kushan Period 
Part II. Amaravati 


Part I. The Hellenistic Sculpture of Gandhara . 
Part II. The Extent of the Foreign Influences 




Part I. Ajanta and Bagh 

Part II. Ceylon ..... 


Part I. Medieval Architecture . 

Part II. Medieval and Modern Sculpture 


Part I. Architecture 

Part II. Sculpture and Bronzes . 


Part I. Singhalese Architecture and Sculpture 
Part II. Singhalese Metal Castings 
Part III. Java .... 

Part I. Chinese Turkistan 
Part II. Tibet and Nepal 

3*9* b 

























Part I. Coins, Gems, and Seals 

Part II. Sculpture 

Part III. Calligraphy and Decorative Reliefs 

Part IV. Lattices 

Part V. Inlay and Mosaic 

Part VI. Tiles . 


Part I. The Gujarati School 
Part II. Mughal Painting 













Elevation of railing. From A. Cunningham, Bhilsa Topes ', 1854 

Plan of Buddhist Church. From A. Cunningham, Bhilsa Topes 

Fan-palm capital, Besnagar. From A. Cunningham, A.S. Rep. y vol. x 

Siva dancing: in Lankesvara temple. From a drawing by H. Cousens in A.S. W.I., vol. v 

Das Avatar Cave, Ellora, c. a.d. 700. From a drawing by H. Cousens in A.S.W.I., vol. 

Early sketch of elephant in Cave X, Ajanta. From Burgess, Notes . 

Raja and women, Early painting, Cave X, Ajanta .... 

Scroll, Hariharesvara temple, Bellary District. From Rea, Chalukyan Architecture 
Makara torana (arch), Malikarjuna temple, Kuruvatti. From Rea, op. cit. . 
Yali, or rampant lion, in Vijayanagar style, at Virinchipuram 
Female figure at Jinji (Gingee) . ..... 

From Kadam Rasul Mosque, Gaur, a.d. 1480. From Ravenshaw, Gaur 
Inscribed principal Mihrab, Jam* i Mosque, Fatehpur Sikri. From E. W. Smith, Fathpur- 
Sikri . ......... 

Vase motive panel, east false gate of Akbar's tomb. From E. W. Smith, Akbar's Tomb 















# Plates in colour 

pp. 20-1 

* Fresco painting. Bagh caves, 6th century 

a.d. British Museum . Frontispiece 

facing page 

i. Inscribed Asoka pillar, Lauriya-Nandan- 
garh. Photograph, Indian Museum, 
Calcutta 1 8 

2. Capital of inscribed Asoka pillar at Sarnath 19 

3. A. Capital of Sankisa pillar. 

Photograph, Indian Museum, 
Calcutta. B. The Heliodoros 
pillar, Besnagar. C. Asoka 
pillar at Bakhira, Muzaffarpur 

4. The great stupa, Sanchi, east side, 

before restoration. 

5. A. Sudarsana Yakshi, Bharhut. B. 

Bharhut; medallion. C. Yakshi, 
Bharhut. D. Bharhut; medal- 
lion. E. Bharhut \jataka scenes 
on coping. 

6. Bharhut, inner view of eastern 


' facing page 

7. A. Chaitya Cave, Nasik. B. Lomas Rishi 

Cave, Barabar ..... 

8. Colossal female statue from Besnagar. 

Photograph, Indian Museum, Calcutta 

9. Reliefs from the railing. Bodh-Gaya. 

Photographs, Indian Museum, Calcutta 

10. A. The great stupa, Sanchi, as restored. 

B. Inside, west gateway, Sanchi . 

11. Inside, east gateway, Sanchi . 

12. A. East gateway, Sanchi. B. East gateway, 

Sanchi 35 

East gateway, Sanchi. 

Portion of frieze in Rani Gumpha 
Cave, Udayagiri, Orissa. Photo- 
graph, Indian Museum, Cal- 

A. Part of frieze on torana beam, 

Mathura. B. Tablet with relief I between 
sculpture of a Jain stupa. \ pp. 36-7 

A. Fragments in Bharhut style; 
Mathura Museum. B. Naga 
statue, with inscription of Hu- 
vishka's reign, from Chhargaon ; 
Mathura Museum. C. Yakshi 
on dwarf; Mathura Museum. 


l S- 



3 1 



facing page 

17. A. Two Yakshis; Indian Museum. B. 

Female, half-back view; Mathura Mu- 
seum. C. A Bodhisattva, from Katra; 
Mathura Museum .... 40 

18. A. Nude female on Jain railing pillar, 

Mathura. From Jain Stupa of Mathura. 
B. Medallion, Mathura. Ibid. C. Brac- 
ket figure, Mathura. Ibid. D. Medal- 
lion, Mathura. Ibid. E. Medallion, 
Mathura. Ibid. F. Medallion, Mathura. 
Ibid 41 

19. A. Female with right arm bent; Mathura 

Museum. B. Female with right leg 
bent; Mathura Museum. C. Female 
and child; Mathura Museum. D. A 
soldier; Mathura Museum. E. Lion 
and rider; Indian Museum . . 42 

20. A. Pali Khera block, front group ; Mathura 

Museum. B. Kuvera, Mathura Mu- 
seum. C. Bodhisattva from Mathura; 
Indian Museum .... 43 

21. Torana arch, Mathura . ... 46 

22. Slab with representation of a stupa, &c, 

from the base of the great stupa, 
Amaravati ..... 47 

23. A. Basal medallion, Amaravati. B. Undu- 

lating roll motive on coping of rail, 
Amaravati 48 

24. A. Man and boy, Amaravati. From Bur- 

gess, The Buddhist Stupas of Amaravati 
and Jaggayapeta, 1887. B. Marble 
Buddhas, Amaravati ... 49 

25. A. Lotus forms, Amaravati. From Rea, 

South Indian Buddhist Antiquities, 1894. 
B. Lotus and makara, Amaravati. 
Ibid. C. A pond, Amaravati. Ibid. . 50 

26. A. Court scene. Medallion; Amaravati. 

B. Enthronement of relics. Medallion; 
Amaravati. British Museum . . 51 

27. Buddha, &c; relief from Muhammad 

Nari 54 

28. Modified Corinthian capitals from Gan- 

dhara 55 

29. Seated Buddha; Berlin Museum . . 56 

30. A. Visit of Indra to Buddha in Indrasaila 

Cave. Photograph, Indian Museum, 


facing page 

87. Durga, Mahishamardini, Mahisa Man- 

dada, Mamallapuram . . . 136 

88. Rock-sculpture ('Arjuna's Penance'), Ma- 

mallapuram 137 

89. A. Siva: cave-temple (Pallava). Trichino- 

poly, 7th century. B. Siva and Parvati; 
on north wall of great temple (Chola). 
Gangaikonda-Cholapuram , Trichino- 
poly, nth century .... 138 

90. A. Siva and Parvati; on north wall of 

great temple (Chola). Gangaikonda- 
Cholapuram, Trichinopoly, nth cen- 
tury. B. Siva. Darasuram, Tanjore 
District, 16th century . . . 139 

91. Ramayana reliefs, Hazara Rama ' 

Temple, Vijayanagar. 

92. Southern Gopuram. Great Tem- 

ple, Madura, 17th century. 

93. A. Siva ; Tirumal Naik's choultry, 

Madura. B. Woman and baby ; 

Great Temple, Madura. C. 

Female carrying male deity; I between 

Ramesvaram temple. f pp. 140-1 

94. Cast brass portrait images of 

Krishna Raya of Vijayanagar 
(a.d. 1510-29) and his Queens; 
in the Sri Nivasa Perumal tem- 
ple on the hill of Tirumalai near 
Tirupati, N. Arcot District. 
Progr. Rep. A.S. Madras and 
Coorg, 1903-4. 

facing page 

95. Southern Indian, copper castings now in 

the Museum of Fine Arts, Boston, 
Mass. A. Uma. B. Siva Nataraja . 142 

96. A. Stone railing at Anuradhapura, as re- 

stored. B. Siva temple No. 1, at Polon- 
naruwa, west wall. C. Stucco reliefs on 
porch of the Hetadage, Polonnaruwa . 143 

97. A. Circular shrine (wata-da-ge) at Polon- 

naruwa ; part of north-eastern quadrant. 
B. The same; western stairs. C. Circu- 
lar shrine (wata-da-ge) at Medirigiriya, 
N.C.P 144 

98. A. and B. Capitals at Abhayagiri Vihare, 

Anuradhapura. C. Column in Ruwan- 
veli area, Anuradhapura ; 9 ft. 4 in. high 145 

99. Sculptured stelae at Abhayagiri dagaba, 

Anuradhapura . .146 

100. A. North stele, east chapel, Abhayagiri. 

B. Naga door-keeper, Ruwanveli vihare 147 


101. A. Dwarf door-keeper, Ruwan- ^ 

veli dagaba. B. Dwarf right- 
handed door-keeper of south 
porch of west chapel of Jetawa- 
narama. C. Part of dado, Ru- 
wanveli dagaba, vihare. 

102. A. Kapila relief, Isurumuniya, 

Anuradhapura. B. Seated Bud- 
dha (5 ft. 9 in.) in situ at Tolu- 
vila, Anuradhapura, now in I between 
Colombo Museum. f pp, 148-9 

103. A. Seated Buddha (6 ft. 9 in.); 

Pankuliya Vihare, Anuradha- 
pura. B. Seated Buddha; from 
Vihare, No. 2, Polonnaruwa, 
now in Colombo Museum. C. 
Colossal Buddha at Awkana, 

104. Colossal statue of 'Ananda', 


facing page 

105. Pattini Devi. Cast brass. Singalese, 

now in the British Museum . .152 

106. Siva Nataraja (3 ft.). Copper casting. 
No. 1, from Polonnaruwa, Colombo 



A. Bodhisattva. Copper casting. " 

Anuradhapura (No. 97), now 
in the Colombo Museum. B. 
Siva Nataraja. Copper cast- 
ing. Polonnaruwa (No. 15). 

108. Copper castings in the Colombo 

Museum. A. Siva (Ht. 1 ft. 
B. Surya, the Sun-god (Ht. 
1 ft. 5J in.), Polonnaruwa 
(No. 18). C. Sundara-murti 
Swami (Ht. 1 ft. 8 in.), Polon- 
naruwa (No. 16). 

109. Copper castings in the Colombo 

Museum. A. Parvati (Ht. 1 ft. 
8 in.), Polonnaruwa (No. 23). 
B. Parvati (Ht. 1 ft. 4$ in.), 
Polonnaruwa (No. 20). 

no. A. Avalokitesvara, or Pad- 
mapani. Singalese. Boston 
Museum. B. Jambhala, or 
Kuvera. Singalese. Boston 
Museum. C. Seated Buddha. 
Copper casting. Colombo 

* PP- 1 S^7 


facing page 
in. Offerings to a Bodhisattva. A. TheBod- 
hisattva's descent to earth. B. The 
story of Prince Sudhana. Relief- 
panels in the first gallery, Borobudur, 
Java. From van Kinsbergen, Cud- 
heden van Java .158 

112. A. Sarasvati enthroned, from Jogyokaita, 

Java. B. Prajna Paramita. Javanese . 159 

113. A. Seated Buddha. Javanese. B. 

Manjusri. Copper casting. Javanese. 
British Museum . . . .160 

114. Persian Bodhisattva; rev. of wooden 

panel, D. vii, 6, from Dandan-Uiliq. 
From Stein, Ancient Khotan, 1907 . 161 

115. Water-sprite, &c; fresco at Dandan- 

Uiliq. From Stein, Ancient Khotan, 
1907 162 

116. A. Mounted princes or saints; wooden 

panel from Dandan-Uiliq. From Stein, 
Ancient Khotan, 1907. B. Chinese 
princess; fresco at Dandan-Uiliq. 
Ibid. C. Bactrian camel; Indian-ink 
drawing on paper from Endere. Ibid. 163 

117. Buddha and worshippers. Tibet. ('Ti- 

bet, N. 5', Institut de France) . . 166 

118. Portrait of young Lama. Tibet. (Hodg- 

son Collection, Institut dc France) . 167 

119. Portrait of a Lama evoking a demon. Tibet. 

(Hodgson Collection, Institut de France) 1 68 

120. Bronze images of Tibetan deities and 

saints. Copper castings. Tibet. A. 
A teacher. Pitt-Rivers Museum, Ox- 
ford. B. The Bodhisattva Manjusri. 
Pitt-Rivers Museum. C. The goddess 
Sarasvati. Ukhtomskij Collection. 
D. Tsong-kapa. Pitt-Rivers Museum 169 

121. Great Mosque at Ajmer. Photograph, 

Indian Museum, Calcutta . . 172 

122. The Kutb Minar, Delhi. Photograph, 

Indian Museum, Calcutta . . 173 

123. A. Arches of the Kutb Mosque, showing 

the Iron Pillar. B. Arch of the Great 
Mosque, Ajmir . . . . 174 

124. A. Gateway of Ala-ud-din Khalji, 

Delhi. Photograph, Indian Museum, 
Calcutta. B. Tomb of Tughlak Shah, 
Tughlakabad, Old Delhi. From Ann. 
Rep. A.S. lndia f 1904-5 . . . 175 

125. Main entrance of Ataladevi Mosque, 

Jaunpur. From Fiihrer and Smith, 
The Sharqi Architecture of Jaunpur . 176 


facing page 

126. Mosque of Mahafiz Khan, Ahmadabad. 

From Burgess, A.S.R. Western India, 
vol. vii 177 

127. A. Gateway of Small Golden (Eunuch's) 

Mosque, Gaur. From Ravenshaw, 
Gaur, its Ruins and Inscriptions, 1896. 
B. Tomb of Abu Turab, Ahmadabad. 
From Burgess, A.S.R. Western India, 
vol. viii ..... 178 

128. A. Tomb in Golkonda style at Bijapur. 

From Ferguson and Meadows Taylor, 
Architecture at Biejapoor, 1866. B. 
Ibrahim Rauza, Bijapur; front view. 
Photograph, Indian Museum, Calcutta 179 

129. A. Gol Gumbaz, or tomb of Mu- 

hammad Adil Shah, Bijapur, 
front view. Photograph, Indian 
Museum, Calcutta. B. Tomb 
of Humayun. c. a.d. 1560. 

130. Tomb of Sher Shah. Sahasram, 

Shahabad District, Bengal. I between 
Photograph, Indian Museum, [ pp. 180-1 

131. The 'Buland Darwaza' of Jam' i 

Mosque, Fatehpur Sikri. 
Photograph, Indian Museum, 

132. The Taj Mahal. 

facing page 

133. A. Tomb of Itimad-ud-daula, near Agra. 

Photograph, Indian Museum, Cal- 
cutta. B. The Moti Masjid, Agra . 182 

134. A. Diwan-i-khas of Delhi Palace. B. 

Shrine of Sayyid Salar, Bahraich . 183 

135. Indo-Muhammadan Coins. Fig. 1. Mu- 

hammad b. Tughlak (British Museum 
Catal. of Coins of Sultans of Delhi, No. 
260). Figs. 2-5. Akbar (British Mu- 
seum Catal. of Coins of Moghul Emperors, 
Nos. 166, 172, 173, 250). Figs. 6-10. 
Jahangir (Ibid. Nos. 315, 319, 324, 341) 184 

136. Mughal Reliefs. A. From Jam* i Masjid, 

Fatehpur Sikri. From E. W. Smith, 
Fathpur-Sikri. B. South end (jalla 
jalalahu) of Akbar's cenotaph. From 
E. W. Smith, The Tomb of Akbar. C. 
Panel in dado of 'false mosque' (jawab), 
at the Taj, Agra. From Ann. Rep. A. S. 
India, 1903-4 .185 

137. Panels from Sarangpur Mosque, Ahma- 

dabad. c. a.d. 1500. From AS. Rep. 
Western India . . .192 




Windows of Sidi 
Ahmadabad. c. 

facing page 
Sayyid's Mosque, 
a.d. 1500. From 





pp. 196-7 



Arch. S. Western India . . -193 

139. Marble verandah screen, tomb of "1 

Salim Chishti, Fatehpur Sikri. | 
a.d. 1571. From E. W. Smith, 

140. Marble screen round the ceno- 

taph of the Taj. 

141. Upper part of a corner turret, 

Itimad-ud-daula's tomb, Agra, 
showing pietra dura inlay and 
marble mosaic. From E. W. 
Smith, Moghul Colour Decora- 
tion of Agra. 
Pietra dura inlay on the cenotaph 
of the Taj. From J. Ind. Art, 
vol. i. 

facing page 

Minaret of Wazir Khan's Mosque, La- 
hore. From J. Ind. Art t vol. x 

Chini-ka-Rauza, Agra. From E. W. 
Smith, Moghul Colour Decoration of 

Mughal tiles in the Victoria and Albert 
Museum. A. Glazed earthenware tile 
from Panjab, 17th century. B. Ena- 
melled earthenware tile from Delhi, 
1 6th century. C. Enamelled earthen- 
ware tile from Lahore, 17th century 202 

146. Enamelled earthenware tiles from La- 

hore, in the Victoria and Albert Mu- 
seum, South Kensington, 17th century 203 

147. Kalpasutra (Jain) MS. Gujarati " 

School, A.D. 1464. Or. 5149. 
British Museum. 

148. Painting on cotton cloth from a 

copy of the Amir Hamzah; 
Victoria and Albert Museum, 
Indian Section, S. Kensington. 

149. Illustration from the Akbarnama, 

painted by Basawan and Chatar ; 
subject — Akbar in an elephant 
fight on a boat bridge across the 
Jhelum; Victoria and Albert 
Museum, Indian Section, 
S. Kensington. 

150. Illustration of the Darabnamah; 

by Bihzad and Abus Samad. 

facing page 

151. Wall-painting: eight men in a boat; in 

Akbar's bedroom, Fatehpur Sikri. 
From E. W. Smith, The Moghul 
Architecture of Fathpur-Sikri . . 208 

152. Picture of a Plane Tree (c. a.d. 1610). 

Johnson Collection, vol. i, India Office 
Library, London 209 

153. Faqirs resting under trees (i. a.d. 

1650). Collection of Baron 
Maurice Rothschild, Paris. 

1 S4- 





Sketch portrait of the Emperor 
Jahangir(c.A.D. 1625). Collec- 
tion of M. Carrier, Paris. 

Portraits of Sher Muhammad 
Nawal, Jahangir, and Shah- 
jahan, by Muhammad Nadir 
of Samarkand. British Mu- 
seum, MS. Add. 18801. 

facing page 

* Jahangir as Prince Salim, anonymous. 
(Fol. 18 of Dara Shikoh's album) . 214 

pp. 204-5 

pp. 220-1 

157. * Wild duck, anonymous. (Fol. 10, Dara 

Shikoh's album) .... 216 

158. Rajasthani School. Late 17th 

century. British Museum. 

159. Pahari School. 18th century. 

Ghose Collection. 

160. Pahari (Kangra) School. Late 

1 8th century. V. and A. 

161. Pahari (Kangra) School. Late 

1 8th century. V. and A. 

facing page 

162. A. Rajasthani (Jaipur) School. 18th 

century. British Museum. B. Patua 
drawing. Calcutta, 19th century. 
Ghose Collection .... 222 

163. # Lady and sunlight effect, by Rao Gobind 

Singh (India Office Library, Johnson 
Collection, vol. xxi) . . 224 

164. # Marble building, &c, by Muhammad 

Fakirullah Khan (India Office Library, 
Johnson Collection, vol. xvii) . . 226 

165. The exiled Yaksha, by Abanindro Nath 

Tagore 230 

Chapter One 

IN discussing Indian studies I am forced to acknowledge considerable Modem 
diffidence arising from a survey of the huge bulk of material to be dealt India - 
with. In the face of this objective complexity I find myself inclined to 
rely on evidence that is subjective and therefore more or less unscientific, 
in which personal experience and interpretation is increasingly stressed. In 
speaking of India, a country that in its wide extent offers more beauty to the 
eyes than any other in the world, a descriptive vein may well be excused, 
but the more graphic the form, the more dangerous does the method become. 
India is multiple; neither geographically, ethnologically, nor culturally can it 
be considered a unity. This being so, I am led to suspect that the India of 
many writers is more imagination than fact, existing rather in pictorial 
expression than in reality. 

The appeal of the pictorial, rising from a craving for colour and movement, 
is general among the generations of the present, continually chaffing against 
narrowed horizons and an experience bounded by Economical Necessity. 
There is magic to be found anywhere between Cancer and Capricorn. There 
the demands of Necessity would seem to be more easily fulfilled and life to 
run more rhythmically, in the train of the tropic alternation of the seasons. 
There bread is to be gathered direct from the rich lap of the earth. There 
colour fills the day with its wealth, leaping to the eye, like the sudden glow 
of fruit and flower caught by the sunlight, or of kaleidoscopic crowds in 
narrow streets. To enter a tropic town is to enter, as in a dream, the life 
of a dead century. 

Modern complexity is apparently to be regarded as successful and there- 
fore not to be deterred by sentimental leanings towards the simplicities of 
Eden or Arcady. Yet the sentimental mood will have its way, not only in 
the West but in the East where the ready acceptance of change at the expense 
of tradition lies at the very root of the problem that is modern India. Modern- 
ism, supported by thorough-going educational propaganda, may overcome 
the great geographical and ethnological obstacles, and result in the crystalliza- 
tion of Indian nationality. The alternative offered seems to be a return to 
the past on an agricultural basis; Arcady in India under the good king Vikra- 
maditya. The movement is not without parallels, and the pictorial and inter- 
pretational play a great part in its exposition; there is, indeed, something of 
the Pre-Raphaelite about it. The materialism of to-day is to be checked by 
Indian Spirituality. Arts and crafts are to flourish everywhere, centred upon 
the social organization of the village. India is to arise from the ashes of 

It might be claimed, therefore, that there could be no better time than the Indian Art. 

3*J* B 


present for the republication of a survey of Indian Fine Arts, that the credit 

and loss of the exchange between the occidental and the oriental may be 

appraised. Indeed this nationalization of the subject has been set forth at 

length by certain authors. It is, however, in contradistinction to the spirit 

of true criticism and full appreciation. The opposition of Eastern spirituality 

to Western materialism is a generalization without support, while the postu-* 

lation of a metaphysical basis for any art is equally as sterile, and in fact as 

inconsequential, as the postulation of the existence of eternal, immutable 

classical standards. Art cannot be localized, at least if the humanities upon 

which our culture is based have any meaning, and geographical differences 

should be no bar to appreciation, but rather an added attraction in these 

days, when for most of us our voyages of discovery do not exceed the bounds 

of the local time-table. It is, however, unfortunate that in the minds of many 

people the East has a certain romantic but quite indefinite lure about it, 

which accentuates the unusual and leads to the substitution of curiosity for 


Modern Art. It is impossible for any one to deny the advance of modern science, with 

its consequent widening of horizons and enrichment of life. Yet aesthetically 

our advance from the station of our fathers is as great, and possibly greater, 

being more radically concerned with that personal interpretation of life 

which is Reason. 

Modern painting and sculpture provide a definite line of advance and 
logical precepts to an extent that almost makes academicians of many of the 
younger school. This process is directly comparable to that of modern 
scientific method; modern art is indeed the result of methodical, aesthetic 
research. From the painting of Manet to that of Cezanne and the men of 
to-day, Matisse, Picasso, Stanley Spencer, Paul Nash, and many others, the 
story can only be told in terms of intellectual adventure and aesthetic dis- 
covery. The effect of the personal vision of the creators of modern art has 
been a widening of the circle of aesthetic interest and a revaluation of things 
unknown or unconsidered — Chinese painting and sculpture, Gothic sculp- 
ture, archaic Greek sculpture, Negro sculpture, the harmony of fine carpets, 
the virility of primitive design, and not least among these, Indian Art in all 
its branches. 1 In the face of these riches, once despised and rejected, the 
dogmas of the past generations with all their complacency, intolerance, and 
ignorance seem wilful in their restriction and impoverishment of life. 
Appreciation So vital is this movement and so well founded that I would choose as the 
° f ln An ^ e P e °? a rey i ew of Indian Art aesthetic discovery rather than archaeo- 
' logical discovery, and for support I would rely upon the word of living artists 2 
whose creative vision and fellow-craft appreciation provides the basis of a 

1 Fry, 'Oriental Art* (Quart. Rev., Jan. 1910). 

2 Eric Gill, preface to Visvakarma; W. Rothenstein, preface to Ancient India. 


criticism of greater precision than archaeological chop-logic or the ulterior 
ends and confused categories of evidence of those who would carry the dis- 
cussion beyond the proper field of art. I cannot believe it is necessary or 
even desirable to prelude the vision of a work of art with many words. Nor 
can I accept as sound criticism a discourse which shifts the foundations of 
a true understanding of art from the visual into the literary or historical or 
metaphysical. And I can but deplore the twisting awry of aesthetic criticism 
and appreciation to local and temporary ends, whatever the circumstances. 

In 1897 Gauguin wrote: 'Ayez toujours devant vous les Persans, les 
Cambodgiens et un peu rEgyptien.' l One wonders what he would have 
written if he had known of the frescoes at Ajanta with their magnificent surity 
of line and delicately rendered plasticity. The placing on exhibition of castes 
of Indian sculpture mainly of the late medieval period, in the Trocadero in 
Paris, may be taken as the first step towards the western appreciation of Indian 
Art. Until then Indian Art had been left to the archaeologists — not altogether 
without results. The work of Prinsep and Cunningham, 3 of Fergusson 3 and 
Burgess 4 is a well-laid foundation-stone for all future research. 

In The Times of 28 February 19 10 appeared the following declaration above 
the signatures of thirteen distinguished artists and critics.* 5 

'We the undersigned artists, critics, and students of art . . . find in the best art of 
India a lofty and adequate expression of the religious emotion of the people and of 
their deepest thoughts on the subject of the divine. We recognize in the Buddha type 
of sacred figure one of the great artistic inspirations of the world. We hold that the 
existence of a distinct, a potent, and a living tradition of art is a possession of priceless 
value to the Indian people, and one which they, and all who admire and respect their 
achievements in this field, ought to guard with the utmost reverence and love. While 
opposed to the mechanical stereotyping of particular traditional forms, we consider 
that it is only in organic development from the national art of the past that the path 
of true progress is to be found. Confident that we here speak for a very large body of 
qualified European opinion, we wish to assure our brother craftsmen and students 
in India that the school of national art in that country, which is still showing its vitality 
and its capacity for the interpretation of Indian life and thought, will never fail to 
command our admiration and sympathy so long as it remains true to itself. We trust 
that, while not disdaining to accept whatever can be wholesomely assimilated from 
foreign sources, it will jealously preserve the individual character which is an out- 
growth of the history and physical conditions of the country, as well as of those ancient 
and profound religious conceptions which are the glory of India and of all the Eastern 

This declaration was directly caused by a paper read before the Royal 

1 Rotonchamp, quoted on p. 165. Housman, (5) E. Lanteri, (6) W. R. Lethaby, 

2 Arch. Survey Reports. (7) Halsey Ricardo, (8) T. W. Rolleston, (9) 

3 History of Ind. Arch. W. Rothenstein, (10) George W. Russell (A. E.), 

4 Arch. Survey of West India. (11) W. Reynolds Stephens, (12) Charles Wald- 

5 The signatories are (1) Fred. Brown, (2) Walter stein, and (13) Emery Walker. 
Crane, (3) George Frampton, (4) Laurence 


The Society of Arts by Sir George Birdwood, the chronicler of Indian industrial 
Criticism arts. 1 As a matter of fact all that was then said had already appeared in 
of Scholars. p r j n t thirty years before, but the moment was not then ripe for the acceptance 
of the challenge. Birdwood can in no way be accused of lack of sympathy 
with Indian life or things Indian. A stylistic analysis of the crafts of modern 
India is illuminating with regard to one's attitude to the country itself, for 
one is forced to acknowledge the predominance of the Muhammadan and 
especially of the Persian culture of the Mughal court. Except in their every- 
day household form, pottery and metal-work are purely Muhammadan. 
Textiles, especially prints and brocades, are very largely Persian in design, 
although the Indian strength of imagination and purity of colour are evident. 
Certain forms of textiles are, however, purely Indian, the darn-stitch Phul- 
karis of the north-west and certain tied-and-dyed and warp-dyed forms. 
Only in jewellery has the Indian tradition been wholly preserved, in the seed- 
and-bead work of the villages as well as in the enamels of Jaipur. Birdwood's 
love of all this delicate and colourful though hybrid craftsmanship, and of 
the complex, changeful life of which it is a part, is expressed in many passages 
from his pen of very great beauty. The arts of Ancient and Medieval India 
were outside his field, and his criticism of them is not deeply considered and 
purely personal. 

In his paper before the Royal Society of Arts he stated with regard to a 
certain Javanese seated Buddha that this Senseless similitude, by its im- 
memorial fixed pose, is nothing more than an uninspired brazen image, 
vacuously squinting down its nose to its thumbs, knees, and toes. A boiled 
suet pudding would serve equally well as a symbol of passionate purity and 
serenity of soul.' This attack, however, may be considered as being equally 
directed against the loose verbiage of those critics of Indian art to whom 
the ideal content of an object is of greater importance than its form, than 
against Indian art itself. 

An earlier statement in the official handbook to the India Section of the 
Victoria and Albert Museum offers a more definite criticism. 2 'The mon- 
strous shapes of the Puranic deities are unsuitable for the higher forms of 
artistic representation: and this is possibly why sculpture and painting are 
unknown j as fine arts, in India. . . . How completely their figure-sculpture 
fails in true art is seen at once when they attempt to produce it on a natural 
and heroic scale, and it is only because their ivory and stone figures of men 
and animals are on so minute a scale that they excite admiration/ Here it 
must be noticed the subject under discussion is modern Indian ivory- 
In his Handbook of Sculpture , Professor Westmacott 3 dismissed Indian art 

1 J. R. Soc. of Arts, 4 Feb. 1910. Industrial Arts 2 Discussedby Hzvtll, Ideals of Indian Art ,p.xvii. 
of India. 1880. 3 Edinburgh, 1864, p. 51. 


in one paragraph, forming his judgement, apparently, from the steel engrav- 
ings and lithographs of the two or three books that were all that was then 

'There is no temptation to dwell at length on the sculpture of Hindustan. It affords 
no assistance in tracing the history of art, and its debased quality deprives it of all 
interest as a phase of Fine Art, the point of view from which it would have to be con- 
sidered. It must be admitted, however, that the works existing have sufficient character 
to stamp their nationality, and although they possess no properties that can make them 
useful for the student, they offer very curious subjects of inquiry to the scholar and 
archaeologist. The sculptures found in various parts of India, at Ellora, Elephanta, 
and other places, are of a strictly symbolical or mythological character. They usually 
consist of monstrous combinations of human and brute forms, repulsive from their 
ugliness and outrageous defiance of rule and even possibility/ 

In the opinion of Dr. Anderson, author of the catalogue of sculpture at 
the Indian Museum, 1 Calcutta, Indian sculptors 'have never risen . . . beyond 
the most feeble mediocrity', although he acclaims the Orissa temple-sculp- 
tures as 'extremely pleasing pieces of art\ A more guarded opinion is that 
of Sir C. Purdon Clarke, who whilst giving Indian art a good place among 
the arts of the world, would not place it in the first rank, except for its 
'eminent suitability to its country and people \ a 

Such were the opinions current among scholars at the end of the last 
century, concerning an art already accepted by artists and acclaimed by its 
influence upon the work of such men as Rodin, Degas, and Maillol. 

The popularization of Indian art has been mainly the work of Dr. Cooma- Inter- 
raswamy and Mr. E. B. Havell. To a certain extent their methods of ex- P re }^ 
position agree, the vein being interpretational, with a stressing of the literary. 1C: 
For Dr. Coomaraswamy 'all that India can offer to the world proceeds from 
her philosophy', 3 a state of 'mental concentration* (yoga) 4 on the part of 
the artist and the enactment of a certain amount of ritual being postulated 
as the source of the 'spirituality* of Indian art. The weakness of this attitude 
lies in its interweaving of distinct lines of criticism, form being dressed out 
in the purely literary with the consequent confusion of aesthetic appreciation 
with religious and other impulses. It is also historically ill-founded, for the 
sentiment and philosophy out of which the web is spun are the products of 
medieval 5 India, as an examination of the texts quoted will show; many of 
the southern authorities quoted can only be classed as modern. The increas- 
ingly hieratic art of medieval and modern India, especially in the south, is 
doubtless closely knit with this literary tradition. But the literary tradition 
is not the source of the art, for iconography presupposes icons. The technical 
formulae of the Sastras resulted in a standardization of production in spite 

1 Vol. i, p. 175 and vol. ii, p. 221. Welfare?', Athenaeum, 1915. 

* J. I. A. & I. f 1890, p. 526. 4 The Dance of Siva, p. 21. 

3 'What has India contributed to Human 5 p os t t fifth century a.d. 


of which genius, which knows no bonds, asserted itself. The bronze Nataraja 

loaned by Lord Ampthill to South Kensington is supreme among a hundred 

examples of mere hack-work. The bones of the literary formulae too often 

remain bones; here they are clothed with life, and beauty of form is achieved. 

The miracle is a perennial one and world-wide; we marvel at the hand and 

eye that shaped this wonder. However, it is evident that many such images 

are not aesthetically worth the metal they are cast in. Their function as 

objects of worship is an entirely different matter. To insist on the necessity 

of burdening the mind with a host of symbolical and psychological adjuncts 

prior to appreciation is to obstruct the vision. Research literary or historical 

may aid vision, but cannot be substituted for it. Aesthetic vision is, of course, 

distinct from the practical vision of everyday life. Those who indulge in it 

are 'entirely absorbed in apprehending the relation of forms and colour to 

one another, as they cohere within the object'. 1 Intensity and detachment 

from the merely superficial and additional are essential to it. This rigid 

detachment may at any moment be broken by interest in all sorts of ' quasi - 

biological feelings' and irrelevant queries: but then the vision ceases to be 

critical and becomes merely curious. 

India Past A further element is apparent in the recent discussion of Indian art. 

and Present. Aesthetically we are not at all concerned with the sub-continent that is known 

as India or its peoples. But our curiosity must needs be strong as to its 

past and future. The pageantry of Indian history is as glorious as that of 

any country in the world. Artistically it falls into two main periods, the first 

of which, ending with the Muhammadan conquest, is an epic in itself. This 

period discloses the development of a great art. From the vividly pictorial, 

strictly popular sculpture of the Early Period, based on a living tradition, 

increased skill and wider vision lead to the classic art of the Gupta century. 

Henceforward it is evident that a literary tradition has come into being which 

may rightly be designated medieval. The art of the great cave-temples gives 

place to the art of the temple-cities of Bhuvanesvar and Khajuraho, where 

the literary tradition crystallizes into the iconographical forms of the Sastras. 

In the South an imposing architecture is found to survive up to the end of 

the seventeenth century, and the art of casting in bronze produces great works 

of art, few of which can, however, be dated in the last century. It is necessary 

to discriminate, and to acknowledge decadence and poor craftsmanship. 

Having taken its place among the arts of the world, Indian art belongs to 

the world. The future of art in India is another matter, chiefly concerning 


Traditions have died and the symbols that embodied them have died with 
them. Regret for the 'creed out-worn' is ineffectual. New traditions and new 
symbols are surely in the making. Proteus and Triton are become empty 

1 Fry, Vision and Design, p, 49. 


names, but the sea remains. Nothing is lost but a dream, or rather the means 
of expressing a dream. 

Indian religious history must be unfolded against a background of primi- The( 
tive savagery and sorcery. The Vedas, in spite of their antiquity, cannot pkxit 
be accepted as the sole source of religious thought in India, or as anything j*** 
but a critical and highly selective representation of this unvoiced and neces- e lgl 
sarily formless background. This relationship between Brahmanism and the 
primitive, between the formulated philosophy of the schools and the worship 
and propitiation born of the vague fears and desires of savages, is present 
throughout the history of India, both religious and political. The Atharva 
Veda was not known to the early Buddhist writers but its practices and beliefs 
were, and they cannot be separated from the more altruistic and poetical 
polytheism of the less popular, more orthodox but not more ancient collec- 
tions. 1 In the same way the powers and manifestations of the Pur anas and 
Epics arc not necessarily modern because they do not appear in the Veda; in 
a sense they are more ancient, being native to the soil. Vedic thaumaturgy 
and theosophy were never the faith of India. The countless Mother-Goddesses 
and village guardians of the South lie closer to the real heart of Indian 
religion, a numberless pantheon, superficially identified with Brahmanism 
but radically distinct and unchanged. 

Among these lesser gods that keep their place on the fringes of the orthodox The L. 
are to be found spirits of the Earth and of the Mountain; the Four Guardians Gods, 
of the Quarters with Vessavana-Kuvera at their head; Gandharvas, heavenly 
musicians; Nagas } the snake-people who have their world beneath the waters 
of streams and tanks, but who sometimes are identified with the Tree- 
Spirits; and Garudas, half men, half birds who by kind are the deadly foes 
of the Nagas. These diminished godlings must be regarded as the last 
remnant of a whole host of forgotten powers, once mighty and to be placated, 
each in its own place. Strange beings of another sphere, they could not 
wholly be passed over either by Brahman or Buddhist. Vessavana-Kuvera 
appears on one of the pillars of the Bharhut railing, as does also Sirima 
Devata. The latter also received acknowledgement at the hands of the com- 
pilers of the Satapatha Brahmana who are forced to invent a legend to 
account for her existence. 2 In the Taittiriya Upanishad 3 she is again fitly 
mentioned in company with the Moon and the Sun and the Earth. At Sanchi 
she is to be recognized exactly as she is still represented in painted and gilt 
marble at Jaipur, seated upon 2 lotos, lustrated by two elephants. 

In the Maha Samaya Suttanta 4 is described a great gathering of all the 

1 Macdonnell, Sanscrit Literature, p. 185; Rhys 3 i. 4. 

Davids, Buddhist India, chap. xii. Also Dialogues 4 Rhys Davids, Dialogues of the Buddha, ii. 

of the Buddha, note to Ambattha Sutta, i. 3. 284. 

a xi. 4 . 3 . 


The Great gods of the ten-thousand world systems to pay reverence to the Buddha in 
Gathering, the Great Forest at Kapilavatthu. Dhatarattha, king of the East, Virulhaka, 
king of the South, Virupakkha, king of the West, and Kuvera, king of the 
North arrive with their Yaksha host and all their vassals. The Nagas come 
from Nabhasa, VesaH, Tacchaka, and Yamuna, among them Eravana. Their 
enemies the twice-born Garudas, too, are there and also the Asuras, dwellers 
in the ocean. Fire, Earth, Air, and Water are present, and the Vedic gods, 
and lastly the powers of Mara who bids creation rejoice at his own defeat 
at the Buddha's hands. 

Another list of the same description, but possibly earlier, is to be found in 
the Atanatiya. Both lists are, patently, the outcome of a priestly attempt to 
bring these hundred and one strange spirits and godlings within the sphere 
of Buddhist teaching, by representing them as gathered in hosts at the 
Buddha's feet. The group of Yakshas, Yakshinis, and Devatas carved upon 
the stone pillars of the Stupa railing at Bharhut fulfil exactly the same func- 
tion. They are manifestly earth-born and possess something of the delicate 
beauty of all forest creatures. They seem beneficent enough, but their mani- 
festation here is admittedly chosen to serve Buddhist ends. Like all primitive 
powers they are exacting in their demands and when neglected or provoked 
their anger is implacable and cruel. They are adorned with earthly jewels to 
represent the treasures they have in their gift, but are to be more closely 
identified with the trees under which they stand and the forest flowers thev 

This primitive cult of trees and tree-spirits has a long history. In the 
sculptures of the early period the Buddhas are represented only by symbols, 
among which are their distinctive trees. Gotama attained enlightenment 
seated beneath the Assatha or pipal-tree ' sacred from of old, for it was from 
pipal wood that the soma vessels were made and also the sacred fire-drill. In 
the Atharva Veda it is said that the gods of the third heaven are seated under 
the Asvattha and it may also be the 'tree with fair foliage' of the Rig Veda 
under which Yama and the blessed are said to pass their time.* In the 
Upanishads the Tree-spirits have definitely materialized. They, like all things, 
are subject to rebirth. If the spirit leaves the tree the tree withers and dies, 
but the spirit is immortal.* In the Jatakas these Tree-spirits play a great 
part, being worshipped with perfumes, flowers, and food. They dwell in 
many kinds of trees but the Banyan seems most popular. The scarlet- 
flowered silk-cotton tree « and the Sal tree as well as the Pipal retain their 
sanctity to-day. The goddess of the Sal is worshipped as giver of rain by 

1 The oldest accounts make no mention of the Sans. Lit., p. 146. 

tree under which the enlightenment took place. 3 Chand, vi. 11 ; Kathaka, v. 7. 

Rhys-Davids, Buddhist India, p. 230. * Sleeman, Rambles and Recollections (1015), p. 

*Zmaner,Altindisches Leben,p.58;Macdoimell, 385. 


the Oraons of Chota Nagpur, 1 and in South Mirzapur the Korwas place 
the shrine of Dharti Mata under its branches.* In the Jatakas more 
than once animal and even human sacrifices are spoken of in connexion 
with tree-worship. 3 To-day the slaughter of cocks and goats is added to 
the more usual offering of flowers and sweetmeats, in extreme cases of 
propitiation. 4 

The character and functions of these deities correspond closely to those of The Mother- 
the Mother-goddesses of Southern India. Among these are Mariamma, god- G °ddesses. 
dess of small-pox, Kaliamma, of beasts and forest demons, Huliamma, a tiger 
goddess, Ghantalamma, she who goes with bells, and Mamillamma, she who 
sits beneath the mango-tree.* However, it is usually made plain that these 
are but different names for the one great goddess. In Brahman hands this 
female pantheon appears as the Ashta Sakti or eight female powers. But 
a more primitive group is that of the Sapta Kannigais or seven virgins, 
tutelary deities of tanks. In Mysore, too, is found a similar group of seven 
sister-goddesses, vaguely identified with the Sivait mythology. However, 
they and all the Mother-Goddesses are distinguished from the true gods of 
Brahmanism by the fact that they are acknowledged to be local in their in- 
fluence warding off or inflicting calamities of various kinds, but strictly 
limited in their sphere of action. Still more limited are the powers of tanks, 
trees, and groves which periodically are alternately propitiated and exorcized, 
but are, as a whole, unsubstantial in personality and short lived. 

It is against this complex background of creed and culture that Indian Indian 
philosophy and Indian art, and all things Indian, must be viewed. Here lies Philosophy, 
the origin of the lovely treatment of flower and fruit at the hands of Indian 
sculptors and painters, and also of the imagination that kindled their vision 
and gave such dynamic power to their designs. 

Indian philosophy begins with Vedic speculations, or rather questionings 
as to existence and the creation. The unformulated philosophy of the 
Upanishads sprang from these and from it the pantheistic Vedanta system 
was evolved. As a foil to this existed from early times the atheistic Sankhya 
system, upon the reasoning of which Buddhism and Jainism were founded. 
At the root of everything lies the acceptance of metensomatosis 6 and a cycle 
of existences [samara], modified only by karma, past action, called adrishta, 
the unseen. At the root is ignorance, avidya. From ignorance comes desire, 
which leads to action, so the wheel revolves within the wheel. The Vedanta 
doctrine derived from the Upanishads taught the absolute identity of the 

1 Dalton, Descriptive Ethn., p. 201. the mango, sacred basil, and jasmin being com- 

2 Crooke, Popular Religion, i. 32. monly chosen. 

3 Nos. 472, 488. 5 Whitehead, Village Gods, p. 23. 

4 Marriages of fruit-bearing trees, and between 6 It appeared during the period of the Upani- 
women and trees, are an extension of this subject : shads. 

32* c 


individual soul with the spirit of the universe — 'That is the Eternal in which 
space is woven and which is interwoven with it. . . . There is no other seer, 
no other hearer, no other thinker, no other knower. . . .' l From this identi- 
fication of the mortal, limited self with the eternal and universal sum of all 
things arose 2 the idea of the illusion [maya] of the world of sensual experience. 
Only when the illusion of experience ceases, as in dreamless sleep, can the 
lesser self reunite with the universal self. This implied duality is in fact 
itself an illusion. Desire and action are inherent in such an illusion and the 
consequence is samsara. But Knowledge disperses the Illusion. 'Whoever 
knows this: "I am Brahma", becomes the All. Even the gods are not able 
to prevent him from becoming it. For he becomes their Self.' 3 

The Sankhya system is atheistic and dualistic, admitting matter and the 
individual soul as eternal but essentially different. In the absoluteness of 
this division lies release. The soul being removed from all matter, con- 
sciousness must cease, and the bondage to pain, in which term pleasure is 
included, be ended. 
Buddhism. Both Buddhism and Jainism presuppose the existence of the Sankhya 
philosophy. But it is evident that the sixth century B.C. when both Gautama 
and Vardhamana lived and taught was a period of extensive mental activity 
of an extremely sophisticated kind. The Brahma-Jala Sutta mentions 
Eternalists, Non-Eternalists, Semi-Eternalists, Fortuitous originists, and 
Survivalists, and also certain recluses and Brahmans who as dialecticians are 
typified as Eel wriglers. Buddhism is as much in revolt against this mental 
complexity as against the ritual complexity of the Brahman priest-craft. With 
regard to generalities its position is Agnostic. The Three Marks of Im- 
permanence, Pain, and Lack of Individuality must be considered as a 
practical summary of the characteristics of life. Upon these the doctrine of 
the Four Noble Truths, the essence of Buddhism, is founded: — Suffering 
exists; ignorance and desire are its causes; release is possible; the means are 
the Eight Points of Doctrine = right knowledge, right aspiration, right speech, 
right conduct, right living, right endeavour, right mindfulness, and right 
meditation. Throughout the teaching uncertain, empirical opinion [ditthi] 
is set apart from true wisdom [panna]. Above all, the cultivation and regula- 
tion of the will is stressed in an entirely new way. 4 

Lastly, as against the changing, foundationless illusions of the unregu- 
lated personal life in a universe that can only be described in terms of change, 
the Buddhist Doctrine [Dharma] is held out as being well-founded in time 
or rather in human experience. It is described as an ancient path well- 
trodden, a claim that paves the way to the conception of not one Buddha 

1 Brihadaranyaka Upanishad, III. viii: trans, by 3 Brihadaranyaka Up., I. iv. 6, trans. Mac- 
Macdonnell . donnell. 

2 Svetasvatara Up., IV. io. « Rhys- Davids, Buddhist Dialogues, i. xxvi. 


but many Buddhas. At Bharhut and Sanchi the seven Buddhas of the canon 
are all found, symbolized by their respective trees. 

This doctrine of wise renunciation was preached by Gautama, a prince Jainism. 
of the Sakya clan, who renounced his worldly heritage in pursuit of truth. 
Much of the adverse criticism which Buddhism has been subjected to has 
been due to a misunderstanding of Nirvana, the goal of all Indian speculation. 
Buddhism has had a complex history. Divided into two main sects, that of 
the Lesser and that of the Greater Vehicle, and changed beyond recognition, 
it exists no longer in the land of its origin. The Jain faith preached by 
Vardhamana, a contemporary and therefore rival of Gautama, 1 still persists 
in India. He, too, was of the Kshattriya race, and renouncing his birth-right, 
eventually attained Wisdom, appearing as the leader of the Nirgrantha 
ascetics. According to Jain tradition Vardhamana, or Mahavira, as he came 
to be known, was the twenty-fourth of a series oijinas or conquerors of the 
world. Like Buddhism the Jain faith opposes the exclusiveness of Brahman- 
ism by a claim to universality. Like Buddhism it is founded upon the teach- 
ing and achievement of Right Faith, Right Knowledge, and Right Action. 
Unlike Buddhism asceticism % is greatly stressed even to the point of volun- 
tary death by the refusal of nourishment on the part of those who have at- 
tained the highest knowledge, the Kevalin. From an early date two Jain 
sects have existed, the Digambara, who regard nudity as indispensable to 
holiness, and the Svetambara or * white-clothed', who do not. Besides these 
two bodies of ascetics, the Faith is extended to a large body of laity, who are 
represented in the history of Indian art, by many sculptures dedicated in 
the Kushan era, and by the magnificent medieval temples at Mount Abu, 
Girnar, and Satrunjaya. Like the Buddhists the Jains founded many 
monasteries. The worship of stupas was also included in their rites. 

The cult of the Upanishads and its forest-dwelling adherents is described Hinduism, 
in the Agganna Suttanta? 

'They making leaf-huts in woodland spots, meditated therein. Extinct for them the 
burning coal, vanished the smoke, fallen lies the pestle and mortar; gathering of an 
evening for the evening meal, they go down into the village and town and royal city, 
seeking food. When they have gotten food back again in their leaf-huts they meditate/ 

But from forest-life and meditation many sank to a mendicant life on the 
outskirts of the towns and to being mere repeaters of the sacred books. 4 
Such were the Brahmans of the Buddha's day. 

Modern Hinduism is divided into two main cults, Vaishnavism and Saivism. 

1 See Buhler, Indian Sect qfthejainas. 4 Comment: 'compiling the three Vedas and 

2 Dialogues of the Buddha, Pt. I, p. 218. See teaching others to repeat them.' See Jataka, 
Kassapa-Sihanada Sutta for the Buddhist view. No. 94: 'Beside no fire, but all afire within, 

3 Dialogues of the Buddha, Pt. Ill, p. 88. Naked the hermit wrestles for the Truth.' 


From the point of view of Indian art the early period is almost entirely 
Buddhist, while the Gupta period, and the succeeding medieval period are 
Brahmanical, the sculpture of the latter period being radically based upon 
Brahmanical iconography. 

Rudra, the storm-god of the Vedas, is made known by many epithets. He 
is called Girisa, 'lying on a mountain', Kapardin, 'wearer of tangled locks', 
and Pasupatihj 'lord of cattle'. When appeased he is known as Sambhu or 
Samkara, 'the benevolent', and as Siva, 'the auspicious', but he remains lord 
of the powers of the universe and is to be feared as well as loved. Yet the 
element of bhakti, of personal adoration and willing self-surrender to the 
deity is not wanting in the worship of the Great Lord as unfolded in the 
later Upanishads. 1 

In a lesser aspect Siva is lord of spirits [bhutas] and his rites are connected 
with snake-worship. In his worship the central object is the phallus [linga]. 
The Siva-littga does not seem to have been known to Patanjali, nor does it 
appear on the coins of Wema-Kadphises on the reverse of which the god is 
represented, holding the trident, with the bull, Nandi, in the background. In 
the Mahabharata, Siva is represented as dwelling in the Himalaya with his 
hosts. His vehicle is the bull and his consort is variously known as Uma, 
Parvati, Durga, and Kali. Having completed the creation, he turned yogi 
and the phallus became his emblem. 

The earliest lingas existing do not ante-date the Kushan period. They 
are of the kind known as Mukha-lingas with one or more faces at the top 
of the member. One of the earliest iconographical representations of the god 
is the Dakshinamurti in relief on one side of the Vishnu Temple at Deogarh 
[Plate 48] which may be dated in the second half of the fifth century a.d. 

The earliest historical records of Vaishnavism are the Besnagar Heliodora 
inscription and the Ghosundi inscription, both of the second century B.C. 
The former testifies to the erection of a Garuda pillar to Vasudeva, god of 
gods. Heliodora , who was the son of Diya and a native of Taxila, was 
ambassador from the Yavana Amtalikita [Antialkidas?] to Bhagabhadra. He 
calls himself Bhagavata. The Ghosundi inscription witnesses to the 
erection of a hall of worship to Samkarshana and Vasudeva. 

Vishnu is a Vedic deity and although he is represented by but few hymns, 
his personality is vividly portrayed. He measures all things with his three 
wide strides, the third passing beyond human discernment to the high places 
of the deity. 2 This conception of the third step of Vishnu as the highest 
heaven and goal of all things, had obviously much to do with his elevation 
as Supreme Being. In the Mahabharata this Supreme Being is addressed as 
Narayana, Vasudeva, and Vishnu. 

1 Svetasvetara Upanishad. a R. V., i. 155. 5. 


Later Vishnu found a more intimate place in popular worship by means of 
his ten incarnations [Avataras]. 

The earliest iconographical presentations of the god are two standing, 
four-armed figures, one on either side of the door-guardians of the Chandra- 
gupta Cave at Udayagiri [a.d. 401]. 

Unlike Buddhism and Jainism, the Hindu sects are not organized into Hindu 
definite congregations. Whatever the shrine be, one of the magnificent archltect we. 
temples of Bhuvanesvar or Khajuraho, or a red daubed stone by the roadside, 
the worship is individual. For certain ceremonial purposes the aid of priests 
is sought, and all the larger temples have their hosts of attendants. But there 
is never a congregation worshipping in unison. Architecturally speaking, the 
Hindu shrine is the dwelling-place of the god, although various pavilions or 
porches dedicated to the preparation of the offerings or to music and dancing 
stand before it. 

The earliest structural Hindu shrines existing are the flat-roofed Gupta 
temples, square in plan with a verandah supported by four pillars, 1 the door- 
way being elaborately carved. At Ajanta the cell in the centre of the back 
wall of the oblong, many pillared caves, is cut on exactly the same plan, the 
doorways corresponding very closely. The introduction of the linga shrine 
at Badami and Ellora eventually altered the plan radically by placing the 
shrine in the body of the hall as at Elephanta. The great medieval temples 
consist of high-towered shrines, each with its entrance-pavilions. 

As portrayed in the Brahma-Jala Sutta, primitive Buddhism gave no place Indian 
to aesthetics, for music, song, and the dance are classed with sorcery and iconography, 
cock-fighting as minor examples of foolishness, unprofitable to the wise. 
Manu and Chanahya also adopt this slighting attitude towards the arts. How- 
ever, that is of little account, and Bharhut and Sanchi are not less fine because 
they are not supported by the argumentative analysis of the schoolmen. The 
art of the Early Period is a spontaneous growth, endued with native virility. 
Essentially narrative, it is vividly perceptive. The history of Indian art must 
be written in terms of the action of a literary, metaphysical mode of thought 
upon this naive, story-telling art, resulting in the formation of an immense 
and intricate iconography. Around this iconography has grown a still more 
abstruse, secondary literature, in which the least variation of detail is seized 
upon to sanction the subdivision and endless multiplication of types of icons. 

Images are roughly divided into two classes, the fixed and the movable 
[Achala and Chala]. They are likewise roughly described as standing 
[Sthanaka], sitting [Asana] or reclining [Say ana]. Also they may further be 
described in terms of the nature of the manifestation: as terrible [Ugra] as 
is Vishnu in his Man-Lion incarnation, or pacific [Santa]. The images of 

1 As at Sanchi and Tigowa. 


Vishnu are further classified according to their natures as Yoga, Bhoga, and 
Vita, to be worshipped respectively according to the personal desires of the 

This classification of gods and devotees according to their innate natures 
refers directly to the classification by natures of the Sankhya philosophy, 
primeval matter being distinguished by the three properties [Gunas] of 
Light [Sattva], Might [Rajas], and Darkness [Tamas]. It is clear that the 
needs of the worshipper specify the type of the image worshipped. Complex 
manifestations, whose many attributes are symbolized by their many hands 
are considered Tamasic in character, and their worshippers of little under- 
standing. To the wise images of all kinds are equally superfluous. 
Indian Indian aesthetics must be regarded as being of late date, a supplement to 
aesthetics, the iconographical literature of the medieval period. Much of the Agamas 
is of great iconographical interest, but these late literary canons have no 
aesthetic light to shed, although they do indicate something of the religous, 
hieratical atmosphere which deadened artistic creation in the last period of 
medieval decadence. 1 Indian aesthetics are based upon the conception of 
aesthetic value in terms of personal response or reproduction. This value is 
known as Rasa, and when it is present the object is said to have Rasa [Rasa- 
vant] and the person to be Rasika or appreciative. Rasa produces various 
moods in the Rasika varying in kind according to the initial stimulus ; from 
these moods emotions spring. The mechanics of this system is worked out 
in detail in the Dhanamjaya Dasarupa and the Visvanatha Sahitya Darpana. 
The whole system is based upon and illustrated by literature, and cannot be 
applied directly to sculpture and painting. 2 

1 Rao, Hindu Inconography> Pt. I, vol. i, p. 31. a See Coomaraswamy, Dance of Siva, p. 30. 

Chapter Two 

A short time after the death of Alexander in 323 B.C. the throne of The 
Magadha or Bihar, then the premier kingdom of Northern India, was Mauryan 
seized by Chandragupta, surnamed The Maurya, known as Sandro- I) y nast y- 
kottos to Greek authors. In the course of a victorious reign of twenty-four 
years this able prince caused his influence to be felt over all India, at least 
as far south as the river Narbada, and acquired from Seleukos Nikator, first 
his enemy and then his ally, the valuable provinces lying between the Indus 
and the Hindu Kush mountains which now constitute the major part of the 
kingdom of Afghanistan. 

Chandragupta was succeeded by his son Bindusara, who, in or about 
273 B.C., transmitted the imperial sceptre to his son, Asoka, the third and 
most renowned sovereign of the Maurya dynasty. For forty-one years (273- 
232 B.C.) Asoka ruled his immense empire with great power and might, main- 
taining friendly relations with his neighbours, the Tamil states of the extreme 
south and also with thf island kingdom of Ceylon and the more remote Greek 
monarchies of Macedonia, Epirus, Western Asia, Egypt, and Cyrene. 

Early in life the emperor became a religious convert and as the years rolled Asoka's 
on his zeal increased. Finally, his energies and riches were devoted almost E at ?^ ageof 
entirely to the work of honouring and propagating the teaching of Gautama Buddhism - 
Buddha. With one exception he abstained from wars of conquest and was 
thus free to concentrate his attention upon the task to which his life was 

The imperial palace at Pataliputra, the modern Patna, the capital of Early 
Chandragupta Maurya is described by Greek and Roman authors as excelling wooden 
in splendour the royal residences of Susa and Ekbatana. Although no vestige arc tecture - 
of such a building has survived, with the possible exception of some brick 
foundations, there is no reason to doubt the statements of the historians. 
The result of much excavation seems to support the literary evidence that 
Indian architects before the time of Asoka built their superstructures chiefly 
of timber, using sun-dried brick almost exclusively for foundations and 
plinths. No deficiency in dignity or grandeur was involved in the use of the 
more perishable material; on the contrary, the employment of timber enables 
wide spaces to be roofed with ease which could not be spanned with masonry, 
especially when, as in India, the radiating arch was not ordinarily employed 
for structural purposes. 

Excavations of widely spread sites dating from the Maurya to the Gupta Stone 
periods, and even later, emphasize the fact that timber and unburnt brick buildings. 
were the standard architectural materials of ancient India, mud being used as 
it still is, for ordinary, domestic work. However, Asoka is credited by the 


literary sources with the use of masonry in the many building activities 
reported of him. It is on record that during his reign of about forty-one years 
he replaced the wooden walls and buildings of his capital by more substantial 
work and caused hundreds of fine edifices in both brick and stone to be 
erected throughout the empire. So astonishing was his activity as a builder 
that people in after ages could not believe his constructions to be the work 
of human agency, and felt constrained to regard them as wrought by familiar 
spirits forced to obey the behests of the imperial magician. Few sites can, 
however, be definitely ascribed to the Asokan or even to the Mauryan period. 
No building with any pretensions to be considered an example of architecture 
can be assigned to any earlier period than this, with which the history of 
Indian architecture as of the other arts begins. 

Beginnings The Mauryan emperors must surely have built palaces, public offices, and 
of Indian temples suitable to the dignity of a powerful empire and proportionate to the 
art ' wealth of rich provinces, but of such structures not a trace seems to survive. 
The best explanation of this fact is the hypothesis that the early works of 
Indian architecture and art were mainly constructed of timber and other 
perishable materials, ill-fitted to withstand the ravenous tooth of time. What- 
ever the true explanation of this may be the fact remains that the history of 
Indian art begins with Asoka. 'But', as Professor Percy Gardner observes, 
'there can be no doubt that Indian art had an earlier history. The art of 
Asoka is a mature art: in some respects more mature than the Greek art of the 
time, though, of course far inferior to it, at least in our eyes/ * 

Its origins. We can affirm with certainty that the forms of Asokan architecture and 
plastic decoration were descended from wooden prototypes, and may also 
discern traces of the influence of lost works in metal, ivory, terra-cotta, and 
painting. The pictorial character of the ancient Indian reliefs, histoires 
sans paroles, is obvious, and the affinity of much of the decorative work 
with the jeweller's art is equally plain. The sculpture on a pier of the southern 
gate at Sanchi was actually executed by the ivory-carvers of the neighbouring 
town of Vedisa (Bhilsa). 8 We may, moreover, feel some confidence in affirm- 
ing that the sudden adoption of stone as the material for both architecture 
and sculpture was in a large measure the result of foreign, perhaps Persian, 
example. The fuller consideration of the foreign influences affecting Indian 
art will be more conveniently deferred and made the subject of a separate 
Personal Whatever the foreign elements of ancient Indian art may have been, great 
^ at * ve weight must be allowed for the personal initiative of Asoka, a man of marked 
so a ' originality of mind, capable of forming large designs and executing them with 
imperial thoroughness. The direction taken by Indian art was like the diffu- 

1 Trans. 3rd Inter. Congress for the Hist, of a -B/>./»rf.,ii.92,378,insr.§200ofStupaL:C.i89: 
Religions (Oxford, 1908), vol. ii, p. 81. Buhler. 


sion of Buddhism, determined in its main lines by the will of a resolute and 
intelligent autocrat. 

Like most of the extant works of early Indian art, the Mauryan columns Early art 
and caves were executed in honour of Buddhism, which became the state nearl y ?U 
religion in the empire of Asoka and is said to have been introduced during his Buddhist - 
reign into independent Ceylon. Although we know that both Jainism and 
Brahmanical Hinduism continued to attract multitudes of adherents during 
the Mauryan period, hardly any material remains of works dedicated to the 
service of those religions have survived. 

The monuments which can with certainty be dated in Asoka's reign are Inscriptions, 
not very numerous, but it is not improbable that more may be discovered, columns, and 
His buildings having perished, our direct knowledge of the art strictly con- scul P ture - 
temporary with him is derived from his inscriptions, the carving and sculp- 
tures on his monolithic columns, certain caves, and a few fragments of 
pottery excavated at Mauryan level. 1 The inscriptions are worthy of being 
mentioned among the Fine Arts on account of their beautiful execution, for 
nearly all are models of careful and accurate stone cutting. The most faultless 
example is the brief record on the Rummindei Pillar, which is as perfect as on 
the day it was incised. 2 The craft of the skilled mason and stone-cutter, so 
closely akin to fine art, reached perfection in the days of Asoka, as appears 
from every detail of their work and especially from an examination of the 
beautifully polished surface of the monoliths and the interiors of the 
cave-dwellings dedicated by him and his grandson, Dasaratha, in the hills 
of Bihar. 

Isolated pillars, or columns, usually associated with other buildings, and Monolithic 
frequently surmounted by a human figure, animal sculpture, or sacred pillars of 
symbol have been erected in India at all times by adherents of all the three so a ' 
leading Indian religions. The oldest are the monolithic pillars of Asoka, who 
set up at least thirty of these monuments, of which many survive in a more 
or less perfect state. 3 Ten of these bear his inscriptions. The Lauriya- 
Nandangarh monument, inscribed with the first six Pillar Edicts, and practi- 
cally uninjured, is shown on Plate i. The shaft of polished sandstone, 32 feet 
9i inches in height, diminishes from a base diameter of 351 inches to a 
diameter of only 22£ inches at the top— proportions which render it the most 
graceful of all the Asoka columns. The uninscribed pillar at Bakhira in the 
Muzaffarpur District, in perfect preservation, and presumably of earlier date, 
is more massive and consequently less elegant. The fabrication, conveyance, 

1 With regard to archaeological method in India research purposes. 

it is to be regretted that the evidence of the a Asoka, the Buddhist Emperor of India, VI. ii. 
pottery of the various sites and levels has not 3 See the author's paper on the subject, 
been given more prominence. Nor is it always Z. D. M. G. for 191 1. 
clear where such lesser finds are deposited for 
3202 d 


and erection of monoliths of such enormous size — the heaviest weighing about 
fifty tons — are proofs that the engineers and stone-cutters of Asoka's age 
were not inferior in skill and resource to those of any time or country. 

The capitals of these pillars provide excellent evidence of the state of 

the art of sculpture, both in relief and in the round, during the period 

between the year 250 B.C. and the end of the reign of the great emperor 

in 232 B.C. 

Sculptured The capital of each pillar, like the shaft, was monolithic, comprising three 

capitals, principal members, namely, a Persepolitan bell, abacus, and crowning sculp- 
ture in the round. The junction between the shaft and the abacus was marked 
by a necking, the edge of the abacus was decorated with bas-relief designs, 
and the crowning sculpture was occasionally a sacred symbol, such as a wheel, 
or more commonly a symbolical animal, or group of animals. Sometimes 
the inanimate and animal symbols were combined. 
Forms of Within the limits thus determined the artists enjoyed considerable latitude, 

abacus, and in consequence the surviving capitals vary widely in detail. The abacus 
might be either rectangular or circular so as to suit the form of the sculpture 
above. The edge of the abacus of the beautiful Lauriya-Nandangarh pillar 
is decorated by a row of flying sacred geese in quite low relief. The abaci 
of the pillars at Allahabad and Sankisa (Plate 3 a) and the bull pillar at 
Rampurva exhibit elegant designs composed of the lotus and palmette or 
honeysuckle. Whatever the device selected, it is invariably well executed, 
and * chiselled with that extraordinary precision and accuracy* which charac- 
terize the workmanship of the Maury a age, and have never been surpassed 
in Athens or elsewhere. 1 

Animal The topmost sculpture in the round was most often one or other of four 
symbolism, animals^-namely, the elephant, the horse, the bull, and the lion. All these 
animals, except the horse, are actually found on the round on extant capitals, 
and it is recorded that a horse once crowned the pillar at Rummindei, the 
Lumbini garden. On the sides of the abacus of the Sarnath capital (Plate 2) 
all the four creatures are carved in bas-relief. 2 

The elephant of the Sankisa capital is well modelled, but unhappily has 
been badly mutilated. The two pillars at Rampurva bear respectively the 
bull and lion. 3 
The The magnificent Sarnath capital discovered in 1905, unquestionably the 

Sarnath best extant specimen of Asokan sculpture, was executed late in the reign 

capi a • between 242 and 232 B.C. The column was erected to mark the spot where 

1 Sir John Marshall, in Ann. Rep. A. S., India, Ceylon. See Vincent Smith: The Monolithic 
1906-7, p. 89. Pillars of Asoka\ in Z. D. M. G. y 191 1. The 

2 These animals may symbolize the four corners lion was also regarded as a symbol of Buddha 
of the world, which explanation of the symbolism himself. 

was suggested by the discovery of rude sym- 3 J. R. A. S. t 1908, p. 1085, PL I. 
bolical bronze figures of the four animals in 

PLATE I. Inscribed Asoka pillar, Lauriya-Nandangarh 

PLATE 2. Capital of inscribed Asoka pillar at Sarnath 


Gautama Buddha first 'turned the wheel of law\ or in plain English, publicly 
preached his doctrine. 1 The symbolism of the figures, whether in the round 
or in relief , refers to the commemoration of that event for the benefit of the 
Church Universal. The four lions standing back to back on the abacus once 
supported a stone wheel, 2 feet 9 inches in diameter, of which only frag- 
ments remain.* 

It would be difficult to find in any country an example of ancient animal Sanchi 
sculpture superior or even equal to this beautiful work of art, which success- edict-pillar 
fully combines realistic modelling with idealistic dignity, and is finished in capitah 
every detail with perfect accuracy. The bas-reliefs on the abacus are as good 
in their way as the noble lions in the round. The design, while obviously 
reminiscent of Assyrian and Persian prototypes, is modified by Indian senti- 
ment, the bas-reliefs being purely Indian. Sir John Marshall's conjecture 
that the composition may be the work of an Asiatic Greek is not supported 
by the style of the relief figures. The ability of an Asiatic Greek to represent 
Indian animals so well may be doubted. 

The only rival to the artistic supremacy of the Sarnath capital is the replica The 
which once crowned the detached pillar at Sanchi engraved with a copy of Bakhira 
the Sarnath edict denouncing schism. The Sanchi capital is decidedly P iIIar ' 
inferior to that at Sarnath, but it is possible that both works may proceed 
from the hands of a single artist. 3 A century or so later, when an inferior 
sculptor attempted to model similar lions on the pillars of the southern gate- 
way at Sanchi, he failed utterly, and his failure supports the theory that the 
Sarnath capital must have been wrought by a foreigner. Certainly no later 
sculpture in India attained such high excellence. 

The perfection of the Sanchi and Sarnath lions on the edict-pillars must 
have been the result of much progressive effort. The uninscribed pillar at 
Bakhira (Plate 3 c) seems to be one of the earlier experiments of Asoka's 
artists. The clumsy proportions of the shaft contrast unfavourably with the 
graceful design of the Lauriya-Nandangarh column (ante, Plate 1), which 
bears a copy of the Pillar Edicts, and may be dated in 242 or 241 B.C., while 
the seated lion on the summit is by no means equal to the animals on the 
edict-pillars of Sarnath and Sanchi erected between 242 and 232 B.C. I am 
disposed to think that the Bakhira column was set up soon after 257 B.C., 

1 The wheel is one of the earliest Buddhist Beginnings of Buddhist Art, p. 70, note 88. 

symbols and with the Tree and the Stupa 2 Discovered by Mr. F. O. Oertel and described 

appears everywhere at Bharhut and Sanchi. by him in Ann. Rep. A. S., India, 1904-5, pp. 

Foucher associates these three symbols which 68-70, PI. XX. His account of the excavations 

obviously represent the First Sermon in the has been reprinted in a separate volume entitled 

Deer Park, the Enlightenment at Bodh-Gaya, and Buddhist Ruins of Sarnath near Benares, 1904-5. 

the Death at Kusinagara with a fourth, the God- 3 Maisey, Sanchi and its Remains, PI. XIX, 2; 

dess Sri and the Bhadra-Ghata (vase with Codrington, Ancient India, PI. I. 
lotuses), representing the birth at Kapilavastu. 


the date of the earliest Rock Edicts. It must also be noted that at Rampurva 
there are two pillars only one of which is inscribed. In the Sahasram inscrip- 
tion it is clearly stated that edicts are to be inscribed on rocks, or on pillars 
wherever a stone pillar is standing, which suggests that some of these 
pillars may considerably antedate Asoka's reign, although their technique 
is obviously one with the inscriptions and caves, and they are clearly 

1 Camb. Hist, of India, p. 501. 

A. Capital of Sankisa pillar 

H. The Hcliodoros pillar, 

C. Asoka pillar at Bakhira, Muxaffarpur District 




■'$;'//*' f 


R. Rharhut ; medallion 

A. Sudarsana Yakshi, 

1). Rharhut; medallion 

C Yakshi, Iiharhut 

Dp**,. fli&^**7i»ir iff ' hf r — .iirrSwB 

» ; 7 mw £, # ■Zzrx**&tf£ 

^ 'V '- V, ■" 

K. Rharhut ; jataka scenes on coping 

*:*&&?: *?* • ' ITr-SfSS^ 

PLATE 6. Hharhut, inner view of eastern gateway 

Chapter Three 


AFTER the death of Asoka the empire broke to pieces, but his Maurya Early 
£A descendants continued to rule the home provinces for about half a architecture. 
x -^century, at the end of which they were superseded by the Sunga kings 
who governed parts of Northern India until the beginning of the first century 
B.C. However, the style of architecture, decoration, and sculpture which 
perhaps first assumed a permanent form under the patronage of Asoka con- 
tinued in use up to about the close of the first century of the Christian era, 
forming a distinct and definite period in the history of Indian art. 

Although Buddhism at this period, approximately extending from 273 B.C. 
to a.d. 100, was by no means the only religion in India, it enjoyed a dominant 
position as the result of the great Buddhist emperor's propaganda, and the 
monuments remaining, therefore, are almost all Buddhist, though few are as 
early as the reign of Asoka. The huge mass of solid brick masonry known as 
the great stupa of Sanchi, later encased with stone, may belong to his reign, 
as well as several other similar structures, but most of the buildings that now 
survive are of a later date. 

The ancient civil buildings having all perished utterly, except the tangle of Architectural 
superimposed foundations that is all that the spade lays bare at most of the forms - 
early sites, the story of Indian architecture must therefore be reconstructed 
from the somewhat one-sided evidence of the temples and shrines, and the 
bas-reliefs that adorn them. The most characteristic early architectural com- 
positions were stupas, with their appurtenant railings and gateways, monas- 
teries, and churches, the ' chatty a-ha\\s } of Fergusson. The monasteries and 
churches include both rock-cut and structural examples. Isolated pillars also 
were frequently set up. 

Stupas or 'topes', the Dagabas of Ceylon — solid cupolas of brick or stone Stupas, 
masonry — were constructed either for the safe custody of relics hidden in a p*gaba, or 
chamber near the base, or to mark a spot associated with an event sacred in topes ' 
Buddhist or Jain legend. Until a few years ago the stupa was universally 
believed to be peculiarly Buddhist, but it is now a matter of common know- 
ledge that the ancient Jains built stupas identical in form and accessories 
with those of the rival religion. However, no specimen of a Jain stupa is 
standing, and our attention may be confined to the Buddhist series. 1 The 
earliest stupas were of unburnt bricks like the Bharhut stupa. The great 
stupa at Sanchi was originally of this type, a casing of roughly trimmed 
masonry and a ramp forming an upper procession-path being added later. 

1 See the bas-relief of a Jain stupa from Mathura, PI. 15 B. 


This Stupa as it appeared before restoration is shown in Plate 4. As time 
went on, the originally hemispherical dome was raised on a high drum or tier 
of drums, and so by a series of gradual amplifications the ancient model was 
transformed first into a lofty tower after the kind of Kanishka's Stupa at 
Peshawar, described by Hiuen Tsiang, and ultimately into the Chinese 

r a el 

tyLjUfetfoV W4W*f5J: 

1 -ft Q * * S *(■« 

Sanchi railing. 

lin stupas The most ancient stupas were very plain. They were usually surrounded 
rai mgs. by a s tone railing, sometimes square in plan, but more often circular, marking 
off a procession-path for the use of worshippers and serving as a defence 
against evil spirits. The earliest examples of such railings, at Sanchi, are 
unadorned copies of wooden post-and-rail fences. The bars of the railing 
were usually lenticular in section, inserted in the posts as shown in the 
diagram. At Besnagar another form of ancient railing has been unearthed, 
consisting of oblong slabs held by grooved uprights. 1 

Bharhut and Sanchi represent two sequent stages in the development of 
the stupa of the Early (post-Mauryan) Period. They and their appurtenances 
had become more ornate. Sculpture was freely applied to every member of 
the railing— to the posts, rails, and coping. Late in the second century of 
the Christian era at Amaravati the railing was transformed into a screen 
covered with stone pictures in comparatively low relief 2 but with the richest 
effect. The openings giving access to the procession-path inside the railing 
were dignified by the creation of lofty gateways (torana) copied from wooden 

1 A. S. /., 1913-14. of rock-cut Viharas as seen in the Gautamiputra 

2 The same development applies to the railings Cave (No. 3) at Nasik. 


models, and covered with a profusion of sculpture. The best examples of 
such gateways are those at Sanchi. 

The origin of the stupa lies in primitive burial ceremonies for they are 
primarily tombs 1 like the 'iron-age' cairns of the south and such tumuli as 
those excavated by Bloch near Nandangarh in the Champaran District. 2 
Originally mounds of earth, the earliest stupas existing are of unbaked brick, 
hemispherical in shape. Although their first object was the enshrinement 
of sacred relics, in later times they acquired a symbolical value and many 
cenotaphs were built, the dedication of miniature stupas of stone or clay being 
customary at the great shrines. This idea of the symbolic value of stupas 
and the merit of sft//>a-building, on the part of the faithful, apart from the 
relics they might or might not contain, is to be found at the root of the 
legendary accounts of Asoka's ten-thousand stupas. Fa-hian says that in 
monasteries it was customary to raise stupas to Mudgalaputra, Sariputra, and 
Ananda, as well as in honour of the Abhidharma> Vinaya, and Sutra, such 
stupas in fact being regarded as altars. 3 The word chatty a is indeed often used 
where a stupa is intended, in the sense of a shrine or holy place. So Anatha- 
pindika builds Sariputra's Chatty a which was four stories high, decreasing in 
size, and which contained a relic vase, and was surmounted by a roof and many 
umbrellas. 4 

In the Dulva, 5 too, it is laid down that a Bhikshu's body is to be covered 
with grass and leaves and a chaitya raised over it. In a still more remote 
sense, the converted but disconsolate Queen Sivali raised stupas at the places 
where her ascetic husband had argued with her and finally convinced her. 6 
In medieval times the stupa with its pyramid of sheltering umbrellas is 
dwarfed in importance by the sculpture that adorns it. At Ajanta and Ellora 7 
and everywhere, in miniature at Bodh-Gaya, it is really nothing but a domed 
shrine, the tier of umbrellas being fused together into a spire. 8 

Stupas y not to speak of miniature votive models, varied greatly in size. Size of 
The very ancient specimen at Piprahwa on the Nepalese frontier, which may stu P as - 
possibly be earlier than Asoka, has a diameter of 116 feet at ground level, 
and stands only about 22 feet high. The diameter of the great Sanchi monu- 
ment at the plinth is I2i£ feet, the height about 77I feet, and the stone 
railing is a massive structure 1 1 feet high. Several monuments in Northern 

1 The use of the stupa as the symbol of the of the domical shape of these structures, con- 
Parinirvana indicates this. siders them to have no connexion with the 

2 Ann. Rep. A. S. y East Circle, 1908-9, p. 3. tumuli. He put forward the suggestion that the 

3 Beal, i. xxxviii. brick stupa was derived from 'the curved roof 

4 Rockhill, Life of Buddha, p. iii. of bamboos built over a primitive circular hut- 

5 xi. 53. shrine constructed of perishable materials' and 

6 Mahajanaka Jataka. quotes the Toda hut. At the same time he refers 

7 Visvakarma Cave. to similar Phoenician tombs at Anvinth (Perrot 

8 Vincent Smith, because of the cenotaphs and and Chipiez, Hist . de VArt, Phinicie, Figs. 94, 95). 


India, some of which were ascribed to Asoka, are recorded to have attained 
a height of from 200 to 400 feet; and to this day the summit of the 
Jetavanarama Dagaba in Ceylon towers 251 feet above the level of the 
ground. The larger monuments afforded infinite scope to the decorative 
Indian On the Bharhut bas-reliefs two types of buildings are to be found. The 
wooden g rst j g domed and round in plan. The second is barrel-roofed and sometimes 
itecture. ^^ stor j es ^jg^ This second type is the origin of the barrel-roofed chaitya- 
caves where the details of the octagonal pillars, the balcony railings and the 
arched doorways and windows are faithfully portrayed. At Sanchi the same 
types appear and also at Amaravati and Mathura. Shrines are shown in three 
instances and are all of one type. At Bharhut the Shrine of the Head-dress 
Relic, is circular in plan, closed in by a low railing but otherwise open on all 
sides. It has the usual ogee doorway, the arch of which is ornamented, above 
its beam-heads, with little rosettes. The semicircular part of the opening is 
filled in with the usual framework which served as a weather screen. The 
roof is dome-shaped and has a pointed finial. It is divided into two by a 
narrow clerestory opening which comes between the dome and the curved 
eave. In the centre on a stone platform technically known as a 'throne' 1 is 
a cushion bearing the sacred relic. The throne is ornamented with pendent 
garlands and is marked with the impressions of the right hands of devotees, 
a custom still common in India. 

The first scene of the conversion of Kasyapa of Uruvilva on the middle of 
the inner side of the left-hand pillar of the East gateway at Sanchi shows 
another shrine of this type (Plate 13 a). 2 This is the Shrine of the Black 
Snake which the Buddha eventually caught in his begging-bowl. Here the 
dome is broken by eight windows and is surrounded by a balcony railing. 

The famous shrine which Asoka built around the Bodhi-tree appears at 
Bharhut, Sanchi, Mathura, and Amaravati. At Bharhut it is sculptured on 
the Prasenajit pillar 3 and seems to consist of a barrel-roofed colonnade, 
circular in plan entirely surrounding the tree. The upper story is provided 
with many windows and a balcony railing. At Sanchi this same building is 
accurately reproduced on the front of the left pillar, and again on the outside 
of the lower architrave, of the East gateway, where it is the centre of a huge 
host of pilgrims. At Mathura it also appears on an architrave of Kushan 
date 4 and again in a slightly amplified form at Amaravati. 5 Here other build- 
ings have arisen around it and to one side is a gateway {tor ana). These gate- 
ways were apparently used everywhere, for secular purposes as well as 
ecclesiastical, for on the middle architrave of the East gateway at Sanchi, 

1 Asana. 3 Cunningham, Bharhut, vi, xii, and xxx. 

2 Burgess and Griinwedel, Buddhist Art in India, 4 Plate XII A and XIII B. 

P- 62. 5 Li 8t f Negatives Ind. Off., §2163, Serial 798. 


one appears as the entrance to a town through which a procession is passing 
beneath crowded windows and balconies. 

A survey of such scenes where buildings of two and three stories abound The use of 
accords with the colourful descriptions of the splendours of such towns of stone pilars. 
ancient India as Vaisali 1 or Pataliputra. Buildings of seven stories in height 
are even spoken of (Satta-Bhumaka-Pasada). Among the most famous of 
these piles was the Kutagara-Vihara at Vaisali, which Buddhaghosa describes 
as a storied building raised on pillars with a pinnacle, and like the chariot of 
the gods. 2 

Civil architecture is described in the Jatakas on almost as lavish a scale. 
The large houses had wide gateways leading into an inner courtyard with 
rooms opening into it on ground level. There were granaries and store-rooms 
and a treasury, but the flat roof, as at all times in the East, played a great 
part in the life of the house, at least during the day, being probably roofed- 
in to form an open-sided, airy pavilion. 

Plaster (chunam) was used everywhere to adorn these buildings, and as a 
base for painting. Yaksha figures were painted as door-guardians and certain 
decorative motives are also mentioned: wreath-work, five-ribbon work, 
Dragon's teeth work, and creeper- work. 3 

As has been said, nothing of these splendours has come down to us in any 
of the various sites that have been excavated. It is obvious, however, that 
the greater part of these structures was of wood and therefore perishable, as, 
indeed, layers of ashes testify in many places. It is noticeable that the pillars 
of the upper stories of the buildings depicted on the bas-reliefs are octagonal, 
usually without capital or base. The pillars on the ground floor are octagonal 
also but have heavy bells surmounted by animal-capitals or brackets, which 
suggests that the lower pillars were possibly of stone. On the right jamb of 
the East gateway at Sanchi are represented six superimposed stories, said by 
Grunwedel to represent the six Devalokas. The pillars of these structures 
are grouped in pairs, the lowest of each having bell-capitals, the upper being 
plain and leading up to the barrel-roof. There is a considerable difference 
between the proportions of the upper and lower pillars, which again suggests 
a difference in material. 4 

1 Fausboll, Dhammapada, p. 390; Lefman, Lali- likening stupas to the early circular shrines, but 
tavistara, p. 21 ; Rockhill, Life of the Buddha, he obviously is considering the medieval stupas 
p. 63 ; For the site : Vincent Smith, J.A.R. S., of Ajanta and Ellora, which he goes on to discuss. 
1902, p. 267 n.; Marshall, Ann. Rep. A. S., I, There the sanctity of the relic has long been 
I 9°3""4» P- 74- confused with the stupa itself as a symbol which 

2 See Law, Ksatriya Tribes, p. 43. again has been subjugated in ceremonial impor- 

3 Vinaya Texts, 3-170. tance b Y the Buddha figures sculptured upon it. 

4 See Rhys-Davids, Buddhist India, p. 66. Stone Hist, of Ind. Arch., VI, i, p. ,66. 

pillars and stone staircases are mentioned in the Simpson developed the idea of the bamboo 
Jatakas and a stone palace in Fairyland (6.269). origin of all Indian architectural forms, citing 
Vincent Smith develops Fergusson's idea of the Toda hut. The early shrines and halls are, 



Although monastic institutions in India were not confined to the Buddhists, 
the Buddhist Sangha attained a height of power and a detail of organization 
to which the Jain and Brahmanical communities never aspired; and in con- 
sequence, the buildings dedicated to the use of the Order were frequently 
designed on a scale of the utmost magnificence. The central and all important 
building of the early monasteries seems to have been the 
Sabha or hall of meeting of the community. Gateways, 
store-houses, kitchens, and well-houses are mentioned, but 
the actual cells of the monks were apparently a group of 
separate buildings. These, it seems, were built by the 
brethren themselves, among whom were many skilled archi- 
tects. 1 In the Jatakas it is said, however, that only the senior 
brethren had their own chambers, while the juniors slept in 
the Hall. Later the Buddha ordained that novices should 
be lodged with their supervisors for three days and then 
sent to their own place. 2 The forest-dweller's leafy hut is 
often portrayed in the early sculpture and many of the lesser 
dwellings of the monastery were probably of this type. The 
meeting-hall or service-hall must have been a common type 
of building in ancient India, for the Buddhist Sangha was 
by no means an innovation and can be directly compared 
to the hundred and one political and social corporations of 
the time. Every village, profession, and craft was organized 
into guilds which had their appointed places of meeting. 3 
The Mote-hall of the Licchavis (Santhagara) must have been a building of 
the same kind as the Assembly-hall of the Buddhists. 

Before the period of the rock-cut halls and cells like those at Bhaja and 
of later Bedsa, in Gandhara and in medieval India generally, the monasteries took 
times. U p a quadrangular form, the cells being built so that they faced inwards on 
the four sides of a courtyard. 

of course, wooden in construction and not lithic, 
but it is a hewn wood construction of joint and 
tenon. The origin of the form of the 'Indo- 
Aryan* steeple (sikhara) cannot be traced to this 
ancient art, still less to bamboos. It lies in the 
nature of the material accessible to the hand of 
the medieval builders, finely burnt brick and 
well-trimmed stone. It is a logical development 
of the Indian corbelling methods. J.R.I.B.A., 
VI, vii, p. 225. 
1 Cowell, Jat. % § 6. 
2 Cowell y Jaf. l §z6. 

3 Omitting the family council (Kula) and the 
probably later organizations of castes, among 
them were the Gana, Puga, Vrata, and Sreni of 

Sanskrit literature. The Sangha being the 
generic term, the Gana is specifically tribal, the 
Puga is based on common interests, financial or 
social, the Vrata is an organization of labour 
outside the crafts, while the Stent is essentially 
a craft or professional guild. The Setthi of 
Buddhist literature was a prominent member of 
these corporations whose officers (Seniyo) were 
the Pamukha (president) and the Jetthakas 
(aldermen). Political confederations such as that 
of the Licchavis are of the same origin, which 
is in fact the source of the whole system of 
political and social life in Northern India. 
Bhandarkar, Carmichael Lectures, 1918, p. 181 ; 
Mookerji, Local Government, p. 36 et seq. 


When such a quadrangle became multiple, through the addition of chapels, 
stupas, refectories, halls, churches, store-houses, and other buildings, the 
greater monasteries covered an enormous area, and offered to the architect, 
sculptor, and painter endless opportunities for the display of art in every form. 
Although no very early monastery has survived in a condition at all complete, 
the ground-plans of many such establishments have been clearly traced, and 
in Gandhara considerable remains of superstructures crowded with statuary 
have been disclosed. Recorded descriptions and extant remains amply attest 
the splendour of the more important monasteries, each of which was a centre 
of secular as well as of religious education, and also a school of art in which 
men were trained in all the crafts needed for the adornment of the holy places. 

Something of this great school of art is preserved for us in the great rock- Cave- 
cut halls and dwelling-caves of Western India. Here, at Bhaja, Kondane, tem P les - 
Pitalkhora, Bedsa, Ajanta, Nasik, Karli, and Kanheri, have been hewn out 
of the very heart of the rock full-scale reproductions of the ancient Assembly- 
halls in all the detail of their wooden construction. In general plan they 
correspond with the barrel-roofed buildings of the early sculpture. They are 
apsidal with side aisles on either hand and are lit by the great horse-shoe 
window at one end. A survey of this series of caves lays bare a stylistic 
advance from purely wooden imitation to definitely lithic forms. At Bhaja 
the plain octagonal pillars rake inwards considerably; the screen that closed 
the lower part of the great window was actually of timber morticed into the 
rock as are the carefully inset roof beams. There is no decoration except 
bands of railing-pattern and tiers of miniature 'chaitya- windows', derived 
from the piled-up stories of the wooden originals. These details apply to the 
caves at Kondane, Pitalkhora and to the earliest at Ajanta (Cave X). Later 
the wooden screen is reproduced in stone and bell-capitals and bases, and 
tiered-up abaci with heavy animal upper-capitals appear, while at Nasik, 
Karli, and Kanheri sculpture is freely used. This sculpture is all obviously 
post-Sanchi. At Karli and Kanheri highly decorated railings of the Amara- 
vati kind are found and also guardian figures which closely correspond to 
the middle phase of Kushan sculpture, found at Mathura. The epigraphical 
evidence coincides with the artistic evidence, dating the last of these early 
caves (Karli and Kanheri) in the second century a.d. The fa9ade of Bhaja is 
so exactly like the bas-relief representations of the wooden original at Bharhut 
and Bodh-Gaya that the earliest of the series may be accepted as second 
century B.C. 

The Lomas Rishi Cave in the Barabar hills belongs to a group of small 
rock-cut cells some of which were dedicated in the reigns of Asoka and 
Dasaratha, his grandson. Like the other caves its interior walls have received 
the fine polish which is so typical of Mauryan work. The original work seems 
to have been discontinued owing to a flaw in the rock. The fa9ade must have 



been a later addition, for it is akin to the work at Bharhut. It, however, offers 
a good example of the close imitation of wooden construction (Plate 7 b). 1 
Structural At Ter, the ancient Tagara, in the Sholapur District, Bombay Presidency, 
there is an example of a structural chattya-hall which has escaped destruction 
by being converted into a Brahmanical temple. 2 It is a long chamber, con- 
structed of brick, 26 feet in length and 12 feet in width on the inside, with 
walls 3 J feet thick, an apsidal end, and a waggon- vaulted ridge roof. The 
bricks, laid in mud cement, with exceedingly fine joints, are of huge size, 
measuring 17x9x3 inches. They are finely burnt. The building is medieval 
and probably belongs to the eighth century. 3 


The art of the times dealt with in this chapter is characterized by frank 
naturalism. It is thoroughly human, a mirror of the social and religious life 
of ancient India, apparently a much pleasanter and merrier life than that of 
the India of later ages, when the Brahmans had reasserted their superiority 
and imposed their ideas upon art and upon every branch of Hindu civiliza- 
tion. The early sculptures, while full of the creatures of gay fancy, are free 
from the gloom and horror of the conceptions of the medieval artists. The 
Buddhism with which nearly all of them are concerned was, as already 
observed, the popular creed of men and women living a natural life in the 
world, seeking happiness, and able to enjoy themselves. The recent critics of 
the ' nationalist' school, in their anxiety to secure adequate recognition for the 
merits of the medieval Brahmanical art, sometimes appear to believe that it 
alone truly expresses Indian thought. It is well to remember that for several 
centuries Indian thought was content to find its artistic utterance in a fashion 
much less sophisticated. 

There has, also, been a tendency to apply certain literary standards, which 
are in essence medieval, to the work of the Early Period, and in fact, to all 
Indian art, wholesale. The various members, mouldings, and motives dealt 
with in the Silpa Sastras cannot be found outside the buildings of the medieval 
period. With regard to the passages dealing with the sculpture the same thing 
applies. The Sastras are in fact technical memoranda based on a literary 
tradition, which may be taken to have crystallized out from the great literary 
activity of the Gupta period. Their import is very great with regard to the 
iconography of medieval and modern India. They can only be applied with 
great circumspection to the earlier art, the inspiration of which is oral and living . 4 

1 Ann. Rep. A. 5., i, 1902-3, p. 197. masters of modern music of popular folk-music. 

2 G. O. Madras Pub. 382, 30, April 1889. He claims that medieval art is the 'expression 

3 The apsidal plan also survived in medieval stone of the Indian consciousness at the height of its 
construction, as is shown at Sanchi and Pattadkal. greatest intellectual, literary, and artistic activity*. 

4 Havell compares the development of Puranic One feels that here there has been some confusion 
(medieval) art to the use made by the great of purely artistic appreciation with the literary. 


The study of the existing monuments of Asoka, scanty as they are, leaves 

one with a clear impression of a definite and distinct school of sculpture, 1 

with great stylistic and architectonic qualities and certain characteristics which 

distinguish it from the sculpture of the Early Period 
and from all other periods of Indian art. Firstly, finely 
stylized as these works are they are essentially naturalistic. 
Secondly, columns, capitals, and caves all have a highly 
finished, polished surface which is unique and unmis- 
takable. Certain sculptures, however, exist which possess 
this distinguishing finish and yet as sculptures are to be 
classed with the work of Bharhut and Sanchi. These 
maybe treated as a link between the two schools. Anyhow 
the Mauryan period, which is historically exact, provides 
a lower limit for the dating of the work of the Early Period. 
Among these sculptures, which are mostly of colossal 
size, is a mutilated standing statue of a male, perhaps 
representing the Yaksha demi-god Kuvera> god of 
wealth, found at Parkham in the Mathura District, and 
now in the Mathura Museum. 2 The material is polished 
grey sandstone similar to that used for the Asoka pillars. 
The height, including pedestal, is 8 feet 8 inches, and the 
breadth across the shoulders is 2 feet 8 inches. The 
excessively massive body, which possesses considerable 
grandeur, is clothed in a waist-cloth {dhoti) held around 
the loins by means of a flat girdle tied in a knot in front. 
A second flat girdle is bound round the chest. The 
ornaments are a necklace and a torque from which four 
tassels hang down on the back. Some praise may be 
given to the treatment of the drapery. 

This is probably the earliest example of * early ' sculp- 
ture as distinct from the Mauryan. In treatment and 
detail it is clearly a forerunner of the sculpture of 
Bharhut and has nothing in common with the art of the Mauryan capitals. 
Several other colossal sculptures, which do not possess the distinctive 
Mauryan polish, emphasize this development. An uninscribed statue of a 
female (Plate 8), 6 feet 7 inches in height, found near Besnagar adjoining 
Bhilsa in the Gwalior State, Central India, a locality associated by tradition 
with Asoka, is to be classed among these on account of the style and costume. 

Its relation 
to the 

1 uu, tui..? 


Fan-palm capital, 
Besnagar (cf. PI. 3 b). 




1 In many senses un-Indian, although the exact 
extent of its un-Indian qualities and their origin 
is very hard to state. 

2 For the inscription see J. B. O. Res. Soc.> vol. 

v, Part IV, Dec. 1919, where this sculpture and 
the Patna sculptures are said to be contemporary 
portraits of kings of Magadha in the fifth 
century B.C. 


The figure wears the heavy head-dress as found at Bharhut and Sanchi and 
also the linked belt of beaded strands and the double breast chain. The finely 
pleated waist-cloth is held at the hips by a belt with a looped clasp and its 
folds are treated in fashion that is reminiscent of the Sanchi bracket-figures 
rather than the Bharhut devatas. The modelling is naturalistic, but the sculp- 
ture has suffered severely from violence and exposure. 

There is a second colossal female at Besnagar, 7 feet high, locally known as 
the Telin or Oil-woman, which has been described by Cunningham. He also 
mentions the existence in his time of a polished sandstone elephant and rider. 
The stupa In 1873 Cunningham discovered at Bharhut, about midway between 
f Bharhut. Allahabad and Jabalpur, the remains of a Buddhist stupa, surrounded by 
a stone railing adorned with sculptures of surprising richness and interest. 
The stupa had then been almost wholly carried off by greedy villagers in 
search of bricks, who treated the sculptures with equal ruthlessness, and were 
prevented from destroying them only by the great weight of the stones. 
During the following years to 1876, Cunningham and his assistant uncovered 
the ruins and saved a large number of the sculptured stones by sending them 
to Calcutta, where they now form one of the chief treasures of the Indian 
Museum. Everything left on the site was taken away by the country people 
and converted to base uses. 1 
ie railing. The railing, constructed after the usual pattern, in a highly developed form, 
was extremely massive, the pillars being 7 feet 1 inch in height, and each of 
the coping stones about the same in length. The sculptures of the coping 
were devoted mainly to the representation of incidents in the Jatakas y or 
tales of the previous births of the Buddha. The carvings on the rails, pillars, 
and gateways, all treating of Buddhist legends, were exceedingly varied in 
subject and treatment. The structure must have been very much like Sanchi. 
The composite pillar of the gateway, made up of four clustered columns 
crowned by a modified Persepolitan capital, is worthy of special notice. An 
inscription records that the Eastern gateway with the adjoining masonry 
was erected during the rule of the Sunga dynasty (185-173 B.C.), but it is 
not possible to determine the date of the monument with greater precision. 2 
The execution of work so costly and elaborate must have extended over many 
years. Certain masons' marks in the Kharoshthi character of the north- 
western frontier suggest that perhaps foreign artists were called in to teach 
and assist local talent. 3 The railing exhibits a great mass of sculptures of a 
high order of excellence. The subjects and style are described by Cunningham 
as follows: 

1 Anderson, Catalogue, Part I, pp. xiii-xx, 1-120. Asoka's inscriptions were inscribed in both 

2 Ind. Art., vol. xxi, p. 227, also vol. x, p. 118 n., scripts and the Maski inscription is signed in 
vol. xi, p. 25 n., vol. xiv, p. 137 n. Kharoshti. It is possible that the scribes of the 

3 Whatever its origin, the use of Kharoshti was day were skilled in both scripts. Lack of evi- 
not necessarily confined to the north-west, dence must at any rate qualify any statement. 

A. Chaitva Cave, Nasik 



B. Lomas Rishi Cave, Barabar 

PLATE 8. Colossal female statue from Bcsnagar 


'The subjects represented in the Bharhut sculptures are both numerous and varied, Subjects and 
and many of them are of the highest interest and importance for the study of Indian style of the 
history. Thus we have more than a score of illustrations of the legendary Jatakas, and sculptures. 
some half-dozen illustrations of historical scenes connected with the life of Buddha, 
which are quite invaluable for the history of Buddhism. Their value is chiefly due to 
the inscribed labels that are attached to many of them, and which make their identifica- 
tion absolutely certain. Amongst the historical scenes the most interesting are the pro- 
cessions of the Rajas Ajatasatru and Prasenajita on their visits to Buddha; the former 
on his elephant, the latter in his chariot, exactly as they are described in the Buddhist 

1 Another invaluable sculpture is the representation of the famous Jetavana monastery 
at Sravasti — with its mango tree and temples, and the rich banker Anathapindika in 
the foreground emptying a cartful of gold pieces to pave the surface of the garden. 

'Of large figures there are upwards of thirty alto-rilievo statues of Yakshas and 
Yakshinis {Yakshis), Devatas, and Naga Rajas, one half of which are inscribed with 
their names. We thus see that the guardianship of the north gate was entrusted to 
Kuvera, King of the Yakshas y agreeably to the teaching of the Buddhist and Brahmani- 
cal cosmogonies. And similarly we find that the other gates were confided to Devas 
and the Nagas. 

'The representations of animals and trees are also very numerous, and some of them 
are particularly spirited and characteristic. Of other objects there are boats, horse- 
chariots, and bullock-carts, besides several kinds of musical instruments, and a great 
variety of flags, standards, and other symbols of royalty. 

'About one half of the full medallions of the rail-bars and the whole of the half- 
medallions of the pillars are filled with flowered ornaments of singular beauty and 
delicacy of execution.' 1 

The medallions on the rail-bars and the half-medallions on the pillars are 
filled with a wonderful variety of bas-relief subjects. The comic monkey 
scenes collected in Cunningham's Plate XXXIII display a lively sense of 
humour, freedom of fancy, and clever drawing. They must, of course, like 
all the early bas-reliefs, be judged as pictures drawn on stone, rather than as 
sculpture. The rollicking humour and liberty of fancy unchecked by rigid 
canons, while alien to the transcendental philosophy and ascetic ideals of 
the Brahmans, are thoroughly in accordance with the spirit of Buddhism, 
which, as a practical religion, does not stress the spiritual to the extinction 
of human and animal happiness. Everything seems to indicate that India 
was a much happier land in the days when Buddhism flourished than it has 
ever been since. The first medallion selected for illustration is a very funny 
picture of a tooth being extracted from a man's jaw by an elephant pulling 
a gigantic forceps, and the second is nearly equally humorous. The stories 
alluded to are presumably of the Jataka class (Plate 5). The spontaneity 
of the work vouches for the popularity of the tradition, stories that must 
have been on every child's lips. 

1 Cunningham, The Stupa of Bharhut (1879), p. 18. 


Coping. PL 5E gives a characteristic and well-preserved specimen of the bas- 
reliefs on the coping. The large fruit is that of the jack (Artocarpus integri- 
folia), and the deer are the spotted hog-deer kind (Aoctsporcinus). The artists 
who could design and execute such pictures in hard sandstone had no small 
skill. Mr. Havell observes that the technique is that of the wood-carver. The 
Chulakoka (Cunningham, Plate XXIII, 3) sculpture is especially interesting 
as the earliest extant example of the woman and tree motive. 1 One of the 
best statues is that of the Yakshi Sudor sana (PL 5 a) which exhibits a good 
knowledge of the human form and marked skill in the modelling of the hips 
in a difficult position. 2 

The large alto-relievo images of minor deities on the pillars vary much in 
Sculptured Besnagar offers an excellent example of a sculptured railing of the same 
railing at t yp e anc j st yj e as Bharhut. 3 The coping-stone is adorned with a frieze repre- 
esnagar. sent i n g a re iigi us procession, with elephants, horses, &c, divided into com- 
partments by the graceful sinuosities of a lotus stem. The pillars exhibited 
various scenes in panels and on the cross rails elegant lotuses are carved. 
Sculptured A better-known example is the often described railing at Bodh-Gaya, which 
railing at US ed to be called 'the Asoka railing', but is stylistically later than Bharhut, 
Bodh-Gaya. t^^gh earlier than Sanchi. Thirty pieces have been found, evidently belong- 
ing to two distinct structures, some pieces being of granite and others of 
sandstone. All are similar in style, irrespective of material. 

Most of the subjects are treated in low relief. Those on the coping are 
purely fanciful; those on the panels and medallions include weird centaurs, 
winged beasts, domestic animals, sacred trees, and sundry scenes of human 
life, all no doubt significant and readily intelligible to ancient Buddhists 
versed in the legends of their scriptures and traditions, but now difficult of 
interpretation. Most of the more interesting sculptures have been published 
more than once; a few are here reproduced from photographs. They are 
simply pictures in stone, and should be criticized as drawings slightly in relief 
rather than from the point of view of a sculptor. They exhibit a lively fancy, 
considerable skill in drawing, and much neatness of execution; both concep- 
tion and execution are purely Indian. 
R Ci <^ C r ° f * n ^ ate ^ * s s ^ own a P art °f an an * ma l frieze on the coping, very similar 
sculptured to w ^ at we s ^ a ^ meet * ater at Amaravati. Fig. A gives an interesting 
picture of an early Buddhist chapel enshrining the symbol of the preaching 
of the Law. Images of Buddha still do not occur. The fa9ade, with its curved 
roof, exactly illustrates the origin of the architecture of the western cave- 
temples, and their wooden proto-types. A frieze on the coping pictures queer 

1 Repeated several times at Bharhut. Indian sculptures. 

2 Vincent Smith comments on the narrow waist 3 Cunningham, A. S. Rep., vol. x, p. 38, PI. XIII. 
and exaggerated hips which 'disfigure' so many 




PLATE g. Reliefs from the railing, Hodh-(Jaya 

A. The great stitpa, Sanchi, as restored 

B. Inside, west gateway, Sanchi 


fish-tailed monsters, which recall many forms in Hellenistic art familiar in 
variant shapes to Asiatic art from very remote times (Fig. e). 1 

The series of illustrations may be closed by two purely naturalistic pictures 
— an excellent buffalo (Fig. g), and a husband and wife seated together 
(Fig. h). The treatment of the lotus is excellent. It is the most character- 
istic and universal of all Indian art motives. Infinite variety in the treatment 
of the conventionalized flower is exhibited in the minute details both at Bodh- 
Gaya and elsewhere. 2 

Fig. B is equally instructive concerning the practice of the early Bud- 
dhist cult. This sacred tree, surrounded by a plain railing, square in plan, 
is an example of the many shrines of Chaityas of ancient India, another 
example being the Deer-and-Lions-Lying-down-together Chaitya of the 
Bharhut coping-stone: the latter is of special interest for its does not seem 
to be Buddhist in origin and may represent the survival of a pre-Buddhist 

The remaining figures on Plate 9 illustrate various fantastical hybrid 
creatures, winged lions and oxen, a centaur, a horse-headed female or kinnara y 
and a frieze of the fish-tailed monsters common at Mathura and in Gandhara. 
These are human-bodied and appear to be half-ragw, \\2\i-makara. These 
strange beasts have a debatable origin. The Naga or snake-godling is usually 
represented in Indian with his snake-hood, but in the Jatakas appears to be 
able to cast off this stigma and is then only to be known by his red-eyes. 
These lesser divinities are by birth Indian and native in the earliest folk- 
lore and sculpture. The makara, too, whose scrolled tail is used so magni- 
ficently to form the volutes of the architraves of tor anas at Bharhut, Sanchi, 
and Mathura, is also well founded traditionally. These with the kinnaris or 
half-bird musicians and the horse-headed kinnaras may be classed together 
as gandharvas or lesser heavenly beings. 4 They are as types paralleled with 
several other motives of early Indian art in the sculpture of West Asia, 
Assyria, and Persia. The bell and frieze design of the Bharhut cope-stone 
and its upper pyramid and lotus band are among these, and also, the bell 
capital surmounted by animal groups. Whatever the distant sources of these 
motives may be, their treatment at Bharhut, Bodh-Gaya, and Sanchi, is wholly 
Indian. As has been said many of them spring directly from the soil. 

The Bharhut sculptures, having escaped the destructive zeal of Muham- The remains 
madan iconoclasts by reason of their situation in an out-of-the-way region, at Sanchl - 
lay safely hidden under a thick veil of jungle until a century ago, when the 

1 Vincent Smith accepts the West Asiatic origin Rep., vol. i, Pis. VIII-XI; vol. iii, Pis. XXVI- 
of these motives, but they have close parallels XXX; and Mahabodhi ; also Rajendralala Mitra, 
in Indian legend : the Makara is undoubtedly Buddha Gaya. 

indigenous. 3 Codrington, Ancient India, p. 31. 

2 For numerous drawings of the sculptures on 4 Burgess and Grunwedel, Buddhist Art in India, 
the Bodh-Gaya railing see Cunningham, A. S. p. 47. 



establishment of general peace and the spread of cultivation stimulated the 
local rustics to construct substantial houses from the spoils of the old monu- 
ments for which they cared nothing. The extensive group of early Buddhist 
buildings at and near Sanchi in the Bhopal State similarly evaded demolition 
because it lay out of the path of the armies of Islam. Although the monu- 
ments of Sanchi have not suffered as much as those of Bharhut from the 
ravages of the village builder, they have not wholly escaped injury. During 
the first half of the nineteenth century much damage was done by the ill- 
advised curiosity of amateur archaeologists. Now, however, the authorities 
concerned are fully alive to their responsibility, and everything possible is 
being done to conserve the local memorials of India's ancient greatness. 
Sanchi to-day is a triumph of archaeological restoration, 
ronology. The importance of Sanchi in the history of Indian art rests chiefly upon 
the four wonderful gateways forming the entrances to the procession path 
between the stupa (Plates 10 and n) and the surrounding railing. A key to 
the chronology of the site is provided by the Asoka column which stands to 
the right of the South gateway. The Mauryan level is marked by a floor of 
pounded earth and clay. Three other levels or floors appear over it, the top- 
most being lime-plastered. Above all is the pavement of large slabs con- 
temporary with the stupa railing. This is a perfectly plain copy of a wooden 
post and rail fence and may be dated in the latter half of the second century 
B.C., since there is 4 feet between the upper pavement and the Mauryan level, 
which could hardly have accumulated in less than a century. 

The four gateways, which are additions to the original railing, fall artistic- 
gateways, a lly i n to pairs, the East and West gates, showing a slight development in 
5 °a r> C i" m °delling and the use of light and shade. A little more than fifty years may 
have elapsed between their execution, the end of the first century B.C. being 
accepted as a general date for all four. The Southern gateway was prostrate 
when visited by Captain Fell in 181 9. The Western gate collapsed between 
i860 and 1880, but the Northern and Eastern gates have never fallen. All 
have undergone thorough repairs during recent years under the able direction 
of Sir John Marshall, the Director-General of Archaeology in India. Sanchi 
has taken on a new lease of life and beauty in his hands, the more important 
remains of this huge site being carefully and exactly restored and preserved, 
istmction The Sanchi gateways, or tor anas > stand 34 feet high, and are all sub- 
e ^ a th s e stantially alike, while differing much in detail: 

'Two massive square pillars, one on either side, 14 feet high, forming as it were the 
gate-posts, support an ornamental superstructure of three slightly arched stone beams 
or architraves placed horizontally, one above the other, with spaces between them. 
The topmosr beam of each gate was surmounted by the sacred wheel flanked by atten- 
dants and the trisula emblem. 

'The faces, back and front, of the beams and pillars are crowded with panels of 

mTtn - J "-''-.-t gatc , av> 


A. East Gateway, Sanchi 


]*. East Gateway, Sanchi 


sculpture in bas-relief representing scenes in the life of Buddha, domestic and silvan 
scenes, processions, sieges, adoration of trees and topes, and groups of ordinary and 
extraordinary animals, among which are winged bulls and lions of a Fersepolitan type 
and horned animals with human faces.' 1 

Plate 11, representing the eastern gateway, will enable the reader to appre- Extent of 
date the wealth of ornament lavished on the four monuments. The same sculptures. 
gateway may be further studied by the aid of full-sized casts supplied to the 
Victoria and Albert Museum, South Kensington, and other institutions, 
some of which, however, do not exhibit the casts. Numerous illustrations, 
more or less accurate and satisfactory, will be found in the works of Fergus- 
son, Maisey, Cunningham, and other writers on Indian archaeology, the best 
of all being the official handbook issued by the Archaeological Survey. 2 In 
January and February 1901, Mr. H. Cousens succeeded in photographing 
the whole mass of sculptures on 225 negatives to a uniform scale of one- 
eighth, but so far little use has been made of the huge supply of material thus 
accumulated. The preparation of a full descriptive and critical monograph 
would be an arduous undertaking, and the work would probably fill several 
large quartos. 

AH critics are agreed that the gateways were built in pairs and that the Capitals of 
southern gateway is one of the earliest of the four. The capitals of its gate- the 80Uthern 
posts are formed by four lions seated back to back, 'indifferently carved', 6 F 
and of the same type as those on Asoka's inscribed pillar already noticed 
(ante, p. 18). The marked decline in skill demonstrated by the contrast 
between the lions on the gate-post and those on the inscribed pillar is sur- 
prising considering the shortness of the interval of time, about a century, 
between the two compositions, or rather the essential difference between the 
Mauryan and the ancient Indian school. The difference is most easily verified 
by comparing the treatment of the lions' paws on the gate-post capital 
(Maisey, Plate XIX) and of the same members on the capital of the inscribed 
pillar, or the similar Sarnath pillar. The paws of the early Asokan sculptures 
are correctly modelled with four large front claws and one small hind claw, 
the muscles also being realistically reproduced. In the later work five large 
claws, all in front, are given to the paws, and the muscles are indicated 
by some straight channels running up and down in a purely abstract 
manner. 3 

The capitals of the gate-posts of the northern gateway exhibit four elephants Other 
standing back to back, and carrying riders. Those of the eastern gateway ca P itals - 
(Plate 1 1) are similar. On the capitals of the latest gateway, the western, four 

1 Cousens, Progr. Rep. A. S., Western India, for 3 Vincent Smith quotes Coomaraswamy's Aims 
year ending 30 June 1900, para. 9. of Indian Art, p. 14, and states his preference 

2 Guide to Sanchi, also see A. S. Report, 1913- for the naturalistic Asokan lions. 
14, and Foucher, Beginnings of Buddhist Art. 



hideous dwarfs, clumsily sculptured, take the place of the elephants or lions 
(Plate 10 b). 1 

S S!S!:!? ^ t ^ ie Sanchi sculptures, like the Ajanta paintings, deal with Buddhist 
subjects. If a composition seems in our eyes to be purely secular, that is only 
because we do not understand its meaning. Genre pictures, whether in paint 
or bas-relief, do not exist in the ancient art of India. The main object of 
the artist was to illustrate his Bible, and if, perchance, the illustration could 
be made into a pretty picture, so much the better; but anyhow, the sacred 
story must be told. 

Worship by In addition to his desire to tell edifying stories in a manner readily intelligi- 
1 creation 6 kk to l ^ e e y es °f ^he ftdthful, the old artist clearly was dominated by the 
feeling that he was bound to impress on all beholders the lesson that the 
dead Teacher, the last and greatest of the long line of Buddhas, had won 
and continually received the willing homage of the whole creation — of men, 
women, and children, of the host of heaven, the water-sprites, and the demons 
— nay, even of the monsters of romance and the dumb animals. And so, in 
all the ancient Buddhist art, whether at Sanchi or elsewhere, weird winged 
figures hovering in the air, snake-headed or fish-tailed monsters emerging 
from their caverns or haunting the deep, offer their silent homage to the Lord 
of all, and the monkeys bow down in adoration before the Master who had 
turned the wheel of the Law and set it rolling through the world. The early 
artists did not dare to portray his bodily form, which had forever vanished, 
being content to attest his spiritual presence by silent symbols — the foot- 
prints, the empty chair, and so forth. 2 But, whether the Master was imaged 
or symbolized, the notion of his adoration by all creation was continually 
present in the minds of the artists and influenced their selection of decorative 
motives. Although concerned in the main with thoughts of religion and wor- 
ship they were not unmindful of beauty, which they often succeeded in 
attaining in no small degree. 

In the early works, like those of Sanchi and Bharhut, the absence of images 
of Buddha has the advantage of saving the stone pictures from the formal 
symmetrical arrangements grouped round the central figure which often 
weary by their monotonous iteration in Gandhara and at Amaravati. 

fi ^r&f * n a 8 enera l wa y> ^ e st y* e °f t ^ ie Sanchi reliefs resembles that of those 

1 These dwarfs have already appeared as transl. Rhys-Davids, Dialogues of the Buddha, 
Atlantes at Bharhut. p. 54). The absence of images of Buddha from 

2 'The outward form, Brethren, of him who has early Indian art does not imply that images of 
won the truth (Tathagata) stands before you, the Hindu gods were then unknown. It has been 
but that which binds it to rebirth is cut in twain, claimed that they were in use as early as the 
So long as his body shall last, so long do gods fourth century B.C. (Ind. Ant. xxxviii (1909), 
and men behold him. On the dissolution of the pp. 145-9). Rao, Hindu Iconography, Introd., 
body, beyond the end of his life, neither gods vol. i, Pt. I. 

nor men shall see him' (Brahmajala Sutta, 




East gateway, Sanchi 








A. Part of frieze on tor ana beam, Mathura 

H. Tablet with relief sculpture of a Jain stupa 

A. Fragments in Bharhut style, Mathura Museum 

B. Naga statue, with inscription of 

Huvishka's reign, from Chhargaon; 

Mathura Museum 

C. Yakshi on dwarf; 
Mathura Museum 



at Bharhut, compensation may be found in the elegant bracket figures, practi- 
cally statues in the round, which are a specially pleasing feature of Sanchi 
art. A good example of such a figure is shown in Plate 11. It is a form 
of the Woman and Tree motive. The beautiful decorative details of the pillar 
are worthy of careful study. No nation has surpassed the Indians in the variety 
and delicacy of the floral designs enriching their sculptures and pictures. 

Plates I2and 13 may be taken as being typical of the Sanchi reliefs. Plate 13 A Reliefs on 
is from the inside of the left pillar of the East Gate. 1 At the bottom stands the East 
the Yaksha guardian of the door in princely dress. His fellow stands opposite S atewa y- 
him on the other pillar. They are comparable with the Bharhut Yakshas, 
but the treatment of figure and ornament is considerably more rhythmic. The 
tree in the background is a Bignonia and the devata holds one of its blossoms 
in his right hand. The upper panel of Plate 13 a represents the Buddha's 
victory over the black snake and the conversion of Kasyapa at Uruvilva. The 
snake and the flames of the conflict and the astonished Brahmans, some of 
whom are attempting to fetch water, are all shown, but the figure of the 
triumphant Buddha is left to the imagination. Below this scene the story of 
the conversion of Kasyapa is continued and the incident of Buddha and the 
Brahman sacrifice is shown. Wood is being split and the preparations made, 
but the fire springs up and dies at the Buddha's command. On the front of 
the same pillar (Plate 13 b) the final incident of the Buddha walking on the 
waters is told and the sequent visit to Rajagriha, King Bimbisara being 
depicted as arriving at the gate of the city in his two-horsed chariot. In the 
top panels of Plates 12 a and 13 A is the Bodhi-tree Shrine already discussed. 

Surveying the work of the Early Period (second century B.C. -early first Other early 
century a.d.) one recognizes certain distinctive common elements: the sculptures. 
absence of the Buddha figure; its replacement by certain simple symbols; and 
the popular quality of the work, the living oral tradition of which is indicated 
by the predominance of Jataka scenes even over the scriptural; 2 the naive 
technique which treats each story as a pictorial entity contained in a single 
panel or medallion, the figures of the protagonist being repeated twice and 
three times according to the demand of the drama to be unfolded. At Sanchi, 
while the method of exposition and the bulk of the decorative motives are 
the same as at Bharhut, the canonical is very definitely to the fore, and the 
technique has advanced considerably. At Mathura and many other sites in 
India sculptures have been found which belong to the Early Period. With 
regard to these it is advisable to take Bharhut and Sanchi as types of sub- 
periods and so arrive at the classification Early Period I and Early Period II. 

From Mathura come the reliefs shown in Plate 16 a. These fragments are Early 

sculpture at 
1 For a detailed discussion of all these reliefs see 2 At Bharhut, eight Jatakas have been recognized, Mathura. 
Foucher, Beginnings of Buddhist Art, p. 97, vols, at Sanchi only five, 
viii and ix. 


respectively i foot 3 inches and 1 foot 4 inches in height. The turbans and 
jewellery and the general treatment of form and features are distinctly of the 
Bharhut kind. Plate 15 a is also of this period and is interesting because of 
its garland-bearers 1 and its three-tiered stupa. 
Sculptures The sculpture in the most ancient cave-temples of Western India, at Bhaja 
ti the cave- an( j Bedsa (Poona District), Pitalkhora (Khandesh District), and Kondane 
temples. ( Kolaba District), offers little of aesthetic interest. The small five-celled 
hermitage at Bhaja is perhaps the oldest. The cornice is supported by male 
figures used as carytids, wearing waist-cloths, large turbans, and much 
jewellery. The statues of the armed door-keepers are similarly clothed. They 
must be compared to Sanchi rather than to Bharhut. The 'horses and ele- 
phants bearing men and women of bold execution' of the Bedsa capitals are 
likewise post-Bharhut. The sculpture at Karli, Kanheri, and Nasik is all later 
than Sanchi and must be compared to Kushan types among which close 
similitudes are to be found. 
Jain bas- The sandstone hills known as Khandagiri, Udayagiri, and Nilagiri, situated 
reliefs m j n t j ie p ur j District, Orissa, a few miles from the Bhuvanesvar temples, are 
nssa. h one y Com b e ci with Jain caves of various dates, probably covering a con- 
siderable period. The local worship appears to have been devoted chiefly 
to the Tirthankara Parsvanath. The elaborate, but ugly and semibarbaric 
sculptures in the Rani Gumpha, or Queen's Cave, are interpreted as repre- 
senting a procession in honour of Parsvanath. This work is unskilled rather 
than primitive and is probably post-Sanchi. 2 
Female At the Jayavijaya Cave on Udayagiri a female statue about 6 feet high, and 
statue, a j most i n t ^ e round, seems to be of early date and to possess considerable 
merit. The goddess, or whoever the personage may be, is represented as 
leaning her weight on the right leg, the left foot being bent in behind the 
right, so that only the toes touch the ground. In her right hand she holds 
up an object, presumably a flower, while the left forearm is bent horizontally 
across her waist. She apparently wears drawers, and is nude above the waist, 
in accordance with the fashion of ancient India, maintained in the south until 
recent days. The head-dress is a peculiar ribbed cap with long lappets. The 
features have been destroyed. The form is naturalistic and the pose easy. 

1 Suparnas and Kinnaras. (1908), s.v. Khandagiri, Ann. Rep. A. S., India, 

2 Sixty-six caves, viz. 44 on Udayagiri, 19 on 1902-3, pp. 40-2. The best account of the caves 
Khandagiri, and 3 on Nilagiri. The inscription is that by Babu Monmohan Chakravarti in Gaz. 
of King Kharavela in the Hathi Gump ha, or Puri District (1908). See also Fergusson, Hist. 
Elephant Cave, is of the second century B.C., of Indian and Eastern Architecture, 2nd ed. 
but not precisely dated, as formerly supposed (1910), vol. ii, pp. 9-18. 

(Fleet, J. R. A. 5., 1910, p. 242). See Imp. Gaz. 

Chapter Four 


Mathura is the chief find-spot of Kushan sculpture and, since it is linked The 
directly to Bharhut and Sanchi by many works from its studios which Kushan 
clearly belong to the Early Period of Indian sculpture, it is advisable d y nast y- 
to discuss the Kushan sculpture of this site by itself, apart from Gandharan 
art and questions of foreign influence. 

The chronology of the Kushan dynasty is still unsettled, and decisive proof 
is lacking of any one of the many rival theories on the subject. Six sovereigns 
of the dynasty are of importance for the history of India and of Indian art. 
The first two are most conveniently cited as Kadphises I and II. The next 
four kings, Kanishka, Vasishka, Huvishka, and Vasudeva I, certainly reigned 
in that order for a century in round numbers. 1 As a working hypothesis I 
revert to Professor Oldenberg's old theory, and assume that Kanishka came 
to the throne in a. d. 78. Thus the first and second centuries after Christ are 
approximately filled by the rule of the leading kings of the dynasty. 2 

In the early centuries of the Christian era Mathura on the Jumna, a city Mathura. 
of immemorial antiquity, and prosperous to this day in spite of many disasters, 
was sacred in the eyes of the adherents of all the three indigenous Indian 
religions — Jainism, Buddhism, and Brahmanical Hinduism. The abundant 
supply of excellent red sandstone at Rupbas and other quarries in the neigh- 
bourhood favoured the development of an active school of sculptors, whose 
workshops supplied all parts of Northern India with idols, much as Jaipur 
does now. The craftsmen, of course, were prepared to supply whatever was 
wanted by their patrons of any religion. The character of the local stone 
is so distinct that the products of the Mathura studios are easily recognized 
wherever they may be found. 3 Wealthy worshippers did not hesitate to 
undertake the cost of transporting heavy, even colossal, statues for hundreds 
of miles. 

Sarnath, like Mathura, was holy ground to the Jains as well as the Bud- Samath. 
dhists, and is connected with Mathura and declared a Kushan site of impor- 

1 Inscribed pillar of Maharaja Shahi Vasishka, is thus followed by the Kushan Period, which he 
dated in 24th year (Mathura Museum, Q 13, applies to everything up to the Guptas (r. a.d. 
CataL, p. 189). In India the reigns of Kanishka, 320), including Amaravati. It is better to avoid 
Vasishka, and Huvishka overlap. Probably such pseudo-dynastic classifications which are 
Vasishka reigned in India only, and Huvishka literary in origin and arbitrary in application, 
succeeded to the whole empire on Kanishka's 3 Vindhyan sandstone, however, often has a red 
death about a.d. i 23 . ti n g c : mistakes have occurred where the evidence 

2 Vincent Smith puts forward a dynastic chrono- of the stone has been set above that of the style, 
logy of Indian Art. His Sunga Period (Sanchi) especially with regard to medieval sculptures. 


tance by finds of fine sculptures of red Mathura sandstone inscribed in the 
Kushan era. Its richly adorned buildings, crowded with sculpture, were 
involved in common ruin by the violence of the fierce hosts of Islam at the 
close of the twelfth century. The Brahmanical Hindus lavished their devotion 
on the neighbouring city of Benares, and shared the misfortunes of their 
rivals. The sculptors of Sarnath ordinarily used the excellent pale sand- 
stone from the quarries of Chunar in the Mirzapur District, which had 
supplied the blocks for Asoka's pillars. But, as already observed, wealthy 
donors sometimes preferred to import red sandstone images from Mathura. 
During the last few years much progress has been made in unearthing the 
buried treasures of Sarnath, but much more remains to be found. Several 
statues of Bodhisattvas, executed in the round on a large scale, are almost 
identical with the Mathura specimen reproduced above (Plate 20c), and one 
of these is dated in the third year of the reign of Kanishka, which may be 
regarded provisionally as equivalent to a.d. 80. The Kushan age of such 
works is thus definitely determined. Halos, when present, are plain, not 
highly decorated as in the Gupta period. 

A finely executed bas-relief, which once decorated a doorway and exhibits 
artistic lotus and vine patterns, besides a picture of an elephant worshipping 
a stupa, is quite in the Mathura style, and may be assigned with some con- 
fidence to the first century of the Christian era. 1 The style of the Sarnath 
works is so closely related to that of Mathura that illustrations may be dis- 
pensed with. 
Relief of As at Bharhut and Sanchi the earlier sculptures at Mathura are derived 
a Jam stupa. f rom stupas. Many of them are pre- Kushan and may be directly compared 
to Bharhut and Sanchi as belonging to the early period. The Lonasobhika 
votive-tablet shown in Plate 15 b may be taken as illustrating the Mathura 
stupas, of which none have escaped the hand of the iconoclast. It must be 
referred to Kushan times, however, being distinguished from the latest work 
of the Early School (Sanchi) by its three superimposed tiers, the form of its 
corner pillars, and the stylized representation of the octagonal railing pillars, 
as well as by the freer treatment of the flying spirit-host. 

This 'tablet of homage', with a relief sculpture of a Jain stupa (2 feet 
4 inches high, 1 foot 9I inches wide), now in the Mathura Museum, was 
found embedded in a wall near the Holi gate, but is said to have come from 
a field near the village of Maholi. It was dedicated by a certain courtesan 
named Lonasobhika to the Arahat Vardhamana or Mahavira, and gives a 
good picture of an ancient Jain stupa, which was constructed and decorated 
on exactly the same lines as the Buddhist edifices of a similar kind. In this 
case the building depicted stood on a high plinth, and was approached by 
nine steps, leading to a torana gateway of the Sanchi type, with a garland 

1 See Ann. Rep. A. S., India, pp. 59-104. 

A. Two Yakshis; Indian Museum 

H. Female, half-back view; 
Mathura Museum 

C\ A iliulhisattva, from Katra; Mathura Museum 

A. Nude female on Jain 
railing pillar, Mathura 

IS. Medallion, Mathura 

C. Bracket figure, Mathura 

D. Medallion, Mathura 


E. Medallion, Mathura 

l\ .Medallion, Mathura 

PLATE 1 8 


hanging from it. The stupa was surrounded by a plain railing, and two 
similar railings were carried round the drum. The posturing females are 
unmistakably nude. The side columns are of the so-called Persepolitan type 
and bear the Wheel and Lion. 1 

Not only are certain of the Mathura sculptures definitely comparable to Early 
Bharhut and Sanchi, but it is evident that the tradition was never broken, sculpture at 
Kushan sculpture springing directly from the older school. As has been Mathura - 
said, most of these sculptures had as their function the adornment of Jain or 
Buddhist stupas and consist chiefly of railing pillars and medallions. Many 
of the ancient motives are preserved such as the bull of Plate 18 E and the 
fish-elephant (Makara) of Plate 18 f. The bracket figure in Plate 18 c is 
a development of the 'Woman and Tree' motive used for the same structural 
purpose as at Sanchi. Here the rendering is a little more schematic and 
architectural but much of the bold sinuous freedom of the East Gateway 
nymphs is preserved. The work of this period shows an increasing schematic 
and patterned quality, well illustrated in the knotted foliage of Plate 18 b. 
This delicate abstract treatment of foliage, suggesting the half unfurled leaves 
of the vine, was afterwards used with great effect in the doorways of Gupta 

The excavations at Mathura have yielded numerous specimens of pillars Sculptures 
of stone railings associated with stupas, both Jain and Buddhist. Most of on railings, 
the Buddhist ones were found on the site of Huvishka's monastery in the 
Old Jail or Jamalpur mound, now entirely removed. The Jain specimens 
came from the Kankali mound, which included the remains of an early stupa 
and two temples. The pillars have high-relief statuettes, usually of females, 
on the front, and other panelled scenes one above the other, or floral patterns 
on the back. 

Plate 1 8 A represents a Jain railing pillar on which is carved a Yakshi in the 
conventional Woman and Tree pose. Her beaded belt, heavy ear-rings and 
anklets are interesting and typical of the period. The sword she holds is of 
the ancient Indian kind which was still in use in Mughal and Maharatta days. 
Such rather immodest females adorning many of the pillars were supposed 
by Cunningham to be dancing-girls, an opinion certainly erroneous. They 
appear rather, as argued by Dr. Vogel, to belong to the Yakshi class, like the 
similar figures of the Bharhut railing. Some of the figures seem to be naked, 
but in others the apparent nudity is merely an artistic convention, the female 
drapery being treated schematically by flowing incised lines. This treatment 
of drapery persists throughout Indian art, and is radically different from the 
deeply undercut naturalistic drapery of certain Gandharan work. 2 

Plate i6crepresentsavariant of the common Woman and Tree motive. The 

1 The palaeographical dates quoted in V. Smith's Jain Stupa cannot be accepted. 
* Marshall, Taxila, PI. XXII. 

3»a G 


female stands on a prostrate dwarf, a male Yaksha. The pose, as in many 
other cases, is easy and graceful. A sculpture in Calcutta shows two females 
together, under a tree. A pillar in the Mathura Museum (Plate 17 b) presents 
a half-back view of a female. The unusual attitudes shown in Plate 19 A and 
B are treated much more skilfully, the first being obviously a dancing pose. 
The male figure, seemingly of a soldier, in Plate 19 D is quite exceptional and 
effectively designed. A well-executed sculpture in the Indian Museum 
(Plate 19 e) represents a youth riding a conventional lion. 1 

There is a dearth of photographs of the magnificent sculptures in the 
Mathura Museum and an illustrated catalogue is urgently needed. 
i Bodhi- A seated Bodhisattva (Plate 17 c) in the Mathura Museum, bearing a dedi- 
sattva. catory inscription, 'for the welfare and happiness of all beings', is of special 
interest as exhibiting the saint seated in the traditional yogi attitude, which 
became general subsequently, with his right shoulder bare, and the right 
hand raised in Abhaya Mudra. 2 The drapery is excessively formal in its 
folds, though the modelling of the figure is very suavely accomplished. The 
two flying spirits are early examples of a motive common in the sculpture 
and painting of later periods. The formal portrayal of their scarves and the 
knotted waist-clothes of the other two attendant figures is typical of Kushan 
work. The ushnisha or skull-protuberance is simply represented in a unique 
manner which must be accepted as the primitive form of this divine sign of 
Buddhahood, afterwards influenced by Gandharan forms. The figure is 
called a Bodhisattva in the inscription, although he is seated underneath the 
Bodhi-tree and wears the orthodox costume of the Buddha. The tree is the 
Pipal [ficus reltgiosa], the proper tree of Gautama. 

This sculpture closely corresponds to the Anyor Buddha [Mathura 
Museum, No. A 2], and is typical of the middle Kushan period to which the 
bulk of Mathura sculpture belongs. 

The standing Buddha of the Mathura school found at Sarnath, mentioned 
above, is the earliest dated Buddha-figure, being inscribed in the Kushan 
third year. It may be compared to a Bodhisattva in the Indian Museum, 
Calcutta (Plate 20 c). In the Sarnath sculpture the ushnisha seems to have been 
inset in the head by means of a tenon or mortice. It is interesting to note 
how naively this divine excrescence is treated by the sculptors who first 
dared to portray the Buddha in stone; quite different is the sophisticated 
attempt at disguisement of the Graeco-Buddhist tradition. The treatment of 

1 Dr. Vogel describes a mutilated statue (height 'It is of interest as the only polycephalic image 

3 feet 10 inches or 1 metre 17) of a male deity which can be attributed to that epoch/ 

standing with his left hand resting on his hip 2 In the Kushan period the hand in this Mudra 

(Mathura Museum, E. 12, Catal., p. 108), which is left en bloc with the shoulder. The Abhaya is 

evidently had three heads, of which that on the the usual Mudra, both at Mathura and Amara- 

proper right has been lost. The style indicates vati. 
that the image belongs to the Kushan period. 

B. Female with right leg bent; 
Mathura Museum 

Female with right arm bent ; 
Mathura Museum 

C Female and child; 
Mathura Museum 

K. Lion and rider; 
Indian Museum 

I). A soldier; Mathura Museum 

A. Pali Khcra block, front group. Mathura Museum 

B. Kuvera, Mathura Museum 

C. Bodhisattva from Mathura; 
Indian Museum 



drapery and jewellery in these Kushan Buddhas and Bodhisattvas is purely 
Indian. However, a distinct type of Buddha-figure is to be found at Mathura, 
which approaches the Gandharan image in its treatment of the clothing and 
its drapery. Most of these figures appear to belong to the later reigns of the 
dynasty. They have a certain clumsiness about them that suggests foreign 

Among the Mathura sculptures of the Kushan period is a rather anomalous Western 
group which is usually considered to be the result of foreign influence. The influence. 
technique of these sculptures is one with that of the purely Indian sculptures 
already discussed. The treatment of the figure is easy and naturalistic, 
although somewhat heavy and lacking in rhythm when compared to Bharhut 
and Sanchi. The drapery is somewhat markedly less stylistic than that of 
the early Buddhas and Bodhisattvas. The foliage backgrounds are also 
absolutely according to the Indian tradition. However, the subjects of this 
group of sculptures do not seem to be either Jain or Buddhist. 

The much discussed group, usually described as 'Herakles and the Nemean Herakles and 
Lion', was discovered by Cunningham serving a lowly purpose as the side the Nemean 
of a cattle-trough and is now in the Indian Museum, Calcutta. It is 2 feet Lion - 
5 inches high. The hero grasps the beast with his left arm and presumably 
threatened it with a club in the missing left hand. He is nude, except for 
a skin hung behind his back, and fastened by the paws round the neck. The 
lion with stylized mane is typically Indian, like the lions supporting Kushan 
thrones. The naturalistic, full modelling of the figure has been considered 
to be the result of Greek reminiscences. 1 

The Herakles-and-Lion motive is of great antiquity, going back to Assyrian 
art,which represented Gistubar,the 'Assyrian Hercules', clubbing and strang- 
ling a lion in the same way. 2 This Indian version is usually dubbed Hellenistic 
with an airy indication of Western Asia as the source of the foreign influence. 

A certain group of sculptures from Mathura or its neighbourhood, all deal- Bacchanalian 
ing with strong drink and intoxication, which may be classed together as images. 
'Bacchanalian', have excited much interest and discussion, in spite of which 
their interpretation is still far from clear. The supposed Greek character of 
these sculptures, when first discovered, was much exaggerated by the early 
commentators. As with the Herakles and most examples of Western influence 
in Indian art this 'Greek character* is difficult to define. 3 

The block discovered in 1836 by Colonel Stacy at Mathura and now Silenus. 

1 The group is M. 17, 1. M., Calcutta: Anderson, p. 136, Fig. 36; Maspero, Ancient Egypt and 
Catal., Pt. I, p. 190. See Cunningham, A. S. Assyria (trans. 1892), p. 302, Fig. 152. 

Rep., vol. xvii, p. 139, PI. XXX. Vincent Smith 3 V. Smith says that many of these groups have 

refers to the bronze figure, 2} feet high, from 'nothing Hellenistic about them'. Like the 

Quetta, Baluchistan: J. A. S. B., Pt. I, vol. lvi, Gandharan sculptures their subjects are clearly 

p. 163, PI. X. Indian though neither Buddhist, Hindu, nor 

2 Bonomi, Nineveh and its Palaces, 2nd. ed., Jain. 


marked M. i in the Indian Museum, Calcutta, was at first supposed to 
represent Silenus, and so became known as the 'Stacy Silenus\ But every- 
body now acknowledges that the subject is Indian, whether the sculptor was 
influenced by the Silenus model or not. The stone is 3 feet 8 inches high, 
3 feet broad, and 1 foot 4 inches thick, with a circular basin on the top 16 
inches in diameter and 8 in depth, seemingly intended to serve as the socket 
for a column. Both this block and its replica, to be described presently, were 
carved on back as well as front, and were evidently designed to be viewed 
from both directions. Apparently they were the bases of columns, which 
stood at an entrance, or entrances. But the difference of dimensions suggests 
that the two blocks may have belonged to distinct buildings. 

The front group comprises four persons in two pairs, each consisting of a 
man and woman standing under an asoka tree in flower. The stout man on 
the right has his left arm round the waist of his female companion, who 
holds his right hand in hers, thus giving him the support rendered neces- 
sary by his intoxicated condition, due to the liquor, pots of which stand on 
the ground. The couple on the left stand facing, in attitudes apparently in- 
different, but their countenances have been destroyed, so that their expres- 
sion is lost. Traces of chaplets may be discerned on the heads of all. 

The reeling man wears nothing except a pair of short bathing-drawers, and 
a scarf or cloak hanging behind his back and fastened round his neck by a 
knot. The slighter and perfectly sober man on the left is decently dressed in 
long drawers extending to his ankles, and a close-fitting tunic reaching below 
his knees. Both of the women are clad in a short tunic coming down a little 
below the waist, and possibly also in a long skirt. Each holds a piece of loose 
drapery, worn as a scarf, across her legs. The woman on the left has it thrown 
over her left arm in the fashion adopted by some of the Gandhara Bodhi- 
sattvas. Both women are adorned with heavy Indian anklets, armlets, and 

The reverse group, much mutilated, comprises five figures, of whom the 

principal is a fat elderly man sitting on a stone seat with his left leg tucked 

up, and so drunk that he has to be supported on his left side by a man and 

a boy, and on his right by a woman dressed like the females in the front 

group. The drunkard does not wear drawers like the merry fellow in that 

composition, but has a waistcloth loosely fastened. In style both reliefs are 

similar, the modelling being life-like, and the action clearly expressed. 

The Pali The companion block of nearly the same dimensions, but somewhat larger, 

block* was discovered many years later by the late Mr. F. S. Growse at Pali Khera, 

' a suburb of modern Mathura included within the limits of the ancient city. 

The reverse group, exhibiting the effects of deep potations, being almost 

identical with the reverse of the Stacy block, need not be further described. 

The front group, however, differs from its companion. Five figures under an 


asoka tree again appear. The principal is a fat man, seemingly nude, seated 
with his left leg tucked up, on a low heap of stones laid in courses, in the 
conventional manner usually used to indicate mountain-heights. He is drink- 
ing from a noggin, apparently of wood, which a male attendant is ready to 
replenish. The proceedings are watched by another man, a woman, and a 
small boy (Plate 20 a). 

Two other Bacchanalian groups, found among the sculptures in the Bacchana- 
Mathura Museum by Dr. Vogel and described by him, throw welcome light lian Kuvera. 
upon the date and meaning of the earlier discoveries described above. One 
of these groups, 1 foot 2 inches high (Plate 20 b), represents a corpulent, 
coarse-looking man, apparently nude, squatted, and holding in his right hand 
a cup, which a female attendant is about to fill from a jar. His left hand 
grasps a long object, presumed to be a money-bag. This last attribute and 
the physique of the obese drinker permit of little doubt that the personage 
represented is Kuvera, the god of riches, whose podgy form has become 
familiar from the many images collected of late years in connexion with 
Buddhist monasteries from the Punjab to Ceylon. This sculpture, however, 
is medieval and closely corresponds to another of reddish sandstone, probably 
of Mathura workmanship, found at Osia, Rajputana. 1 Kuvera (also called 
Vaisravana and Jambhala) was king of the Yaksha demi-gods or sprites, and 
forms of his effigy are closely related to certain images from Gandhara. Dr. 
Vogel probably is right in associating all the Bacchanalian sculptures of 
Mathura with Yaksha worship. 2 

Mr. Growse also published a mutilated statue, 3 feet 1 inch high, lying at Naga 
Kukargrama in the Saadabad pargana of the Mathura District — a singularly sculptures, 
graceful figure of a Naga youth with a canopy of seven cobra heads, holding 
his right hand above his head, while his left grasps a cup similar in shape 
to that seen on the Pali Khera block, but apparently without the curved 
handle. A garland of wild flowers is twined round his body, and he wears 
a high head-dress of a pattern commonly found in Kushan sculptures. The 
worship of the Nagas, the spirits of the waters, was much favoured by the 
ancient inhabitants of the Mathura region in Kushan times. 3 This drinking 
Naga is related to another fine life-size statue of a Naga water-sprite from 
Chhargaon, near Mathura, now in the Mathura Museum, the approximate 
date of which is fixed by an inscription on the back, recorded in the fortieth 
year during the reign of Huvishka. According to the chronology provisionally 
adopted in this work, the statue (Plate 16 b), which is 5 feet high, may be 
ascribed to the year a.d. 117 or 118. The modelling is good. The arrange- 

1 'Etudes de Sculpture bouddhique' (Bull, de 60. The second group in the Mathura Museum 

I'&ole francaise d'Extreme-Or., t. viii (1908), differs little from the one figured. 

Nos. 3, 4, Fig. 2) ; Am. Rep. A. S., India, 1906- 2 See A. S. Rep., 1906^7. 

7, "The Mathura School of Sculpture*, pp. 137- 3 loc. cit. 


ment of the waistcloth in a twisted roll is typically Kushan. The broken left 
hand probably held a cup, 1 

ibjectsof Besides the Kushan Buddhas or Bodhisattvas and the Nagas, various 
Kushan canonical scenes are found in bas-relief. A common representation is the 

:ulptures. v ^ t ^ j ndr2i to B u( jdha in the Indrasila cave. The mountainous locality is 
conventionally indicated by 'rock-work' and its desolateness by birds and 
beasts looking out from their lairs. Plate 15 A represents a three-tiered stupa 
with trees on either side of it and pairs of harpies (Suparnas) and centaurs 
(Kinnaras) bringing offerings and garlands. These ' offering-bearer ' scenes are 
very common and, of course, are also to be found at Bharhut and Sanchi. At 
Mathura and in Gandhara they develop into processions and pageants as in 
the archway spandril (Plate 21). The figure-sculpture here is excellent, the 
garland-bearers of the middle band being portrayed with a fine rhythmic effect. 
The floral-bands are very simply treated and are typical of a common style 
of Kushan decoration. 

It is to be noticed that just as there are fewer Jataka scenes at Sanchi than 
at Bharhut, there appear to be still fewer at Kushan Mathura. The canon is 
fast crystallizing into a literary form, to the exclusion of the ancient popular 
parables. The Jatakas, which are to be recognized at Ajanta, are on the whole 
of a different class, most of them being definitely literary. 


onaravati The sculptures from the stupa of Amaravati and its surrounding railing or 

iTknowiT screen °f mar ble may claim the distinction of being the most accessible 

' specimens of early Indian art. No visitor to the British Museum, however 

indifferent to Indian curiosities, can help seeing the spoils of the stupa and 

railing displayed on the walls of the grand staircase. 

^miction The small town of Amaravati on the south bank of the Krishna (Kistna) 

e stupa. r j ver ^ j n ^ Guntur District, Madras, represents a more important ancient 

city called Dharanikota, a place of considerable note from at least 200 B.C. 

1 Vogel in Prog. Rep. A. S., N. Circle, 1907-8, expressive, as Mr. Growse suggested, of a little 

p. 38;^. jR. A. S., 1910, p. 1313 n. These so- understood sensual form of popular Buddhism, 

called 'Bacchanalian' sculptures of Mathura can- not indicated by literature until a time seemingly 

not be at all understood if considered by them- much later than the second century. But when 

selves. They evidently belong to a large class the true history of Indian Buddhism comes to be 

of Buddhist works of art, represented by the written it must be based on the evidence of the 

'scenes bacchiques' of Gandhara, which fill two sculptures and pictures as much as on the books, 

plates of M. Foucher's book (Foucher, VArt M. Roller's question, addressed to Christian 

gr&o-bouddhique du Gandhara, Figs. 127-33 b), ecclesiastical archaeologists with reference to 

several reliefs on railing pillars at Mathura, the the art of the Catacombs, may be repeated to 

'Indian Bacchus' of the Tank silver dish, and Indianists: 'La pierre ne servirait-elle pas k con- 

the festive scenes depicted in the Aurangabad troler le manuscrit?' (Les Catacombes de Rome 

and Bagh Caves. All such works appear to be (1881), Preface, p. ii.) 

PLATE 21. Torana arch; Mathura 

PLATE 22, Slab with representation of a stupa, &c, from the base of the great stupa, Amaravati 


A richly decorated stupa, known to have been in good repair and still venerated 
in the twelfth century, continued to exist to the south of the town up to the 
close of the eighteenth or the beginning of the nineteenth century, when it 
was utterly destroyed by a greedy local landholder, eager to obtain cheap 
building material and convinced that marble slabs, plain or carved, formed 
excellent food for a lime-kiln. About a century ago Colonel Mackenzie 
visited the place and had drawings made of numerous slabs, now no longer 
in existence. Various archaeological explorers have salved remnants of the 
sculptures, which are now mostly housed in either the British Museum or 
the Central Museum, Madras. Our knowledge of the extraordinary richness 
of the decoration of the stupa and its railing is derived from the poor remnants 
thus rescued and Colonel Mackenzie's drawings, which have been published 
fully by Mr. Fergusson and Dr. Burgess. 

The stupa in its earliest form was of high antiquity, dating, as inscriptions Date of the 
prove, from about 200 B.C. 1 But the great mass of the sculpture is much principal 
later, and belongs to the Kushan period. The authority of the Kushan kings, scuI P tures - 
however, did not extend as far south as Amaravati, which was then within 
the dominions of the powerful Andhra dynasty of the Deccan. By the help 
of two inscriptions mentioning Andhra kings, the construction of the great 
railing may be assigned to the half-century between 150 and 200 after Christ. 
The highly ornate slabs which cased the stupa itself may be a little later. We 
are almost certainly safe in saying that all the sculptures of the railing and 
casing fall within the hundred years between a.d. 150 and 250. Originally it 
was believed that there used to be two railings, and all the printed descrip- 
tions give details of an 'outer' and an 'inner* railing. But Dr. Burgess later 
stated that he and everybody else were mistaken, the fact being that no more 
than one railing, the so-called 'outer* one, ever existed. The slabs supposed 
to have belonged to an 'inner' railing really formed a casing applied to the 
body of the stupa. 2 However, two types of sculpture clearly belonging to two 
different periods are distinguishable. In the first the Buddha figure is not 
found: in the second it is. The latter is stylistically also very much easier 
and richer. The bulk of the sculptures belong to this second period. 

The railing, by far the most magnificent known example of such structures, The railing. 
was 192 feet in diameter, about 600 in circumference, and stood 13 or 14 feet 
high above the pavement. It was constructed of upright slabs connected by 
three cross-bars between each pair of uprights, which stood upon a plinth 
and supported a coping about 2 feet 9 inches in height. On the outer face 
each upright was adorned with a full disk in the centre and a half-disk at 

1 These inscribed slabs arc unsculptured. The not earlier than second century a.d. and most 

characters are described as being of 'Mauryan of them later— on palaeographical evidence, 

type*, a rather inexact phrase. Konow doubts 2 Fergusson, Hist, of Ind. and E. Archit., 2nd 

whether any sculpture is pre-Christian. Buhler ed. (1910), vol. i, p. 119. 
seems to regard the inscribed sculptures as being 


top and bottom, minor sculptures filling the interspaces. Similar but ever- 
varying disks decorated the cross-bars, and the coping was ornamented with 
a long wavy flower-roll carried by men, numerous figures being inserted in 
the open spaces. The plinth exhibited a frieze of animals and boys, often in 
comic or ludicrous attitudes. The decorations on the inner face were even 
more elaborate; the coping presenting a continued series of bas-reliefs, and 
the central disks being filled with delicate sculptures, treating every topic of 
Buddhist legend. Thus every part of the structure, with a surface of about 
16,800 square feet, was covered with sculptured reliefs. 
The casing. The slabs forming the casing of the lower part of the stupa> i6z\ feet in 
diameter, were carved more richly even than the inner face of the railing, 
if that be possible. Apparently there were twelve in each quadrant, the 
principal object depicted on each slab being a highly decorated stupa with 
its railing, the rest of the surface being covered with an infinite variety of 
figures. Study of Plate 22, reproducing the best preserved of such slabs, 
will dispense with the necessity for detailed description, and at the same time 
give a good notion of what the appearance of the Amaravati stupa must have 
been in the days of its glory. When fresh and perfect the structure must 
have produced an effect unrivalled in the world. However much severe taste 
may condemn the characteristic Indian lavishness of decoration which 
scorned to leave an inch of plain surface, the vast expanse of sculpture in 
white marble gleaming in the brilliant sunshine cannot have failed to exhibit 
a scene of unequalled splendour. 
Details. While abstaining from minute description of Plate 22, which serves as a 
synopsis of the sculptures generally, I may invite the attention of the reader 
to a few points. In the relief picture the sculptured decoration is carried high 
up the dome, but the extant slabs seem to have been attached only to the 
lower part of the Amaravati stupa. It is possible that higher bands of decora- 
tion may have existed and been wholly destroyed. The railing in the relief 
has four cross-bars, and not only three as in the real monument. The 'moon- 
stone* at the entrance agrees in form, though not in design, with the Ceylonese 
examples. The lions and some of the architectural forms are survivals of the 
Assyrio-Persian patterns of the Asokan age. The meaning of the five stelae 
or pilasters on the face of the stupa is not known. The worshippers in the 
central scene adoring the chair occupied only by an object which may be the 
sacred head-dress relic, might have appeared in a Sanchi or Bharhut relief, 
where images of Buddha are unknown; but here, at the top of the picture, 
we also find Buddha seated in the conventional yogi attitude. The frieze at 
the top of the slab contains nearly fifty figures, and the general effect, like 
that of nearly all the reliefs, is excessively elaborate. But the skill of the artist 
in design and drawing, and his technical powers of execution, are beyond 

A. Basal medallion, Amaravati 

B. Undulating roll motive on coping of rail, Amaravati 

A. Man and boy, Amaravati 

H. Marble Buddhas, Amaravati 



The infinite variety of the patterns used in the medallions and bars may Medallions. 
be realized by study either of actual examples or of the relief pictures. 
Plate 23 A is an excellent and well-preserved example of a charming decorative 
design based on the lotus-flower motive. The beauty and delicacy of the floral 
devices in the border and plinth deserve special notice and admiration. They 
will repay minute examination with a magnifying glass. 

The treatment of floral and animal decorative motives has been illustrated Decorative 
above by photographs on a small scale. Three specimens may be added from motives - 
Mr. Rea's drawings on a larger scale, which have not been published except 
in his book (Plate 25). 

A few separate images have been found at Amaravati. Two large marble Buddhas. 
statues, 6 feet 4 inches in height, are illustrated in Plate 24 B. The opaque 
drapery is treated in a formalized style, quite different from the smooth 
transparent robes of the Gupta period, to be discussed in the next chapter, 
but to a certain extent resembling Gandhara work and the Mathura figures 
discussed above. These images may date from the third or fourth century, 
or even later; they closely correspond to the Buddhas painted on the columns 
in Cave X, Ajanta. 

Fergusson's opinion that the sculptures of the Amaravati school mark 'the Criticism. 
culmination of the art of sculpture in India', which was generally accepted 
until recently by English writers, including myself, does not now command 
such ready assent. I will not presume to say which work marks the 'culmina- 
ting point', but it is certainly safe to affirm that the pre-eminence claimed for 
the Amaravati reliefs may be effectively challenged by compositions of later 
date, at least in some respects. All critics, however, can agree with Mr. 
Havell that the marbles of Amaravati offer 'delightful studies of animal life, 
combined with extremely beautiful conventionalized ornament', and that 
'the most varied and difficult movements of the human figure are drawn and 
modelled with great freedom and skill'. The obvious overcrowding of the 
compositions unfortunately is a defect common in Indian art. Historically, 
the sculptures are interesting as an academic development of the style of 
Sanchi and Bharhut. Mr. Havell may be right in believing that originally 
the effect of the Amaravati marbles was heightened by colour, and in holding 
that technically they should be regarded as 'painted rilievos' rather than as 
true sculpture. But whether they were painted or not, they must have formed , 
when perfect, one of the most splendid exhibitions of artistic skill known in 
the history of the world. 1 

1 Considering the geographical and political in the seventh century, did not really describe 
separation of the Kushan and Andhra empires, the stupa as being 'ornamented with ail the mag- 
I think the presumption is that the sculptors of nificence of the palaces of Bactria (Tahia)', as 
Amaravati had not direct knowledge of the Gand- Fergusson and Burgess suppose him to have 
hara school, although it is possible that they may done. A slight slip of the pen in the Chinese 
have had it. Hiuen Tsiang, the Chinese pilgrim, text used by Julien introduced the word mis- 

320a H 

Chapter Five 


Discovery ^tf Indian art as a whole may complain of undeserved depreciation and 
of Indo- I neglect, one branch of it, the Hellenistic sculpture of the regions on the 
Hellenic art. JL n0 rth-western frontier, anciently known as Gandhara, has received its full 
share of attention in Europe and been the subject of voluminous discussion. 
The existence of an Indo-Hellenic school of sculpture was not recognized 
generally until 1870, when the late Dr. Leitner brought to England a con- 
siderable collection of specimens, to which he gave the name of Graeco- 
Buddhist. But so far back as 1833 Dr. Gerard had disinterred the first known 
example, a circular relief of Buddha, from the chamber of a ruined stupa near 
Kabul. 1 In 1836 James Prinsep published his account of the so-called 
'Silenus' discovered by Colonel Stacy at Mathura, which has been already 
discussed; and in 1848 Cunningham examined the ruins of Jamalgarhi to the 
north-east of Peshawar. His observations, however, were not published until 
many years later. The first description of a selection of the Jamalgarhi sculp- 
tures was that printed by Sir E. C. Bayley in the Journal of the Asiatic 
Society of Bengal for 1852,* with illustrations so miserably rude that they gave 
little notion of the aesthetic value of the objects described. The sculptures 
thus imperfectly illustrated, having been subsequently brought to England, 

translated as 'Bactria'. The pilgrim really This material correction and Mr. Watters's com- 
praised two monasteries in the Deccan as ments on the current 'identification* of the pil- 
'having all the artistic elegance of a great man- grim's monasteries with the Amaravati stupa have 
sion and all the beauty of natural scenery'. The been overlooked in the revision of Fergusson's 
assumption made by Dr. Burgess and other book. 'It is hard', Mr. Watters observes, 'to 
authors that the account of two monasteries given understand how any one could propose to 
by Hiuen Tsiang should be applied to the stupa identify a large monastery among hills and 
of Amaravati is far from being established. Thus streams, and having spacious chambers and great 
disappears the basis for Fergusson's argument corridors, with a building which is only a remark- 
that the school of Amaravati should be con- able tope situated on a plain.' The error con- 
sidered the offspring of the marriage of the art curred in by Julien, Fergusson, and Dr. Burgess 
of the North — that is to say, Bactria as repre- will not readily disappear from books on Indian 
sented by Gandhara — with that of interior India art and antiquities. 

as represented by Sanchi and Bharhut (Fergus- l K. 1 of Indian Museum; Anderson, CataL, 

son, Hist. Ind. and E. Arch., reprint of 1899, P- ^ art h P« 2 ^ J • 

103; new ed. by Burgess (1910), vol. i, p. 123). 2 The approach to the subject by means of a 

'Instead of the ta-hsia, a "great mansion", here, postulated Hellenic artistic tradition progres- 

the B text, used by Julien, has ta hsia, which sively Indianized is by no means scientific. A 

is a Chinese name for the country called Bactria. period of initial Indianization, during which the 

But this is evidently a slip of the pen, and the whole body of the Indian canon was appropriated, 

proper reading is that of the other texts which must necessarily have preceded this so-called 

means a "great mansion" ' (Watters, On Yuan decline. 
Chwang's Travels in India (1905), vol. ii, p. 218). 

A. Lotus forms, Amaravati 

ii - >■■!! n ■ n ii ■■■ ■■mi — ■■! rr»i^rir-rn— i ■ - — ^ a * t ^-^HrT"^rT M ntirMHTr^^r^FTl^^^^^^Hg 

&.-. «~>7) ) ;■•> /j,yj> /J OUjiiJj i ;UD ; ijTu UJJjJj Jll 

B. Lotus and makara, Amaravati 

C. A pond, Amaravati 

A. Court scene. Medallion ; Amaravati 

R. Enthronement of relics. Medallion ; Amaravati 


perished in the fire at the Crystal Palace which also destroyed Major Gill's 
copies of the Ajanta frescoes. Thus it happened that, as already observed, 
Dr. Leitner is entitled to the credit of having first convinced the learned 
world of the fact that during the early centuries of the Christian era North- 
western India was the home of a school of Hellenistic sculpture of consider- 
able artistic merit. 

The fact was so novel and surprising that one distinguished antiquary, European 
Mr. W. Vaux, F.R.S., was bold enough to dispute it, and to declare his interest in 
inability to perceive any manifest traces of Greek art on the sculptures pro- sub J ect - 
cured by Dr. Leitner and other collectors in the neighbourhood of Peshawar. 
In a short time, however, evidence accumulated so rapidly that no possibility 
of doubt remained, and Professor Curtius was able to announce that the dis- 
coveries opened 'a new page in the history of Greek art'. That is the explana- 
tion of the keen interest taken in them by European scholars, who are eager 
to follow out in its most minute details the story of Greek art, on which that 
of modern Europe is based, while they usually remain indifferent, or even 
contemptuous, towards manifestations of artistic power in the nations of the 
East developed independently of the Hellenic tradition. 

During the last forty years thousands of Indo-Hellenic sculptures have Abundance 
come to light, while considerable numbers, including most of the choicest examples, 
specimens, have been catalogued, described, and photographed. The num- 
ber, indeed, is so great that it is difficult to make a small selection thoroughly 
representative. Most of the examples chosen to illustrate this chapter have 
been selected in virtue of their conspicuous aesthetic merits, and may be 
regarded as evidence of the highest attainment of a school of artists working 
on Indian soil, and applying more or less modified Greek methods of com- 
position and technique to Indian subjects. A few of the figures mark the 
gradual disappearance of the Hellenic tradition and the progressive Indianiza- 
tion of the treatment. 1 

The country from which comes this wonderful wealth of semi-foreign The Gan- 
sculpture may be described in general terms as the North- Western Frontier. dhara terri " 
It includes the modern District of Peshawar, the valley of the Kabul river, tory ' 

1 Certain facts may be brought together which Mathura sculptures in Gandhara style bear 

connect the sculptures at several localities with Kushan inscriptions. For reasons stated already 

the Kushan kings: (i) coins of Kanishka in I now take the most probable date of the ac- 

foundation deposit of Sanghao monastery (Cole, cession of Kanishka to be a.d. 78. With re- 

Second Report, p. cxx) ; (2) coin of Huvishka with gard to these coin-finds and the chronology that 

apanelofbeststyleatTakht-i-Bhai^.-R.-^.iS., has been built upon them, it must be realized 

1899, P- 4 22 )i (3) seven coins of Vasudeva with that they actually only provide a lower limit 

Jamalgarhi sculptures (Cunningham, Reports, v. for the dating of Gandharan art. The coins of 

194); (4) coin of Huvishka in good condition at several reigns and even dynasties are often to- 

Ahmposh stupa, along with coins of Sabina, &c. gether. 
(Proc. A. S. B. % 1879, p. 209); (5) some of the 


Swat, Buner, and other tribal territories, as well as the western portion of 
the Punjab between the Indus and the Jhelum. The kingdom of which 
Peshawar (Purushapura) was the capital having been known in ancient times 
as Gandhara, the sculptures are usually described by that territorial name, 
although Graeco-Buddhist finds in Khotan and in the vicinity of Kabul render 
this title rather meaningless. 

The richest sites as yet explored are those crowded together in the Yusufzai 
country to the north and north-east of Peshawar, comprising Jamalgarhi, 
Sahri-Bahlol, Takht-i-Bhai, and many more which it would be tedious to 
enumerate. Some of the best sculptures come from Swat, but the hostility 
of the tribes prevents systematic exploration of the antiquities beyond the 
British frontier. 
Arrange- Even within the frontier most of the exploration done until recently has 
mC ?cts y alone k een ^ wor ^ °^ amateurs > conducted in a haphazard fashion, without the 
JC possible 6 formation or preservation of adequate detailed record. Consequently, many 
buildings have been utterly destroyed, and the value of the large collections 
of sculptures found by many public institutions and private persons is 
seriously impaired by the lack of information concerning the provenance of 
the specimens. M. Foucher, the most learned and authoritative commentator 
on the sculptures, declares that it is impossible in the present state of know- 
ledge to arrange them in chronological order. As a general rule, no doubt, 
the most Greek may be considered the oldest, and the most Indianized the 
latest, but the practical application of this principle presents many difficulties. 
Arrangement by localities is equally impracticable, because nobody knows 
where many of the best examples were found, and also because there is no 
distinct evidence of local variations in style. The general style over the whole 
region is fairly uniform. The result is that the only practicable arrangement 
is one by subjects. In this chapter it will not be possible to illustrate more 
than a few of the multifarious subjects treated by the artists, and students 
who wish to examine the whole field must be referred to special treatises. It 
is hoped, however, that the specimens reproduced will suffice to enable the 
reader to judge of the aesthetic qualities of the sculptures, and to place them 
in their due relation to Greek art on the one hand and to indigenous Indian 
art on the other, subject to a certain amount of vagueness in the chronology 
of the school. 
Chronology. Whenever the date of Kanishka, the celebrated king of Gandhara, shall be 
determined, that of the best period of the Hellenistic sculpture will also be 
known. Many of them undoubtedly are contemporary with him, though 
some are earlier and others later. Without going into complicated anti- 
quarian discussions, it may suffice to say here that none of the sculptures 
are later than a.d. 600, few, if any, later than a.d. 400, and that in all prob- 
ability extremely few are earlier than the Christian era. The culmination of 


the art of the school may be dated from about a.d. 50 to a.d. 150 or 200. It 
is quite safe to affirm that the works of good quality belong to the first three 
centuries of the Christian era. Thus the best productions of the Gandhara 
Indo-Hellenistic school nearly synchronize with the art of the Flavian and 
Antonine periods in Western Asia and Europe, and in India with the reliefs 
on the great rail at Amaravati in the Deccan, as well as with many sculptures 
at Mathura on the Jumna, both of which will be discussed in the next chapter. 

Without exception, all the sculptures come from Buddhist sites and were All the 
executed in the service of the Buddhist religion, so far as is known. No trace sculptures 
of works of the pure Gandharan school dedicated to either Jainism or Brah- Buddhist - 
manical Hinduism has been discovered. Moreover, the subjects treated are 
not only Buddhist but purely Indian. Buddha may appear in the guise of 
Apollo, the god Brahma in that of St. Peter, or a door-keeper in that of Pallas 
Athene, but however Greek may be the form, the personages and incidents 
are all Indian, and centre round the person of Buddha, whose image dominates 
the compositions. 

Herein lies the most obvious, and at the same time, perhaps, the most Dominance 
important difference between the ancient schools of interior India at Sanchi, ? f the 
Bharhut, or Bodh Gaya, and the school of Gandhara, and the contem- ^dha. 
porary art of Mathura and Amaravati. In Gandhara art, as M. Foucher 
observes, Buddha is everywhere ; and whatever be the form which he assumes, 
as Prince Charming, emaciated ascetic, or ideal monk, or by whatever name 
he may be called, whether it be Siddhartha, Sramana Gautama, or Buddha 
Sakyamuni, he dominates almost every composition, so that the preparation 
of a full list of the sculptors 1 subjects is equivalent to writing an illustrated 
life of the Master. The early schools of Indian art, as we have seen, were 
content to indicate his supposed presence by mere symbols, and did not pre- 
sume to imagine his bodily likeness. 

The material of the sculptures is usually a blue clay-slate, described as Material, 
'horn-blende-schist'. The stone was finished with fine plaster, like the rock 
sculptures of Ajanta and many other localities in India and Ceylon, and the 
effect was heightened by the free use of colour and gilding, traces of which 
are still nearly always discernible. 

Great numbers of detached heads, made sometimes of stucco and some- piaster 
times of terra-cotta, have been found, varying in dimensions from tiny objects heads, 
two or three inches high to life size. These heads, as various in character as 
in dimensions, are often of high artistic merit. One mode of their use is 
explained by an observation of Masson, who noted that at Hidda, near 
Jalalabad, in the upper Kabul valley, 

'idols in great numbers are found. They are small, of one and the same kind, about 
six or eight inches in height, and consist of a strong cast head fixed on a body of earth, 
whence the heads only can be brought away. They are seated and clothed in folds of 


drapery, and the hair is woven into rows of curls. The bodies are sometimes painted 
with red lead, and rarely covered with leaf-gold; they appear to have been interred in 
apartments, of which fragments are also found.' * 

A period of work in stucco and clay seems to have succeeded the best period 
of work in schist. The latter work at Taxila is all stucco and clay. 2 Moulds 
were used for the wholesale reduplication of these heads. Buddhists consider 
the multiplication of sacred images an act of merit, and the practice of making 
the bodies cheaply with clay enabled the pious donor to accumulate a credit 
balance of numerous good works without undue expense. Mr. J. P. Rawlins, 
who was stationed for a considerable time in the Hazara District, now in the 
North- Western Frontier Province, informs me that in that country he has 
seen numbers of perfect plaster casts, 'for the most part only of heads, of all 
sizes and descriptions, fastened to the walls in appropriate groupings or 
singly. Many of them seem to be portraits of living people at the time, full 
of expression, and with many and varied head-dresses/ My informant 
believed the practice to have 'come down from Greek times'. The age of 
the heads actually seen by Mr. Rawlins does not appear, but, whatever it 
may be, the practice referred to by him proves that the ancient stucco and 
terra-cotta heads might have been used to fix on walls as well as on clay 
images. When objects of this class were exhibited before the Asiatic Society 
of Bengal, Sir J. B. Phear remarked that similar heads from the neighbour- 
hood of Peshawar preserved in the Indian Museum, Calcutta, 

'obviously had been attached to masonry, and no doubt formed part of a subject 
worked out in high relief upon the frieze of some building. It was also remarkable 
that every one of them was unsymmetrical, i.e. compressed or flattened either on the 
right side or on the left side. The purpose of this must have been to adapt them to 
being seen with the greater artistic effect from a particular point of view; and it 
indicated considerable advance in knowledge of the peculiar conditions necessary for 
the success of sculptural ornament.' 3 

Examples. The British Museum possesses about forty such detached heads, mostly 

from the Peshawar District, purchased in 1861, fifteen of which have been 

published by Dr. Burgess. Two of those are here reproduced (Plates 38 A 

and b). Terra-cotta heads, somewhat similar in character, have been found in 

excavations at Sahet-Mahet in Oudh, supposed to be the site of Sravasti. 4 

No Greek No trace of the existence of Greek architecture in either India proper or 

architecture ^ borderland has ever been found, that is to say, no building yet examined 

' was designed on a Greek plan, or with an elevation exhibiting one or other 

of the Greek orders, Doris, Ionic, or Corinthian. But the Indo-Hellenic 

1 Ariana Antigua, p. 113. PI. XXVIII. For other Gandhara stucco heads 

2 Marshall, Taxila Guide, p. 108. from Sahri Bahlol see Ann. Rep. A. S., India, 

3 Proc. A. S. J3., 1870, p. 217. 1906-7, p. 107, Fig. 2 and PI. XXXV. 

4 J. A. S. B n Part I, vol. Ixi (1892), extra No., 

PLATE 27. Buddha, &c; relief from Muhammad Nari 

PLATE 28. Modified Corinthian capitals from Gandhara 


architects freely used certain debased Greek architectural forms — columns, 
pilasters, and capitals — for decorative purposes, much in the same way as 
English architects of a century ago often applied a Greek pediment to the 
front of an English dwelling-house. The Ionic column has been found in two 
temples on the site of Taxila, associated in one case with coins of Azes I, 
who is supposed to have reigned between 90 and 40 B.C. 1 Growse noted the 
occurrence of a 'niche supported by columns with Ionic capitals' on a frag- 
ment of sculpture at Mathura, 2 and Simpson found the plaster fragment of 
a capital with corner volutes of the Romano-Ionic kind in the Ahinposh stupa 
near Jalalabad in the valley of the Kabul river. 3 More recently two more 
quasi-Ionic capitals have been discovered, one at Patna and the other at 
Sarnath, but they are really only variants of the Indian bracket-capital. 4 
The Kashmir columns are often denominated 'Doric', but there is no real 

The abundance of modified Corinthian columns, pilasters, and capitals in Indo- 
the art of Gandhara contrasts strongly with the total lack of Doric and the Corinthian 
extreme rarity of Ionic forms. Most of the Gandharan friezes exhibit repre- fo^™* 1 ™ 
sentations of columns or pilasters with capitals more or less related to those 
of the Corinthian order, and which may be fairly called Indo-Corinthian. The 
shafts, whether round or square, are never fluted, and resemble those of the 
second or third century after Christ at Palmyra and Baalbec. The bases of 
structural pillars have been found at Jamalgarhi, and show that the shaft 
might be either cylindrical or square. The conviction of the architects that 
the form of column used concerned merely the decoration of a facade is well 
illustrated by the often-published slab from Muhammad Nari, on which 
Persepolitan columns are mixed up with Indo-Corinthian pilasters (Plate 27). 

The Indo-Corinthian capitals vary widely in detail, but all may be described 
as agreeing generally with the luxuriant cosmopolitan style in vogue through- 
out the Roman Empire during the early centuries of the Christian era. Six 
good specimens, believed to be from Jamalgarhi, are grouped together in 
Plate 28. The introduction of figures of Buddha in two cases may be illus- 
trated from Graeco-Roman art of the time of Augustus, and again, two 
centuries later, at the Baths of Caracalla. The shell canopy is found in the 
art of both Alexandria and Asia Minor. Even the modillions of cornices are 
sometimes made in the form of miniature Corinthian pilasters. 5 All capitals 
of the Indo-Corinthian class seem to be post-Christian, and their introduction 
appears to have been associated with the Kushan conquest of Kabul and the 
Punjab during the first century of the Christian era. 

1 Cunningham, A. S. Rep., vol. ii, p. 129; vol. v, * Hist. Ind. andE. Archit., 2nd ed. (1910), vol. i, 
pp. 60, 72, 100 ; vol. xiv, p. 9, PI. VII : Early Hist. p. 207, note and woodcut. 
of India, 2nd ed., p. 227. s The purely Indian architecture of the earliest 

* Mathura, A District Memoir, 3rd ed., p. 171. cave-temples makes great use of beam-ends, 
3 Proc. A. S. B., 1879, P- 20 9> PL XL here decorated m ** Corinthian manner. 


Two classes The figure sculptures, as distinguished from detached heads and from 
of figure mere iy decorative motives, may be grouped in two classes, as detached statues 
scu pture. Qr smB \i groups, often completely or nearly completely in the round, and 
relief pictures illustrating sacred stories in successive scenes. The reliefs, 
commonly spoken of as 'bas-reliefs', are, as a matter of fact, more often in 
high relief. 
Infinite The statues and small groups represent Buddhas, Bodhisattvas, or saints 
variety of on t h e wa y t0 become Buddhas, besides minor deities of the populous Bud- 
subjects. jkj gt p ant j ieon The stone pictures, like the later painted pictures at Ajanta, 
deal with the infinite variety of subjects presented by the scriptures, legends, 
and traditions of the developed system of Buddhism, known as the Mahay ana, 
or 'Great Vehicle*. That system practically deified Gautama the Buddha, as 
well as other Buddhas, and surrounded them with a crowd of attendant 
deities, including Indra or Sakra, Brahma, and other members of the Brah- 
manical heavenly host, besides a multitude of attendant sprites, male and 
female, of diverse kinds and varying rank, in addition to human worshippers. 
All the elements making up this motley retinue appear in the reliefs, and 
offer infinite opportunities for the exercise of fancy by the artists, who did 
not feel bound by strict rules, such as those of the Silpa-sastras. Although 
the accessible sculptures amount to only a small fraction of those which once 
existed, or even of those known to exist, they are thousands in number, and 
so varied in subject and treatment that several bulky volumes would be re- 
quired for their adequate description and illustration. In this work it 
is not possible to give more than a small selection, representative so far as 
Historical The Gandhara sculptures suggest problems and speculations of many 
interest of ki n( j s# Regarded as an authentic expression of an obviously literary religious 
C nires" tradition, they control and illustrate the testimony of the Buddhist scriptures, 
throwing much fresh light upon the beliefs and practices of the early followers 
of the Great Vehicle. Viewed as a collection of sacred effigies they serve as 
a guide to the iconography of Buddhism, an aspect of the study specially 
attractive to Dr. Burgess and M. Foucher, which must be almost ignored in 
this volume. 

Considered as pictures of human life, they present as in a mirror a vivid 
image of almost every phase of the life of Northern India, lay and clerical, 
during several centuries. The artists cause to pass before our eyes landscapes, 
towns, domestic interiors, streets, fields, trees, and animals, with unlimited 
realistic detail. All the material objects of the civilization of the times — 
furniture, vehicles, arms, tools, and the rest, are depicted as they were used 
by the ancients, and numberless illustrations of the manners and customs of 
the times bring clearly before our imagination the way in which those ancients 
passed their days. Every class of the population from prince to pariah is 

PLATF 29. Seated Buddha, Berlin Museum 










represented, and, in short, no subject of human interest was regarded as 
material unsuitable for the sculptor's chisel. 1 

Just as the sculptures and paintings of the Catacombs and the writings of Develop- 
the early Christian Fathers prove that no trustworthy tradition concerning ment of 
the person of Jesus survived in the Church, and that artists for several cen- ?^ dha 
turies felt themselves at liberty to give free scope to their fancy in delineating 
His image, even so, during the first two or three centuries of the Christian 
era, Buddhist sculptors had not arrived at any settled convention as to the 
correct way of representing the effigy of Gautama the Buddha, whose real 
appearance in the flesh had been utterly forgotten. A long course of experi- 
ment was needed before Buddhist orthodoxy, guided by the later sculptors 
of the Gandhara school, settled down to the monotonous and insipid con- 
ventionality of the figures of Buddha now manufactured by the thousand, 
and adopted, with rare exceptions, in all Buddhist lands. Ultimately, the con- 
ception of the Indian yogi ascetic as worked out in Mathura, Amaravati, and 
Gandhara became dominant, and passed through Khotan to the Far East. 

A Buddha with long hair and moustaches, although not unknown even now Buddha with 
in Japan, would seem strange and improper to most modern Buddhists. It moustaches, 
is, indeed, essentially un-Indian. In Gandhara such a presentation of the 
Master long continued to be legitimate, and the legend of the cutting of his 
locks when he dismissed the charioteer, although known, was usually ignored 
in sculpture. 

The remarkable figure, which recurs frequently in variant forms at the The Thun- 
Buddha's side, requires explanation. His characteristic attribute is the derbolt- 
thunderbolt (Sanskrit vajra, Tibetan dorje) held in his left hand. The older 
writers on Buddhism wrongly identified the Thunderbolt-Bearer as Deva- 
datta, the heresiarch enemy of Gautama Buddha; or as Mara, the Buddhist 
Satan; or as the god Sakra, the Indra of Brahmanical mythology. Dr. Vogel 
has developed a fourth theory, ingenious but not proved, that he should be 
regarded as a personification of Dharma, the Law. The best-supported 
hypothesis is that which treats him as a Yaksha, or attendant sprite, insepar- 
able from the person of the Buddha. Probably the sculptors intended that 
he should be considered invisible to spectators, in accordance with a well- 
understood convention. The figure occurs on one relief of the Mathura 
school, a fragment of a stele found to the south of the city of Mathura (Catal. 
Archaeol. Museum, Mathura, No. H. 5, p. 127). 

A seated Buddha in the Berlin Museum (Plate 29) is one of the finest 

1 This attitude may be criticized on two grounds: inordinately clumsy in portraying many of the 

firstly, that Gandhara, the birth-place of this most ordinary eastern subjects — their lotuses 

hybrid art, is not in India proper; the North- are often almost unrecognizable, as also are their 

West has always been a land of mixed races and trees, 
traditions: secondly, the Gandharan artists are 

3292 t 


examples of the early Buddha type, with coiled hair, moustaches, and the 
robe falling over the feet. 
The visit One of the most elaborate and beautiful products of Gandhara art is the 
of Indra. relief panel from Loriyan Tangai in Swat (3 feet 10 inches x 2 feet 8 inches), 
representing the visit of the god Sakra (Indra) to Buddha while seated in a 
cave near Bodh Gaya (Plate 30 a). Here the central figure has a sweet, calm 
dignity, while the numerous subordinate figures and the scenery are rendered 
with much grace and beauty. The device of exhibiting wild beasts looking 
out from their dens as a conventional indication that the scene is laid in a wild 
mountain country is common in early Indian art, and occurs more than once 
in sculptures of Gupta age. 
The meaning of the composition is explained by Griinwedel: 

'The Swat sculpture represents the visit of Sakra and his retinue, with the Gand- 
harva harper Panchasika, to the Buddha while he was living in the Indrasailaguha, 
a cave near Bodh-Gaya. The entrance of the cave is surrounded by flames to represent 
the glory of the Teacher, "resplendent with a halo of many colours, extending to 
a fathom's length all round his person/' Above and below, the birds, beasts, and trees 
indicate the isolation of the place. Indra appears as a royal personage on the right, 
doing reverence to the ascetic, with his parasol-bearer close behind, and the Devas 
[minor deities] of his train beyond on both sides. His peculiar crown or head-dress 
is very similar to what we find also in the Mathura sculpture. The figure of the Gand- 
harva musician on the other side has been much damaged by the fracture of the stone, 
but his harp is still visible.' 1 

Sundry Four various representations of Buddha are shown in Plate 32. In Fig. A 
Buddhas. the Master is depicted with flames issuing from his head and the water of life 
from his feet. This represents the fire and water miracle ( Yamaka-Prattharva) 
mentioned in Jataka, No. 483.* A remarkable parallel occurs in the Cata- 
combs of Rome, where we find similar representations of the water of life 
streaming from the feet of Christ. 3 Fig. B shows Buddha seated under a tree. 
Fig. C is a good specimen of Buddha seated on the 'diamond throne', closely 
resembling the Berlin figure seated on a 'lion throne* (ante, Plate 29). 
The remaining figure D is interesting as a distinctly more Indian Buddha 
type, on a 'lotus throne', and with the soles of the feet turned up in yogi 
fashion. The right shoulder is bared. This represents the latter part of the 
Great Miracle at Sravasti when the Buddha multiplied his person in the air 
and was heard preaching on all sides/ 4 
The It is impossible to omit notice of the remarkable sculpture, 2 feet 8| inches 
high, representing the Emaciated Buddha, or, more accurately, Bodhisattva, 

1 Grttnwedel-Burgess, Buddhist Art, p. 142. 4 The 'Indianized' style of a number of these 

2 Roller, Les Catacombes de Rome (1881), vol. ii, Great Miracle reliefs and the fact that the sub- 
p. 291, PI. LXXXVII, Figs. 2, 3, 4. ject does not appear in India proper until post- 

3 See Rhys-Davids, Buddhist Birth Stories, p. Gupta times, suggests a closer reliance of Gan- 
105; Spence Hardy, Manual, p. 331 ; Foucher, dhara upon India for inspiration than is usually 
Beginnings of Buddhist Art, p. 1 53 . acknowledged. 







*{*i ;> ;JT> 


i. 'Tfc *^ 














_ r -*** jr. 

A. The first miracle at Sravasti 

IJ. Buddha seated under tree 

C. Buddha seated 

I). The Great Miracle at Sravasti 



in the Lahore Museum, excavated from the ruins of a monastery at Sikri in 
1889, which is the most notable known example of the treatment of a repul- 
sive subject. It depicts the Master as he sat at Bodh-Gaya making the vain 
attempt to attain by the severest austerity that supreme knowledge which did 
not come to him, according to the story, until he abandoned the practice of 
self-torture (Plate 30 b). 1 The subject is sometimes treated by Chinese and 
Japanese artists in another fashion, as may be seen in the South Kensington 
Museum and the Musee Guimet. The Brahmanical parallel is Bhringi, an 
attendant of Siva, who 'was a model ascetic, and fasted so continuously that 
he became not only emaciated, but a living skeleton. He is so represented in 
the sculptures of the caves of Elephanta near Bombay/ 2 

We cannot linger over the Buddha figures, or attempt to follow the personal Bodhisattvas, 
history of Gautama from his conception and infancy to the funeral pyre and 
the distribution of his relics, as depicted in a long series of reliefs; but must 
pass on to another class of images, formerly described as 'kings' or 'royal 
personages', but now recognized as Bodhisattvas, or saints destined to become 
Buddhas. AH considerable collections include specimens, and many have 
been published. 

An image in the Lahore Museum (No. 0239), with finely sculptured 
drapery, is a beautiful work, and typical of its class (Plate 31 a). The small 
relief on the pedestal follows the tradition of the Early School in the interior 
by abstaining from all attempt to image the dead Master, his presence being 
symbolized by the empty seat. 

A larger statuette found near Peshawar, and generally regarded as the most Kuvera. 
striking piece in the large collection of sculptures in the Central Museum, 
Lahore , represents a royal personage seated in European fashion on a throne, 
with his left foot on a footstool and his left hand grasping a spear, his attitude 
being obviously reminiscent of that of the Zeus of Phidias (Plate 33). This 
notable figure, at one time believed to be the portrait of an Indo-Scythian 
monarch, is now recognized as Kuvera or Vaisravana, god of riches and king 
of the Yakshas, who played a very important part in Indian Buddhism, and 
will be met with again in medieval times. The image is free from the tinge 
of effeminacy which mars some of the best finished works of the school, and 
must always command admiration for its virility and dignity. 3 

Excavations at Sahri-Bahlol yielded another figure of the throned Kuvera Kuvera and 


1 Lahore Museum Guide, PI. V; Senart, 'Notes 3 Found at Tahkal on the old road from Pesha- 
d'fipigraphie Indienne,' iii, PI. II (Journal As., war to the Khyber Pass. A cast is in the Indian 
1890). Section of the Victoria and Albert Museum, 

2 Monier Williams, Religious Thought and Life in South Kensington. J. A. S. 2?., Part I, vol. 
India, p. 441. 'A muni or sage— perhaps Bringi lviii (1889), p. 122, PI. VIII; Lahore Museum 
[sic] — very lean, with a long beard, and an offer- Guide, PL III ; Vogel, 'Note sur une Statue du 
ing in his left hand' (Burgess, The Rock Temples Gandhara conserve au Musee de Lahore' 
of Elephanta or Gharupuri, Bombay (1871), p. 23). (B. E. F. E. O., Avril-Juin, 1903). 


with the goddess Hariti as his consort seated beside him (Plate 31 b), which 
is one of the most delicately modelled works of the Gandhara school, and is 
presumably of early date. Hariti, in one of her aspects, was the protector of 
children from the dangers of epidemics. A standing figure from Sikri (Plate 
31c) presents her in the same aspect of her character, but posed in quite 
another fashion. The clever and unusual treatment of the drapery may be 
Pallas One of the most interesting statuettes is the well-known image of Pallas 
Athene. Athene in the Lahore Museum (Plate 34 a). The goddess is represented 
standing, facing front, wearing Greek costume, chiton and himation, and 
holding a spear across her body. Both hands have been lost. Probably the 
right hand grasping the spear was raised to her head, as was the right hand 
in the Pallas type of the coins of Azes I (? first century B.C.), while the left 
hand held the aegis. The late Dr. Bloch seems to have been right in inter- 
preting the image as that of a foreign female guard set over the women's 
apartments of a palace, and forming part of a court scene. 1 
Panels from A panel from the Dames collection, now in Berlin (Plate 34 c), is an un- 
CoUectkm S common variant of the ' Woman and Tree* motive, which will be discussed 
" later. The panel seems to be part of a larger composition, and is apparently 
of tolerably early date, although the figure is very Indian. Plate 34 d, also 
from the Dames collection, is not equal in merit to the preceding, the 
drapery being treated in a more formal and commonplace manner. A man 
stands under a tree playing the vina, or lyre which, however, does not 
correspond to the Indian kind of to-day. These three figures apparently 
formed parts of a frieze or larger composition. The trees, necessarily treated 
conventionally in order to bring them within the limits of the panels, have 
a fine decorative effect. 
Adaptation One Hellenistic group, known from at least five or six specimens, is of 
of the Rape special interest as being demonstrably adapted from a masterpiece of Leo- 
mede" c ^ ares > a famous Attic artist of the fourth century before Christ (372-330 B.C.). 
' His bronze work, praised by Pliny (d. a.d. 79), but long since lost, inspired 
many later copyists, who translated the theme into marble, with variations. 
One of the marble copies, or imitations, is in the British Museum, another 
at Thessalonica, a third at Venice, and the fourth and finest is in the Museo 
Pio Clementino at the Vatican. The subject is the carrying off of the beautiful 
boy Ganymede by an eagle, represented sometimes as the messenger of Zeus, 
and sometimes as the god himself transformed. In the Vatican copy the 

1 A cast is in the Indian Section of the Victoria J. A. S. £., Part I, vol. lviii (1889), p. 121, PI. 

and Albert Museum, South Kensington. For VII; and Lahore Museum Guide, PI. VI. For 

the coin of Azes referred to see Gardner, B. M. Bloch 's remark see No. 1195 in Indian Museum 

Calais Coins of Greek and Scythic Kings, PI. List of Negatives. 
XVIII, 4. The statuette has been published in 

PLATE 33. Kuvera; Lahore Museum 

A. Pallas Athene ; Lahore Museum 

B. Garuda and the Xatfini, 
from Sanghao 

C. Woman and tree, from 

Yusufzai (L. Dames, 


I). Man playing lyre 

(vina) % from Yusufzai 

(L. Dames, Berlin) 


A. Bovs armed as soldiers 

15. Hindu ascetic 

C. Buddha attended by Vajrapani, 

from Vusufzai, Dames Collection, 









eagle is shown as supported by the trunk of a tree in the background, with 
wings expanded and neck stretched upwards, grasping with tender firmness 
the nude youth, whose feet have just ceased to touch the receding earth. His 
robe, disclosing the nude figure, is so disposed as to protect his back from 
injury caused by the bird's talons. A dog, seated below, howls piteously for 
his vanishing master, as described by Virgil. 1 Nobody can look at Plate 34 b, 
reproducing the best of the Buddhist adaptations, obtained from the monas- 
tery at Sanghao in the Yusufzai country, and compare it with the Vatican 
copy of the Attic artist's composition, without perceiving that the composi- 
tion is essentially the same as that of Leochares, made familiar to the Hellen- 
istic world in marble replicas. All the Buddhist adaptations omit the dog, 
and so agree with the groups preserved at Venice, Thessalonica, and in the 
British Museum, while in the pose of the eagle and the introduction of the 
trunk of the tree they resemble the Vatican example. The subject, although 
retaining the essentials of the Greek myth, has been thoroughly Indianized, 
both in general treatment and by the substitution of a heavily draped female 
for the nude boy. The notion once held that the woman should be regarded 
as Maya, the mother of Buddha, is erroneous. The better opinion is that 
the group was intended to represent to Indian minds the carrying off of a 
female Naga, or snake sprite, by a monstrous Garuda, the implacable enemy 
of the snake tribe. As in all the Gandhara sculptures, the subject is absolutely 
Indian, no matter how foreign the presentation of it may be in outward form. 2 

Plate 35 A is a remarkable panel in the Lahore Museum (CataL, Plate VII, 
3), showing two boys of Greek appearance armed with the old Indian broad- 
sword, as described by Megasthenes and represented in the Bharhut and 
Sanchi sculptures. The work is artistic and attractive, and, as Professor 
Gardner reminds me, recalls the Pergamene style. 

I now proceed to illustrate a few representative relief scenic pictures of high The 
quality, beginning with the Dames specimen of the Nativity, unpublished, Nativity, 
and the finest example known to me of that favourite subject (Plate 36). 
According to the legend, Gautama Buddha was born in a pure fashion by 
springing from his mother's side as she stood under a tree in the Lumbini 
Garden, the modern Rummindei, to the east of Kapilavastu. The composi- 
tion is arranged in a perfectly symmetrical manner. On the left of the picture 
the god Indra, or Sakra, with his characteristic high head-dress, receives the 
child, behind him stands Brahma, and two other unnamed gods complete 
the divine party. The woman who supports the mother is her sister, and 

1 Tuer . . . quem praepes ab Ida found in ^7. A. S. 2?., Part I, vol. lviii (1889), p. 
Sublimem pedibus rapuit Iovis armiger uncis; 134. The Vatican group is reproduced in Vis- 
Longaevi palmas nequiquam ad sidera tendunt conti, Museo Pio-Clementino, vol. iii, p. 149, in 
Custodes, saevitque canum latratus in auras.' the histories of sculpture by Winckelmann, 

(Aen. v. 252-7.) Liibke, and Perry, and EncycL Brit., nth ed. 

2 Full references to the marble groups will be 'Greek Art', PL I, Fig. 53. 


three attendants balance the gods on the other side. The figures are 
thoroughly naturalistic men and women, cleverly modelled, and ingeniously 
arranged so as not to interfere one with the other. The draperies are treated 
with freedom and variety. On the whole, I am disposed to regard this group 
as the finest of the more complex stone pictures produced by the school of 
The 'Great The story of the c Great Renunciation' of domestic joys and the splendours 
Renuncia- f princely life by the young Gautama or Siddhartha when he went forth 
tlon " from his father's palace to take up the career of an ascetic, as told in both the 
books and the sculptures, comprises many incidents, which were treated in 
art with much freedom and variety of detail. Here I select for reproduction 
a rare representation of the groom Chandaka leading out the horse Kanthaka 
ready saddled for his master's use (Plate 38 c). The modelling of the horse 
is better than that of the animal in Indian sculpture generally, which often 
fails with the horse, while almost always successful with the elephant. This 
minor accident is intended to serve as a symbol of the whole story. 
Symbol Plate 38 D represents the worship, by shaven monks, of the trisul symbol, 
worship, signifying Buddha, the Law, and the Church. It closely resembles the repre- 
sentation of the adoration of the labarum in the Catacombs. 1 
Demon The well-known unique relief representing a group of figures with de- 
hosts. m oniac faces attended by three soldiers (Plate 37) has puzzled the inter- 
preters, who usually assume the demons to be a part of the host by which 
Buddha was assailed in the Temptation. It was Dr. Leitner who remarked 
that the so-called demons are simply monks wearing masks for a 'devil 
dance', such as those now worn by Tibetan Lamas. The equipment of the 
soldiers has been described sometimes as Greek and sometimes as Roman. 
But it is neither. The men evidently belong to the Himalayan region, and 
wear the dress and armour used in that region about the time of Kanishka, 
say a.d. 100. The arrangement of the scales of the armour, probably made 
of leather, with the curved ends uppermost, is explained by Sir Aurel 
Stein's discoveries of similar scales at Dandan-Uiliq in Khotan, and by 
a suit of Tibetan mail preserved in the British Museum. The Khotan scales 
date from the seventh or eighth century, but there is no difficulty in believing 
that the fashion of armour may have remained unchanged for ages. 2 
Frieze of An imperfect frieze in the British Museum, about 16 inches long by 6| 
deities! j nclies Wg* 1 (Plate 38 e), which puzzled Dr. Burgess, has been convincingly 
interpreted by M. Foucher as a representation of marine deities in a quasi- 
Greek fashion. The character of the personages as tritons or marine deities 

1 Roller, Les Catacombes de Rome, PL LXXXVII. may be as old as the second or third century, and 

2 Stein, Ancient Khotan, pp. 252, 41 1 , PI. II, and approximately contemporary with the Gandhara 
Addenda, p. xvi. The stucco relief statue of a relief. 

warrior in similar scale armour shown in Plate II 

PLATE 37. Procession of maskers and soldiers 

A. Head of Iiodhisattva 

H. Head of old man 

r '>. -T^i'CA 




I). Worship of Irisiilu s\inbol 
hv monks 

C. Gautama riding away; Lahore Museum 

K. Frieze of marine deities: H. M. 
PT-ATK 38 


of some kind is established by the paddles which they carry and their kilts of 
fins cut in the shape of vine-leaves. The object borne in the right hand 
of the figure the second from the right end appears to be a dolphin, indicating 
that the holder was intended for Poseidon. The figure on the extreme left 
is in the familar pose of Herakles. The Corinthian pillar on the right is in 
the style of Palmyrene work of the second or third century. The modelling 
of the forms would deserve praise but for the disfiguring exaggeration of 
the abdominal muscles. The bearded faces resemble that of an unmistakable 
triton, also in the British Museum, who has a fin and a curly tail (Foucher, 
Fig. 123). 1 

The general impression produced by study of the Gandhara sculptures is Apartness 
that they form a class standing to a considerable extent apart from the main of Gan- 
current of the evolution of art within the limits of India. M. Foucher has dharan art - 
succeeded, I think, in demonstrating that the Gandhara school has no direct 
filial relations with the earlier art of Maurya and Sunga times, notwithstand- 
ing the appearance in both of certain elements common to the Hellenistic 
art of Western Asia. The artists of the north-west, who were masters of the 
technique of Asia Minor, had no need to copy tritons, centaurs, and so forth, 
from the works of their humbler predecessors in the interior. The true view 
seems to be that, whatever may be the sources and extent of foreign influence 
on the work of early Indian sculptors, the rapid development of the Gandhara 
school during the first century of the Christian era was the direct result of 
a fresh importation into the frontier regions, by accomplished artists intro- 
duced from outside, of Hellenistic ideas expressed in the forms then current 
throughout the Roman Empire. 

According to Cunningham such importation of artists and ideas appears Association 
to have been closely associated with and dependent on the extension of the with the 
foreign Indo-Scythian and Kushan empires, as they gradually advanced ? ush !? 
their borders from the Oxus to the Ganges, and possibly as far as the Nar- ynas ™ 
bada. Unfortunately, as already observed, the chronology of those times is 
uncertain; and until the chronological question, summed up as the problem 
of the date of Kanishka, shall be definitely solved, the exact relations of the 
art of Gandhara with that of the Graeco-Roman world and India proper can- 
not be elucidated with all the precision desirable. 

It is, however, safe to affirm both that the Kushan kings had become lords 
of Kabul, with at all events part of the Punjab, before a.d. 100, and that 
sometime after that date the character of the Gandhara style was fixed. Much 
of the better sculpture of the Gandhara school undoubtedly was produced 
during the reigns of the later Kushans. The characteristic of this work is 
the modified Corinthian capital, similar in style to the capitals fashionable 
throughout the Roman empire in the early centuries of the Christian era. 

1 Foucher, UArt grico-bouddhique du Gandhara, p. 244, Fig. 126. 


Origin of The appearance in sculpture of that specially Graeco-Roman form co- 
the ^ an - incides with the introduction of the Kushan gold coinage, agreeing in weight 
school! with the Roman aureus, though somewhat debased in standard. 1 All the 
evidence leads to the inference that the rapid development and extension of 
the distinct Gandhara school, with its characteristic Indo-Corinthian capitals, 
were effected under the patronage of the great Kushan kings, who may even 
have imported foreign artists. Such foreign artists, accredited by royal 
authority and the fashion of the court, would have been readily accepted as 
teachers by the local Indian sculptors, who, after their accustomed manner, 
would have proceeded to adapt the new methods to their own purposes, 
sometimes, perhaps, bettering the instructions of their masters. 2 
Parallel case The popularization of the Persian style of painting in India under Akbar 
of Indo- j n the sixteenth century, and the immediate development of a prolific Indo- 
^ntin n P ers * an school, surpassing its prototype in certain respects, while inferior in 
' others, offer an almost exact parallel to the events which happened, as I 
believe, in the kingdom of Gandhara during the first century of the Christian 
era. The parallel fails in so far that the Persian style of painting, being con- 
genial to Indian taste, readily admitted of certain modifications which may 
be reasonably regarded as improvements, whereas the ultimate models of 
the Gandhara sculptors having been the masterpieces of Attic and Ionic art, 
alien in spirit to the art of India, were usually susceptible of modification by 
Indian craftsmen only in the direction of degradation. 
Indianiza- It is obvious that the foreign elements in the art of Gandhara tended to 
tion as a diminish as time went on, and that, generally speaking, the sculptures with 
test of age. most c i ear jy marked Greek character should be considered early, and those 
most Indianized as comparatively late. But, as already pointed out, this 
criterion affords no infallible test of age. Some of the best finished works 
in Hellenistic style may have been executed by clever Indian imitators long 
after the introduction of the style, just as among the Mughal paintings we 
find close imitations of Persian models side by side and contemporary with 
paintings profoundly Indianized. 
Decadence Many European critics, convinced of the unapproachable excellence of the 
° r im ment? h*&h est tyP e °f Greek art, the model of the less excellent Hellenistic art, see 
in the process of Indianization a decadence. But the critics of the Nationalist' 
school are persuaded that this view is erroneous, and that the process of 
Indianization is in itself an artistic improvement. Mr. Havell, in general 
agreement with Dr. Coomaraswamy, teaches that the earliest Gandhara 
sculptors were no better than mechanical craftsmen, hirelings following more 
or less impure Hellenistic traditions, engaged by the frontier kings in the 

1 For details see Cunningham, Coins of Mediaeval does not take into consideration the part played 
India, p. 16. by Parthian and Sassanian Persia, intervening 

2 This explanation of the rise of Gandhara art between East and West. 


manufacture of inferior objects of handicraft, which are mere 'soulless pup- 
pets, debased types of the Greek and Roman pantheon posing uncomfortably 
in the attitudes of Indian asceticism 1 , and tarred with the vices of com- 
mercialism, insincerity, and want of spirituality, most conspicuous in the 
earliest examples. The indictment continues: 

'The insincerity and want of spirituality typical of nearly all the art of Gandhara are, 
as I have said, most conspicuous in the earliest examples, or those which are attributed 
to the first century of our era, when the Roman influence was strongest. Two centuries 
later, in the sculptures of the Loriyan Tangai Monastery, which Professor Griinwedel 
describes as belonging to the best period of Gandhara, 1 the art has become more 
Indian, more national, and more spiritual, but it has not yet achieved the true ideal of 
Indian art. Since, however, it is Indian influence, Indian thought, which has so far 
perfected the style, it is surely incorrect to say that the ideal of Indian Buddhist art 
has been created by foreigners. Foreign hands may have held the tools, but the 
influences which have dominated the art have been throughout Indian. . . . The 
perfected ideal of Indian art is as far in advance of the Gandharan type as the art of 
the Parthenon surpasses the art of Gandhara. Neither artistically nor technically is 
it possible to place the best Gandharan sculpture in the same plane with that of 
Borobodur, Elephanta, or Ellora, or even with the best modern Nepalese metal-work, 
such as the Buddha in Plate VI.' 2 

The critic then proceeds to liken Gandharan art to ' cheap, modern Italian Alleged 
plaster work', and to extol the later medieval sculpture and bronzes as exhibit- faults - 
ing * quiet restrained dignity, calm conviction, and eflFacement of physical 
detail . . . the embodiment of a great national tradition, a synthesis of Eastern 
philosophy and religious art\ We are further told that the Brahmanical art 
of the eighth and ninth centuries expresses 'the true Indian conception of 
divinity in a superhuman, spiritualized body', or, as elsewhere phrased, 'the 
idea of a purified, transcendental body formed by the practice of Dhyana 
[meditation] and Yoga [ascetic restraint]'. So Dr. Coomaraswamy declares that 

'just as through all Indian schools of thought there runs like a golden thread the 
fundamental idealism of the Upanishads — the Vedanta — so in all Indian art there is 
a unity that underlies all its bewildering variety. This unifying principle is here also 
Idealism, and this must of necessity have been so, for the synthesis of Indian thought 
is one, not many'. 3 

The substance of these criticisms seems to mean that all high-class Indian The spirit 
sculpture must be an expression of Brahmanical metaphysics, nothing else ° f ^*? 
being truly Indian or national. But the Gandhara artists, who certainly did JJ 
not worry about a 'superhuman, transcendental body', or take any interest 
in the Upanishads, agreed in those respects with the artists of all the early 
Buddhist schools, who were, nevertheless, just as Indian and national as any 
ninth-century Brahman could be. Although the technique of Gandhara 

1 Now in the Indian Museum, Calcutta. See 3 The Aims of Indian Art (Essex House Press), 
Plate 30 A. i9°8. 

2 Indian Sculpture and Painting, pp. 45-50, 

3292 K 


differed widely from that of Bharhut, Sanchi, and the rest, all the early 
Buddhist schools alike, that of Gandhara included, were animated by the 
Buddhist kindly humanistic spirit, as different as possible from the Tantric 
notions dominating medieval art, both Brahmanical and Buddhist, but 
equally Indian. We are not entitled to denounce Gandharan art as 'lacking 
in spirituality', and so forth, merely because it does not express the ideas of 
Ellora and Elephanta. As a matter of fact, many of the good Gandhara 
sculptures may be fairly held to express with admirable feeling and sincerity 
the ideal of a saintly Indian man, and to be not lacking in 'restrained dignity'. 
For instance, the beautiful Bodhisattva (Plate 31 a) is very far from being a 
'soulless puppet'; the Lahore Museum Kuvera (Plate 33) has a good share of 
'restrained dignity'; and many of the Buddhas are quite equal to any of the 
Javanese or Ceylonese images. Much credit is given by the new school of 
critics to the achievements of medieval sculptors in the representation of 
gesture and strenuous action; but, without depreciating their work, it is per- 
missible to insist on the similar merits of the Gandharan heads and Atlantes. 
Restricted Political conditions seem to have been responsible to a great extent for the 
influence of failure of the art of the north-western frontier to penetrate deeply into the 
an^Indfa 1 ^ nter ^ or - The Kushan empire apparently broke up in the time of Vasudeva I, 
" the successor of Huvishka, and was followed probably by a time of unrecorded 
anarchy. The next empire, that of the Guptas, who completed the conquest 
of the Gangetic valley about the middle of the fourth century, did not include 
the Punjab, and so was separated from Gandhara by foreign territory. 
Gandhara But outside India the Gandhara school achieved a grand success by 
the parent becoming the parent of the Buddhist art of Eastern or Chinese Turkistan, 
° f art^n Ae Mongolia, China, Korea, and Japan. The stages of the transmission of the 
Far East. s tyl e to l ^ e F ar East have been clearly disclosed by the abundant discoveries 
of sculptures and paintings in the manner of Gandhara throughout Chinese 
Turkistan, both to the north and south of the Taklamakan (Gobi) Desert. 
Through China the imported forms of Buddhist art passed to Korea, and 
thence to Japan. Pious pilgrims, like Fa-hian and Hiuen Tsiang, played a 
large part in determining the course of Buddhist art in China by bringing 
back from the Indian Holy Land multitudes of images and pictures which 
became the authoritative models for Chinese monastic artists. The Indian 
influence, it must be clearly understood, affected the art of China and Japan 
only in its application to Buddhist uses. In other departments Chinese art, 
and its daughter in Japan, developed independently of Indian teaching. 

The pursuit of the eastern ramifications of Indian Buddhist art lies beyond 
the scope of this work, but a slight sketch in outline of the process by which 
the Gandhara style became the basis of the art devoted to the service of 
Buddhism in the Far East is an almost indispensable supplement to an ac- 
count of the Gandhara school, and may be presented in few words. 


Communications between China and the western countries were first The pro- 
opened up during the time of the Early Han Dynasty (226 B.C. to a.d. 25) 1 f e *? of 
by means of the mission of Chang-Kien, who was sent as envoy to the Oxus Buddhist 
region, and died about 114 B.C. That mission resulted in the establishment art east- 
of regular intercourse between China and the Scythian powers, but did not wards. 
involve contact with India. In the year a.d. 8 the official relations of the 
Chinese government with the western states came to an end, and when the 
first Han dynasty ceased to exist in a.d. 25 Chinese influence in those 
countries had vanished. But in a.d. 73 a great general named Pan-chao 
reduced the King of Khotan to subjection, and from that date continued 
his victorious career until his death in a.d. 102, when the power of China 
attained its greatest western extension. In the last decade of the first century 
Pan-chao inflicted a severe defeat on the Kushan king of Kabul somewhere 
beyond the Pamirs in the Yarkand or Kashgar country. Most probably that 
king was Kanishka. After Pan-chao 's death the Kushan king retrieved his 
defeat and occupied Khotan, at some time between a.d. 102 and 123. To 
that Indo-Scythian conquest of Khotan I would attribute the rapid spread 
of Indian languages, scripts, religion, and art in Chinese Turkistan, as 
disclosed by the discoveries of recent years. Kanishka 's defeat of the Chinese 
and conquest of Khotan afford an adequate explanation of the archaeological 
facts. 2 Probably the Indo-Scythian occupation of Khotan did not last very 
long, but no documentary evidence on the subject has yet been discovered. 
During the third century Buddhism effected considerable progress in China, 
and from the beginning of the fifth century to the eighth a constant stream 
of learned pilgrims devoted themselves to the task of saturating Chinese 
Buddhism with Indian ideas and Indian art. Early in the seventh century 
Bajna and his son, Wei-tschii I-song, distinguished painters from Khotan, 
visited the Chinese court, and founded an Indo-Chinese school of painting. 
China transmitted the Indian forms of Buddhist art to Korea, whence they 
passed to Japan. That is the outline of the facts. 3 During all the centuries 
mentioned there is no indication of a reflex action of Chinese on Indian art, 
the supposed Chinese influence on the Ajanta paintings a little before or after 
a.d. 600 being very doubtful. 

The fact that the prevalent existing forms of Buddhist art in the Far 
East originated in Gandhara has been fully proved in detail by Professor 
Griinwedel and other authors, whose finding on that point is generally 

1 Chinese dynastic dates are given according to sischen Kunst (Munchen und Leipzig, 1896), 
Tchang, le Pere Mathias— Synchronismes chinois p. 83. For art of Gandhara style in Turkistan, 
(Chang-hai, 1905). see Stein's works, and the German publica- 

2 See Ox. Hist. Ind., p. 128 for V. Smith's theory tions giving the results of the first German ex- 
on this point. pedition to Turfan, as enumerated by Dr. v. Le 

3 Hirth, F., Ueber fremde Einflusse in der chine- Coq in J. R. A. 5., 1909, p. 301. 



The The isolation of India, so apparent on the map, has never been absolute. 

channels jj er inhabitants from the most remote ages have always been exposed to the 

LfluenSl act i° n °f foreign ideas conveyed by one or all of three ways — by sea, through 

' the passes of the north-eastern frontier, or through the more open passes of 

the north-west. The only foreign art which could influence India from the 

north-east being that of China, which certainly produced no considerable 

effect on Indian art prior to the Muhammadan conquest, the ingress of 

foreign artistic ideas through the north-eastern passes may be left out of 


Long before the dawn of history traders from distant lands had brought 
their wares to the ports of India, and in all probability introduced the alphabet 
and art of writing. But in those ancient days the sea, although open to the 
passage of adventurous merchants, was not the bond of union between 
distant lands which it has become in these latter times for a great naval power, 
and the influence exercised upon the art of the interior by small bodies of 
traders at the ports must have been comparatively trifling. The constant 
invasions and immigrations from the continent of Asia through the north- 
western passes had more effect; 1 and one prehistoric immigration, or series 
of immigrations, which brought the Vedic Aryans, ultimately settled the 
future of all India for all time by laying the foundations of the complex, 
exclusive, religious, and social system known as Hinduism. When history 
opens in the sixth century B.C., Northern India, at all events, was already 
largely Hinduized, and in the third century, when the earliest extant monu- 
ments came into existence, the Hindu system stood firmly established. In 
attempting to estimate the nature and extent of foreign influence on Indian 
art, as conveyed by sea and through the north-western passes, we must 
assume the existence of Hinduism as an accomplished fact, and acknowledge 
that nothing positive is known about Hindu art before the age of Asoka. 
Early In his days the dominant foreign influence may be designated Persian, 
Persian traceable clearly in his monolithic columns, in the pillars of structural build- 
in uence. [ n g S ^ anc j j n architectural decoration. Capitals, crowned by recumbent bulls 
or other animals, are found at Bharhut, Sanchi and elsewhere, in the Gan- 
dhara reliefs, and at Eran in Central India, even as late as the fifth century 
of the Christian era, but these do not very exactly correspond with the true 
Achaemenian type. The capitals of the monolithic columns, likewise with their 
seated and standing animals, although distinctly reminiscent of Persia, differ 

1 Suggested indications of Babylonian influence 241-88). This view does not accord with the 

include the earliest Indian astronomy, the know- ethnological facts. Any 'Babylonian' influence 

ledge of iron, urn-burial, and the marriage-mart that left its mark on the land and its people must 

at Taxila. See Kennedy, 'The Early Commerce have been pre-Sumerian. 
of India with Babylon' (J. R. A. S., 1898, pp. 


widely from Persian models, and are artistically far superior to anything pro- 
duced in Achaemenian times. Sir John Marshall, as already observed, can 
hardly be right in ascribing the beautiful design and execution of the 
Sarnath capital (ante, Plate 2) and its fellows to Asiatic Greeks in the service 
of Asoka. 1 

We are thus led to consider the second foreign element in the most ancient The 
schools of Indian art, that is to say, the Greek element, expressed in Asiatic Hellenistic 
Hellenistic forms. In Asoka's age the chief schools of Greek sculpture were clement - 
in Asia Minor at Pergamum, Ephesus, and other places, not in Greece, and 
the Hellenistic forms of Greek art had become largely modified by Asiatic 
and African traditions, reaching back to the ancient days of Assyria and 
Egypt. It is consequently difficult to disentangle the distinctively Greek 
element in early Indian art. The acanthus leaves, palmettos, centaurs, tritons, 
and the rest, all common factors in Hellenistic art, are as much Asiatic as 
Greek. The art of the Asokan monoliths is essentially foreign, with nothing 
Indian except details, and the fundamentally alien character of its style is 
proved by the feebleness of later attempts to copy it. I think that the 
brilliant work typified by the Sarnath capital may have been designed in its 
main lines by foreign artists acting under the orders of Asoka, while all the 
details were left to the taste of the Indian workmen, much in the same way 
as long afterwards the Kutb Minar was designed by a Muhammadan architect 
and built by Hindu masons, under the orders of the Sultan Altamsh. 2 

Our knowledge of the fine art of Asoka's reign (273-232 B.C.) is restricted Post- 
to the monolithic columns almost exclusively. The other sculptures of the Asokan 
Early Period probably are all, or nearly all, of later date. They present a great tid\y Tndian. 
contrast, being essentially Indian, with nothing foreign except details, and 
they presuppose the existence of a long previous evolution of native art 
probably embodied in impermanent materials, and consequently not repre- 
sented by actual remains. 

Are we to regard these sculptures, and especially the reliefs of Bharhut, 
Sanchi, and Bodh Gaya, as purely Indian in origin and inspiration, or as 
clever adaptations of foreign models? The sudden apparition simultaneously 
of stone architecture, stone sculpture, and stone inscriptions during the reign 
of Asoka, when considered in connexion with the intimate relations known 
to have existed between the Maurya empire and the Hellenistic kingdoms of 
Asia, Africa, and Europe, raises a reasonable presumption that the novelties 
thus introduced into the ancient framework of Indian civilization must have 
been suggested from outside. That presumption is strengthened by the 

1 See J. R. A. S., 1907, p. 997, PL III; ibid., ont paru si directement empruntes a la Perse que 
1908, p. 1092, PI. IV, 6. l eur importation ne s'explique guere autrement 

2 M. Foucher', writing of the Sanchi reliefs, ob- que par une immigration d'artisans iraniens'. 
serves that 'quantite de motifs decoratifs nous (La Porte Orientate du StUpa de Sdncki, p. 34.) 


foreign style of the monolithic columns, which undoubtedly were a novelty 
brought into being by the command of an enlightened despot in close touch 
with the outer world. It must be remembered, however, that Early Indian 
architecture was essential wooden. No sudden transition can be traced dating 
from Asoka's age. The small, square Gupta shrines are the earliest stone 
structures in India proper. 
Probable Although I do not now feel justified in expressing as confidently as I once 
Alexandrian ^id my theory of the Alexandrian origin of Indian bas-relief sculpture in 
of ston^b^ stone , I am still disposed to believe that such reliefs would never have been 
reliefs, executed if works essentially similar had not previously existed in the Hellen- 
istic countries, and especially at Alexandria. 1 The Indian reliefs certainly 
are not modelled on those of Persia, which are utterly distinct in character; 
and it seems unlikely that the Indians should have suddenly invented the full- 
blown art of stone bas-relief out of their own heads without any foreign sug- 
gestion. The Alexandrian reliefs were available as indications how stone 
reliefs should be executed, and the clever Indian artists and craftsmen, once 
they had seized on the main idea, would have had no difficulty in transmuting 
it into purely Indian forms, just as the Hindu play-writer, mentioned by 
Weber, transformed the Midsummer Night's Dream into a piece thoroughly 
Indian in character, showing no trace of its English source. 2 Complicated 
relief pictures, like those of Bharhut and Sanchi, placed in exposed positions, 
could not have been satisfactorily executed in wood or ivory; but the trained 
wood and ivory carvers, who existed in India from time immemorial, could 
easily have applied their skill to making stone pictures as soon as the novel 
material had become the fashion. Carvers in wood and stone often are the 
same people and use tools substantially identical. The truth seems to be 
that the Indians illustrated the Jatakas with Indian scenes just as the Alex- 
andrians illustrated pastoral poems with Greek scenes, and that the Indians 
got from abroad the idea of so doing. But the theory must be admitted to be 
incapable of decisive proof, although to my mind it appears to be highly 
probable. The subject-matter and treatment of the post-Asokan reliefs are 
certainly on the whole Indian, and such obviously foreign details as they 
exhibit are accessory rather than integral. 
Indirect M. Foucher, however, may be right when he discerns in the Sanchi sculp- 
of Greek tures more su ^ e indications of Hellenistic influence in certain examples of 
influence. b°fcl fore-shortening, in clever presentations of the three-quarter face figure, 
and in the harmonious balancing of groups. It is, indeed, inconceivable that 
the Indian sculptors of Asoka's time should have failed to learn something 

1 Imp. Gaz. (1908), vol. ii, p. 105; Hastings, that remained of the original was the order of the 
EncycL of Religion and Ethics, s.v. f Amaravati'. contents and the substance of the examples. 

2 Ind. Ant., vol. xxx (1901), p. 287, note 59, All the rest was Indian.' The Japanese treat 
Weber relates a similar transformation in a European plays and tales in the same way. 
Sanskrit adaptation of Euclid's Elements. 'AH 


from the Greek art if it was accessible to them. The exact channel of this 
foreign influence is not historically or geographically clear. And its extent 
is debatable if it existed at all. A great deal must be allowed for natural, 
purely native development, when discussing such improvement of skill as is 
visible in the East and West gates of Sanchi when compared with the North 
and South. But whatever was borrowed the Indian craftsmen made their 
own, so that their work as a whole is unmistakably Indian in character and 
original in substance. 

I proceed to discuss in some detail certain motives of ancient Indian sculp- 
ture which seem to be of foreign origin, and in some cases lend support to 
the theory of specially Alexandrian influence. 

The first to be considered is that which may be conveniently designated The 
the 'Woman and Tree' (Plate 39). The form which may be regarded as <Wo ™ an , 
normal represents a woman standing under a vine or other tree, with her legs motiveln 
crossed, the left arm twined round a stem, and the right hand raised to her sculpture, 
head. Many variations, however, occur. Occasionally, the left hand is raised 
above the head, as in an example from Mathura, in which also the right arm 
is not twined round a stem. Sometimes the legs are not crossed. The woman, 
in some cases, is more or less clothed, but frequently, and especially at 
Mathura , is unmistakably and aggressively naked . Very often , but not always , 
she stands on a dwarf, animal, or monster (Plate 16 c). 

The attitude is well calculated to display the charms of the female form, 
and, as M. Foucher observes, is frequently described in Sanskrit poetry, so 
that it may be regarded as 'la pose plastique par excellence* of India. 1 The 
dates of Sanskrit literature are so uncertain that it is quite possible that the 
descriptions may have been suggested by the statuary. It seems to me highly 
probable that the plastic rendering was a foreign introduction. Dates seem 
to forbid the suggestion that Western art might have borrowed it from 

The earliest Indian example known to me is the Bharhut draped figure of The motive 
the Yakshi Chanda, who is represented in what I call the normal manner. m India - 
That may be dated about 200 B.C. 2 The pose is in exact keeping with her 
character as a primitive godling. The lady also appears on the Sanchi gate- 
ways, and in Gandharan art over and over again with many variations. I can- 
not find her at Amaravati, but at Mathura she is specially characteristic of 
the local art, both Jain and Buddhist, and is often represented with lascivious 
suggestiveness in a manner to which the Mathura school was too much 
inclined. Slightly modified she becomes Maya, the mother of Buddha, in 
the Nativity scene (Plate 36). The latest example that I can quote is a Brah- 

1 VArt grico-bouddhique du Gandhara, p. 229. 2 Stupa of Bharhut, PI. XXII: Griinwedel- 
A sculpture in the Mathura Museum exhibits a Burgess, Buddhist Art, Fig. 16. 
male figure in the same pose. 

in Greek art. 


manical sculpture of the period at Vijayanagar Tadpatri in the Anantapur 
District, dating from the sixteenth century. Thus, it is established that in 
Indian sculpture the motive had an history of more than 1,700 years. 

The motive J n Greek art it occurs in the fourth century B.C., a century or two before its 
first appearance in India at Bharhut. The Hellenistic artists transported the 
motive to Egypt, where, by reason of contact with native Egyptian sensual 
notions, its treatment acquired a lascivious tinge, agreeing strangely with the 
Mathura presentation, the nude figure, however, in Egypt being often male 
instead of female. M. Strzygowski gives the name of Copto-Alexandrian to 
the mixed or mongrel art produced by the intermingling of Hellenic and 
Coptic ideas. The art of Gandhara does not share with that of Mathura the 
reproach of lasciviousness. It deserves credit, as M. Foucher points out, for 
its 'irreprochable tenue* in dealing with the relations of the sexes. 1 

The Aachen The most striking illustration of the close resemblance between the 
Bacchus. Madura presentation of the Woman and Tree motive and the Copto- 
Alexandrian form is found in an unexpected place, the cathedral of Aachen 
or Aix-la-Chapelle in Rhenish Prussia. Six remarkable ivory panels on the 
sides of the cathedral pulpit have been examined in a special disquisition by 
M. Strzygowski, who has proved to my satisfaction that the Aachen ivories 
are of Egyptian origin, and should be considered as examples of the Copto- 
Alexandrian school. They may have reached their resting-place by way of 
either Ravenna and Milan or Marseilles. 2 

Two figures, one on the right and one on the left of the pulpit, identical 
save in certain minor details, are known conventionally as 'Bacchus*. Each 
represents a nude young man facing, standing with the right leg straight and 
the left leg crossed over it. The body is supported by the left arm, which is 
twined round the stem of a vine overtopping and surrounding the youth with 
its foliage. His right hand is raised to the crown of his head (Plate 39 a). 
The pose is precisely the same as that of the Woman and Tree motive in 
Indian art, and the resemblance between the Mathura and Aachen figures is 
so close that, in my judgement, it cannot be accidental. Both must have a 
common origin, which should be sought in Syria or Asia Minor, from which 
Egyptian Hellenistic art drew its inspiration. The motive was variously 
treated in Egypt, and, at least in one case, a woman takes the place of the 
youth. There is no difficulty in believing in the transference of Alexandrian 
ideas to India either before or after the Christian era. From Asoka's time 
for several centuries intercourse between the ports of Egypt and India was 
continuous. The cupids, birds, and beasts interspersed in the foliage of the 

1 VArt grico-bouddhique du Gandhara, p. 248. 2 Strzygowski, J., 'Hellenistische und koptische 

The 'orgy' relief (ibid., Fig. 130: Gandhara Kunst in Alexandria' (Bull, de la Soc. Archiol. 

Sculptures, PI. XXII. 7) is the only one open to d Alexandria (Wien, Vienne), 1902). 
a charge of impudicity. 

A. 'Bacchus,' on left side of 
Aachen pulpit 

15. 'Woman and Tree, 1 as caryatid, 

from Upper Monastery, Nathu, 


C\ Part of jamhof N./ffl/wrff, 
Tadpatri, Anantapur District 

PL All* 39 

A. From Lower Monastery, Nathu, Yusufzai 


**"***+ ... 

•• • * ■---*■ '<%, 



■—■ • 

H. Part of a frieze from Mathura, Kushan Dynasty 

C. Frieze from Upper Monastery, Nathu, Yusufzai 


Aachen ivory are also often found in India. Compare, for instance, the 
Garhwa pillar (Plate 45 b) and various Mathura sculptures. 

The female figure in the Woman and Tree design used to be described as The 
a ' dancing-girl \ But, whether nude or clothed, she is never represented as woman not 
dancing, and Dr. Vogel certainly appears to be right in maintaining that she * 'jkncing- 
should be interpreted, not as a dancing-girl, but as a Yakshi, or female sprite. 1 p " 
The Yakshas and Yakshis played in ancient popular Indian Buddhism a 
prominent part comparable with that played by the Nats in modern Burmese 

Other motives must be discussed more briefly. At Amaravati and in Gan- The Rider 
dhara a favourite subject is the departure of Gautama Buddha as Prince motive. 
Siddhartha from Kapilavastu on horseback. Generally the horse is shown 
in profile, but occasionally is represented as emerging from a gateway, and 
facing the spectator, fore-shortened. This latter form of the design especially 
seems to be connected with the Rider motive as seen in the Barberini ivory 
diptych in the Louvre, of the fourth century, and in one of the Aachen panels, 
the origin of both being traced back by Strzygowski to the Egyptian repre- 
sentations of Horus triumphing over the powers of evil represented by a 

The Indian sculptures usually show earth-spirits, or Yakshas, male or The earth- 
female, holding up the horse's hoofs. As Griinwedel and Strzygowski point spirits. 
out, the sculptures illustrate the Buddhist legend that the earth-goddess dis- 
played half her form while she spoke to the departing hero, and also are a 
reminiscence or translation of the Greek motive of Gaia rising from the 
ground, familiar to Hellenic art from the fourth century B.C. Similar earth- 
spirits are seen in the Barberini diptych. 2 The Rider motive is used on the 
uprights of the Sanchi gateways, and there is a large Kushan Horse and 
Rider in the Mathura Museum. 

The use of a long undulating stem, band, garland, or roll to break up a long Undulating 
frieze into sections was familiar to Indian sculptors from early days. As seen garland or 
on the Bharhut coping, the device used is a lotus stem with jack fruits ro11 - 
attached. The stem is not carried by anybody. This design seems to be purely 

But the later forms of the motive must be compared to the garland carried 
by amortni, Erotes, or cupids, which was constantly used in the later Hellen- 
istic and Graeco-Roman art. In Gandhara an imbricated roll, quite in the 
Graeco-Roman fashion, carried by boys, equivalent to cupids or Erotes is 
substituted for the Indian lotus stem. At Mathura and Sarnath we find a 

1 Ann. Rep. A. S., India, 1906-7, p. 146. - J. I. A. /., vol. viii (1898), PI. XXII, Fig. 1, 

2 Griinwedel-Burgess, pp. 98-103, Figs. 50-4. where the spirits are female. 
See also The Gandhara Sculptures, PI. XX, Fig. 1 

3292 1 


smooth roll carried by men, not boys (Plate 40 b), and at Amaravati a bulky 
tinsel roll with Indian decoration, also carried by men (Plate 23 b). 1 
Pergamene The hippocamps, tritons, centaurs, and other weird creatures, which cer- 
mfluence. tainly were borrowed from Western art, occur, as we have seen in Chapter II, 
at Bodh Gaya and other places in the sculptures of the Early Period. 
It does not much matter whether we call them Hellenistic or Western 
Asiatic. Forms more or less similar recur at Mathura and Amaravati 
and in Gandhara. The strongly-marked muscles of some of the Gandhara 
figures and the snake-tailed monsters suggest the notion that the sculptors 
of the north-west felt the influence of the vigorous Pergamene school. 
The Atlantes of Jamalgarhi especially seem to be reminiscent of Perga- 
mum; from the Buddhist point of view they may be regarded as Yakshas. 
A few of these Western Asiatic Hellenistic forms are shown together in 
Plate 41. Atlantes occur in later Hindu art in the form of dwarfs, usually 
Some archi- Certain architectural details represented in ancient sculptures, in addition 
detdk. t0 l ^ e well-known Corinthian and Ionic capitals, may be mentioned as being 
common to Indian and Western Asiatic Hellenistic art. The fluted spiral 
column, frequently met with on the sarcophagi of Asia Minor and in later 
Roman work, does not seem to occur at Amaravati or in Gandhara, but is 
found at Mathura in sculptures which are difficult to date, but which seem to 
be post-Kushan. 2 Subsequently it was freely used in the cave-temples of 
Western India. The scallop shell of 'shell-niche* canopy, often seen on 
Asiatic sarcophagi and in Egyptian art, occurs in India, so far as I am aware, 
only in the details of the Corinthian capitals at Jamalgarhi. M. Strzygowski 
holds that the form probably originated in Mesopotamia, and that it was 
ultimately developed into the characteristic Muhammadan mihrab? But that 
suggestion seems to be of doubtful validity. The rectangular incised panel 
frequently found on pilasters in Gandhara reliefs is specially characteristic 
of the Roman architecture of Palmyra (a.d. 105-273). 4 Much of the Gandhara 
art resembles that of Palmyra and Baalbec more closely than that of any 
other specific locality. The buildings at Baalbec date from the second century. 
It is, of course, unnecessary to point out in detail the numerous echoes of 
Greek art in the Gandhara sculptures. I have confined myself to noticing 
certain points of particular interest. 
The vine. The introduction of the vine into Indian bas-reliefs used to be considered 
as in itself evidence of copying from Hellenistic models. But that view is 

1 Anderson, Catal. I. AT., Part I, p. 241 ; Griin- 2 Jain Stupa of Mathura, PI. XLVIII, 3. 

wedel-Burgess, p. 148, Figs. 99, 100; Foucher, *J. Hellenic Studies, 1907, p. 114, Fig. 11. 

UArt grdco-bouddhique du Gandhara, p. 239, 4 In Gandhara this 'panel' is a misunderstood 

Figs. 1 16-18 m ,J.A. S. B., Part I, vol. lviii (1889), conventional rendering of the octagonal railing- 

p. 158, citing a Roman parallel. pillar. Codrington, Ancient India, Fig. 22. 


not tenable. The vine is still largely grown in India proper, and until 
the recent Afghan conquest was freely cultivated in Kafiristan. Sir George 
Watt believes that the plant is indigenous on the lower Himalayan ranges, 
and is even inclined to think that its cultivation may have been diffused into 
Europe from that region. However that may be, it is certain that Indian 
artists had ample opportunities of studying the forms of vine-growth at first 
hand, and were under no necessity to seek foreign models. 

In certain cases, however, Indian sculptors chose to treat the vine motive Different 
after the European or West Asiatic manner. 1 The best example of such treat- st yk of 
ment is the well-known frieze from the Upper Monastery at Nathu, Yusufzai, treatment - 
which is almost a replica of a similar work at Palmyra, executed in the third 
century after Christ (Wood, Palmyra, Plate 41). The design (Plate 40 c) 
consists of a vine stem knotted into five circles forming small panels, the 
first of which, to the left, contains leaves only; the second is occupied by 
a boy or 'genius' plucking grapes; the third exhibits a boy playing with a goat; 
the fourth displays a crudely executed goat nibbling the vine; and the fifth 
represents another boy plucking grapes. Plate 42 a reproduces a Mathura 
sculpture treating the vine after the Indian manner, and admirably executed. 

The motive consisting of a vine or other conventionalized plant springing Plant forms, 
from a vase is common to Egyptian and ancient Indian art. M. Strzygowski 
gives three Egyptian examples in the essay cited above. The motive is found 
everywhere at Bharhut, Sanchi, and Amaravati, and is the basis of the later 
vase-and-foliage capital. 

The Indian treatment of indigenous animals in both sculpture and painting Animal 
is as original and artistic as that of plant motives. motives. 

'You have only', Sir George Watt writes, 'to look at the plants and animals employed 
in the most ancient designs to feel the strong Indian current of thought there con- 
ventionalized, which must have involved centuries of evolution. The treatment of the 
elephant, monkey, and serpent is Indian, and in no way Greek. No Greeks (as few 
Englishmen to-day) could give the life touches of those animals seen on all the oldest 
sculptures and frescoes.' 2 

Those observations are perfectly true, and in all discussions of the foreign 
elements in Indian art we must remember that in certain respects Indian 
artists were not only free from obligation to the Greeks, but actually superior 
to them. The illustrations in this work bear abundant testimony to the Indian 
power of delineating indigenous living forms, both vegetable and animal. 
The Gandhara treatment of the elephant is inferior to that of the same subject 
by the artists of the interior, who were more familiar with that wonderful 
beast, which is not easy to model or draw well. 

The general result of examination of the foreign influences upon Indian Substantial 
pre-Muhammadan art, whether sculpture, painting, or architecture, is to JJg^JJ[ t of 

1 Letter dated 6 Nov. 1909. 2 Les Monuments de VInde, pp. 12-15. 


support the opinions of those who maintain the substantial originality of 
Indian art. It may be true that the general use of stone for architecture and 
sculpture was suggested by foreign example, and that the notion of making 
story-telling pictures in stone came from Alexandria; but, even if both those 
hypotheses be accepted, the substantial originality of the Indian works is not 
materially affected. The principal forms of Indian architecture, so far as 
appears, were developed in India, and it is impossible to connect them with 
Western forms. They have, as M. Le Bon observes, a character of 'frappante 
originalite'. The actually proved borrowings by India are confined to details, 
such as Persepolitan columns and capitals, and a multitude of decorative 
elements, some of which continued in use for many centuries. 
Incompati- M. Le Bon is well supported by facts in his opinion that India, 
I dianand <ma ^8 r ^ un contact assez prolong^ avec la civilisation grecque, ne lui a emprunt6, et 
Greekldeals ne P ouva ^ h" emprunter aucun de ses arts. Les deux races itaient trop diflterentes, 
' leurs pens&s trop dissemblables, leurs genies artistiques trop incompatibles pour 
qu'elles aient pu s'influencer. . . . Le g&iie hindou est tellement special que, quel que 
soit Tobjet dont les n6cessit£s lui imposent limitation, Taspect de cet objet se trans- 
forme imm£diatement pour devenir hindou/ The same author continues: — 'Cette 
impuissance de Tart grec k influencer l'lnde a quelque chose de frappant, et il faut bien 
Tattribuer k cette incompatibility que nous avons signage entre le g£nie des deux races, 
et non k une sorte d'incapacit6 native de llnde k s'assimiler un art etranger.' 1 

The readiness of India to assimilate suitable foreign material is shown by her 

proved willingness to borrow freely from Persia in ancient times and again 

after the Muhammadan conquest. 

The end Whatever influence Greece had exercised on Indian art was practically ex- 

of Greek hausted by a.d. 400. After that date the traces of Hellenistic ideas are too 

in uence. tr jflj n g to fe wor th mentioning. The medieval Brahmanical and Buddhist 

schools have nothing in common with Greek art, and the strange artistic 

forms introduced by the Muhammadan conquerors at the beginning of the 

thirteenth century were equally alien to Hellenic feeling. From the fifth 

century the art of India, whether Hindu or Muslim, must stand or fall on 

its own merits, without reference to Hellenic standards. The medieval 

Hindu revival and the advance of Islam, in large part synchronous, both 

involved a revolt against Hellenic ideas and a reversion to ancient Asiatic 

modes — a Renaissance aux depens des influences helleniques'. 2 

1 Brihier, L, 'Les Origines de TArt musulman' p. 108, Fig. 12) ; V. A. Smith (J. A. S. £., Part I, 
(La Revue des Iddes, No. 75, Mars 1910, p. 190). vol. lviii t p. 160); Foucher, VArt grico-boud- 

2 Simpson (J. Roy. Inst. Brit. Architects, 1894, dhique du Gandhara, p. 240, Fig. 290). 

:*« t^4P»^»«i 

A and H. Atlantes from Jamal^arhi 

C 'Cii^antoinachia 1 from 

I). Garland from Sarnath 
PLATE 41 JL) rr <J0 . ('l-bOtf 




Hi < 


i» = 


•i nr " ■r'Hf" <•■■ 

'iiiii'ii^i^^iiilliiiii!!!!!!!! 51 























PLATE 44. Seatrd Budilha, Sarnath 

Chapter Six 

The displacement of the Arsacid by the Sassanian dynasty of Persia u 
A.D. 226, the approximately simultaneous downfall of the Andhra kings in third 
who had ruled the Deccan for four-and-a-half centuries, and the dis- centur y- 
appearance of the Kushan or Indo-Scythian sovereigns of Northern India 
about the same time, unquestionably must have resulted in violent political 
and social disturbances on Indian soil during the third century. But hardly 
any record, archaeological or literary, has survived of that stormy interlude. 

The rise in a.d. 320 of the Imperial Gupta dynasty, with its capital at The Gupta 
Pataliputra (Patna), the ancient seat of empire, marks the beginning of a new em P ire - 
epoch. Under a succession of able and long-lived monarchs the Gupta 
dominions rapidly increased, until in the first quarter of the fifth century 
they comprised in modern terms Central and Western Bengal, Bihar, the 
United Provinces of Agra and Oudh, part of the Central Provinces, and the 
whole of Malwa and Gujarat, with the peninsula of Surashtra or Kathiawar. 
We know from the contemporary testimony of the Chinese pilgrim Fa-hien 
that the compact empire thus formed was then well governed by Chandra- 
gupta II, surnamed Vikramaditya. 

During the last quarter of the fifth century the Gupta empire was shattered Hun in- 
by the inrush of swarms of fierce Huns and allied nomad tribes from Central y^ ™* 
Asia. The short-lived Hun power was broken in India by a decisive victory ™* a ' c " 
gained by native princes about a.d. 528, but a long time elapsed before new 
political combinations of any stability could be formed. In the seventh 
century a great king named Harsha (606-47) reduced India north of the 
Narbada to obedience, while the Deccan submitted to his able contemporary 
Pulakesin II Chalukya, and the far south was governed by a powerful Pallava 
king. The Chalukya fell before the Pallava in 642, and five or six years later 
Harsha died childless, leaving the empire which he had won a prey to anarchy. 

During the seventh and eighth centuries the foreign settlers had become Limits of 
Hinduized, tribes developing into castes. When the ninth century opens we the Gupta 
find a new distribution of power among kingdoms mostly governed by so- peno ' 
called Rajputs, in many cases the descendants of chieftains belonging to the 
foreign tribes of Hunas, Gurjaras, and the like. The Huna or Hun invasions 
with the subsequent readjustments mark the division between the history of 
Ancient and that of Medieval India. 

All students of Indian literature now recognize the fact that during the Gupta 
reigns of Chandragupta II and his next two successors, from about a.d. 375 S^ re ^ d 
to 490, every branch of Hindu literature, science, and art was vigorously j^. ' 
cultivated under the stimulus of liberal royal patronage; and there is general 


agreement that Kalidasa, the greatest of Indian poets, graced the Gupta court 
and produced his masterpieces in the later years of the fifth century. 1 The 
plastic and pictorial arts shared in the good fortune of literature and science. 
In painting we have the frescoes of Ajanta and Bagh, and also those of 
Sigiriya in Ceylon. In coinage a marked improvement took place during the 
reigns of the earlier Gupta kings. 2 
Gupta Until quite recently the merits of Gupta sculpture were not generally or 
sculpture. f ree i y recognized. Owing to the destruction wrought by iconoclast Muslim 
armies and kings who overran and held in strength almost every part of the 
Gupta empire, few remains of the period exist above ground, except in out- 
of-the-way localities, and our present knowledge of Gupta art is largely the 
result of excavation. Sarnath, especially, has proved to be a rich treasure- 
house of Gupta, as well as of Kushan and earlier art. 3 The ravages of the 
Huns did not wholly stop the practice of the arts of civilization, and one of 
the surprises of recent exploration has been the discovery of many large Bud- 
dhist monasteries at Sarnath and other places in Hindustan dating from the 
fifth and sixth centuries. The sculpture of the period is mainly Buddhist 
and Brahmanical, the Jain works being few and of little artistic interest. 
E Gu"ta Exce P t certain coins of high artistic quality, as judged by an Indian 
works* stanc *ard, no wor k of art yet discovered can be referred to the reign of Samu- 
' dragupta (a.d. c. 335-75), the victorious general, and accomplished poet and 
musician, who has recorded his achievements on Asoka's pillar at Allahabad. 
The earliest known Gupta remains date from the beginning of the fifth 

archi Gu P ta In the fifth century were built the earliest stone buildings that have 
1 ec ure. surv i ve( j t They are chiefly tiny shrines situated in out-of-the-way places. 
Cunningham treated those little edifices as examples of the 'Gupta style', 
and enumerated seven characteristics of that style, namely, 

(1) flat roofs, without steeples of any kind; 

(2) prolongation of the head of the doorway beyond the jambs; 

(3) statues of the personified Ganges and Jumna guarding the entrance; 

(4) pillars with a massive square capital, surmounted by two lions back 

to back, often with a tree between them; 

(5) bosses over the capitals, and peculiar friezes; 

(6) continuation of the architrave of the portico as a moulding round 

the building; and 

(7) deviation of the plan from the cardinal points. 

1 For the history in detail see Early Hist. India, 'Greek influence has been suggested in the 

f m k ui 8 '^ 1 '^' drama of KaKdasa- Vincent Smith finds Euro- 
Marshall, JR. A. S., 1907, p. 1000. How- pean influence in the Gupta coins, but owns to 
ever, the supply of sculpture of the actual Gupta a lack of proof. The Gupta coinage may best 
fifth century is small when compared with the be taken as an orderly development of the numis- 
bulk of medieval sculpture. matic art of the Kushans. 


A characteristic example exists at Tigowa in the Jabalpur District, Central 

These small shrines are really the prototypes of much of the architecture 
of the great cave temples at Ajanta and Ellora. All the known examples are 
Brahmanical. At Udayagiri caves are to be found cut in the rock on exactly 
the above plan. At Ajanta the Buddhist rock-cut Vihara, which was originally 
nothing but a large pillared hall with cells for dwelling purposes leading into 
it on the three inner sides, was converted to titual purposes by cutting a shrine 
exactly corresponding to the Gupta structural shrines in the back wall. The 
doorways, with their pilasters and river-goddesses are reproduced in detail, 
proving the near relation of Ajanta architecture to the Gupta. Some of these 
shrines actually stand free, having a circumambulation passage cut around 
them. In the Saiva caves at Ellora and Elephanta the shrine is pushed forward 
into the body of the hall directly in front of the main entrance. These Linga 
shrines have doorways and door guardians on all four sides. The river- 
goddesses of the true Gupta shrines are placed on the level of the lintel on 
either side of the door. Ganga stands on her Makara and Yamuna on her 
tortoise. At Udayagiri, on the doorway of the Chandragupta Cave excavated 
in a.d. 401-2, the goddesses are represented without their vehicles. Here 
and elsewhere they stand beneath trees usually in the Woman and Tree 
posture. It seems that originally they were tree-spirits, like the Yakshis 
at Bharhut, and only became river-deities later. 1 

In the Ajanta frescoes it is evident that the palace and town architecture 
was entirely of wood, beautifully carved and painted. 

Although in the matter of style no distinctions based on the religious Siva and 
destination of particular images can be drawn, it will be convenient to finish Parvati. 
the description of selected Brahmanical stone sculptures before proceeding 
to the discussion of the Jain and Buddhist works and the metal castings. 

The Indian Museum, Calcutta, possesses a remarkable group of Siva and 
Parvati (Km. 40) from Kosam in the Allahabad District, bearing an inscrip- 
tion dated a.d. 458-9. The consorts stand side by side, each with the right 
hand raised and the open palm turned to the front. The head-dress of the 
goddess is described as a most elaborate construction, which recalls that 'of 
some Dutch women, and consists of a huge, transverse, comb-like ornament 
projecting beyond the side of the head, and terminating on both sides in 
large wheel-like ornaments, from the centre of which depends a large tassel. 
There are huge ear-ornaments and very massive bangles.' 2 

A temple at Deogarh, in the Lalitpur subdivision of the Jhansi District, Siv j*as 


1 Cunningham, A. S. Rep., vol. x, pp. 48-56; a Anderson, Catal., Part II, p. 286; Cunning- 

Ibid., vol. ix, pp. 42-6 ; Bloch, Progr. Rep. A. S. t ham, A. S. Rep., vol. x, p. 3 ; photo. 669 in I. M. 

East. Circle, 1907-8 ; Cousens, Progr. Rep. A.S., List. 
West. Ind. % 1903-4. 


U. P., is adorned with sculptures of exceptionally good quality in panels 
inserted in the plinth and walls, which may date from the first half of the 
sixth century. That region probably escaped the Hun troubles owing to its 
remote situation. A panel on the eastern fa9ade, representing Siva in the 
garb of an ascetic (mahayogi), attended by another yogi and various heavenly 
beings hovering in the air (Plate 48), may claim a place among the best efforts 
of Indian sculpture. The principal image is so beautifully modelled and so 
tastefully posed that we almost forget the inartistic excrescence of the extra 
pair of arms. The flying figures are admirably designed so as to give the 
^appearance of aerial flight. The modelling of the feet and hands deserves 
particular notice, and the decorative carvings are in good taste. The close- 
fitting garments of all the figures and the wigs of some of the attendants are 
characteristic of the period. 
Vishnu on Another panel from the south facade of the same temple is equally good 
Ananta. (pj ate ^ The su bject is Vishnu as the Eternal, reclining on the serpent 
Ananta, the symbol of eternity, with the other gods watching from above. 
The principal image is beautifully posed, and the extra arms most dexterously 
arranged. The wig-like dressing of the hair is very prominent in this fine 
group. 1 
A Rajgir The little-known ruins at Rajgir, the ancient capital of Magadha, include 
sculpture. a re jj e f f a f ema i e) facing front, which is of the Gupta age (Plate 46 c). 2 The 
sculptures at Nachna-Kuthara in Ajaigarh State are very fine, especially the 
doorways of the two shrines. Cunningham describes them as 'being much 
superior to all medieval sculptures, both in ease and gracefulness of their 
attitudes, as well as in real beauty of the forms'. 
Fifth-century Several ancient sites in the south-western part of the Allahabad District 
sculptures 2 ^ ave y^^ed to slight excavation many remarkable Buddhist sculptures in 
' stone, proved by dated inscriptions to be assignable to the reigns of Chandra- 
gupta II, his son Kumaragupta I, and his grandson Skandagupta in the fifth 

The vigorous, and at the same time refined, sculpture adorning the ruins 
of a Gupta temple at Garhwa, twenty-five miles south-west of Allahabad, is 
illustrated on Plate 45, giving back and side views of one pillar. The panels 
on the front are arranged according to the ancient Indian fashion, and the 
style is related to the art of Sanchi and Bharhut much more closely than to 
medieval art. There is no trace whatever of Gandharan influence. The 
figures are well drawn, and modelled on purely naturalistic principles. 

The beautiful ornament on the side is described by Cunningham as con- 
sisting of — 

1 For Deogarh antiquities see Cunningham, A. previously. 

S. Rep., vol. x, pp. 100-4, P1 - XXXIII-VI. The 2 Cunningham, Rep., vol. xxi, p. 96. 

groups of sculpture have not been published 

A. Sculptured door- 
jainh, (iarwha 

IJ. Sculptured door- jamb, 

C Buddha; Mathura 
Museum [A. 5] 


^,.^£ ■*'•£*'" . 

A. The rivcr-jjoddess, OariLja. 
l 7 dayai*iri, lihopal 

B. The river-ijoddess, (ianj^a. 

C. Female imaj^e, stucco. Rajpir 


'the undulating stem of a creeper, with large curling and intertwining leaves, and small 
human figures, both male and female, climbing up the stem, or sitting on the leaves 
in various attitudes. The whole scroll is deeply sunk and very clearly and carefully 
carved ; and ... is one of the most pleasing and graceful specimens of Indian archi- 
tectural ornament.' 1 

The commendation is fully justified; nothing better can be found in the 
earlier work at Mathura, and the Garhwa design would do credit to an Italian 
fifteenth-century artist. 

Among the numerous excellent sculptures of Gupta age, disclosed by Seated 
recent excavations at Sarnath, the most pleasing, perhaps, is the seated Bud- Buddha 
dha in white sandstone, 5^ feet in height (Plate 44). n r at ™ 

The deer-park at Sarnath having been the place where the Wheel of the 
Law was first turned, or, in other words, the doctrine of the Buddhist way 
of salvation was first publicly preached by Gautama Buddha, his effigy is 
naturally represented with the fingers in the position (mudra) associated by 
canonical rule with the act commemorated. The wheel symbolizing the Law 
and the five adoring disciples to whom it was first preached are depicted on 
the pedestal. The woman with a child on the left probably is intended for 
the pious donor of the image. The beautifully decorated halo characteristic 
of the period is in marked contrast with the severely plain halos of the 
Kushan age. The style, marked by refined restraint, is absolutely free from 
all extravagance or monstrosity. Allowance being made for the Hindu canon 
prohibiting the display of muscular detail, the modelling must be allowed to 
display high artistic skill. The angels hovering above may be compared with 
the similar figures at Dcogarh. The close-fitting smooth robe is one of the 
most distinctive marks of the style, which is singularly original and absolutely 
independent of the Gandhara school. 2 The composition is so pictorial that 
it might have been designed after the model of a painted fresco. 

An excellently inscribed standing Buddha of the fifth century in the Standing 
Mathura Museum, height 7 feet z\ inches, while clearly related to the Sarnath ?? d ? ha at 
seated image in several respects, differs widely in the treatment of the drapery, at ura " 
which at Mathura shows a reminiscence of Hellenistic forms. The skill with 
which the body is shown through the transparent garments is characteristic 
of the best Gupta sculpture. 3 

1 The Garhwa remains are fully described and tion set in. This period is characterized by quite 
illustrated by Cunningham, A. S. Rep.* vol. iii, an extraordinary amount of intercourse between 
pp. 53-61 ; vol. x, pp. 9-15, with plates. India and foreign countries' (Kaye, J. R. A. £., 

2 Foreign influence has been discerned in the 1910^.759). Many Indian 'embassies' to China 
Gupta coins. The conquest of Western India and the Roman empire are recorded during this 
undoubtedly opened the western ports to North- period. All that is of value in the Hindu mathe- 
ern India, which suggests a channel of influence matics of the time, according to Mr. Kaye, is 
via the Red Sea. 'The period when mathematics Greek. 

flourished in India commenced about a.d. 400 3 Careful study of an adequate number of ex- 
and ended about a.d. 650, after which deteriora- amples might disclose the existence of several 



Colossal The unique copper colossus of Buddha, about 7 J feet high, now in the 
copper Museum and Art Gallery, Birmingham (Plate 47 b), is, perhaps, more closely 
Buddha. a ^ n to t ^ e g arna th than to the Mathura image, the robes being almost 
smooth, with the folds marked very faintly. The transparency of the gar- 
ments is clearly marked. The statue was excavated by certain railway 
engineers in 1862 from the hall of a ruined monastery situated between the 
modern mart and the railway station at Sultanganj, on the Ganges, in the 
Bhagalpur District, Bengal. One of the discoverers brought it home, and 
some years later presented it to the Birmingham Museum. The image was 
found lying on the ground, having been wrenched from its massive granite 
pedestal; but was practically perfect, except that the left foot was broken off 
above the ankle. The earliest possible date is indicated by the discovery 
in an adjoining stupa of a coin of the last Western Satrap of Surashtra, accom- 
panied by one of his conquerors, Chandragupta II, Vikramaditya, who an- 
nexed his dominions about a.d. 390. The statue, therefore, may be dated 
early fifth century. 
Technique According to Rajendralala Mitra, the material is 'very pure copper', cast 
of the i n two distinct layers, the inner of which was moulded on an earthy, cinder- 
co ossus. uk e core ^ composed of a mixture of sand, clay, charcoal, and paddy (rice) 
husks . The segments of this inner layer were held together by much corroded 
iron bands, originally three-quarters of an inch thick. The outer layer of 
copper seems to have been cast over the inner one, presumably by the are 
perdue process. It was made in several sections, one of which consisted of 
the face and connected parts down to the breast. 

Lumps of copper ore found close by indicate that the smelting and casting 
were done on the spot. The hand of another large copper statue was picked 
up, and three small Buddhas of the same metal were discovered. One, nearly 
destroyed by rust, was seated, the three others were standing, with halos 
broken and detached. A large Bihar image of carboniferous shale was 
found near by : this image is also in the Birmingham Museum. 1 
TheManku- At the adjoining village of Mankuwar a very perfect seated Buddha of 
war Buddha, unusual type was found (Plate 47 a), bearing a dedicatory inscription dated 
in the year 129 g.e. = a.d. 448-9. The peculiar head-dress, if it be a head- 
well-marked local schools of sculpture during J. A. S. 2?., vol. xxxiii (1864), pp. 361 seqq. with 
the Gupta period; but it would be premature lithograph: Cunningham, A. S. Rep. y vol. x, 
at present to attempt such refinement in the p. 127; xv, p. 126: Anderson, Catalogue, I. M., 
treatment of a subject which needs to be first Part II, p. 481. The modern cire perdue pro- 
sketched on broader lines. Vincent Smith ac- cess of casting bronze over a core made of 
centuates the division into local schools of modelling clay mixed with pounded brick and 
Indian art too much. The development of plaster of Paris is described, ibid., p. 85. The 
Indian art is a matter of periods rather than Sultanganj process seems to have been essentially 
geography, as the affinities between Mathura the same, with the addition of a second layer of 
and Amaravati show. copper. 

1 The Sultanganj discoveries are described in 

A. The Mankuwar Buddha 

B. Sultanganj Buddha, copper- 
casting now in the Birmingham 


PLATE 48. Siva as an ascetic (mahayogi) ; Dcogarh 


dress, is, as Cunningham remarked, like that now worn by the Abbots of 
Bhutan, and the image may be the work of a northern artist. 1 The webbed 
hand was one of the traditional marks of a Buddha, according to some schools. 
The wheel below symbolizes the turning of the Wheel of the Law, that is to 
say, the preaching of the doctrine destined to traverse the world like the 
chariot wheels of a conquering monarch. The expression of the face differs 
from that of most images, and the work undoubtedly is a notable example of 
fifth-century sculpture. The clothing is merely the Indian waist-cloth, quite 
different from the robe of the ordinary Buddha. 

The existence of the Sultanganj Buddha, weighing nearly a ton, is good The Iron 
evidence of Indian proficiency in metallurgy at the beginning of the fifth Klhr of 
century. Still stronger testimony to that skill is borne by the celebrated Iron 
Pillar of Delhi, set up about a.d. 415 by Kumaragupta I in honour of his 
father, Chandragupta II, Vikramaditya. The total length of the pillar 
from the top of the bell capital to the bottom of the base is 23 feet 8 inches, 
and the diameter diminishes from 164 inches below to 1205 inches above. 
The material is pure malleable iron of 766 specific gravity welded together, 
and the weight is estimated to exceed six tons. 'It is not many years since 
the production of such a pillar would have been an impossibility in the 
largest foundries of the world, and even now there are comparatively few 
where a similar mass of metal could be turned out.' The statue originally 
surmounting the pillar having disappeared, this marvellous metallurgical 
triumph does not further concern a history of fine art. 2 

The old Asokan practice of erecting isolated monumental columns, usually Gupta 
monolithic, was revived in Gupta times. Samudragupta, perhaps the most m o noll * lc 
brilliant of an able dynasty, does not seem to have erected pillars of his own, 
and was content to record the history of his reign on a pillar of Asoka, now 
at Allahabad (Prayaga), which, apparently, has been removed from Kau- 
sambi. The earliest extant stone pillar of Gupta age is that erected at Bhitari 
in the Ghazipur District, U. P., by Skandagupta about a.d. 456 to com- 
memorate his wars with the Huns and Pushyamitras. The next, set up at 
Kahaon in the Gorakhpur District, U. P., in a.d. 460-1, early in the reign of 
the same king, by a private member of the Jain community, is adorned with 
the images of five Jain saints, one in a niche at the base, and four on the sum- 
mit. The statues, as usual with the Jains, are conventional and of little artistic 

The third in date is the fine monolithic pillar, 43 feet high, set up at Eran 
in the Sagar District, C. P., as 'the flag-staff of four-armed Vishnu', in 

1 According to Dr. Bloch (J. A. S. B. % vol. lxvi, (Mihrauli)* (J. R. A. S., 1907, pp. 1-18). The 

Part I, p. 283), what looks like a close-fitting cap passage quoted is from V. Ball, Economic 

really is a conventional arrangement of the hair. Geology of India, p. 338, 1st ed., 1881. 
2 V. A. Smith, 'The Iron Pillar of Delhi 


a.d. 484-5. The statue now on the top is a two-armed male figure with two 
faces and a radiated halo— a form not easy of interpretation. 

Two great monolithic columns, the better preserved of which is 39 feet 
5 inches long, excluding the detached abacus, lie at Sondani or Songni, near 
Mandasor in Sindhia's Dominions, and bear inscriptions recording the 
decisive defeat of the Huns by King Yasodharman about a.d. 528. Several 
specimens of good contemporary sculpture adjoin. 

Another great monolithic column, with a worn inscription of late Gupta 
age, and 47 feet high, stands at Pathari, about thirteen miles to the south- 
west of Eran. 

Furthermore, Cunningham's illustration of the Sanchi torso, now in the 
Victoria and Albert Museum, South Kensington, proves it to have stood at 
the base of a column. It is to be dated c. a.d. 500. 1 
Gupta forms The Gupta form of capital is generally characterized by a large square 
capita . a b acus f twice the breadth of the shaft, surmounted by two lions sitting 
back to back, sometimes with a tree or human figure between them. The 
Budhagupta column has four lions, one at each corner. The process by which 
the medieval capital was evolved from the Persepolitan through the Gupta 
forms is explained by Cunningham as follows: 

'The old bell-capital of the Asoka period has now been considerably altered by 
bands of ornament and the addition of foliated turn-overs. In later times these turn- 
overs were greatly increased in size, while the body of the bell was lessened until it 
resembled a water-vessel or kumbha, which eventually became its well-known designa- 
tion. This curious change from the old bell-capital of Asoka to the water-vessel of 
the medieval temples is very clearly traceable in the different examples of the Gupta 
period.' 2 

A transition The foregoing select illustrations will, it is hoped, be considered sufficient 
sculpture. tQ establish ^e claims of the Gupta sculpture of Northern India to favourable 
consideration on its merits as art. It is, as Sir John Marshall observes, en- 
dowed with 'freshness and vitality', while the designs are singularly refined 
and the technical execution of the best pieces is exquisite. Students who 
desire to pursue the subject further will find more material in the publications 
noted below. 3 

1 A. S. I., vol. x. Ant., 1908, p. 107, with plates: Fleet, op. cit., 

2 A. S. Rep., x. 88. References for the pillars Nos. 33-5. Pathari— Cunningham, Rep., vol. 
are:— Bhitari— Cunningham, A. S. Rep., vol. i, vii, p. 67; x, p. 70: not in Fleet. Smaller pillars 
p. 38, Pis. XXIX, XXX : Fleet, Gupta Inscr., No. or stelae of Gupta age exist at several places. 
13. Kahaon — Cunningham, Rep., vol. i, p. 92, *J. R. A. S., 1907, pp. 996-1000: Attn. Rep. 
PI. XXX: Fleet, op. cit., No. 15. Eran — Cun- A. S., India, 1903-4, pp. 213-26; 1904-5, pp. 
ningham, Rep., vol. vii, p. 88; x, p. 81, PI. 43-58 and 59-104; 1905-6, pp. 61-85 *&& l ZS~ 
XXVI: Fleet, op. cit., No. 19. Sondani— Ind. 40; 1906-7, pp. 44-67 and 68-101. 

Chapter Seven 

While the most characteristic and distinctive sculptures of Gupta age Ajanta 
occur in Northern India, the rock-cut shrines and monasteries of the sculptures. 
Deccan are adorned with numerous sculptures more or less closely 
related to those of the north. These as a whole are later and must be con- 
sidered as intermediate between the Gupta work and the later medieval. 
At Ajanta, interest having been concentrated chiefly on the paintings, the 
accounts of the sculptures are meagre and detailed photographs are scarce. 

The numerous sculptures in Cave XXVI include a gigantic recumbent The Temp- 
Dying Buddha, 23! feet in length, bearing a general resemblance to the fifth- tation - 
century image at Kasia in the Gorakhpur District, U*. P. The most notable 
sculpture on the walls is the large and crowded composition representing 
the Temptation of Buddha, which Dr. Burgess describes as 'beautiful', add- 
ing that 'several of the faces are beautifully cut'. The subject is also treated 
at Ajanta in fresco and at Borobudur, Java, in sculpture. The fantastically 
dressed hair, characteristic of the period, worn by several of the figures in 
the Ajanta sculpture should be noted. The elephants are well drawn, as usual. 

In Cave I, supposed to be the latest of the completed excavations, a great Chase of 
quantity of rich sculpture exists, dealing chiefly with incidents in the lives of w"* bul1 - 
Buddha. A scene depicting the chase of the wild bull is praised as being 
'spiritedly carved'. 

The sculptures in the Bagh caves, Gwalior State, until recently known only Sculptures 
through drawings prepared for Dr. Burgess, have since been officially of tilt Ba S h 
photographed. The best images, representing Buddha, or possibly a Bodhi- caves ' 
sattva, with two attendants, are the south-western group in the Gosain's Cave, 
No. II. The style connects them with the Gupta rather than the medieval 
period, and especially with the sculptures in Cave IX, Ajanta. They may 
have been executed in the latter half of the sixth century. The pose is easy 
and the modelling good. 1 

The late Buddhist caves at Aurangabad in the Nizam's dominions, not far Aurangabad 
from Ellora, are supposed to date from the 'seventh century of our era, and caves - 
perhaps towards the end of it'. z Whatever their exact date may be, the sculp- 
tures are related more closely to those of the Gupta age than to the Tantric 
works of the medieval period. 

The principal cave, No. Ill, contains many columns most elaborately The 


1 The Bagh caves (India Society, London). seems more nearly related to Ajanta than to ro 8 re8s • 

1 Burgess, in Hist. Ind. and E. Arckit., and ed. Ellora or Badami. It is probably mid-sixth 
(1910), vol. i, p. 205. The Aurangabad work century. 


decorated with figure sculpture as well as complex patterns. On certain of 
these columns a sixteen-sided portion is 

'carved with sixteen scenes which may be an anticipation of Cruikshank or John 
Adam, for they seem intended to picture the "Drunkard's Progress". The number of 
figures varies from two to four in each. Two persons are represented sitting together, 
apparently drinking in the most friendly way, then staggering along, then dancing 
with their backs to each other, then quarrelling; one is being dragged along helpless 
between two men, and so on in successive panels/ l 

It is a pity that no reproductions of these lively stone pictures have been 
published. The subjects recall the much earlier ' Bacchanalian ' sculptures of 
Mathura, and suggest speculations concerning certain varieties of Buddhism 
in practice. 
A frieze. In the same cave an architrave bears on the front a long frieze of fourteen 
scenes of the Jataka kind in relief, including an impalement, a battle in a 
forest, and other incidents, the meaning of which is not known. The drawing 
in Dr. Burgess's volume is on such a small scale that it is impossible to judge 
fairly the quality of the art, but, so far as can be seen, the action is vigorously 
Egyptian- Certain groups of kneeling worshippers in the same excavation are ex- 
like heads, tremely curious. The mode of hair-dressing has quite an Egyptian appearance. 
The later cave At Ajanta much of the sculpture is reminiscent of the Gupta fifth-century 
temples. s tyl e . In the later caves the work is definitely medieval, being based on the 
iconography of the time. It is almost entirely hieratic. It is distinguished 
from the earlier work, also, by its richly crowded design. The bands of masks 
('face of fame', kirtti-mukha), the grotesque animals with foliated tails, and 
many motives based on jewellery designs, distinguish it from the Gupta. For 
the purpose of illustration it will suffice to reproduce a few select specimens 
from the shrines at Badami, Ellora, and Elephanta, with two sculptures from 
temples of later date. The cave sculptures of interest range in date from the 
sixth to the eighth century. 
The Brah- The works of art are shared by all the three indigenous Indian religions — 
™mo?t im" Brahmanical Hinduism, Buddhism, and Jainism. The Buddhist and Brah- 
portant. manical works are both numerous and very much alike in spirit and style. 
The spirit of the new art will be most easily understood from study of the 
Brahmanical sculptures, to which the few illustrations for which there is space 
will be restricted. In those days Buddhism was a dying faith, slowly perishing 
by absorption into the enveloping mass of Hinduism. The Brahmanical 
works of art exceed the Buddhist, not only in number but in merit. To Mr. 
Havell and Dr. Coomaraswamy the compositions in the cave temples are 
'examples of the finest period of Hindu sculpture, from about the sixth to 
the eighth century, when orthodox Hinduism had triumphed over Buddhism' ; 

1 Burgess, A. S. W. /., vol. iii, p. 67. 

PLATE 49. Vishnu on Ananta; Deogarh 

PLATE 50. Verandah pillar. Cave III, Uadami 


but most European observers experience difficulty in appreciating the artistic 
qualities of those compositions. Mr. Fry, commenting on Mr. HavelPs book, 
is more appreciative than many writers: 'The free and picturesque composi- 
tion from Ellora', he says, 'representing Ravana under the mountain of 
Kailasa, complicated though it is, is held together by the dramatic beauty 
of movement of the figures of Siva and Parvati. The same dramatic vitality 
is apparent in the struggle between Narasinha and Hiranya-Kasipu> also from 
Ellora. Indeed, all the Ellora sculptures here reproduced appeal to the 
European eye by a relatively greater observance of the laws of co-ordination, 
and by an evidence of dramatic force which indicates that Indian art did not 
always convey its meaning in a strange tongue/ l To be judged fairly the 
sculptures should be seen in the mass and among their solemn surroundings. 
They undoubtedly suffer grievously by being excerpted in bits and repro- 
duced in illustrations a few inches square. While fully conscious of the 
difficulties inherent in the attempt to illustrate the colossal and fantastic 
creations of the cave sculptors within the limits of an ordinary page, I have 
tried to select fairly a small number of examples generally recognized as 
among the best. 

The cave temples at Badami in the Bijapur District, Bombay, exhibit Sculptures 
among other decorations long sculptured story-telling friezes, extremely at Badami. 
curious, but so clumsily executed as hardly to deserve the name of works of 
art. They date from the closing years of the sixth century. 2 From an artistic 
point of view the bracket figures of a god and goddess on the top of a pilaster, 
as shown in Plate 50, are by far the best things at Badami. 

There are four cave-temples, all linga-shrtnes, at Badami, all cut on the same 
plan and at more or less the same time. As has been said, at Ajanta many 
of the shrines inset in the back wall of the so-called Viharas are simply repro- 
ductions of the flat-roofed, structural Gupta shrines of the fifth century, with 
doorway and four-pillared verandah accurately reproduced. At Badami the 
shrine is cut in the same position but is simplified into a plain cell without 
verandah. At Ellora this cell was cut away from the rock by means of a cir- 
cumambulation passage. The stylobate of Cave 1 is carved with a distinctive 
frieze of dancing dwarfs, which also appears at the base of sculptured panels 
and in the other caves. The sculpture of Caves 2 and 3 is Vaishnava, and 
contains magnificent sculptures of the Man-Lion and Boar incarnations, 
and a fine Bhogasanamurti. Cave 3 contains an inscription of the Chalukyan 
king, Mangalesvara, dated in a.d. 578. Cave No. 4 is Jain. 

The iconographical nature of the subjects chosen by the cave sculptors is Bhairava 
well exemplified by the Bhairava and Kali group in the Das Avatara, or ^J^ ali ' 

1 Imp.Gaz., s.v.Abu. Quart. Rev., 1910, p. 235. 66. Also Burgess, Arch. Rep., Belgam and 
1 Described, and illustrated with other sculp- Kaladgi; Rea, Chalukian Architecture, and the 
tures, by seven plates in Ind. Ant., vi, pp. 354- monograph, Arch. Sur. Memoirs. 


'Ten Incarnations' temple at Ellora, dating from about a.d. 700 (Plate 51), 
described by Dr. Burgess as follows: 

'Beginning on the north side with the Saiva sculptures — the first from the door is 
Bhairava or Mahadeva in his terrible form ; and a more vivid picture of the terrific 
a very diseased imagination only could embody. The gigantic figure lounges forward 
holding up his elephant-hide, with necklace of skulls (mundmala) depending below 
his loins ; round him a cobra is knotted, his open mouth showing large teeth, while 
with his trisula [trident] he has transfixed one victim, who, writhing on its prongs, 
seems to supplicate pity from the pitiless ; while he holds another by the heels with 
one of his left hands, raising the damru [small drum] as if to rattle it in joy, while he 
catches the blood with which to quench his demon thirst. To add to the elements 
of horror, Kali, gaunt and grim, stretches her skeleton length below, with huge mouth, 
bushy hair, and sunken eyeballs, having a crooked knife in her right hand, and reaching 
out the other with a bowl, as if eager to share in the gore of its '[sic] victim ; behind her 
head is the owl [one species is called Bhairava], the symbol of destruction, or a vampire, 
as fit witness of the scene. On the right, in front of the skeleton, is Parvati; and higher 
up, near the feet of the victim Ratnasura, is a grinning face drawing out its tongue. 
Altogether the group is a picture of the devilish ; the very armlets Bhairava wears are 
ogre faces.' l 

Rescue of A subject rarely represented in sculpture, the rescue by the god Siva of his 
Mar dTa wors hipper Markandeya from the clutches of the messenger of Yama, god of 
death, 2 appears twice at Ellora, and is treated with less grimness than the 
Bhairava group. The earlier composition in the Das Avatar a Cave is more 
vigorous than that at the Kailas, half a century or more later in date. 
Siva The sculptures in the Lankesvara section of the Kailasa temple are com- 
ancmg. menc j ec j as having been 'executed with great care and minute detail'. The 
best known, and perhaps the most meritorious, is that exhibiting Siva per- 
forming the Tandava dance (see p. 89), a work remarkable for the good model- 
ling of the principal image, and the scrupulous exactitude of the carving. The 
river-goddess from this cave is especially fine (Plate 53). 
A good Vishnu at Ellora is shown in Plate 52 A. The god is imagined as 

1 Fergusson and Burgess, The Cave Temples of Tantric Hinduism (including late Buddhism) 

India, p. 346 (1880). concerns itself little with ethics. The earlier 

Vincent Smith's attitude to Brahmanical Buddhism, as a religion, busied itself mainly 

Indian art was not sympathetic: 'The religion with morals, and consciously aimed at "the wel- 

which finds expression in imagery so truly devil- fare and happiness of all creatures", a noble 

ish is not a pleasant subject of contemplation, ideal which found its utterance in art. In Brah- 

and no amount of executive skill or cleverness manical Hinduism of all varieties each man seeks 

in the production of scenic effect can justify, on at the most his own personal salvation, and so 

aesthetic grounds, such a composition, which is Brahmanical art seldom exhibits a trace of 

frankly hideous. Its claim to be considered a human sympathy, a defect dearly purchased by 

work of art rests solely upon its display of power its much praised idealism.' 

in a semi-barbaric fashion. The horror of the 2 This is a common subject in the later sculpture 

subject and its treatment is not redeemed by any of the south, 
apparent ethical lesson. Indeed Puranic and 

PLATE 51. Bhairava. Dasavatara Cave, Ellora 

A. Krishna. Kailasa, Kllora 

B. Marriage of Siva and Parvati. Klephanta 


C. Siva as the Great Ascetic. Elephanta 


striding through the seven regions of the universe in three steps, and is here 
shown as taking the third step. 1 

The famous caves on the island of Elephanta in Bombay Harbour are Siva at 
usually supposed to date from the eighth century. The colossal sculptures Elephanta. 

%'yj; -,>fii 

Siva dancing ; in Lankcsvara temple. 

are most imposing and effective when viewed in the recesses of the caverns. 
The first of the two specimens selected is the favourite subject of the marriage 
of Siva with Parvati; and the second is the representation of Siva as the Great 
Ascetic (Plate 52 b and c), which may be compared with the far finer Gupta 
treatment of the same subject. The most imposing of the Elephanta sculp- 

1 The sculpture is unfinished. It is more probably a rendering of Krishna lifting up Govardhan 
and sheltering the flocks. 



tures is the gigantic Trimurti or Trinity, which is the first thing discerned as 
the eyes become accustomed to the gloom, on approaching the cave through 
the present main entrance. The original main entrance is to one side and 
leads direct to the square Linga shrine. 1 

I — - ■ M^~ .-. " '. ~ -_- It in 1 ■ 

of medieval 


Das Avatdra Cave, Ellora. 

The chronology of these caves must be deduced from the following facts: 

I. The likeness of the Ajanta shrines to the Gupta fifth century shrines, 

taking into consideration their plan, the sculptured doorways, and 

the Vakataka epigraphy. 

II. The Visvakarma Cave at Ellora is linked with Ajanta by the style of its 

sculptured stupa. The seated Buddha on it is in the style of certain 

1 For the survival of Buddhist traits in the Languages, &c, of Nepal, 133 seqq., as quoted 
Saiva cults see Or. Qu. Mag., vii. 218 seqq.; by Sir R. Temple in Ind. Ant., xxii. 363. 


later sculptural additions at Ajanta such as the bas-relief Temptation 
of the Buddha in Cave XXVI. It is also linked with Badami by the 
little frieze of dancing dwarfs in the bas-relief pavilions on either side 
of the chaitya-mndovr. It may be dated in the second half of the sixth 
century. The sculpture of the Buddhist caves at Ellora corresponds 
most exactly with the Ajanta frescoes in style, especially the doorway 
of Cave VI. 1 

III. At Ellora two other styles of sculpture exist, which may be respectively 
typified by the dynamic Brahmanical sculptures of the Das Avatara 
Cave and the later Kailasa sculptures. The Kailasa is accepted as 
having been excavated in the second half of the eighth century. The 
Brahmanical caves at Ellora would seem to belong to the earlier half 
of the same century. 

1 Fergusson and Burgess, Cave Temples, PI. LXI. 

Chapter Eight 


Long history T^ew, very few, people realize that the art of painting in India and Ceylon 
of painting wA has a long history, illustrated by extant examples ranging over a period 
in India A exceeding two thousand years, and that during the so-called Dark Ages 
eyon. ^ j nc u an anc j Ceylonese painters attained a degree of proficiency not 
matched in Europe before the fourteenth or fifteen century. Nevertheless, 
such are the facts. In this chapter and the next following the history of the 
art in India and Ceylon, so far as its practice was dominated by Hindu ideas, 
will be traced from the earliest times of which there is record until the 
present day; but, unfortunately, the incompleteness of the record compels 
the historian to leave many gaps in his narrative. The widest of those gaps 
lies between the close of the Ajanta series in the seventh and the introduction 
of the Indo-Persian style by Akbar in the sixteenth century. During that long 
period of more than nine hundred years hardly anything definite is known 
concerning the productions of Indian and Ceylonese painters. 
Literary The ancient literature of India and Ceylon contains many references to 
evidence, pictorial art, the earliest, perhaps, being those in books of the Pali Buddhist 
canon dating from some three or four centuries before the Christian era. 
Several passages in those books tell of pleasure-houses belonging to the kings 
of Magadha and Kosala in Northern India as being adorned with painted 
figures and decorative patterns, presumably similar to the earliest known 
frescoes in Orissa and at Ajanta. 1 Painted halls are also mentioned in the 
Ramayana; and allusions to portraits are frequent in the dramas of Kalidasa 
and his successors from the fifth to the eighth century after Christ. The 
Ceylonese chronicle, the Mahavamsa, composed probably in the fifth century, 
tells of the mural paintings decorating the relic-chamber of the Ruwanweli 
dagaba constructed by King Dutthagamini about 150 B.C. The testimony of 
native writers is confirmed by that of the Chinese pilgrims in the fifth, sixth, 
and seventh centuries, who notice several examples of celebrated Buddhist 
pictures; and by Taranath, the Tibetan historian of Buddhism, who, when 
writing at the beginning of the seventeenth century, ascribes the most ancient 
pictures to the gods, and declares that they were so marvellous as to bewilder 
beholders by their realism. 2 

1 Rhys-Davids, Buddhist India, p. 96; citing Schiefner, p. 278. His testimony will be dis- 
Vin., ii. 151 ; iv. 47, 61, 298; Sam., 42, 84. cussed in the next chapter. 

2 Geschichte des Buddkismus, ch. xxiv, trans] . 


The literary evidence thus summarily indicated would alone amply prove Range in 
the early and continuous practice of the painter's art in both India and Ceylon ; time of 
but it is unnecessary to labour the proof from books, because evidence of a extan * 
more satisfactory kind is furnished by the considerable surviving remains of 
ancient painting from the second century before Christ; which, even in their 
present fragmentary and mutilated state, enable the modern critic to appraise 
the style of the early Indian artists, and to recognize the just claim of the art 
of India and Ceylon to take high rank among the ancient schools of painting. 
We will now proceed to give in this chapter an account of the extant remains 
of Indian and Ceylonese painting from the second century before Christ to 
the middle of the seventh century of the Christian era. 

The oldest Indian pictures are found in the Jogimara Cave of the Ramgarh The oldest 
Hill to the south of the Mirzapur District now attached to the Central Provinces. P aintin gs- 

These pictures, apparently executed in the customary Indian method of 
fresco, which will be explained presently, are divided into concentric circles 
by bands of red and yellow, sometimes enriched with a geometrical design, 
these circles seemingly being again subdivided into panels. The general 
nature of the subjects of the four best preserved panels can be understood 
from the following brief description: 

A. In the centre a male figure is seated under a tree, with dancing girls and Description, 
musicians to the left, and a procession, including an elephant, to the right. 

B. This panel exhibits several male figures, a wheel, and sundry geo- 
metrical ornaments. 

C. One half of this panel merely shows indistinct traces of flowers, horses, 
and clothed human figures. 

In the other half is seen a tree having a bird and apparently a nude child 
in its branches, while round the tree are grouped other nude human figures, 
wearing their hair tied in a knot on the left side of the head. 

D. The upper part of one half of this panel contains a nude male figure 
seated and attended by three clothed men standing, with two similar seated 
figures and three more attendants on one side. In the lower part are depicted 
a house with the horseshoe or so-called chaitya window, an elephant, and 
three clothed men standing in front. Near this group are shown a chariot 
drawn by three horses and surmounted by an umbrella, and a second elephant 
with an attendant. In the second half of the panel the figures are generally 
similar in character. 

The early date of the paintings, which are fairly well preserved, is attested Date and 
by inscriptions, evidently contemporary, and by the style, which recalls that st y le - 
of the sculptures at Sanchi and Bharhut. They probably date from the second 
century, and cannot well be later than the first century before Christ. The 
subjects cannot be interpreted at present, but the nudity of the principal 
figures suggests a connexion with the Jain rather than the Buddhist religion, 


if the cave and paintings had any religious significance, which is doubtful. 
As regards technique, the designs are painted usually in red, but occasionally 
in black, on a white ground. The outlines of the human and animal figures 
are drawn in black. Clothing is white with red outlines, hair is black, and 
eyes are white. Yellow appears in the dividing bands only, and blue does not 
seem to occur. 1 
Topography. The story of the art of painting in India is continued by the celebrated 
frescoes of the Ajanta caves in the west, ranging in date from about a.d. 50 
or earlier, to about the sixth century, a period of some six or seven centuries, 
and constituting the most important mass of ancient painting extant in the 
world, Pompeii only excepted. The caves, twenty-nine in number, are 'ex- 
cavated in the face of an almost perpendicular scarp of rock about 250 feet 
high, sweeping round in a curve of fully a semicircle, and forming the north 
or outer side of a wild and lonely glen, down which comes a small stream \ 
This glen or ravine, a scene of great natural beauty and perfect seclusion, 
admirably adapted for a monastic retreat, is situated about three and a half 
miles south-west from Phardapur, a small town in the Nizam's Dominions, 
standing at the foot of a pass across the Indhyadri Hills, which divide the table- 
land of the Deccan from the Khandesh District in the Tapti valley, and four 
miles WNW. from the town of Ajanta, not far from the battle-field of Assaye. 
The caves. 'The caves extend for a distance of about 600 yards from east to west round 
the concave wall of amygdaloid trap which hems in the stream on its north 
or left side, and vary in elevation from about 35 to 100 feet above the level 
of the torrent/ The numbers by which authors have agreed to designate 
them begin at the east end. Four of the excavations, Nos. IX, X, XIX, and 
XXVI, are churches (the so-called chatty as) , the rest being monastic resi- 
dences, the viharas of English writers. Some have never been completed. 
The principal works are elaborate architectural compositions, executed in the 
solid rock , the nature of which is very inadequately expressed by the term * caves ' . 
Extent and i n 1879 paintings to a greater or lesser extent remained in sixteen caves, 

pain?S^ N ° S - *' H > IV > VI > VI1 ' IX > X > XI > XV > XVI > XVII > XIX > XX ' XXI > XXII > 
and XXVI. The most important fragments were then to be seen in nine 

caves, Nos. I, II, IX, X, XI, XVI, XVII, XIX, and XXI, those in Cave XVII 

being the most extensive. 2 The most ancient excavations, Nos. VIII, XII, 

and XIII, have no paintings. No. XIII, perhaps the earliest of all, has 

polished walls, and may date from 200 B.C. Six of the caves, Nos. VIII, IX, 

X, XI (with some sculpture possibly later), XII, and XIII are concerned with 

1 The only information on the subject is re- able remains only in Caves I, II, IX, X, XVI, 

corded by the late Dr. Bloch in Ann. Rep. A. S., and XVII (Burlington Magazine, vol. xvii, June 

Bengal Circle, 1903-4, pp. 12-14; and Ann. Rep. 1910, pp. 136-8, with two Plates). See also her 

A. S., India, 1903-4, p. 130. remarks in Catalogue and Guide to the Indian 

% The amount remaining is now much reduced. Court, Festival of Empire, published in July 

In 1909-10 Mrs. Herringham found consider- 191 1. 


the early Hinayana form of Buddhism, and may be considered to cover a 
period of about three and a half centuries from 200 B.C. to a.d. 150. All the 
others were dedicated to the Mahayana forms of worship. Nos. VI and VII 
may be assigned to the century between a.d. 450 and 550. The rest, namely 
Nos. XIV to XX, XXI-XXIX, and I-V seem to have been excavated between 
c. a.d. 500, several having been left incomplete. No. I was held by Fergusson 
to be the latest of the completed works. 

The paintings are not necessarily of exactly the same age as the caves which 
they adorn. The most ancient unquestionably are certain works in Caves IX 
and X, partially overlaid by later pictures. These earliest paintings are so 
closely related to the Sanchi sculptures that they may be referred to approxi- 
mately the same age, about the beginning of the Christian era, or earlier. They 
may, perhaps, be credited to the patronage of the powerful Andhra kings of 
the Deccan, who, even if not themselves Buddhists, certainly put no obstacle 
in the way of Buddhist worship. So far as appears, no paintings were executed 
for centuries afterwards. 1 

The bulk of the paintings unquestionably must be assigned to the time of 
the great Chalukya kings (a.d. 550-642) and of the earlier Vakataka kings of 
Berar. A Vakataka inscription exists in Cave XVI. It is unlikely that any 
can have been executed later than the second date named, when Pulakesin II 
was dethroned and presumably killed by the Pallava king of the South. The 
resulting political conditions must have been unfavourable for the execution 
of costly works of art dedicated to the service of Buddhism, the Pallava kings 
having been, as a rule, ardent worshippers of Siva. The related paintings at 
Bagh in Malwa may be dated at some time in the sixth century, or the first 
half of the seventh. A close relation exists between the frescoes and certain 
sculptural additions at Ajanta such as the Temptation scene in Cave 26 as 
well as with the earliest work at Ellora which is also Buddhist. 

The Ajanta paintings first became known to Europeans in 1819, but failed Recent 
to attract much attention until 1843, when Mr. James Fergusson, the historian h^tpry 
of architecture, published a description of them and persuaded the Directors p a i nt ings. 
of the East India Company to sanction the preparation of copies at the public 
expense. In pursuance of the orders of the Court, Major Gill, a competent 
and conscientious artist, was deputed some years later, and continued at 
work until the outbreak of the Mutiny in 1857. 2 The copies then executed, 
thirty or more in number, were sent home from time to time, and with the 
exception of five, the last executed, perished in 1866 in a fire at the Crystal 

1 The Buddhas on the pillars of Cave X have scribed by Burgess in 1879 have disappeared, 

been considered to antedate the bulk of the 2 His portrait appears in Plate 34 of Fergusson's 

paintings, but they show signs of poor crafts- scarce octavo work entitled The Rock-cut Tem- 

manship rather than of antiquity. According to pies of India, illustrated by 74 photographs taken 

Mrs. Herringham, these are the only paintings on the spot by Major Gill (Murray, 1864); photo- 

now left in Cave X. The wall-paintings de- graph No. 616 in India Office List of Negatives. 


Palace, where they were exhibited. Nothing remains of the lost copies except 
a few small-scale outline engravings in Mrs. Speir's Ancient India (1856), and 
reproductions of them in Ancient and Medieval India (1869) by the same lady 
under the name of Manning, and also in the Notes on the Bauddha Rock- 
Temples of Ajanta (1879) by Dr. Burgess. 

Since then fresh copies have been prepared between 1872 and 1885 by Mr. 
Griffiths of the Bombay School of Art, and his pupils, which have been pub- 
lished in two magnificent atlas folio volumes entitled The Paintings of the 
Buddhist Cave Temples of Ajanta, Khandesh, India (1896). The India Office 
also possesses a fine volume of photographs arranged by Dr. Burgess. 

The Crystal Palace fire did not exhaust the ill-luck of these famous paint- 
ings. A subsequent fire at the South Kensington Museum destroyed or 
damaged many of Mr. Griffiths's copies, as shown in detail in the Appendix 
to volume ii of his work. The copies, more than a hundred in number, which 
escaped the fires are exhibited in the Indian Section of the Victoria and Albert 
Museum at South Kensington, but many of them have been damaged. 
Injuries Publicity has been fatal to the originals, and the Government of H.H. the 
Sl lf cr f d Nizam, in whose territories the caves are situated, for some time showed little 
paintings! concern for their preservation. Indeed, thirty years ago one of his sub- 
ordinate officials wrought much damage by cutting out heads to present to 
visitors; and, shameful to say, Dr. Bird, a Bombay archaeologist, was guilty 
of the same crime with the intention of benefiting the Museum at Bombay. 
Of course, all the fragments of plaster thus abstracted crumbled to dust and 
were lost irretrievably. Much injury also has been done by smoke from the 
fires of Hindu ascetics camping in the caves, by the folly of irresponsible 
scribblers of various nationalities, and by the unchecked action of bats, birds, 
and nest-building insects. In 1903-4 wire screens were fixed up in all the 
more important caves, and a good deal of cleaning was done. In 1908 the 
Department submitted a scheme for further conservation to the Government 
of the Nizam. Since then the caves have been amply protected and a curator 
appointed. Exact copies have also been made of the frescoes by means of trac- 
ing and photography and the frescoes themselves have been finely preserved. 

The long-continued neglect of these precious remains offers a painful con- 
trast to the vigorous and effective action taken by the Government of Ceylon 
to preserve the fifth-century paintings at Sigiriya which will be described in 
due course. At Ajanta the result of neglect and wilful injury is that the exist- 
ing paintings are only a small fraction of those visible in 181 9, when the 
caves were first brought to notice. Nevertheless, in spite of all mischances, 
enough either remains or has been recorded to indicate the course of Indian 
pictorial art for some six centuries or more. 

Many of the paintings referred to in this chapter, which existed in 1879, 
when Dr. Burgess wrote, have since disappeared. 

PLATE 53, The Goddess Yamuna. Lankesvara shrine, Ellora 



p *r* x AJANTA AND BAGH 97 

The Ajanta pictures may be correctly termed frescoes, although the process The 
used is not exactly the same as any practised in Europe. paintings 

'The Indian practice of wall-painting at Ajanta, as elsewhere\ Mr. Griffiths observes, % e J" rescoes 
is in fact a combination of tempera with fresco. The hydraulic nature of Indian lime, JJ^ kind 
or chunam, makes it possible to keep a surface moist for a longer time than in Europe, 
and the Indian practice of trowelling the work — unknown in Europe — produced a 
closer and more intimate liaison between the colour and the lime, and a more durable 
and damp-resisting face than the open texture of European fresco. The art has been 
practised all over India since the time of the Ajanta frescoes, and to this day houses, 
mosques, and temples are thus decorated. The modern method is first to spread a 
ground of coarse mortar (chunam) of the thickness of from half to one inch on the wall. 
This is allowed to stand for a day. If on the next day the ground is too dry, it is 
moistened, and then tapped all over with the edge of a small piece of wood of triangular 
section, to roughen it and give it a tooth. Then, with a coarse brush a thin coating of 
fine white plaster (chunam) is applied, and the work is allowed to stand till the next day, 
being moistened all the time. If the painting is to be highly finished, the ground is 
carefully smoothed with a small flat iron trowel about the size of a dessert spoon, which 
produces a surface on which the design is first sketched, or transferred by pouncing 
from a perforated drawing on paper, and then painted. 

The outline is usually put in first in brown or black; local colour is filled in with 
flat washes, on which the details aie painted. 

The colours are ground with rice or linseed-water with a little coarse molasses (gur)> 
and water only is used in painting. Then, when the painting is completed, it is again 
rubbed over with the same small trowel. ... It is considered absolutely necessary that 
the work should be kept damp from beginning to finish, so that the plaster is not 
allowed to set until the completion of the picture. When once the smoothly trowelled 
surface is dry, it bears a distinct sheen or gloss and the colours withstand washing. 

Between the methods of modern India and that employed at Ajanta, the only differ- The process 
ence is that instead of a first coat of mortar, a mixture of clay, cow-dung, and pulverized used at 
trap rock was first applied to the walls and thoroughly pressed into its [sic] surface, Ajanta. 
when the small cavities and air-holes peculiar to volcanic rock and the rough chisel 
marks left by the excavators served as keys. In some instances, especially in the ceilings, 
rice husks were used. 

This first layer — which, according to our modern notions, promises no great per- 
manence — was laid to a thickness varying from one-eighth to three-quarters of an inch, 
and on it an egg-shell coat of fine white plaster was spread. This skin of plaster, in 
fact, overlaid everything — mouldings, columns, carven ornaments, and figure sculp- 
tures, but, in the case of carved details, without the intervention of the coat of earthen 
rough-cast; and, from what remains, it is clear that the whole of each cave was thus 
plaster-coated and painted. The texture of the volcanic rock, which is at once hard, 
open, impervious to damp, and yet full of air-holes, is especially suitable for this treat- 
ment. Great pains were taken with the statues of Buddha; one in the small chamber 
to the right of the first floor of Cave VI is covered with a layer of the finest plaster one- 
eighth of an inch thick, so painted and polished that the face has the smoothness and 
sheen of porcelain. 1 

1 In Cave IX the early picture H which Mr. of finest plaster, ^ inch thick, applied directly 
Griffiths exposed and copied, after removing a to the rock, and polished like porcelain, 
later damaged painting, was executed on a coat 

3292 O 


It will be seen that a parallel to the technique of the Ajanta paintings is scarcely 
to be found in the Italian frescoes. But it is evident from specimens of the Egyptian 
work in the British Museum that loam or clay mixed with chopped straw formed the 
substratum over which, as at Ajanta, a layer of fine plaster was laid to receive the final 

It may not be impertinent again to point out the exceeding simplicity of the Indian 
and Egyptian methods, which have ensured a durability denied to more recent attempts 
executed with all the aids of modern chemical science.' l 

The foregoing description of the technique of the Ajanta paintings, based 
upon Mr. Griffiths 's patient study for thirteen years on the spot, may be 
accepted with confidence as authoritative, although Mr. Havell may be right 
in adding that the pictures were sometimes touched up in tempera after the 
surface had dried. Italian workers in true fresco (fresco buono) often permit 
themselves the same liberty. 2 
Observations But it will be well to supplement Mr, Griffiths^ account by the recent 

of Mrs. observations of Mrs. Herringham, also an expert artist, who writes: 
Hernngham. m . . . 

'The technique adopted, with perhaps some few exceptions, is a bold red line- 
drawing on the white plaster. Sometimes nothing else is left. This drawing gives all 
the essentials with force or delicacy as may be required, and with knowledge and 
intention. Next comes a thinnish terra-verde monochrome showing some of the red 
through it; then the local colour; then a strengthening of the outlines with blacks and 
browns giving great decision, but also a certain flatness ; last, a little shading if necessary. 
There is not much definite light and shade modelling, but there is great definition given 
by the use of contrasting local colour and of emphatic blacks and whites.' 

Mr. Griffiths, it will be observed, does not mention the first outline in red. 
Pigments. The nature of fresco-painting in any of its forms implies the use of a limited 
range of pigments capable of resisting the decomposing action of lime, and 
consequently composed of natural earths. At Ajanta and Bagh the colours 
most freely used are white, red, and brown in various shades, a dull green, 
and blue. The white is opaque, mainly composed of sulphate of lime; the 
reds and browns derive their tints solely from compounds of iron; the green 
is a silicate, similar to the mineral now known as terre verte; and the blue is 
ultramarine, which was obtained in ancient times by grinding calcined lapis- 
lazuli, a costly semi-precious mineral usually imported from either Persia or 
Badakshan. AH the other pigments are to be found locally. The long panels 
of the ceilings in Cave II, dating from about A.D. 600, offer well-preserved 
examples of charming floral decorations in blue (Griffiths, Plates 123-5). I* 1 

1 Griffiths, op. cit., p. 18. good style, is true fresco-painting, the buono 

2 For a good summary account of the European fresco of the Italians, and, like the inlaid ceramic 
processes see the article 'Fresco' in Chambers's work, is now no longer practised, modern native 
Encyclopaedia (1905). Mr. J. L. Kipling, a com- decoration being usually fresco secco, or mere 
petent authority, states that the fresco-painting distemper painting' (Lahore Guide, 1876; quoted 
on the walls of the mosque of Wazir Khan at in Birdwood, Industrial Arts of India (1880), 
Lahore, 'which is very freely painted and in p. 228). 


fV : ^~p*-. Jfo'' 




» E 

A, IJ, C, I), and K. Small panels from ceiling of Cave I, A junta 

F. Figures in spandril of central ceiling panel, Cave I, Ajanta 


B. Buddha, Cave X, Ajanta 

' ;M $? 

A. Seated woman, Cave IX, 

C. Female figure, Cave II, 



the early paintings of the Ramgarh Hill, Orissa (ante, p. 273), and the fifth- 
century works at Sigiriya in Ceylon {post, Sec. 5 of this chapter), blue never 
occurs. At Ajanta, yellow, so largely used at Anuradhapura in Ceylon, appar- 
ently is very rare. The yellow of ancient painters is believed to have been 
always orpiment, a natural arsenic sulphide. 

The subjects of the pictures, as distinguished from the purely decorative Subjects of 
devices, are almost exclusively Buddhist. They include, of course, numerous pictures. 
figures of Buddha and representations of sacred objects and symbols. The 
more complex compositions for the most part deal with either the incidents 
of the life of Gautama Buddha or those related in the Jataka stories, which 
narrates the events of his former births. In at least two cases the Jataka story 
is indicated beyond dispute by a painted label, but the fragmentary condition 
of the pictures renders difficult the identification of most of the scenes. 1 
There is, however, no difficulty in recognizing in Cave X the tale of the six- 
tusked elephant, and a few other legends may be identified with more or less 
certainty. 2 Miscellaneous edifying Buddhist subjects, not taken from the 
Jataka collection, include the Litany of Avalokitesvara and consecutive scenes 
from the life of the Buddha; the Wheel of Life, formerly miscalled the 
Zodiac. 3 

The high achievement of the Ajanta artists in decorative design executed Decorative 
with masterly skill is most freely exhibited in the ceiling panels of Cave I, desi s ns - 
painted in the first half of the seventh century (Plate 55). Mr. Griffiths, 'who 
took so much pleasure in copying the designs, describes their variety as 
infinite, carried into the smallest details, so that repetition is very rare; fancy 
is given full play, and the simplest objects of nature, being pressed into the 
artist's service, are converted into pleasing and effective ornament. 

'The smaller panels*, he observes, 'are ornamented with designs as varied and grace- 
ful as they are fanciful. Some with grotesque little figures, rich in humour and quaintly 
dressed in Persian turbans, coats, and striped stockings ; gambolling amid fruits and 
flowers; dancing, drinking, or playing upon instruments; or chattering together; some 
with animals combined with the lotus, drawn with remarkable fidelity and action : as 
the elephant, humped bull, and the monkey; parrots, geese, and conventional birds 
singly and in pairs, with foliated crests, and tails convoluted like heraldic lambrequins, 

1 Close upon twenty Jatakas may be identified. Ajanta', Nachr. d konigl Gesellschaft d. Wissen- 
Guide Ajanta Frescoes. Arch. Dep. Hyderabad schaften zu Gottingen, Phil.-Hist. Klasse, 1902, 
State. P- 758). The story of the six-tusked elephant is 

2 In Cave XVII the story of Sibi Raja, who gave No. 514 (Cowell and Francis, vol. v, p. 20). 
his eyes to the beggar (No. 499; Cowell and 3 Fergusson identified one of the scenes in 
Rouse, transl., vol. iv, p. 250) is labelled. In Cave I portraying foreigners in pointed caps as 
Cave II the Kshantivadin and Maitribala;^fl*a the reception of an embassy from Khusru 
pictures are accompanied by quotations from Parviz, King of Persia, to Pulakesin II, c. a.d. 
the Jataka Mala of Arya Sura, inscribed in 626: the representation of such a secular scene 
characters of about the sixth century, the former is contrary to the general trend of Ajanta paint- 
being also labelled by name (Heinrich Liiders, ing and Indian art as a whole. 

'Arya Sura's Jataka-mala und die Fresken von 


showing the upper and under surface of the ornament. Some contain the large pink 
lotus, full-bloom, half-bloom, and in bud, as well as the smaller red and white; some 
with the mango (Mangifera indicd), custard apple (Anona squamosa) ; a round fruit which 
may be the bel (Aegle marmebs) or the lime (limbu) ; another that looks like the brinjal 
or aubergine (Solatium melongena), and many others. 

The ornament in these panels is painted alternately on a black and red ground. The 
ground colour was first laid all over the panel, and then the ornament painted solidly 
upon this in white. It was further developed by thin transparent colours over the 
white.' J 

Examples The reader who desires to realize fully the justice of Mr. Griffiths's 

of small panegyric must study his numerous plates, or the full-sized copies in the 

pane s. j nc jj an s ec ti n of the Victoria and Albert Museum at South Kensington; or, 

best of all, Mrs. Herringham's recent copies. Here it is not possible to give 

more than a few specimens. 

Panels in Cave II presents some very good work. The circular panels (Griffiths, 

Cave II. pj ates u^ 1 17-19 coloured and 120, 121 uncoloured) are very fine, the 

figures in the spandrils being particularly good and full of movement. These 

circular panels have a distant resemblance to the carved moonstones of 

Ceylon. The long ceiling panels (Griffiths, Plates 122-31 coloured and 132 

uncoloured) are admirable. 

Picture of The decorative designs in Cave I include a minor picture of considerable 

^n 8 * nterest P a * nte d on a bracket capital (Griffiths, Plate 114). The subject is 

that of two bulls fighting, and its treatment proves the artist's knowledge of 

animal form and his power of expressing vigorous action. The same subject, 

with variations of detail, is treated in a sculpture at the ancient cave of Bhaja, 

dating from about the beginning of the Christian era or earlier, and again in 

a sixteenth-century painting at Akbar's capital, Fatehpur Sikri. It occurs also 

in a well-known sculpture in the Louvre, brought from the Doric temple at 

Assos in the Troad, and dating from about 500 B.C. 2 

Cave XVII. In the sixth-century Cave XVII, the charming floral designs combined 

with human figures on the panels of the pillars (Griffiths, Plates 144-9) are 

closely related to the slightly earlier sculptured work on the Garhwa pillars 

in Northern India (ante, p. 166). The kirttimukha grinning faces in Plate 

146 are common throughout medieval Indian art. As chaste decoration it 

would be difficult to surpass the frets in Griffiths, Plates 143 and 149. 

Spandril The pair of lovers in a spandril of the central panel of the ceiling of Cave I 

pl Cave T * s admirably drawn, and although forming only a subordinate member of 

' a decorative design, is worthy of reproduction as a cabinet picture (Plate 55 f). 

Wall- w e now proceed to describe, so far as space permits, characteristic ex- 

1 Griffiths, op. cit., pp. 41, 42. The Fathpur Sikri frescoes are reproduced by 

2 For Bhaja see Fergusson and Burgess, The Cave E. W. Smith, Fathpur-Sikri, vol. i, PI. XI-XIII , 
Temples of India (1880), p. 536; for Assos, Texier XV a, 6, c; CIX-CXX, and also partly in the 
and Pullan, Principal Buildings in Asia Minor, J. L A. I. (July 1894), and Griffiths, op. cit. 
PI. I ; or Texier, Asie Mineure, vol. ii, pp. 112-14. 


amples of the larger pictures on the walls of the caves in chronological order. 
But the pictures being too large to admit of intelligible reproduction as com- 
plete compositions, except on a scale far beyond the dimensions of this book, 
the illustrations will be confined ta extracts from the paintings, which are 
generally overcrowded and lacking in the unity derived from skilled com- 

The earliest works, as already stated, are certain paintings in Caves IX and Early 
X, closely related to the Sanchi sculptures. The seated woman is a pleasing P ictur *s in 
example from the painting H in Cave IX (Plate 56 a), which Mr. Griffiths Cave 1A ' 

Outline of elephant in Cave X, Ajanta. 

exposed by removing a later and damaged picture. The old composition was 
painted on a thin porcelain-like skin of fine plaster applied direct to the rock. 

In Cave X the remains of early paintings are, or were thirty years ago, more Early 
extensive. The fragments on the right-hand wall then consisted chiefly of ^ tur J. in 
elephants drawn in outline 'in a strikingly bold and true style'. elephants. 

On the left 'was a procession of men, some on foot, some on horseback, 
variously armed, some with halberts, and differently dressed; and behind 
were groups of women; but all have been defaced by native visitors within the 
last twenty years or less', that is to say, prior to 1879. Numerous heads and 
figures in these scenes, admirably drawn and full of spirit and character, are 
reproduced in Plates VIII-X of Dr. Burgess's Notes, from drawings pre- 
served at the India Office, made by a Hindu student of the School of Art, 
Jayrao Raghoba. The group shown in his Plate X, a Raja in the midst of 
eight female attendants (see illustration on p. 103), is unusually well com- 
posed. The perspective of the numerous figures is satisfactory, and the 
drawing of the hands and arms is particularly good. 


Figures of I am disposed to think that the figures of Buddha painted on the pillars of 

Buddha. £ ave ^ (Griffiths, Plates 42, 43, and cover) are the next in date, and should 

be assigned to the fifth century, but they might be later. The nimbus and 

draperies recall early Christian art and the sculptures of Gandhara. The best 

is shown in Plate 56 B. These are now (1910) the only paintings left in Cave X. 

Cave XVI. The whole interior of Cave XVI was once covered with paintings of high 

merit, but even thirty years ago many of them had been destroyed. The 

plates in Mr. Griffiths's work include little from this cave, although his 

copies, except three burnt, are preserved at South Kensington. 

The 'Dying The scene known as the 'Dying Princess', reproduced by Mr. Griffiths in 

Princess . jgy^ was deservedly praised by him in glowing language, endorsed by Dr. 

Burgess and Mr. Fergusson, which merits quotation: 

' A lady of rank sits on a couch leaning her left arm on the pillow, and an attendant 
behind holds her up. A girl in the background places her hand on her breast and looks 
towards the lady. Another with a sash across her breast wields the pankha [fan], and 
an old man in a white cap looks in at the door, while another sits beside a pillar. In 
the foreground sit two women. In another apartment are two figures; one with a 
Persian cap has a water-vessel (kalasa) and a cup in the mouth of it ; the other, with 
negro-like hair, wants something from him. To the right two kanchukinis [female 
servants] sit in a separate compartment. . . . For pathos and sentiment and the unmis- 
takable way of telling its story this picture, I consider, cannot be surpassed in the history 
of art. The Florentine could have put better drawing, and the Venetian better colour, 
but neither could have thrown greater expression into it. The dying woman, with 
drooping head, half-closed eyes, and languid limbs, reclines on a bed, the like of which 
may be found in any native house of the present day. She is tenderly supported by 
a female attendant ; whilst another with eager gaze is looking into her face, and holding 
the sick woman's arm as if in the act of feeling her pulse. The expression on her face 
is one of deep anxiety as she seems to realize how soon life will be extinct in the one 
she loves. Another female behind is in attendance with a pankha, whilst two men on 
the left are looking on with the expression of profound grief depicted in their faces. 
Below are seated on the floor other relations, who appear to have given up all hope and 
to have begun their days of mourning, for one woman has buried her face in her hand 
and apparently is weeping bitterly/ l 

Persian Other figures wearing the Persian cap appear in a second painting (No. 6 
figures. f Burgess) in the same cave, and may be compared with the representation 
of the so-called Persian embassy and connected minor pictures in Cave I. 
Cave XVII. Cave XVII, which is little later in date than Cave XVI, and thirty years 
ago, whatever may be the case now, could show more painting than any of 
the others, may fairly be considered the most interesting of the series. 2 No 
less than sixty-one distinct scenes are described in Dr. Burgess's Notes. The 

1 Ind. Ant., vol. iii, pp. 25 seqq. y with uncoloured 2 Mrs. Herringham notes that 'in Cave XVI, 

plate. The text is quoted in Burgess, Notes, p.58. slightly the earlier, nearly everything is obscured, 

He numbers the painting as 5. The picture is but in Cave XVII many interesting subjects still 

not included in Mr. Griffiths's special work. remain intelligible'. 









two large pictures, reproduced in outline in his Plates XVIII and XIX, are 
so excessively crowded with figures and so deficient in unity of composition 
that they cannot be presented satisfactorily except on an enormous scale. 
The 'Wheel The representation in the left end of the verandah of the Buddhist Wheel 
of Life', &c. £ Lj£ e ^ commonly miscalled the Zodiac, is interesting rather as an illustra- 
tion of popular Buddhist teaching in the sixth century than as a work of art. 
Similar pictures are still frequently exhibited in Tibetan monasteries and 
used by the Lamas for purposes of instruction. The dimensions of the Ajanta 
painting, now a mere fragment, are 8 feet 7 inches by 5 feet 1 inch. 1 The 
huge painting indicated in Burgess's Plate XIX was supposed to represent 
the legend of the landing of King Vijaya in Ceylon and his coronation as 
described in the Pali chronicles, but is actually a faithful rendering of the 
Simhala Avadana. Painting No. LIV (Griffiths, Plate 82) gives the story of 
Sibi Raja, already mentioned. 
Cave XIX. Among the later caves the Chatty a or church, Cave XIX, which is elabo- 
rately carved throughout and has its porch and whole front covered with 
beautiful sculpture, was considered by Mr. Fergusson to be 'one of the most 
perfect specimens of Buddhist art in India*. The paintings include many 
effigies of Buddha (Griffiths, Plate 89), and some exquisite panels on the 
roof of the front aisle, as well as rich floriated patterns on the roofs of the side 
Cave II. We now pass to Caves I and II, No. I being probably the latest of the 
completed works. 

Mr. Griffiths has devoted a large number of plates (Nos. 20-35 and 1 15-32) 
to Cave II, besides nine text illustrations. The individual figures are remark- 
able for clever drawing, the artist having apparently gone out of his way to 
invent specially difficult poses. Mr. Griffiths's figure 8, a woman prostrating 
herself, and figure 16, snake-hooded Nagas, or water-sprites, are good ex- 
amples of such tours deforce. The woman standing, with her left leg bent up 
(Plate 56 c), is capital, the feet being as well drawn as the hands; and the 
woman in the swing (Fig. 66) is pleasing and life-like. Fig. 5 of Griffiths 
is reproduced (Plate 56 c). 
Cave I. The elegant decorative designs of Cave I have already been described. The 
numerous large wall-pictures include the Temptation of Buddha, a subject 
also effectively treated in sculpture in Cave XXVI, not far removed in date. 
In this cave is also the so-called Persian embassy scene. The identification 
is based (a) upon the pointed caps which are considered to be Persian; (b) 
upon the statement of an Arab historian that an embassy was sent by 
Pulakesin II to the Persian court in A.D. 626. 

1 The picture and its Tibetan counterparts are J. A. S. B. $ vol. lxi (1892), Part I, pp. 133-55, 
discussed fully by Col. Waddell in The Bud- with plates. His PI. VII corresponds with 
dhist Pictorial Wheel of Life (Zodiac) ' in Griffiths, PI. 56. 


Four smaller pictures placed symmetrically at the corners of the central Khusru 
square of the principal design of the roof, and all replicas of one subject, with Parviz - 
variations, evidently have some connexion with the other 'Persian* pictures, 
which measures 15 by 6 J feet. The best of these small compositions has been 
illustrated by Mr. Griffiths both from a photograph (Plate 95, Fig. 4) and 
from a water-colour drawing (Plate 94, Fig. 4). The colours of the latter 
seem to be too brilliant, and a more faithful reproduction by Mr. Griggs was 
published by Fergusson, which is here reproduced uncoloured by permission 
of the Council of the Royal Asiatic Society (Plate 57 b). Fergusson, developing 
the Persian myth, assumed that the principal personages depicted must be 
King Khusru and his famous consort, Shirin, but this attractive hypothesis 
cannot be said to be proved. 1 

The foregoing descriptions and illustrations will enable the reader to form Aesthetic 
a judgement concerning the aesthetic value of the Ajanta paintings, and I value - 
trust that nobody will be found to agree with the opinion expressed in Sir 
George Watt's book that they 'can hardly be classed among the fine arts 1 . 2 
The pictures and decorative designs in the caves, when compared with Egyp- 
tian, Chinese, or other ancient paintings, which did not profess to show the 
relief effect of modern pictures, are fairly entitled to high rank as works of 
fine art. In judging them the critic should remember that the wall-paintings 
were executed on an enormous scale, some being more than 20 feet in 
diameter, and that they were intended to be looked at in the mass from a 
distance, and not in minute detail. Small reproductions on a page a few 
inches long cannot possibly give a just idea of the effects aimed at by the 
artists. Moreover, those artists were much concerned to tell sacred stories, 
and make their pictures serve for the edification of devout worshippers as 
instructive illustrations of the Buddhist Bible; whereas all the religious senti- 
ment in the spectator on which they relied for sympathetic understanding is 
wanting in the modern European critic. Yet, in spite of the disadvantages 
inherent in small-scale reproductions and criticism by judges out of touch 
with the spirit of the artists, the paintings stand the unfair test wonderfully 
well, and excite respectful admiration as the production of painters capable 

1 J. Fergusson, *On the Identification of the por- by the frescos on the walls of the caves of 
trait of Chosroes II among the Paintings in the Ajanta. . . . The first mentioned is more decora- 
Caves at Ajanta' (J. R. A. S., April, 1879); tive than pictorial, so that it can hardly be classed 
Rajendralala Mitra, 'On Representations of among the Fine Arts, and is therefore omitted 
Foreigners in the Ajanta Frescos' {J. A. S. B. 9 from a description of what is intended to be an 
vol. xlvii (1878), Part I, pp. 66-72, and four un- account of painting in the pictorial sense only, 
coloured plates). His PI. IV corresponds with The earliest true pictures, therefore, of which 
Fergusson 's plate and the small outline copy in we have any record are the productions of the 
Burgess, Notes, PI. IV, Fig. 2. old Moghul painters' (Sir George Watt, Indian 
3 It is only fair to quote this dictum in fall:— Art at Delhi (1904), p. 454). The opinions re- 
' Painting.— This may be said to be divided into corded in the book are partly those of Mr. Percy 
three distinct styles. The Buddhist, exemplified Brown. 

3292 p 


of deep emotion, full of sympathy with the nature of men, women, children, 
animals, and plants, and endowed with masterly powers of execution. 
Mr.Griffiths's The considered verdict of Mr. Griffiths, the artist who spent thirteen years 
verdict. j n ^ c \ osc ^ loving study of the paintings, may be accepted as a sound general 
criticism, not attempting to distinguish periods and styles: — 

'In spite*, he writes, 'of its obvious limitations, I find the work so accomplished in 
execution, so consistent in convention, so vivacious and varied in design, and full of 
such evident delight in beautiful form and colour, that I cannot help ranking it with 
some of the early art which the world has agreed to praise in Italy. . . . The Ajanta 
workmanship is admirable ; long subtle curves are drawn with great precision in a line 
of unvarying thickness with one sweep of the brush ; the touch is often bold and vigor- 
ous, the handling broad, and in some cases the impasto is as solid as in the best Pompeian 
work. . . . The draperies, too, are thoroughly understood, and though the folds may 
be somewhat conventionally drawn, they express most thoroughly the peculiarities of 
the Oriental treatment of unsewn cloth. . . . For the purposes of art-education no better 
examples could be placed before an Indian art-student than those to be found in the 
caves of Ajanta. Here we have art with life in it, human faces full of expression, limbs 
drawn with grace and action, flowers which bloom, birds which soar, and beasts that 
spring, or fight, or patiently carry burdens; all are taken from Nature's book — growing 
after her pattern, and in this respect differing entirely from Muhammadan art, which 
is unreal, unnatural, and therefore incapable of development/ l 

Comparison Whatever be the value of the incidental criticism on Muhammadan art — 
TaT ^rt 7 a su kJ ect t0 k e discussed in due course — Mr. Griffiths's hearty appreciation 
ian " of the Ajanta frescoes is, in my judgement, just and well deserved. 

In support of his comparison with the performance of the early Italians, he 
aptly cites the fragment of a fresco with heads of nuns by Ambrogio Loren- 
zetti, executed in the fourteenth century, and now in the Sienese Room of 
the National Gallery, as being 'singularly like the Ajanta work in colour, 
execution, and treatment; the forms being drawn with a delicate brown out- 
line, and the flesh-tints and drapery flatly put in with very little modelling*. 
The obvious comparison with ancient Italian art was also made by Mr. 
Fergusson, who considered the Ajanta paintings to be better than anything 
in Europe before the time of Orcagna in the fourteenth, or even Fiesole (Fra 
Angelico) in the fifteenth century. Similarly, Mr. Havell, another trained 
artist, who selects the charming Mother and Child in Cave XVII (Griffiths, 
Fig. 76) as the most attractive specimen of Ajanta art, finds in the frescoes 
'the same intense love of nature and spiritual devotion as are evident in the 
sculptures of Borobodur', and compares the 'exquisite sentiment' of the 
picture selected with the wonderful Madonnas of Giovanni Bellini'. 2 

1 Griffiths, The Paintings of the Buddhist Caves used for the decoration of pottery made at the 

of Ajanta, pp. 7, 9; Ind. Ant., iii. 28. The work Bombay School of Art. Examples are shown in 

done by the Bombay students shows that they the Indian Section of the Victoria and Albert 

were capable of appreciating the ancient models Museum, 

set before them. Many of the designs have been z Indian Sculpture and Painting (1908), p. 164. 


Mr. Fergusson was of opinion that while the art of Ajanta resembled that Comparison 
of China in flatness and want of shadow, 1 he had never seen Anything in with Chinese 
China approaching its perfection 1 . Forty years ago so little was known in art ' 
England about Chinese art that this sentiment might pass muster, but Fer- 
gusson's dictum could not now be accepted in the light of fuller knowledge. 
It is interesting to set against it the deliberate judgement of Mr. Laurence 
Binyon, a learned connoisseur in the art of the Far East. 

'The art of Ajanta\ he observes, 'is characterized by the strong outline which marks 
the early Asiatic style; the colouring appears to have been heavy and hot; the figures 
and faces are animated— there is force and individuality in them, a strong sense of life. 
We feel that the painters were possessed by their subject ; they worked with fervour and 
devotion This, and the scale of the frescoes, make a forcible and imposing impres- 
sion. Yet the art of Ajanta has not passed the primitive stage. With all the feeling for 
life in individual figures that the painters show, they betray as yet little of that instinct 
by which an art develops — the instinct towards unity, towards the conception of a 
subject as a synthetic whole. Their compositions are crowded and incoherent. In 
details and in single groups and forms, on the other hand, there is grace, dignity and 
character. . . . What is lacking in the Ajanta paintings, what is so signally manifest in 
Chinese painting throughout its history, is that powerful creative instinct and aesthetic 

Eerception which make for synthetic unity in art, that sense of controlling rhythm and 
alance which inspires all fine design/ 2 

The expert criticisms above quoted all agree in being general in their terms. Various 
Lady Herringham, in the too brief article already cited more than once, carries periods and 
the aesthetic valuation of the paintings farther by distinguishing various sty 
periods and styles. She holds that the frescoes 'fall into about six distinct 
groups, representing various schools and periods rather than the steady 
development of one school'. Going a little into detail, the critic proceeds: 

( I have already alluded to several styles and classes of painting in Caves 1 and 9, 
16 and 17. There are, besides, later developments of the narrative style of Cave 17, 
which we find in Caves 1 and 2. These are (1) a more emphatic and stylistic manner, 
with more formalism in the drawing, more action and less tenderness; (2) a more 
popular, lively, and forcible dramatic narrative, with more incidents and less idealism. 

In Cave 2 are three more distinct styles : on both the side walls of a secondary shrine 
we find four or five elaborately posed, nearly nude life-size figures. These are sinuous 
in outline, quite Cimabuesque in proportion, attitude and general feeling; the arrange- 
ment suggests bas-relief. The late date of this cave indicates the period of the painting. 
In a similar shrine on the opposite side are corresponding decorations, and the figures 

1 Actually the painting at Ajanta is not at all flat Horiuji, which was repaired or built between 
but renders the contours delicately and faithfully. a.d. 708 and 715, is quite Indian in character, 

2 Laurence Binyon, Painting in the Far East and 'there seems no doubt that it is modelled 
(1908), pp. 35, 50. See also the same author's upon the Ajanta frescoes' (Painting in the Far 
article, 'A Chinese Painting of the Fourth Cen- East, p. 87). Anderson gives the date as 607, 
tury' in Burlington Magazine, Jan. 1904, p. 44. but other critics date it a century later. 

One Japanese work, the fresco in the temple of 


on the main west wall might, but for the type, be an assemblage of Chinese sages ; they 
are drawn with a magnificent bravura. There is not much colour left, but the some- 
what caligraphic drawing in forcible blacks and reddish browns is so freely executed 
that one scarcely regrets the destruction which has laid bare such vital work. On a 
separate part of this west wall there is a subject of men and white geese in a water-lily 
pool, which, though closely linked to the earlier definitely Indian types of painting, 
suggests the freedom and at the same time the perfect balance of the very best Chinese 
period. The colour scheme is very beautiful — brilliant white, deep purple-brown, a 
vivid but rich malachite-green, with touches of a clear red. 
Three Further, in Cave 17 there are three paintings by one hand very different from all 
notable the rest. They are (1) a hunt of lions and black buck ; l (2) a hunt of elephants ; 2 and 
paintings. (3) an elephant salaaming in a king's court— the companion picture to No. 2. These 
pictures are composed in a light and shade scheme which can scarcely be paralleled in 
Italy before the seventeenth century. They are nearly monochrome (warm and cool 
greys understood), except that the foliage and grass are dull green. The whole posing 
and grouping is curiously natural and modern, the drawing easy, light and sketchy, 
and the painting suggestively laid in with solid brush strokes— in the flesh not unlike 
some examples of modern French painting. The animals — horses, elephants, dogs and 
black buck — are extremely well drawn/ 

Pam ^2 S ^ e development of criticism on the lines indicated by Lady Herringham 
^Bagh. w °uld require a bulky monograph based on detailed notes taken on the spot 
by a competent expert. It is impossible to work out the differences of the 
supposed schools merely from the fragmentary published reproductions. 3 

The vigorous school of art which produced the Ajanta frescoes did not 
confine its operations to the caves at that place. Several similar excavations 
near Bagh, a village or decayed small town in the Gwalior State, situated on 
an ancient road connecting Gujarat with Malwa, exhibit traces of a set of 
works resembling in general style the Ajanta paintings, and at one time of 
almost equal importance. Unfortunately, the crumbling of the rock, and 
absolute neglect, combined with the effects of the smoke from vagrants' 
fires, have left hardly anything of compositions which once covered thousands 
of square feet. 

The principal group of caves contains eight excavations, the largest being 
94 feet square. The whole of the roof, walls, and columns of this great cham- 
ber was coated with fine stucco and decorated with paintings of high merit 
and infinite variety. Smaller remnants of painting may be still discerned in 

1 Burgess, Notes, Cave XVII, No. 28. p. 92, Lady Herringham states that 'there are at 

2 Ibid., Nos. 36, 37. I cannot trace the 'com- least twenty different kinds of painting. Some 
panion picture'. Burgess does not notice the pictures recall Greek and Roman composition 
distinctions of style. and proportions, a few late ones resemble the 

3 Lady Herringham has generously presented Chinese manner to a certain extent, but the 
her copies of the frescoes to the India Society, majority belong to a phase of art which one can 
They were exhibited in the Indian Section of call nothing except Indian, for it is found no- 
the Festival of Empire (191 1). In the Catalogue where else.' 

and Guide to the Indian Court, Festival of Empire, 

A. Woman carrying child; Cave XVII, Ajanta 

B. Group of foreigners; from ceiling of Cave I, Ajanta 


















two other caves, and there is reason to believe that the work is not all of one 
period. 1 

The paintings appear to have rivalled those of Ajanta in variety of design, Subjects 
vigorous execution, and decorative quality, life being treated in both places of . th ? 
with equal gaiety and hardly a trace of asceticism. Two of the Bagh groups P amtin 6 8 - 
illustrate the performance of the hallisaka, a kind of operetta or musical play, 
acted by a troupe of women led by a man. According to the books the female 
performers should number seven, eight, or ten. At Bagh they are six in one 
case and seven in the other. They are represented as elaborately dressed, 
singing, and performing with much enjoyment on drums, cymbals, and other 
instruments. Our surprise at finding such gay scenes depicted on the walls 
of a Buddhist monastery may be lessened when we consider the nature of 
many of the sculptures at Mathura and in the Aurangabad caves; but we do 
not know quite enough about the real nature of the later popular Buddhism 
in India to understand fully the significance of such frivolous sculptures and 
paintings. 2 

The Bagh caves do not contain an inscription of any kind, and their date Chronology, 
can be determined only by considerations of style. The hair-dressing of 
many of the male figures and the transparent close-fitting robes connect the 
sculptures with the later Gupta rather than with the medieval period. The 
general character of the paintings is sufficiently known to make it certain that 
they are not earlier than the late works at Ajanta. Probably the paintings 
may have been executed between the middle of the sixth and that of the 
seventh century. The paintings include patterns executed in black and white 
with touches of Indian red, as well as works executed in 'excessively vivid' 
colours, with 'marked contrasts in blue, red, and yellow'. The two styles may 
belong to different ages. 


Having been constrained to comment upon the long-continued neglect of Sigiriya. 
the Ajanta and Bagh paintings, and the failure of the authorities to take the 
simple measures needed to save priceless works from destruction,itisapleasure 
to turn to Ceylon and recognize the well-considered and successful policy of 
the island government with regard to the closely related frescoes at Sigiriya. 3 
1 The existing paintings have since been copied, plans and plates). The Bagh Caves (Ind. Soc.> 
References are: — Dangerfield, Capt., 'Some Ac- London). 

count of the Caves near Baug, called the Panch 2 For the definition of hallisaka see Sylvain Levi, 
Pandoo, with three drawings' (Trans. Lit. Soc. t Thi&tre Indien, App. p. 30. 
Bombay, vol. ii (1820), pp. 194-204). Impey, 3 The name, in all its forms, means 'Lion-hill', 
Dr., 'Description of the Caves of Bagh, in Rath' with reference to the passage connecting the 
(J. Bo. Br. R. A. 5., vol. v, pp. 543-73). Bur- galleries, which was wrought in the shape of a 
gess, J., Notes on the Bauddha Rock-Temples of gigantic lion. The hill stands in the Inamaluwa 
Ajanta (1879), pp. 94, 95. Luard, Major C. E., Korale of the Matale District, Central Province, 
'The Buddhist Caves of Central India' (Ind. about twenty miles almost due west from the 
Ant., vol. xxxix, August 1910, pp. 225-35, with medieval capital, Polonnaruwa. 


The marvellous citadel at that place, perched upon the summit of an 
isolated, tower-shaped hill, 600 feet high, and rising abruptly from the plain, 
was constructed as an impregnable refuge by the parricide king, Kasyapa I, 
who reigned from a.d. 479 to 497. The rock-cut galleries leading to the 
tyrant's aerie having crumbled away in the course of ages, the summit had 
become inaccessible save to occasional adventurous cragsmen. The work of 
excavation, repair, and restoration undertaken in 1895 by the Government 
of Ceylon was carried on systematically under the capable guidance of Mr. 
H. C. P. Bell, Archaeological Commissioner, until its completion some ten 
years later, as recorded in Sessional Paper XX of 1909. The paintings, with 
which alone we are now concerned, have been secured by wire nettings and 
other devices in such a way that 'they can be examined closely, without 
difficulty, and in perfect safety; from one end of the caves to the other they 
are for ever secure from further damage'. The story of the operations, as 
related in Mr. Bell's Reports, terminating with the document cited above, is 
a most interesting record of successful wrestling with formidable engineering 
difficulties, and of the completion of a well-devised plan, without parsimony 
and without extravagance. 
Position The paintings are found in two irregular rock-chambers, usually described 
of the as 'pockets', situated on the western cliff, about fifteen yards above the floor 
pamtmgs. Q £ ^ 80ut | iern en( j f t h e gallery. Six such 'pockets' exist, but the remains 
of painting are confined to four, and those of any importance exist only in 
'pockets A and B' — two rough, natural chambers forming a cave 67^ feet in 
length, divided into two sections by a cramped ledge. Tocket B', 41 \ feet long, 
is comparatively roomy, whereas Tocket A', 26I feet in length, is cramped. 
The paintings comprise twenty-one half or three-quarter-length female 
portraits, besides the hand of another figure. Seventeen of these are in 
Tocket B' and only five in Tocket A'. The figures in the more spacious 
chamber B are mostly above life-size, while those in chamber A, where space 
was limited, are below life-size. 
Copies. In 1889 Mr. A. Murray succeeded with great difficulty in obtaining copies 
of thirteen figures in either pastel or coloured photographs, now preserved in 
the Colombo Museum. His meritorious work, performed when the 'pockets' 
were all but inaccessible, has been superseded by a magnificent series of 
facsimile copies made in oils on canvas by Mr. Perera, which also are ex- 
hibited at Colombo. These copies, which are described as reproducing with 
minute accuracy every detail of the originals in size, colour, and all other 
respects, have been carefully photographed. Some of the photographs have 
been reproduced in Mr. Bell's Reports and Mr. Havell's book, and a selection 
is now given from copies liberally supplied by the Government of Ceylon. 
Technique. The paintings were executed on a carefully prepared surface formed by the 
application of fine lime-plaster from a quarter to half an inch thick laid on 

Part II CEYLON in 

a bed about half an inch in thickness, composed of tempered clay mixed with 
kaolin, and strengthened by the admixture of rice-husks, with, perhaps, some 
coco-nut fibre. Mr. Bell believes that the pictures were wrought in tempera 
on a dry surface. The process, possibly, did not differ much from that used 
at Ajanta. Except that Fig. 14 in 'pocket B' has a black background, the 
range of colours is confined to three — red, yellow, and green. The blues, so 
conspicuous at Ajanta and Bagh, are absent. 

The subject is a procession of noble ladies carrying flowers, and attended by Subject. 
female servants, all moving in the direction of the Pidurangala Buddhist 
temple to the north of the hill, as if about to make offerings at that shrine. 
All the figures are fully clothed from the waist downwards in coloured 
kambaiyas, and above the waist in short-sleeved jackets made of the finest 
material, and in some cases barely indicated by a line of deeper colour. 

The noble ladies are painted in pale yellow or orange, their attendants 
being distinguished by a greenish complexion. All the women are decked 
with a profusion of ornaments. Each ends below in a cloud-like mass, a 
peculiarity best explained by Mr. Bell's suggestion that it is due to the irregu- 
lar form of the cramped rock space available, on which the artist could not 
have drawn the legs without unsightly distortion. The suggestion made by 
another author that the clouds are intended to indicate the divine character 
of the personages appears to be incorrect. In accordance with the usual 
Indian practice, the figures were first outlined in red and black, and then 
painted in, not necessarily by the same hand. In one instance it is apparent 
that the outline was not exactly followed. 

The date of the frescoes in the closing years of the fifth century is fixed Chronology 
with sufficient accuracy by the known limits of the reign of Kasyapa I , A. D. 479 an . d . . 
and 497. They are, therefore, practically contemporary with the paintings cntlclsm - 
at Ajanta; all critics recognize the fact that the art of Sigiriya is closely related 
to that of Ajanta. For instance, the lady carrying a lotus in Plate 59 may be 
compared with the similar figure in Cave II at Ajanta, as reproduced in 
Griffiths, Plate 31. But the limitation of the colours and the total absence 
of blue in the Ceylonese paintings are important differences, and I do not think 
that the Sigiriya work equals the best at Ajanta. Mr. Havell is bold enough 
to credit the ladies of Sigiriya with 'Botticellian grace*, a criticism which may 
not meet with universal acceptance. But, whatever may be the final verdict 
of experts as to the intrinsic merits of the Ceylonese paintings, there can be 
no doubt that they are extremely remarkable productions of their age, and 
well deserving of careful study and serious criticism. There is nothing to 
indicate who the Ceylonese artists were, whence they came, or how they 
learned their skill. 

The Sigiriya figures, although by far the most important and interesting, Paintings at 
are by no means the only remains of ancient painting in the island. Numerous ^ radha " 


traces of early wall-paintings have been detected at Anuradhapura, of which 
the best preserved are those on the walls of the detached building ('frontis- 
piece* of Smither) on the eastern side of the Ruwanweli dag aba. Besides 
white, three primary colours, yellow, red, and blue, are used, the yellow and 
blue being sometimes combined to produce green. Yellow in various shades 
is the favourite, and was obtained from the natural arsenic sulphide called 
orpiment. The blue is indigo, not lapis lazuli. 

The style of the specimens reproduced in colour by Mr. Smither is dis- 
tinctly antique and closely allied to that of the later Ajanta paintings, being 
characterized, as they are, by bold free-hand execution of curves, with a 
truthful and at the same time decorative treatment of plant motives. Two 
examples are offered (Plate 61), which may be dated at any time from the 
sixth to the eighth century. The date of the building of the dag aba, of course, 
gives no clue to the date of mural decorations, which, in all probability, were 
retouched from time to time on the old lines. The colours are white and tints 
of brownish yellow. 
Cave Ancient paintings are necessarily so rare that a work hitherto unpublished 
paintmg at canno t be passed over, although it is of but slight intrinsic importance. Mr. 
kaduwa". ^ e ^ discovered two caves at a place called Tamankaduwa (Pulligoda gal- 
kanda), in a southerly direction from Kuda Ulpota and Dimbulagala, North 
Central Province, one of which contains a painting of five men, with halos 
and conical head-dresses, seated in an attitude of adoration. The colours are 
said to be 'well preserved', but no further details are recorded, and the 'short 
inscription* in the adjoining cave does not appear to have been deciphered. 
The age of the painting, therefore, is doubtful, but, so far as can be judged 
from a photograph, it must be of early date, possibly of the seventh century. 
It may, however, be later. 



A. Kinnara and lotuses : Ruwanwcli, Anuradhapura 

B. Dwarf: Ruwanwcli, Anuradhapura 
PLATE 6 1 

A. Muktesvara temple, Bhuvanesvar, Orissa. Cir. 900 a.d. 


, rf% v «*' 

j t/iM ijfti Eilrai ifc 

B. Details of Rajarani temple, Bhuvancsvar. C7/\ 1000 a.d. 


Chapter Nine 



Throughout India, except Buddhist remains, there is hardly anything Jain and 
standing which can be dated earlier than a.d. 450. No early examples Hindu 
of civil architecture exist. After the date named Buddhist structures tem P Ies - 
become scarce. The styles of Indian architecture in the medieval period, 
therefore, must be deduced from Brahmanical and Jain temples, or from the 
buildings represented in the Ajanta frescoes. 

It is now admitted that the variety of styles which may be distinguished 
depends not on differences of creed, but on date and, to a certain degree, on 
locality. At Khajuraho, for instance, Jain and Brahmanical temples are built 
in the same style. 

All authors who treat of Indian architecture notice, and are embarrassed Early stages 
by the fact, that each style when it first comes to our knowledge is full-grown of Indian 
and complete. The earliest specimens betray no signs of tentative effort, and styles 
in no case is it possible to trace the progressive evolution of a given style 
from rude beginnings. The extensive destruction of ancient monuments, 
especially those built of brick, no doubt supplies a partial, though not 
adequate, explanation. 1 I am convinced that the more fundamental explana- 
tion is to be found in the assumption that all the Indian styles are derived 
from prototypes constructed in timber, bamboos, and other perishable 
materials. 2 We have seen how easily the stupa railings can be accounted 
for in this way, and by the extension of the theory an adequate reason for 
the non-existence of the missing links in the chain of architectural evolution 
is supplied. In the essay previously cited, Mr. Simpson has quoted from the 
Satapatha Brahmana (S.B.E., vols, xii, xxvi) a long description of an early 
Brahmanical temple as constructed some five or six or seven hundred years 
before the Christian era. That temple consisted simply of two sheds, which 
were 'merely formed of posts and beams, covered with reeds and mats, and 
could only be described as belonging to the "thatch period" in architecture'. 
From such an edifice to the temples of Mount Abu and Tanjore the distance 
is great, but there seems to be little reason to doubt that the intervening 
stages were worked out for the most part by experiments with evanescent 
materials. Brick, the intermediate stage between the 'thatch period' and the 

1 Cunningham, A. S. Rep., vol. xi, pp. 40-6, pis. Circle, 1907-8, p. 31 . 

XIV-XVII; Vogel, Ann. Rep. A. S. Northern 2 The Ajanta frescoes make it plain. 

3*9* Q 

'stone period', offers such a ready prey to the spoiler that it may be reckoned 
as only ' semi-permanent ' material. 1 Whatever be the validity of this theory, 
we must take the styles ready-made as we find them, and briefly consider 
their several peculiarities, so far as may be necessary for the intelligent 
appreciation of the ancillary fine arts, which form the main subject of this 
Essentials In an ordinary Hindu temple the essential part is the rectangular cell or 
of a temple. s hrine containing the image or symbol of the god, and such a plain cell con- 
stitutes the simplest form of temple. The small shrines of the Gupta period 
have already been described. In the medieval period dignity was gained by 
the addition of a high roof or steeple, and by prefixing a porch, or nave with 
or without side-aisles, transepts, and subsidiary steeples, until an architectural 
composition of extreme complexity was evolved. Another type, built fre- 
quently by Jains and occasionally by Brahmanists, is a modification of the 
monastery, the monks' cells round the quadrangle being replaced by niches 
enshrining images. The modifications of both ground-plan and superstruc- 
ture are, indeed, endless. 2 All forms offer abundant opportunity for artistic 
Two leading In the crowd of varieties two leading styles of temple architecture — the 
styles. Northern or Indo- Aryan of Fergusson, and the Southern or Dravidian — 
may be readily distinguished. 3 If it be possible to amend the nomenclature 
so long established by Fergusson's authority, it would be preferable to give 
territorial names to all styles, calling the Indo-Aryan style that of Aryavarta 
or Hindustan, the great plain between the Himalayas and the Narbada. 4 
The term Dravidian is free from objection, Dravida being the ancient name 
of peninsular India. The two styles may more simply be denominated 
Northern and Southern. 
Aryavarta The Aryavarta^ or Northern style, examples of which to the south of the 
or Indo- Narbada are rare, is characterized by the bulging steeple with curvilinear 

1 The earlier temples, I believe, were built abundant the brick stage may not have inter- 
wholly in brick. At Aunda we find a small one vened. For N. Gujarat see Burgess, vol. ix, 
built almost entirely of that material, while the A. S. W. J., vol. xxxii of New Imperial Series, 
star-shaped plan and sharp crisp mouldings are 2 At Aihole, Bijapur District, Bombay, 'we have 
maintained as well almost as if built in stone, an unbroken sequence in the styles from the 
Remains of some of these early brick temples fifth to the fourteenth century — from the early 
are found in North Gujarat and the foundations cave to the latest mediaeval temple' (Cousens, 
and platforms on which the older stone ones are Progr. Rep. A. S. W. /., 1908-9, p. 35). 
erected are frequently constructed with a brick 3 This classification does not apply to Ceylon, 
core. Brick was, without doubt, the prevalent 4 The term Indo-Aryan implies a disputable 
building material before stone came into general theory. The two styles of architecture, although 
use, and probably immediately succeeded the rising from different origins, are related chrono- 
more primitive wooden structures whose [sic] logically, for the Southern style reached its high- 
forms are reproduced in many of the earliest est development after decadence had set in in 
caves' (Cousens, Progr. Rep. A. S. W. J., 1894- the North. 
5, p. 6). In some regions where stone was 




vertical ribs, placed over the sanctuary, and frequently reproduced on other 
parts of the building. Miniature repetitions of the form are often used with 
good effect as decorations of the steeples themselves. In spite of theories as 
to the bamboo origins of the curvilinear spire, its form is obviously inherent 
in the Indian corbelling methods of building. It appears to have been evolved 
first of all in brick as in the Great Temple at Bodh-Gaya. 

The best early examples are found at Bhuvanesvar in the Puri District, Temples at 
Orissa, where the temples, numbering several hundreds, illustrate the history Bhuvanes- 
of the style from the ninth or tenth to the thirteenth century. The earliest van 
specimens have steeples comparatively low and squat, but pleasing to an eye 
which has become accustomed to the design. The porch is a walled chamber 
with a low, massive roof, and internal pillars are wholly wanting. The com- 
bination of vertical and horizontal lines is skilfully arranged so as to give 
dignity to buildings of moderate height. This early astylar form of temple 
is best illustrated by the Muktesvara shrine, which Fergusson called 'the gem 
of Orissan art' (Plate 62 a). 1 

A second, and later, variety of the style is adequately represented by the The Great 
Great Temple, which has a high steeple tower, with sides vertical for the Tem P le - 
most part, and curving only near the top. The roof of the porch has con- 
siderable elevation, and in many details the design differs from that of the 
earlier variety. Sculptures of remarkable merit are introduced in panels on 
the basement and elsewhere. 

The third, or 'decorated', variety of the Bhuvanesvar style, in which The Raja- 
columns become prominent, dates from the twelfth or thirteenth century, ram temple. 
The most charming example is the Rajarani temple. Some exquisite details 
of this building are illustrated in Plate 62 b. 

The most renowned achievement of the vigorous Orissan school of archi- Temple of 
tects is the temple of the Sun at Konarak {vulgo 'Kanaruc') on the coast, * e Sul J at 
known to sailors as the Black Pagoda, in order to distinguish it from the Konara • 
White Pagoda, or temple of Jagannath at Puri. The remains of the main 
steeple, never completed, which had been overwhelmed long ago by the 
drifting sand, have been lately exposed by excavation. The porch, which 
stands practically perfect, is covered by a beautifully designed pyramidal 
roof, justly praised by Fergusson, and described by the Workmans as the 
most perfectly proportioned structure which they had seen in the course of 
years of study devoted to Indian temples. The temple, when in better con- 
dition than it now is, was admired enthusiastically by Abul Fazl, the minister 
and historian of Akbar in the sixteenth century. It is said to have been built 
by King Narasimha, who reigned between a.d. 1240 and 1280, a time when 
high-class work was not often produced. Considering its exceptional ex- 

1 See Ann. Rep. A. S., India, 1902-3, pp. 46-50; ibid., 1903-4, pp. 46-8; Progr. Rep. E. Circle, 
1908-9, p. 18. 

cellence, it is strangely late in date. A large book might be devoted to the 
description and illustration of this building and its sculptures. Plate 64, 
from a photograph kindly supplied by the Director-General of the Archaeo- 
logical Survey, shows the recently excavated remains of the steeple, as well 
as the porch, seen from the north-west. 
Temples at The Bhuvanesvar group of temples stands first in importance among the 
Khajuraho. exam pl es f the Aryavarta style by reason of the immense number of build- 
ings, usually in fairly good condition, and their variety, which marks the 
stages in the history of the style for at least three centuries. The group next 
in importance, situated at Khajuraho in the Chhatarpur State, Bundelkhand, 
although far inferior in both numbers and variety, includes some admirable 
buildings designed on a grand scale and richly adorned with sculpture. The 
temples, in more or less satisfactory preservation, numbering between twenty 
and thirty, were all erected by order of the Chandel kings c. a.d. iooo. They 
are executed in a fine sandstone, which offers great facilities to the sculptor. 
Several of the domes, constructed in the Indian manner with horizontal 
overlapping courses of stone, are remarkable achievements, the largest being 
22 feet in diameter. The cusps hanging from the centre of some of the domes 
are beautiful, although, of course, not so elaborately carved as the similar 
works executed slightly later and in more manageable marble at Mount Abu. 
Plate 65 gives a good notion of one of the best of the Khajuraho temples. 
The steeple is nearly 100 feet high. 
TheGuja- A beautiful variation of the Aryavarta or Indo-Aryan style, found in 
ra jj» ° r T s ?" Rajputana and Gujarat, is characterized by a free use of columns carved with 
style! a ^ imaginable richness, strut brackets, and exquisite marble ceilings with 
cusped pendants, at least equal to the best Tudor work of the kind. 
By an unfortunate error Fergusson described this Western or Gujarati 
style as the 'Jain style*. In reality it has no concern with any special kind of 
religion, and is Jain merely because Jains were numerous and wealthy in 
Western India in the late medieval period as they are still. When power 
passed into Muslim hands the so-called Jain style, that is to say the local 
style, was applied with the necessary modifications to the needs of Muham- 
madan worship. 
Temples on Two temples at Mount Abu, built wholly of white marble, are famous as 
Mount A u. unsur p assec j models of this wonderful style. The earlier, dedicated to 
Adinathy was built by a minister or governor named Vimala in a.d. 103 i; 
the later was consecrated by Tejpal two centuries afterwards, in a.d. 1230. 
Notwithstanding the considerable difference in age both temples are very 
similar in style. Illustrations are given showing half of the ceiling in Vimala 
Saha's temple (Plate 66) and some of the columns in the upper hall of Tejpal's 
temple (Plate 67). It is needless to comment on the beauty and delicacy of 
the carving and the richness of the design in both cases. 

PLATE 66. Part of ceiling of Temple of Yimalasaha, Mt. Abu. a.d. 1031 

PLATE 67. Pillars of upper hall of TcjpaFs Temple, Mt. Abu. a.d. 1230 


PLATE 68. Temple of the Sun at Osia, Jodhpur State, Rajputana. Late 9th century 


It would be easy to fill many pages with more or less similar specimens of Temples at 
work in the medieval style. I am tempted, however, to add a photograph O sia - 
(Plate 68) of a charming temple of the Sun at Osia in the Jodhpur State, 
Rajputana, brought to notice by Mr. D. R. Bandarkar, and treated in a much 
simpler fashion — an example of the originals of the huge piles at Khajuraho and 
Mount Abu, probably dating from the ninth century. Osia possesses no less 
than twelve large ancient temples, some Jain and some Brahmanical, and all, 
apparently, dating from the eighth and ninth centuries. The residents of 
the town show their appreciation of these works of art by using them as 
public latrines. 1 

Northern India is full of examples of the style, ancient, medieval, and Brick 
modern, mostly in stone, but occasionally in brick. The oldest brick specimen temples, 
in preservation sufficiently good to allow of the recognition of the style is that 
at Bhitargaonin the Cawnpore District, which is probably of the fifth century. 
With it must be classed the great temple at Bodh-Gaya. Another well- 
preserved ancient brick temple, referred doubtfully to the eighth century, 
stands at Konch in South Bihar. 2 There are many fine brick-temples in the 
Central Provinces, the finest of which is at Sirpur. These temples have 
massively carved stone door-posts, lintels, and pillars. The beautifully 
decorated burnt-brick Stupa at Mirpur Khas must also be mentioned as 
belonging to the first half of the medieval period. The art of these sites is 
the forerunner of the art of Khajuraho and Bhuvanesvar. There is reason 
to believe (as already observed) that the transition from wooden to stone 
architecture was made through brick, and that the scarcity of old brick build- 
ings is due to the facility with which the material could be utilized for other 
constructions. The decorations of brick buildings were carried out in terra- 
cotta, and carved as well as moulded bricks were used. Such bricks of good 
design are often seen built into later structures. The art of carving brick 
appears to be extinct. 

The late medieval Bengal variety, showing signs of Muhammadan influence Bengal 
is characterized by the use of the bent cornice, obviously copied from the !? n ^ °* m 
bamboo eaves of an ordinary Bengal hut, and by a peculiar arrangement of v * rta ^ t 
the curvilinear steeples; one lofty steeple placed over the centre being sur- 
rounded by four, eight, or sixteen smaller towers of the same form. Fergus- 
son has described the temple at Kantonagar in Dinajpur District, finished in 
1722, and decorated with applied terra-cottas of slight artistic merit. This 
variety of the Aryavarta style is peculiar to Bengal. 3 The only example re- 

1 Progr. Rep. A. S. W. /., 1906-7, p. 36. XVIII. 

2 Cunningham described and illustrated both 3 Manmohan Chakravarti, 'Bengali Temples and 
temples : Bhitargaon, A. S. Rep., vol. xi, pp. 40- their General Characteristics' (J. A. S. B. t vol. v, 
4, Pis. XIV-XVII; and Konch, ibid., vol. viii, New Ser. (1909)). 

p. 54, PI. VI; vol. xvi, pp. 50-8, Pis. XVI- 

corded outside that province is one at Bilhari, Central Provinces, built to the 
order of a Bengali immigrant. 
Modem I n the modern temples of Northern India the tendency is to reduce the 
temples. curva t ure of the steeple, and to make the form approximate to that of an 
English slender spire. The effect is sometimes pleasing, but lacking in the 
massive dignity of the best designs at Bhuvanesvar and Khajuraho. The con- 
temptible sculptured and painted decorations of the modern buildings testify 
plainly to the general lack of artistic feeling. 

Numerous recent buildings, sacred and secular, combine the Muham- 
madan dome with the Bengali cornice, omitting the steeple. Such buildings 
are erected freely by Hindus for purely Hindu purposes, as, for instance, the 
elegant mausoleum built at Benares to the memory of the lately deceased 
saint, Swami Bhaskaranand, which looks like a Muslim building. 
The Kash- The peculiar styles of architecture prevalent in the Himalayan kingdoms 
mir style. £ Kashmir and Nepal demand brief notice. 

The Kashmir style proper is restricted to the Valley, although a modifica- 
tion of it is found in the Salt Range region of the Panjab. The temples in this 
style, varying in date from about a.d. 750 to 1200, are all of small size, but 
in some cases the dignity of magnitude is attained by the addition of a walled 
quadrangle of imposing dimensions. 

The best-known example is the temple of Martanda or Martand — a local 
name of Vishnu as the Sun-god — which was erected about the middle of the 
eighth century by Lalitaditya (a.d. 724-60), the most powerful sovereign of 
Kashmir. This building, although the largest of its kind, is of modest dimen- 
sions, being a rectangle measuring 60 feet long by 38 feet wide. The width 
of the fa9ade, however, is increased to 60 feet by the addition of wings, and 
the walled enclosure measures internally 220 by 142 feet. The colonnade 
lining the wall is composed of eighty-four pillars, with intervening niches 
surmounted by the trefoil arches and triangular pediments or gables character- 
istic of the style. The cell, or chapel, which occupied the centre of each face 
of the enclosure, originally reached a height of about 30 feet. All the roofs 
have disappeared completely, so that it is uncertain whether they were of 
wood or stone. 
Peculiarities Plate 69 b clearly illustrates most of the peculiarities of the architecture, 
o the style, ^[q^ ma y fe SU mmed U p as consisting of pyramidal roofs, gables, trefoil 
arches, quasi-Doric columns, and dentil ornaments. 
Various The temple at Buniar (Bhaniyar), of uncertain date, which resembles that 
examples. f Martand in being surrounded by a colonnade, differs by being of smaller 
dimensions and in almost perfect preservation. The central shrine is now 
covered with wooden shingles, which may or may not have been the original 
form of roof. 
The more ornate temples at Vantpar (Avantipura) were erected during the 

A. The Council Hall, Yijayanagar. inth century 

II. Details of temple of Martand, Kashmir. 8th century 
(From a drawing by W. Simpson) 










reign of Avantivarman (a.d. 855-S3). The well-known little shrine at Payer, 
which Fergusson assigned to the thirteenth century, is older than he supposed, 
and probably dates from the tenth century. 1 The notion, started by Cunning- 
ham and accepted by certain other authors, that the quadrangles of the more 
important temples were designed to be filled with water, so that the shrines 
might be placed more immediately under the protection of the Nagas, or 
water-sprites, is absolutely baseless. 2 

Two peculiarities of Kashmir architecture — the trefoil arch and the quasi- Trefoil 
Doric columns— have given rise to much discussion. The trefoil arch recurs «ches and 
in certain temples at Malot, Katas, and other places in the Salt Range, which SlumiM 
was subject to the crown of Kashmir in the seventh century; 3 and when 
employed structurally, appears to be derived from the similar form frequently 
used as a canopy to a statue. 4 

The columns of the Kashmir temples are usually described as Indo-Doric 
on the assumption that their design is derived ultimately from Greek models. 
Mr. Tavenor Perry has thrown doubt upon this assumption because the 
Kashmir columns have sixteen flutes and are associated with very unclassical 
gables and trefoil arches. 5 As usual in India, the stages of the evolution of 
the Kashmir style cannot be traced in detail. It is possible that the Salt 
Range temples alluded to, and others at Gop, Sutrapada, and Kadwar in 
Kathiawar, which resemble the Kashmir buildings in certain respects, may 
be older than those in the Valley, but no clear evidence on the subject is 
available. 6 

The small valley of Nepal proper, measuring about 20 miles by 15, is said Nepalese 
to contain more than two thousand temples. 7 Most of them are designed in a style " 
style differing but slightly from the familiar Chinese pattern, in which the 
roof is the main element, the walls being mere screens set between pillars. 
An excellent illustration of this style is afforded by a temple built at Bhatgaon 
in 1703. 

Certain temples and tombs of Jain priests in the South Kanara District on Temples 
the western coast of the Madras Presidency, built in a style obviously derived ^ d j^ 8 - 
from wooden originals, possess a surprising and unexplained resemblance to 
the buildings in distant Nepal. 8 

1 Miscalled Payech by Vigne and many subse- A. S., India, 1903-4, PI. XXII a). 

quent authors (Stein, transl. Rajatarangini, vol. s Trans. Roy. Inst. British Architects, 3rd Ser., 

ii, p. 473). vul - »» P- x 5 8 - 

2 Ibid., Bk. iv, v. 192 note. ' For the Kathiawar temples see A. S. W. India, 
J Fergusson, Hist. Ind. andE. Architecture, 2nd vol. ii, p. 187, Pis. LI-LIII; Cousens, Progr. 
ed., i. 270; Cunningham, A. S. Rep., vol. v, pp. Rep. A. S. W. I., 1898-9, pp. 14-18. 

85-92, Pis. XXV-XXVII; vol. xiv, p. 35, PI. 7 Illustrations of styles used m Nepal will be 

XV; Beal, Buddhist Records, i. 143 J Walters, On found in Le Bon, Us Monuments deVInde. 

Yuan Chwang's Travels, i. 249. Fergusson, Hist. Ind. and E. Arc/at., 2nd ed., 

* It is so used at Konarak in Orissa (Ann. Rep. Figs. 303, 304, 307. 


Contrast The Gupta period may be regarded as one of transition between ancient 
between an( i medieval art, as it was between the polities of ancient and medieval India, 
medieval ^ rom the sixth cei } tu JJ we & n d m sculpture few traces of the kindly, human 
sculpture, spirit and naturalistic treatment which distinguished the ancient schools, 
mainly devoted to the service of Buddhism; and we pass into a world of 
art which scorns to represent the daily life of men and women, concerning 
itself almost exclusively with either asceticism of the self-contained yogi type 
or the weird imaginings of the later Hindu mythology, including that of the 
Mahayanist Buddhists, almost indistinguishable from that of the Brahmans. 
The beautiful story-telling reliefs of Borobudur in Java form a delightful 
exception to this generalization, and carry on the spirit of the old Bharhut 
and Sanchi artists with a delicacy and refinement of style peculiar to them- 
selves. The Jain sculpture is so strictly conventional that it may be almost 
left out of consideration. The spirit of medieval sculpture is chiefly expressed 
in Brahmanical and Buddhist works, which alike exalt the ascetic ideal and 
reflect the teachings of Puranic and Tantric literature. 
The ascetic Buddha no longer appears as the sympathetic human teacher moving about 
ideal. am0 ng his disciples and instructing them in the Good Law. His image is now 
generally made to conform to the ideal of the passionless yogi, as described 
in the Bhagavad-Gita: — 

'Who fixed in faith on Me, 
Dotes upon none, scorns none; rejoices not, 
And grieves not, letting good or evil hap 
Light when it will, and when it will depart, 
That man I love ! Who, unto friend and foe 
Keeping an equal heart, with equal mind 
Bears shame and glory ; with an equal peace 
Takes heat and cold, pleasure and pain ; abides 
Quit of desires, hears praise or calumny 
In passionless restraint, unmoved by each; 
Linked by no ties to earth, steadfast in Me, 
That man I love!' 1 

The representation of 'passionless restraint', however true to Hindu nature, 
affords a strictly limited field for the exercise of the sculptor's powers, and 
there is necessarily much monotony in the images, whether of Buddha or 
other personages, which are devoted to the expression of the ascetic ideal. 
Expression Another dominant note in medieval sculpture is struck by the endeavour 
of passion. f the artists to express violent superhuman emotion or demoniac passion, as 
represented by the whirling dances of Siva, the strivings of Marichi, the 
struggling of Ravana beneath his mountain load, and many other icono- 

1 Bhagavad-Gita, Bk. XII, transl. Edwin Arnold. 


graphical compositions. Multitudes of sculptures are simply the formal 
images of innumerable gods and goddesses, adorned with all the attributes 
and accessories prescribed by various scriptures. 

The sculpture of the early Indian schools makes an appeal far more Medieval 
universal than that of medieval times, which demands from the spectator sculpture 
a certain amount of recondite knowledge of the ideas underlying the later ggjJJ rly 
mythology. Its enthusiastic admirers never weary of extolling its 'idealism', m U * 
and of glorying in the fact that it is so peculiarly and exclusively Hindu as to 
be often unintelligible to the ordinary well-educated critic. The feelings 
which prompt such eulogies appear to be largely influenced by the modern 
nationalist movement. 

The Brahmanical (including later Buddhist) art, as evolved during the Modern art 
seventh, eighth, and subsequent centuries, continues to this day. No clear continuous 
line of demarcation can be drawn between medieval and modern sculpture, JJjJj 1 medi " 
although, unfortunately, modern work of any considerable degree of excel- 
lence is very rare. This chapter, therefore, deals with both medieval and 
modern art as being essentially one, the outcome of the Brahmanical reaction 
by which Buddhism was slowly strangled. 

The selection of medieval sculptures reproduced in this chapter will, it is 
hoped, be adequate to enable every reader to form his own judgement con- 
cerning the merits of the compositions as works of art. The first part of 
the medieval period is illustrated by the great cave-temples of Ajanta, 
Badami, and Ellora. 

Apart from the great shrines of Rajputana, Khajuraho, and Mount Abu, Two art 
late medieval sculpture falls into two main territorial divisions, namely, (i) ^? vm ^ s * 
Bihar, both North and South, with certain adjoining districts of Bengal and an/orissa. 
the Agra Provinces, which collectively formed the dominions of the Pala 
dynasty for more than four centuries from about a.d. 775 to 1193, the date 
of the Muhammadan conquest; and (2) Orissa, on the coast of the Bay of 
Bengal, which never was included in the Pala realm. 

The Pala kings having been devout Buddhists to the last, Buddhism con- Late sur- 
tinued to be the dominant religion in their territories long after it had become I 1 ^ 1 ,?* 
either extinct or moribund in most parts of India; and the Buddhist monas- in u B ihaJ m 
teries of Bihar, especially the wealthy foundation at Nalanda (modern Bar- 
gaon), were crowded with thousands of monks, who cultivated with success 
the arts required for the decoration of the sacred buildings. In consequence, 
a large proportion of the sculpture in Bihar and the neighbouring regions is 
Buddhist. The later Buddhism, as we have occasion to remark more than 
once, was of the Mahay ana or 'Great Vehicle' kind, delighting in the use of 
images, and closely related to Hinduism. The Brahmanical faiths, of course, 
never died out, and their votaries contributed their share to the art pro- 
3»* R 

Brahmani- During the first half of the seventh century, when the Chinese pilgrim 
cal sculpture j£ uen T sa ng recorded his invaluable notes, the Buddhists of Orissa out- 
run. num b ere( j ^g Brahmanical Hindus, but notwithstanding that fact, Buddhist 
sculpture is rare in the province, and the extant specimens, often of a high 
class, are mainly Brahmanical. From the point of view of the historian of art, 
as already observed, religious distinctions in the medieval period are un- 
important, sculptors making use of the style of their own age and country, 
irrespective of the creed to the service of which their works were dedicated. 
Hindu art i n Bihar the Muslim onslaught at the close of the twelfth century over- 
h d M tr T d threw Buddhism suddenly, and scattered all over India those few monks who 
conqueS! survived the indiscriminate massacres committed by the iconoclast armies of 
Islam. The rich monasteries of Sarnath near Benares soon shared the fate 
of the communities in Bihar, and layers of ashes in the ruins testify to this 
day the violence of the conquerors. Hindu art of all kinds, Buddhist included, 
was practically stamped out in the north-eastern provinces by the Muham- 
madan conquest. It lingered, however, in Orissa longer than in Bihar, and 
some of the best Orissan work dates from the thirteenth century. The con- 
quest of Orissa was not completed until Akbar's time in the sixteenth century, 
but it may be said that from the fourteenth century the history of art in all 
the north-eastern provinces is concerned only with Muslim forms. 

In quite recent days a slight revival of Hindu art may be discerned. 
Practically the history of Hindu sculpture in Bengal, Bihar, and Orissa closes 
with the thirteenth century. 
Abundance The innumerable ancient sites in Magadha or South Bihar and the neigh- 
of sculpture, bouring districts are full of well-executed images, mostly dating from the 
times of Pala rule, between the eighth and twelfth centuries. The destruction 
due to Muhammadan hatred of images has been less complete than in the 
upper provinces. Medieval Buddhism in its Tantric forms approximated so 
closely to the Brahmanical Hinduism that even a skilled observer may some- 
times hesitate to decide as to the religion for the service of which the image 
was destined — the Buddhist Tara, for instance, is not easil distinguishable 
from the Hindu Lakshmi. Although the style of the sculptures is always 
dominated by the formalism of ritual prescription, artists of exceptional 
ability and skill could make their powers more or less clearly apparent, and 
so raise compositions mainly conventional to the rank of works of art. A few 
specimens which possess merit greater than ordinary have been selected from 
the mass. 
A ninth- An elaborately decorated seated Buddha, in basalt, from Kurkihar in the 
BwddhZ sam ? re 8 ion > similarly proved by its inscription to date from the ninth century, 
' carries on the history. The folds of the drapery are marked by formal lines, 
and the resemblance to work of the Gupta period has disappeared (Plate 70 a). 
The details are wrought with the highest possible finish, but the type was too 

-*\-\ ** 



■»* V/Ml 
















■•TV ' 



















fc 1 






P S 


en 5 

<*> 5 

•73 O 


PLATE 72. Indrani, Jajpur, Bengal. 10th century 


rigidly determined by rule to allow the sculptor much scope for the exercise 
of his taste. 

The Tantric image of Marichi, goddess of dawn, a weird form with three Marichi. 
heads and six arms (Plate 71 a), offers greater opportunities to an artist in the 
delineation of active exertion. The goddess is supposed to be standing in a 
chariot drawn by seven boars, but the chariot and team are treated merely 
as formal accessories, the spectator's attention being invited solely to the 
sculptor's attempt to express the idea of radiant energy in the person of the 
goddess. The pose is that technically called the 'archer' attitude. 

One of the best and most characteristic examples of Bihar sculpture is the The 
large group of the Sun-god and his attendants now in the Indian Section of Rajmahal 
the Victoria and Albert Museum, which stands 5! feet high, and is in nearly Sun -g° d - 
perfect preservation (Plate 71 b). The god is represented standing in a lotus- 
shaped chariot drawn by seven horses, and driven by the legless Aruna, the 
Dawn. The artist, like the sculptor of Marichi, has concentrated his attention 
on the effigy of the god, reducing the chariot, horses, and charioteer to the 
position of minor accessories, in such a way that a casual spectator might fail 
to perceive their significance. The body of the principal figure is carefully 
modelled with considerable regard to realism, and the same commendation 
may be bestowed on the two female attendants with fly-whisks. The decora- 
tive framework is skilfully treated, and the whole composition produces an 
imposing and very pleasing effect. The mechanical execution of the carving 
is perfect, and the design is more restrained than that of much Hindu sculp- 
ture of the same period. The material is a black carboniferous shale, or clay 
slate, well adapted to the sculptor's purpose, and the twelfth century may be 
assigned as an approximate date. The Rajmahal Hills, where this remarkable 
work was excavated, lie to the south of Monghyr, and, although outside the 
limits of Bihar, were doubtless subject to the Pala rulers of that province. 

One more illustration of the medieval art of Bihar may suffice — a beauti- A Rajgir 
fully modelled and exquisitely finished seated Buddha in black Monghyr B "ddha. 
stone found by Mr. Grierson near Rajgir (Plate 70 b). The standing figures 
are the Bodhisattvas Avalokitesvara and Vajrapani. The seated goddesses 
are the two forms of Tara, the Green and the White. The composition as a 
whole is a compendium of the symbolism of Mahayanist Buddhism. As a 
work of art its interest lies chiefly in the careful modelling of the principal 
figure. The script of the inscription, the usual 'Buddhist creed', indicates 
that the work is approximately contemporary with the Rajmahal Sun-god. 

It may be well to mention the existence of other excellent specimens of the Other good 
medieval Bihar style, without detailed description or illustration. (1) Sir John images. 
Marshall notes as the most beautiful of the later finds at Sarnath, dating from 
the eleventh or twelfth century, a tiny figure of Avalokitesvara, 3I inches 
high, the carving of which, though somewhat stereotyped in character, is said 


to be executed with a delicacy and refinement which would do credit to a 

Chinese artist; l (2) the large Buddha called Mata Kunwar at the famous 

site near Kasia, Gorakhpur District; 2 (3) a fine Vishnu at Devathala, Din- 

ajpur District, Bengal; 3 and (4) sundry Buddhist sculptures from Kurkihar 

and Bishanpur, especially a remarkable relief of a Bodhisattva teaching, as 

described and illustrated by Stein. 4 The list might be largely extended from 

the collections in the Indian Museum, Calcutta, the Provincial Museum, 

Lucknow, and at other places. 

Localities The medieval sculptures of Orissa are chiefly associated with the Brah- 

of Onssan m anical temples of three localities — Bhuvanesvar, Konarak, and Puri — all in 

scu ptures. ^ p^. j)j str j ct ^ an( j ran gj n g [ n d a te from perhaps the ninth century to the 

thirteenth. The peculiarities of the architecture have already been noticed. 

The oldest sculptures, usually in sandstone, are at Bhuvanesvar; the best 

statues, mostly in chlorite, are at Konarak. 

Sculptures The temples and shrines at Bhuvanesvar, said to be five or six hundred in 

at Bhuva- num b er> are usually richly decorated, and so offer a wide field for selection, 

nesvar. |j m j te( j t0 some extent by the fact that many of the sculptures are grossly 

obscene, constituting, it is said, a complete set of illustrations of the Sanskrit 

Kamasastra, or erotic treatises. 

The sculptures, both decorative and statuary, are well represented in the 
Indian Museum, Calcutta, by a series of 128 casts taken in 1869, under the 
supervision of the Principal of the Calcutta School of Art at the time. 'The 
Orissa carver of those days', Dr. Anderson observes, 'went direct to nature 
for his designs, and the results of his labours in combining groups of animals 
with foliage show that he must have been a keen observer. They are ex- 
tremely pleasing pieces of art, not only on account of the beauty of their 
execution, but by reason of their truthfulness to nature/ 5 
Examples of In justification of this criticism a few examples from Rajendralala Mitra's 
decorative wor j t ma y b e given, beginning with a scroll on the Parasuramesvara temple, 
wor ' one of the oldest, possibly dating from the eighth or ninth century (Plate 73). 
Another scroll, including birds, &c, is from the small Rajarani temple of 
later date. A frieze of antelopes from the Muktesvara temple, perhaps of the 
ninth century, illustrates the successful realistic treatment of animal forms. 
Minor The Great Temple is supposed to date from the tenth century. Some of 
figures, the minor accessory figures on it are pleasing, the sculptor having more liberty 
for the exercise of his fancy and taste in treating them than he had when 
modelling the canonical imagesof the gods. Plate74A is from the BaitalDewal, 
a barrel-roofed shrine, like a Southern Gopuram, of about the same period. 

1 J. R. A. S., 1908, p. 1093, not reproduced. 4 Ind. Ant., xxx (1901), pp. 85, 90, 91, with 

2 Martin (Buchanan-Hamilton), Eastern India, photographic plates. 

ii> 357, with sketch. 5 Catal. Archaeol, Coll. I. M., Part II, p. 221. 

3 Cunningham, A. S. Rep., vol. xv, PI. XXVII. 

A. Scroll on Parasuramesvara temple, Bhuvanesvar 

B. Scroll with birds, &c, Kajarani temple, Bhuvanesvar 

C. Antelope frieze, Muktesvara temple, Bhuvanesvar 

A. Dancing girl on Baihil Dewal 
temple, Bhuvanesvar 

B. Bhagavati, Great Temple, 
Bhu vanes var 


A. Two horses; Konarak 

r\ ^SV^N^*' j&F^ <™Ty& *o&tt'-k^ 

U. A wheel, Konarak 

pi ait t: 

A. A colossal horse; Konarak 

H. Colossal elephant ; Konarak 


The chlorite Bhagavati, 7 feet high, on the tower of the Great Temple a goddess. 
(Plate 74 b), is an excellent example of the numerous elaborate and carefully 
carved statues of deities modelled according to strict rule. Such images are 
exhibitions of the skill of the stone-cutter rather than of creative sculpture. 

At the famous temple of Jagannath, Puri, built about A.D. 1100, a well- Mother and 
executed group representing a Hindu mother with her baby (Plate 77 a) offers child - 
a welcome change on gods and goddesses. Human sentiment is painfully rare 
in Indian medieval sculpture. This group seems to me to be of great merit. 

The unfinished temple at Konarak, dedicated to the Sun, and erected wheel at 
between a.d. 1240 and 1280, was designed to simulate a gigantic solar car Konarak. 
drawn by horses. Eight great wheels, each 9 feet 8 inches in diameter, ac- 
cordingly are carved above the plinth, and remarkable statues of seven horses 
stand outside. The wheels, the most perfect of which is shown in Plate 75 b, 
are carved with wonderful patience and admirable skill. 

Two of the detached colossal horses are shown in Plate 75 a, and one of Colossal 
them on a larger scale in Plate 76 a. It is the best preserved. Another, placed hor8es - 
outside the southern facade, is described by Mr. Havell as 'one of the grandest 
examples of Indian sculpture extant'. Mr. Havell's judgement of these works 
is as follows: — 

'here Indian sculptors have shown that they can express with as much fire and passion 
as the greatest European art the pride of victory and the glory of triumphant warfare; 
for not even the Homeric grandeur of the Elgin marbles surpasses the magnificent 
movement and modelling of this Indian Achilles, and the superbly monumental horse 
in its massive strength and vigour is not unworthy of comparison with Verrocchio's 
famous masterpiece at Venice'. 1 

The elephant colossi are also finely executed. One, shown in Plate 76 B, Elephant 
renders with mastery the character of the creature. colossus. 

The recent explorations carried out under Sir John Marshall's direction Chloriti 
have revealed many finely executed chlorite statues in addition to those statues, 
previously known. Two of the most noticeable of these discoveries are here 
reproduced. The image of Vishnu standing, equipped with all his canonical 
attributes, and attended by earthly and heavenly worshippers (Plate 77 b), 
may be fairly credited with no small degree of beauty, notwithstanding the 
hieratic style and the four arms. The flying figures are good, and the carving 
is perfect. 

The effigy of Bala-Krishna, the god as a boy in a swing, on the contrary, 
is ugly (Plate 77 c), and chiefly of interest as a tour deforce in stone-cutting. 

1 Mr. Havell freely admits the defects of the published by that author as 'a superb colossal 

statues in 'equine anatomy'. Verrocchio died in figure', possessing 'in the highest degree the 

A.D. 1488. His masterpiece is the equestrian qualities of great monumental design' (Quart. 

statue of Bartolomeo Colleoni. Mr. Fry so far Rev., 1910, p. 236). 
agrees with Mr. Havell as to describe the horse 

Nobody but a Hindu would think of making such chains in stone. The 
trefoil arch may be noted. 
Modern Orissan art practically ceases with Konarak. A small tract by Mr. Havell 
Orissan art. p roves that the artist families have never died out altogether, nor have they 
wholly lost their ancient skill. The author holds, and gives reasons for hold- 
ing, that 'there are carvers still to be found, whose work, in spite of all the 
discouraging conditions which surround them, is hardly inferior in artistic 
perception and technical skill to that of their predecessors'. He considers 
the men of Orissa to be superior to the north-western workers in sandstone, 
because they have c not hampered themselves by the limitations of a wood- 
carver's technique, but have fully realized the technical possibilities of their 
material for producing bold effects of light and shade suitable for architectural 
work*. I have no doubt that some of the living Orissan stone-carvers possess 
artistic feeling and could produce sculpture of considerable merit, if they 
received adequate patronage. At present their abilities are usually frittered 
away on pretty trifles in soapstone. 1 
Iconoclasm. In the Panjab and the United Provinces of Agra and Oudh thousands of 
Hindu temples and other edifices must have been destroyed by the Muslim 
conquerors during the seven centuries intervening between the raids of 
Mahmud of Ghazui and the death of Aurangzeb in 1 707 . The detailed records 
of the devastation wrought at Kanauj, Mathura, Benares, and many other 
notable cities fully justify the assertion that the buildings and monuments 
destroyed must have been numbered by thousands. Medieval sculpture, 
consequently, is scarce in the territories strongly held by the Musalman 
powers. 2 The more considerable remains are to be found only in regions 
lying remote from the track of the Muslim armies, such as Khajuraho in the 
Chhatarpur State of the Central India Agency, and the more inaccessible 
parts of Rajputana and the Central Provinces. 
Sculpture at Plate 65 gives some slight indication of the sculptured wealth of the 
Khajuraho. greater temples at Khajuraho erected during the tenth and eleventh centuries 
by the kings of the Chandel dynasty. I visited the temples many years ago 
and can testify that the crowd of figures is far more numerous than would 
appear from the photograph. But this 'peuple de pierre', as M. Le Bon calls 
it, was designed for the purpose of architectural decoration in the mass, not 
as an assemblage of individual works of fine art. The group of medieval 
temples at Khajuraho is the largest and most important in Upper India. At 
minor sites we find the same lack of individual works of artistic distinction 

1 E. B. Havell, Stone Carving in Bengal, thin madans have made a clean sweep of the district, 
quarto, 16 pp., 5 plates (Bengal Secretariat and razed to the ground every building, secular 
Dipot, Calcuttta, 1906). or religious, that had been erected by its former 

2 The case of the Bulandshahr District, U. P., Hindu rulers' (Growse, J. A. S. B., Pt. I, vol. Hi 
illustrates what happened. 'As might have been (1883), p. 280). 

expected from its nearness to Delhi, the Muham- 

A. Mother and child; tfitipU* of 
Jatjannath, Puri 

I*. Vishnu; Konarak 

C. Bala-Krishna; Konarak 

PT-ATF 7; 

A. Sculptures on a wall of Mokalji's temple, Chitor 

B. Face in wall of temple, Vasantgarh 
platk 78 


and, as a rule, the same absence of detailed record. The temples of Mount 
Abu in Rajputana undoubtedly exhibit masses of sculptured decoration of 
the most marvellous richness and delicacy, but there does not seem to be 
anything deserving of isolation from the mass for study as a separate work 
of art. 

The Tower of Victory, over 120 feet in height, at Chitor in Rajputana, Tower of 
built in the fifteenth century to commemorate the military successes of a local XjJ? ory at 
chieftain, is covered from top to bottom, inside and out, with an infinite 
multitude of images, representing, so far as may be, all the denizens of the 
Hindu pantheon, with their names attached, and constituting an 'illustrated 
dictionary of Hindu mythology'. Besides the effigies of the more ordinary 
deities, there are images representing the seasons, rivers, weapons, and other 
things as yet unpublished. Whenever this series of sculptures shall be repro- 
duced it will be invaluable as a key to Brahmanical iconography, but is not 
likely to contribute much to the history of art. 1 The better class of art in 
Rajputana dates from an earlier period, ending with the twelfth century. 

If the description recorded by Mr. Garrick, Sir A. Cunningham's assistant, Relief- 
can be depended on, certain relief sculptures at the Mokalji temple on the sculptures 
famous rock of Chitor posseoS high merit as works of art. The darkness of at Chltor - 
the chamber in which they are placed unfortunately frustrated attempts to 
photograph them. The temple, originally erected in the eleventh century, 
was reconstructed in the fifteenth century during the reign of Mokalji (a.d. 
1428-38). The pillars bearing the reliefs evidently belong to the earlier 
building. 2 The bas-reliefs, sixteen in number, are carved on octagonal bands 
of the eastern pair of pillars supporting the principal chamber of the temple, 
eight scenes on each pillar. The first scene on the southern column of the 
pair, according to Mr. Garrick, 

'depicts five human figures, of which two are large and three small; one of the former 
represents a woman carrying a water-jar on her head, and a man standing before her 
with hands joined in an attitude of adoration. The minor figures are much broken. 
This sculpture, along with the others of this set, is remarkable for the elaborate detail 
and technical excellence of its workmanship, the woman's hair being most minutely 
delineated. . . . The third carving is very well modelled and proportioned, and depicts 

two standing figures, male and female The fifth scene is filled with vigorous action, 

and consists of a musical festival; six male figures play six musical instruments . . . 
the sixth and last figure of this interesting group is seen full to the front, blowing a flute 
(murali or bansi) in a very animated position as if he were dancing. . . .' 

On the northern column of the pair — 

'the seventh scene is in all probability the most interesting of the whole series, and in 
its half a dozen figures gives us both a duel and an execution. The upper pair of men 
fight with shields and sabres, and their armour, accoutrements, &c, even to the knobs 

1 Cousens, Progr. Rep. A. S. W. /., 1900-1, p. 9. 2 Ibid., 1903-4, p. 38. 

and bosses on their shields, are most carefully delineated, and show that the manu- 
facture of these articles has altered as little during the last eight centuries as that of 
the musical instruments figured elsewhere. The lower portion of this comprehensive 
and instructive scene shows a pair of kneeling figures bound hand and foot, while an 
executioner holds his knife to the neck of the male figure to our left; but the female 
with him may possibly be a mere witness, though it is pretty clear from the general 
distribution of action in this trio that she awaits her turn for immolation/ l 

Mokalji's temple, as a whole, is decorated with an extraordinary wealth of 
sculpture, very effective in the mass, but not of quality sufficiently high to 
permit of small excerpts appearing to advantage. In order to give some notion 
of the powers of Rajputana sculptors in the first half of the fifteenth century, 
a specimen from the later sculptures of the temple, in high relief, with the 
images almost detached, (Plate 78 a) is presented. 

Face in a The most artistic object discovered by Mr. D. R. Bhandarkar during his 
window, rambles in Rajputana — so fruitful in additions to historical knowledge — is 
the face looking out from a stone window in a wall of an old temple of the 
Sun at Vasantgarh in the Sirohi State (Plate 78 b). Mr. Bhandarkar supposes 
it to date from the seventh century, 2 but, whatever its exact age, it is a beauti- 
ful work, and unique, to the best of my knowledge. The surrounding orna- 
ment is in an excellent style. 

Image of The ancient town of Osia in the Jodhpur or Marwar State possesses no less 
Kuvera. t j ian twelve old temples. In one of these, No. 9, known as the shrine of Devi, 
is the image of Kuvera, the god of riches, which may be compared with the 
effigies of the same deity in Gandhara and elsewhere (Plate 79 a). 3 

Vishnu at A beautifully wrought figure of Vishnu in the Mathura Museum, about 

Mathura. 2 6 inches in height, and presumably produced in the local workshops (Plate 
79 b), may be compared with the Konarak Vishnu (Plate 77 b). The two 
images, while largely in agreement, differ in a multitude of details. 4 The 
Mathura figure is not likely to be later than the tenth century, the temples 
of the city having been burnt by Mahmud of Ghazni at the close of a.d. 1018. 

1 Garrick, in Cunningham's ArchaeoL Survey they can best be compared to Khajaraho and 

Reports, vol. xxiii, pp. 120-3. Mr. Garrick's tour cannot be much earlier than a.d. 900. 

took place in 1 883-4. 4 At Panthia near Mandhata, the ancient Mahish- 

z Progr. Rep. A. 5. W. /., 1905-6, pp. 51, 52. mati, on the Narbada, there are twenty-four 

It is, however, typically medieval, c. A.D. tenth different forms of Vishnu duly labelled and dis- 

century. tinguished by variations in the attributes and 

3 Ibid., 1906-7, p. 36. Bhandarkar dates most position of the hands, 
of the Osia shrines in the eighth century, but 































Chapter Ten 


The Dravidian or Southern style of architecture is sharply distinguished 
from the Northern by the fact that its tower or spire is straight-lined 
and pyramidal in form, divided into stories by horizontal bands, and 
surmounted by either a barrel-roof or a dome derived directly from the 
ancient wooden architecture. The central shrine originally stood alone, but 
in later times it was enclosed in an immense walled court, usually including 
numerous subsidiary temples, tanks, and sculptured halls or cloisters. The 
quadrangle is entered by lofty gateways (gopuram), which in later temples 
overtop the central shrine, and so spoil the effect of the architectural com- 
position. But the great temple of Tanjore, its smaller replica at Gangaikonda- 
puram, and some of the earlier temples at Conjeeveram (Kanchi) are designed 
on correct principles, with the central mass dominating the composition. 
Sometimes there are several quadrangles, one within the other. 

The history of the style begins in the seventh century with the Dharmaraja Rathas of 
Ratha, the earliest of the rock-cut rathas at Mamallapuram, thirty-five miles Mamalla- 
south of Madras, commonly known as the Seven Pagodas, which were ex- P uram - 
cavated in the reigns of the Pallava kings of the South during the seventh 
century. I give an illustration of the Ganesa Ratha (c. a.d. 680), with a ridge 
roof (Plate 81 a). Some of the others are crowned by domes. 

The next stage in the development of the style is marked by the structural Structural 
temples at Conjeeveram (Kanchi), the Pallava capital, which became known temples at 
only a few years ago, and have been described in detail by Mr. A. Rea. Kanchl - 
Six temples of the Pallava period exist in or close to the town. 1 Inscrip- 
tions prove that the two principal edifices, the Kailasanatha and the Vai- 
kuntha-Perumal) were erected by the sons of King Rajasimha, great-grandson 
of Narasimha-varman. The Muktesvara temple of about the same date, say 
a.d. 700 to A.D. 750, with a domical roof, is a typical example. 2 

Further development was effected under the patronage of the powerful 
Chola kings, Rajaraja and his son Rajendra (985 to 1035), the builders 
respectively of the Great Temple at Tanjore and its fellow at Gangaikonda- 
puram in the Trichinopoly District. At this period the shrine was designed 
on huge proportions, towering above the subsidiary gateways and pavilions. 

1 Rea, Pallava Architecture, 4to, Madras, 1909, Pallavas. 

being vol. xxxiv oiArchaeoL S. Rep., India, New 2 For relation of Pallava caves and temples see 

Imp. Series, and vol. xi of the Southern India Venkayya, 'The Pallavas' (Ann.Rep.A.S.Jndia, 

Series. The standard work is Jouveau-Dubreuil, 1906-7, pp. 226-35), *&& Dr. Hultzsch (Ep. Ind. 9 

Architecture du Sud de Vlnde. See also his vol. x, pp. 1-14). 

3»* S 


Later The gigantic South-Indian temples, with vast quadrangular enclosures and 
temples. j f t y gopurams overtopping the central shrine, extend in date from the six- 
teenth century to the present day. Fergusson speaks of 'upwards of thirty 
great Dravidian temples, or groups of temples, any one of which must have 
cost as much to build as an English cathedral — some a great deal more'. 
Several such edifices, at Ramesvaram, Tinnevelly, Madura, and other places, 
are described in his book. The buildings at Madura are of special interest 
because they can be dated closely, having been erected by Tirumal Naik, 
a local chieftain, who reigned from 1623 to 1659. Plate 81 c gives a general 
view of the Madura temple, a typical example. The corridors or cloisters 
connected with such temples are of wonderfully large dimensions — those of 
Ramesvaram, for instance, aggregating nearly 4,000 feet in length — and are 
filled with weird, fantastic sculpture. Perhaps the most marvellous of all 
Dravidian temples is the well-known rock-cut Kailasa temple at Ellora, ex- 
cavated from a hill-side by a Rashtrakuta king in the eighth century. In style 
the Kailasa is a development of the Pallava shrines, but its sculpture is finer 
than anything produced in the South. At Badami and Pattadkal in the 
Bijapur District are other shrines of the same type; these are all structural. 
The Vijaya- The immense ruins of the city of Vijayanagar, dating from the fifteenth and 
nagar style, sixteenth centuries, surrounding Hampi village in the Bellary District, 
Madras, present numerous examples of a special local variety of Dravidian 
architecture. The royal palaces and apartments here show signs of Islamic 
influence. The temples are purely southern Indian in style with high gate- 
ways and many-pillared pavilions. 
The Deccan The style intermediate in both locality and character between the Northern 
k° r Ch n!i U " anc * Southern styles is that which received from Fergusson the inappropriate 
" name of Chalukyan. It is true that the Chalukya clan supplied one of the 
leading royal families of the Deccan from the middle of the sixth to the middle 
of the eighth century, and again from a.d. 973 to the Muhammadan conquest, 
but the typical examples of the style are the work of Hoysala, not Chalukya 
kings; and, if a dynastic designation be given, the style should be named 
Hoysala rather than Chalukya. Territorial designations are, however, prefer- 
able to dynastic, and if it be practicable to modify Fergusson 's established 
nomenclature, the style may be better described either as that of the Deccan, 
or that of Mysore, in which province the finest specimens occur, at Halebid, 
the ancient capital, Belur, and many other localities less known to fame. 
The Belur This style, whatever name be bestowed upon it, is characterized by a richly 
^ m ot J^ r carved base or plinth, supporting the temple, which is polygonal, star-shaped 
' in plan, and roofed by a low pyramidal tower, surmounted by a vase-like 
ornament. The temple of Vishnu in the village of Nuggehalli, in the Tiplur 
Taluk, Mysore, as shown in Plate 83, from an unpublished photograph, gives 
a good notion of this extraordinarily ornate style. The stellate plan appears 


clearly in the view of the Somnathpur temple (Plate 85 a). The Belur temple 
is known to have been erected in a.d. 1 117 by a Hoysala king named Bettiga, 
converted from Jainism to faith in Vishnu. The more magnificent temples at 
Halebid, the Hoysalesvara and Kedaresvara, are somewhat later in date, and 
necessarily must have been under construction for many years. Not long ago 
the disintegrating action of the roots of a banyan tree unfortunately reduced 
the Kedaresvara to a heap of ruins. 1 

Plate 84, showing a small portion of the sculptures on the eastern end of Sculptures 
the Hoysalesvara temple, will give the reader a faint notion of 'one of the at Halebid. 
most marvellous exhibitions of human labour to be found even in the patient 
East'. The architectural framework, it will be observed, is used mainly as a 
background for the display of an infinity of superb decoration, which leaves 
no space uncovered and gives the eye no rest. 

'The building', Fergusson writes, 'stands on a terrace ranging from 5 to 6 feet in 
height, and paved with large slabs. On this stands a frieze of elephants, following all 
the sinuosities of the plan and extending to some 710 feet in length, and containing 
not less than two thousand elephants, most of them with riders and trappings, sculp- 
tured as only an Oriental can represent the wisest of brutes. Above these is a frieze 
of sardulas y or conventional lions — the emblems of the Hoysala Ballalas who built the 
temple. 2 Then comes a scroll of infinite beauty and variety of design ; over this a frieze 
of horsemen and another scroll ; over which is a bas-relief of scenes from the Ramayana, 
representing the conquest of Ceylon and all the varied incidents of that epic. This, 
like the other, is about 700 feet long. (The frieze of the Parthenon is less than 550 feet.) 
Then some celestial beasts and celestial birds, and all along the east front a frieze of 
groups from human life, and then a cornice, with a rail, divided into panels, each con- 
taining two figures. Over this are windows of pierced slabs, like those of Belur, though 
not so rich or varied.' 

The Hoysalesvara and several other buildings of its class are twin temples 
consisting of two distinct shrines set side by side and joined together. The 
beautiful building at Somnathpur is a triple temple. A special feature of 
interest in these Mysore temples is the record of the names of the Kanarese 
artists, who executed individual statues. At Belur there are twelve such 
signatures, and at the Hoysalesvara fourteen, all different. Eight signatures 
on the Somnathpur temple have been noted, among them that of Malli- 
tamma, who executed forty images. 3 

Certain temples near the Tungabhadra river situated in the western part The Ballari 
of the Bellary District, Madras, wedged in between Mysore territory on the temples. 

1 The principal temples in this style range in Hoysala kings, but as part of the canonical 
date between a.d. 1117 and 1268 (Rice, Mysore scheme of decoration— elephants, lions, horses, 
and Coargfrom the Inscriptions, Constable, 1909, men. 

p. 194). See Workman, Through Town and 3 In Epigraphia Carnatica, vol. v, Part I, pp. 

Jungle (1904), ch. v, with many excellent illustra- xxxvi, xxxviii, Mr. Rice describes and illustrates 

tions. several temples. See also Ann. Rep. Archaeol. &., 

2 The lions are there, not as the emblem of the Mysore, 1909-10, para. 25. 


south and the Nizam's Dominions on the north, form the subject of an ex- 
cellent monograph by Mr. Rea, entitled Chalukyan Architecture. The title 
is so far justified that the buildings were erected to the order of Chalukya 
kings in the twelfth century. But the style is a modification of the Dravidian 
or Southern, not of the Deccan or Mysore style called Chalukyan by Mr. 
Fergusson. The plans are rectangular, not star-shaped, and the towers are 
distinctly Dravidian in design. The buildings, as Mr. Rea correctly observes, 
'exhibit a preponderance of Dravidian forms. They might best be described 
as an embodiment of Chalukyan details engrafted on a Dravidian building.' 
Although the statues, individually regarded, are not of high merit, and present 
much of the grotesqueness of commonplace Hindu sculpture, the ornament, 
considered as a whole, is superb. It is impossible, we are assured, to describe 
the exquisite finish of the greenstone or hornblende pillars, or to exaggerate 
the marvellous intricacy and artistic finish of the decoration in even the 
minutest details. The ornament is generally completely undercut, and is 
sometimes attached to the solid masonry by the most slender of stalks, pro- 
ducing the effect of an incrustation of foliage on the wall. Both the intricate 
geometrical patterns of the ceilings and the foliated work covering every other 
part of the building exhibit the greatest possible exuberance of varied forms 
boldly designed and executed with consummate mastery of technical details. 
No chased work in gold or silver could possibly be finer, and the patterns to 
this day are copied by goldsmiths, who take casts and moulds from them, 
although unable to reproduce the sharpness and finish of the originals. 
A ceiling. Opinions may differ as to the propriety of employing such jewellers' work 
as architectural decoration, but concerning the beauty of the result and the 
high standard of executive skill no two opinions are possible. The annexed 
plan of a ceiling in the Suryanarayanaswami temple at Magala may suffice to 
give some notion of the exquisite carving characteristic of the Bellary variety 
of the Dravidian style, as favoured by Chalukya Kings. 

^ ha ™ rter The arts of sculpture and decorative carving in stone continued to be 
° sculpture! P ractis . ed in India to the sou *h of the Narbada under the patronage of many 
' dynasties throughout the medieval period, and even to this day are cultivated 
with considerable success whenever encouragement on an adequate scale is 
offered. But, excepting certain Chola statuary of the eleventh century, which 
is pre-eminently excellent, the Southern figure sculpture does not often 
attain high quality. In quantity it is enormous, the gigantic temples and halls 
characteristic of the Dravidian kingdoms being commonly overloaded with 
sculptured ornament on every member. Mythological subjects from the 
Puranas and Tantras are the favourites, and the tendency is to treat the con- 
ceptions of a luxuriant mythology with exuberant fancy. The result too often 


A. Ganesa Ratha, Mamallapuram 

B. Muktesvara Temple, Kanchi, from 
the south-west 





,7V .- ■ --.* 

C. Madura Temple, general view 
PLATE 8 1 













PLATE 84. Hoysalesvara Temple, Halebid; sculptures on east end 


is merely grotesque, and very few of the individual images can claim to be 
beautiful. The sculpture of the South is really the successor of the medieval 
art of the North. The figure-sculpture is purely iconographical, and executed 
exactly according to the literary canon. 

The purely decorative designs carved on the twelfth-century Chalvhya and 
Hoy sola temples and elsewhere are unsurpassed, but the statuary of the same 
buildings is too often conventional and rarely of much merit. 

During the seventh century the kings of the Pallava dynasty of Kanchi The Pallava 
(Conjeeveram) succeeded in making themselves the dominant power in dynasty- 
Southern India, overshadowing the ancient Chola, Chera, and Pandya 
dynasties of the Tamil region, and, for a time, obscuring the glory of the 
powerful Chaluhya sovereigns of the Deccan. The Pallava king named 
Mahendra-varman I (c. a.d. 600-25), a g reat builder, is responsible for many 
rock-cut temples in the North Arcot, South Arcot, Chingleput, and Trichino- 
poly Districts. The earliest rathas, or monolithic shrines, at Mamallapuram, 
or the Seven Pagodas, also probably should be ascribed to his reign. His son, 
Narasimha-varman I, surnamed Mahamalla, the most mighty prince of his 
line, gave his name to Mamallapuram, and constructed or rather caused to 
be excavated, some of the rathas at that place. The family taste for architec- 
ture survived in the descendants of Narasimha-varman, the so-called 'Shore 
Temple' at Mamallapuram and the early structural temples at Kanchi being 
ascribed partly to his great-grandson, Rajasimha, and partly to Rajasimha's 

The most notable remains of Pallava art are those dating from the seventh Remains at 
and eighth centuries at Mamallapuram, which include, besides the well-known Mamalla- 
rathas, numerous less familar monuments, comprising temples, statues in the puram# 
round, and gigantic sculptures in relief carved on the face of the rocks. 
Among the sculptures in the round mention may be made of a lion, seven 
feet in length, which is said to be well-proportioned and of a noble appearance. 

Several authors concur in the opinion that the most artistic of the reliefs is Durga and 
the great composition depicting the victory of the Good, represented by the Mahisha- 
goddess Durga mounted on a lion, over evil personified in the buffalo-headed 
demon, Mahishasura (Plate 87). The scene undoubtedly is full of life and 
movement, and the goddess is a dignified figure. 

The great bas-relief at Mamallapuram covers a sheet of rock 96 feet in The Great 
length and 43 feet in breadth. Around a central figure, now missing, all Bas-Relief. 
creation, heavenly and earthly are gathered in worship. Before the great deity 
even the animals do penance, while seers and lesser gods and the spirits of 
the air unite in adoration. This gigantic sculpture was erroneously identified 
as representing Arjuna's Penance, after the story in the Mahahharata. 

Another and smaller relief of Pallava age at Trichinopoly (Plate 89 a) seems ^chint 
to be of earlier date and is in a better style of art. This group, consisting of poly< 


five large figures, in addition to the crouching dwarf on whose hand the 

central deity, apparently a form of Siva, rests his right foot, is symmetrically 

composed, due prominence being given to the god, who stands in a natural 

and easy attitude. He has four arms, but only two are prominent, and all the 

other figures are quite free from monstrosity. The kneeling worshippers are 

excellently modelled and pleasing in appearance. The style, in fact, is much 

more akin to that of Northern India, and especially to the work at Badami, 

than to the sculpture commonly seen in the South. 

TheBegur Two spirited bas-relief sculptures from Mysore territory, now in the 

2nd relief r Bangalore Museum, although too crude to rank as fine art, perhaps deserve 

' passing mention. The first, on the Bagur stone, dating from about a.d. 934-8, 

gives a vivid picture of a battle between the force commanded by a chief 

mounted on an elephant and another led by a rival on horseback. The second, 

on the Atakur stone dated a.d. 949-50, commemorating a set fight between 

a mighty hound and a great boar in which both combatants were killed, 

represents an incident in the struggle, the hound having his teeth fixed in 

the boar's snout. 1 The design is better than the execution. 

The Chola The C kolas, who succeeded the Pallavas as the paramount power in the 

dynasty. g ou th, may be said to have filled the principal places in the Tamil countries 

with their edifices, religious and secular, all richly sculptured. Rajaraja the 

Great (985-1018), the most famous king of a capable dynasty, extended his 

power over nearly the whole of the Madras Presidency, Ceylon, and a large 

part of Mysore, while his navy ranged as far as the Laccadive and Maldive 

islands. A king so powerful and wealthy naturally spent freely on building, 

and the world owes to him the temple at Tanjore, his capital, the best designed 

of all the great South Indian temples. 

Gangai- His son and successor, Rajendra-Choladeva I, surnamed Gangaikonda 

konda- (2018-35), continued and extended RajarajYs victories by sea and land. In 

" memory of the subjugation of the Ganga territory in Mysore, or, as others 

say, to commemorate his march northwards as far as the Ganges, Rajendra 

built a new capital, Gangaikonda-Cholapuram, in the Trichinopoly District, 

and constructed there an enormous artificial lake with an embankment sixteen 

miles long. The principal temple, designed on the noble model of the 

Tanjore temple, enshrined a huge monolithic lingam, thirty feet high, and 

the precincts of the city included a palace and many other notable buildings, 

now either vanished or in complete ruin. The sculptures in panels on the 

walls of the great temple are remarkable for their elegance and beauty (Plates 

89 B and 90). 

The excessively exuberant, and yet fascinating, massed architectural sculp- 

1 Fleet, 'Three Western Ganga Records in the graph of the Begur stone in Ep. Carnatica, vol. 
Mysore Government Museum at Bangalore' xi, frontispiece. 
(Ep. Ind., vi, p. 40, with plates). A larger photo- 


A. Somnathpur Temple, Mysore. A. I). izhtt 

H. Plan of ceiling in Suryanarayanaswami Temple at Magala 


PLATE 86. Bracket statuette; Karvati Temple 


ture of the Mysore temples built by the Hoy sola kings in the twelfth century Twelfth- 
has been already illustrated sufficiently. The artists who designed such ^"^ 
enormous sheets of rich sculpture aimed at producing an imposing effect by ofMysSe 
the splendour of a mass of carvings of the highest complexity, rather than by temples, 
inviting attention to individual figures. Nevertheless, the individual figures 
will bear examination in detail, the elephants especially being exquisitely true 
to nature. As already observed many of the larger statues of the Mysore 
temples are signed by the artists. 

The approximately contemporary temples erected in the Bellary District, Chalukyan 
Madras, under the patronage of the Chalukyan kings are remarkable for the future 
unequalled richness and delicacy of their deeply undercut decorative carving. Statrirt 
The figure sculpture is far inferior, and, notwithstanding the perfection of 
its mechanical execution, is generally conventional in design and semi- 
barbarous in style. 

In the year 1336 two Hindu brothers established a principality with its The king- 
capital at Vijayanagar on the Tungabhadra river, which rapidly developed d ?| n at 
into an empire comprising all Southern India beyond the Kistna. The state ^2*" 
attained the height of its prosperity early in the sixteenth century during the 
reign of Krishna Deva Raya, the contemporary of Henry VIII of England, 
who stoutly maintained the Hindu cause against the Muslim Sultans of the 
Deccan until 1565, when he was utterly defeated by the combined forces of 
the Muhammadan princes, and his capital taken. The victors devoted their 
energies for five months to the deliberate destruction of the city, heaping up 
bonfires round the principal monuments, and hacking and mutilating the 
graven images. They succeeded in converting one of the richest and most 
splendid capitals of Asia into the abode of wild beasts, which has remained 
desolate to this day, save for the huts of a tiny hamlet nestling amidst the ruins. 

The actual site of the city covers an area of nine square miles, but the Site of the 
fortifications and outposts include a space far larger. In the days of its great- Clt y- 
ness the capital was filled with magnificent granite edifices erected by forced 
labour, and adorned in the most lavish manner with every form of decoration 
agreeable to the taste of a semi-barbaric court. The extant detailed accounts 
of the glories of Vijayanagar in the sixteenth century recall the familar stories 
of the Aztec capital as it was seen by its Spanish conquerors, the administra- 
tion of both courts combining unbridled luxury with ferocious cruelty. 

The semi-barbarism of the court is reflected in the forms of art. The giant Style of art. 
monolithic Man-lion (Narasiwha) statue, 22 feet high, and the huge Monkey- 
god Hanuman, although wrought with exquisite finish, are hideous inartistic 
monsters; and the sculpture generally, however perfect in mechanical execu- 
tion, is lacking in beauty and refinement. Bas-reliefs 

In the palace enclosure the most striking building is the temple known as tHazara 
Hazara Ramaswami, 'the Thousand Lord Ramas', used by the old kings as Ramaswam. 


their Chapel Royal. The walls of the courtyard of this edifice are covered 
with bas-reliefs depicting scenes from the Ramayana, described by Mr. Rea 
as being beautifully executed and carved with great life and spirit*. The 
specimens illustrated in Plate 91 will show how far such praise is justified. 1 

Scroll, Hariharesvara temple, Bellary District. 

Makara tor ana (arch), Malikarjuna temple, Kuruvatti. 

s ^Ipjures One of the most notable of the ruins is the temple of Vishnu under the 
swami temple name °^ Vitthalaswami, begun early in the sixteenth century, and still un- 
and throne, finished when the city fell in 1565, never to rise again. The great hall in front 
of the shrine 

'rests on a richly sculptured basement, and its roof is supported by huge masses of 
granite, 15 feet high, each consisting of a central pillar surrounded by detached shafts, 

1 For the history see Sewell, A Forgotten Empire Meadows Taylor and Fergusson, Architecture in 

(Vijayanagar), a Contribution to the History of Dharwar and Mysore (atlas fol., 1866). The 

India (1900), a valuable and deeply interesting whole of the Ramayana reliefs is given in Pis. 

book. A photograph of the Man-lion faces p. LXVIII, LXIX of that work. 
163. The Monkey-god forms the frontispiece to 













«& .* 

y *■■ 4 * " lj* * /' » $* 

>f > 

W il 

PLATE 88. Rock-sculpture ('Arjuna's Penance'), Mamallapuram 


figures mounted on demons, and other ornament, all cut from a single block of stone. 
These are surmounted by an elaborate and equally massive cornice; and the whole is 
carved with a boldness and expression of power nowhere surpassed in the buildings 
of its class, showing the extreme limit in florid magnificence to which the style advanced. 
This beautiful building has been grievously injured by the destroyers of the city. 

Several of the carved pillars have been attacked 
with such fury that they are hardly more than 
shapeless blocks of stone, and a large portion 
of the centre has been destroyed utterly. (Imp. 

The sculptures on the walls of the 
throne are also commended, but no illus- 
trations of the works referred to have been 

The best examples of the Vijayanagar 
style are to be found, perhaps, not at the 
capital, but at Tadpatri (Tarpatri), Ananta- 
pur District, Madras, in gateways erected 
during the sixteenth century by a prince 
subordinate to the kings of Vijayanagar. 
Fergusson, who devoted two full -page 
plates to the illustration of the Tadpatri 
greenstone sculptures, judged them to be 
'on the whole, perhaps, in better taste than 
anything else in this style'. 1 

The Margasahayar temple at Virinchi- 
puram in the North Arcot District, 7 J miles 
to the west of Vellore, is believed to have 
been erected late in the fifteenth century, 
while the district was included in the dominions of Vijayanagar. One of the 
columns offers a good example of the yali> or conventional rampant lion, an 
effective, bold form of decoration very fashionable and characteristic of 
the country in both South India and Ceylon during medieval times. The 
lion, about 5! feet in height, is designed and executed with spirit. 

The statue of a goddess on the entrance of the temple of Venkata-ramana- 
svami at the famous fortress of Gingee in the South Arcot District, 
probably built during the time of the Vijayanagar rule a little before or after 
a.d. 1500, is of special interest as proving, like the Tadpatri figures, the 

of Tarpatri 

Yali, or rampant lion, 

in Vijayanagar style, 

at Virinchipuram. 

Female figure at 
Jinji (Gingee). 

Yaliy or 
lion, at 

motive of 
figure at 

1 A new temple at Tadpatri is adorned with 
elaboration equal to that of the old one; but, 
although the decorative carving is good, the 
figure sculpture is grotesque and contemptible. 


The work has been fully illustrated by Mr. Rea 
in his book, Stone Carving and Inlaying in 
Southern India. 


persistence of a very ancient motive, common in Gandhara and Mathura 
art (see illustration p. 137). This late southern example preserves all the 
essentials of the design — the female figure, the crossed legs, the raised right 
arm, and the left arm twined round the stem. 
Sculptures The palace of the Udaiyarpalaiyam zamindar in the Trichinopoly District 
at Udaiyar- contains some good figure and decorative sculpture associated with Indo- 
palaiyam. Miihammadan architecture, and evidently not older than the seventeenth 
century. It is executed in a rather soft stone. My attention was drawn to 
the sculptures by the remarks of Mr. J. P. Bedford, I.C.S., who made a com- 
munication to the Archaeological Survey and wrote: 

'One of the big halls is in general design something after the fashion of Tirumal 
Naik's famous hall in Madura; but the spandrels of the arches are one mass of carving 
of birds, flowers, &c., showing extraordinary fancy and spirit, while the arches them- 
selves are worked out in the most exquisite tracery, with a niche above each column 
containing some god or saint. Above the level of the spandrels is a deep colonnade 
running round the whole hall, corresponding to the clerestory of an English cathedral 
— also a mass of spirited carving in relief. The effect of the whole is, so far as the 
writer's experience goes, absolutely unique so far as an Indian building is concerned ; 
but it is very suggestive of Northern European Gothic, say the porches of Chartres 
Cathedral.' l 

Seventeenth- The numerous gigantic temples of Southern India in the Dravidian style, 
century erected from the sixteenth century to the present day, with their appurtenant 
scu pture. ^^dors and 'halls of 1000 columns', are covered with sculpture, mostly of 
a fantastic and outre character. The most famous princely builder was Tiru- 
mal (Trimul) Naik, who ruled at Madura from 1 621 to 1657. His celebrated 
pillared hall, or choultry, at that city is 333 feet long and 105 feet wide, with 
four ranges of columns, all different, and all most elaborately sculptured. 

Fergusson's The facade of this hall/ Fergusson observes, 'like that of almost all the great halls 
criticism, in the south of India, is adorned either with yalis — monsters of the lion type trampling 
on an elephant — or, even more generally, by a group consisting of a warrior sitting 
on a rearing horse, whose feet are supported on the shields of foot-soldiers, sometimes 
slaying men, sometimes tigers. These groups are found literally in hundreds in Southern 
India, and, as works exhibiting difficulties overcome by patient labour, they are un- 
rivalled, so far as I know, by anything found elsewhere. 

'As works of art they are the most barbarous, it may be said the most vulgar, to be 
found in India, and do more to shake one's faith in the civilization of the people who 
produced them than anything they did in any other department of art. Where these 
monstrosities are not introduced, the pillars of entrances are only enriched a little 
more than those of the interior, where the ornamentation is in better taste, and generally 
quite sufficiently rich for its purpose.' 2 

Fergusson's criticism fails to give the Southern sculptors due credit for 

1 Ann. Progr. Rep. A. S. Madras and Coorg, 2 Hist, oflnd. andE. At chit. (1899), P- 3 6 3J ed - 
1604-5, p. 44. 1910, vol. i, p. 389. 

A. Siva: cave-temple (Pallava). Trichinopoly. 7th century 

I*. Siva and I arvati; on north wall of great temple (Chola). (Janpaikonda- 
Cholapuram, Trichinopoly. nth century 


A. Siva and Parvati; on north wall of threat temple (C hola). 
Oanu;aikonda-Oholapurain, MVichinopoly . iith century 

B. Siva. Darasuram, Tanjore District. 16th century 

PI-ATE 90 


their power of expressing vigorous movement, and, in my judgement, is too 
harsh. Such figures appear to be unknown elsewhere, and it is not apparent 
how they became so much favoured in the Tamil country. Fergusson prob- 
ably was right in his suggestion that the rampant horses, yalis, and heavy 
cornices with double curvature, characteristic of the Dravidian temples in the 
South, were derived from primitive terra-cotta forms. 1 

The Southern sculpture, remarkable, as already observed, for its enormous Character 
quantity, fantastic character, often degenerating into the grotesque, and mar- ° f lat ^ 
vellous elaboration, rarely, if ever, exhibits the higher qualities of art. The sc ^p t u™ 
sculptures being designed to be viewed in the mass, not as individual works, 
reproductions of a few separate figures cannot do full justice either to the 
sculptors' intention or to the general effect. But, subject to that caution, 
a few specimens may be cited to give some idea of the style. The best of this 
class of work dates from the seventeenth century, while the most recent is 
the worst; indeed, modern figure sculpture, as a rule, hardly deserves to be 
called the work of artists. 

Examples of seventeenth- and eighteenth-century sculpture might be multi- Examples, 
plied indefinitely. Selected specimens from buildings in the Madura District 
will suffice as typical illustrations (Plate 93). One of the best images among 
the crowd at Tirumal Naik's choultry (1623-65) is that of Siva in an unusual 
attitude as a supplicant to some other deity. The effigy of the woman holding 
a doll-like baby, from the Great Temple at Madura, is welcome as introducing 
a rare touch of human sentiment, but is far inferior to the treatment of 
a similar subject at Puri (ante, Plate 77 a). The blotchy appearance of the 
photograph is due to the whitewash or paint with which the statue has been 
smeared. The sculptures from the Ramesvaram temple are somewhat later, 
dating from the close of the seventeenth or the beginning of the eighteenth 
century. The image of the female carrying a male deity on her back is 
characteristically grotesque. It too has been smudged with paint or white- 
wash. The modelling of the woman is not destitute of merit. 

The capabilities of modern sculptors in the South are best proved by the Modern 
decorations of the new palace in the town of Mysore, executed to the order ^^ptuw. 
of H.H. the Maharaja and described and illustrated by Mr. A. Rea. Skill is 
not confined to the members of any one caste, and the Maharaja has been 
willing to employ capable men from any district. The material used is some- 
times soapstone and sometimes stone of considerable hardness. The soap- 
stone is employed in fairly large masses, a clever figure of Vishnu, for instance, 
being two feet in height. The drapery of that figure looks as if it had been 
imitated from photographs of Gandhara work. The style throughout is 
frankly eclectic and imitative, and it is obvious that the artists have studied 
models of various periods and schools. One decorative motive is admittedly 
1 However, the use of large terra-cotta figures to decorate gateways, &c, is comparatively modern. 


borrowed from a picture by Ravi Varma, and the more direct influence of 
modern European art can be clearly traced. A relief representing the marriage 
of Rukmini looks as if it had been suggested by study of photographs of the 
Borobudur bas-reliefs. Some of the female figures are very pretty. Artisti- 
cally, the best things are certain decorative soapstone panels wrought with 
floral and other designs, thoroughly Indian in character and of first-rate 
quality. 1 
South Indian Many images cast in copper by the cire perdue process exist and also a few 
brass and castings in brass. In modern times casting in brass has been carried on 
copper m ainly in Mysore and Western India, and not in the South. The better 
cas mg . S p ec j mens £ these castings seem to range in date from the twelfth to the 
eighteenth century. The modern work is usually on a small scale and of 
very poor craftsmanship. 
Vijayanagar Exceptional interest attaches to the brass images reproduced in Plate 94, 
brass w hich are certified by inscriptions on the shoulders to be portraits, apparently 
statuef contemporary, of Krishna Raya, the famous king of Vijayanagar in the early 
" years of the sixteenth century, and two of his queens. They stand inside a 
temple on the sacred hill of Tirumalai or Upper Tirupati, and were photo- 
graphed by a high-caste Hindu, no European or Musalman being permitted 
to enter any temple on the hill. The town of Tirupati is famous for the skill 
of its workers in brass. 2 The images, although formal in design, are defective 
in expression. 
Siya Numerous figures of Siva Nataraja exist, some of which have been 
Nataraja. illustrated in the works of Dr. Coomaraswamy and Mr. Havcll. The figure 
lent by Lord Ampthill to the Indian Museum, South Kensington, which was 
shown at the India Section of the Festival of Empire in 191 1, is perhaps the 
finest of all (Plate 106). The explanation of the symbolism of these repre- 
sentations of the dancing god will be reserved until the Ceylonese examples 
are discussed. 
The date of The best of these images, such as the Nataraja, described above, are 
Southern directly comparable with Pallava and Chola sculpture, and are probably pre- 
castinT e ^ event ' 1 century. It is very difficult to date the later works. As a whole, the 
" scale of the castings is very much reduced. The jewellery and costume is 
also over-emphasized, the waist-cloths of the goddesses being shown round 
the legs and not merely indicated by tooling on the legs. Many of the large, 
early figures are fitted with rings at the base for transport in processions. 

1 A. Rea, Monograph on Stone Carving and In- India, 1902-3, p. 227; citing Hultzsch (Progr. 
laying in Southern India, with thirty-one plates; Report, 1903, in Madras G. O. Public, Nos. 655, 
Madras Government Press, 1906; quarto in 656, dated 24 July 1903). Dr. Hultzsch's recom- 
paper covers. The half-tone blocks cannot be mendation to have the images photographed by 
reproduced. Some of the best objects are shown a high-caste Hindu was carried out by the 
in Pis. XXIV-XXVI. Survey in the following year, but no description 

2 Imp. Gaz., s.v. Tirumala ; Annual Report, A.S., of the statues was recorded. 






PLATE 92. Southern Gopuram. Great Temple, Madura. 17th century 


Siva; Tinmuil Naik's 
choultry, Madura 

■I B 

-V Woman and hahv ; (Jrcat I emple, 

if Madura 


Female carrying male deity ; 
Kamesvaram temple 


PI VTF 04 Cast brass portrait images of Krishna Raya of \ uayanaga 
£i i?£ 20 and hisQuecns; in the Sri Nivasa IVruma 1 empk- on 
( the hill of Tirumalai near Tirupati, N . Arcot District 


The later figures, for the most part on a small scale and much tooled, are 
probably to be associated with Tanjore. 

The image of Parvati, now in the Boston Museum (Plate 95 A), is not very 
dissimilar in style from the Polonnaruwa bronzes, and may, perhaps, date 
from about the same period, the twelfth century. It is well modelled ; the 
hands are specially good. 

Chapter Eleven 


The two fT-^HE principal architectural remains in Ceylon are found at the two most 
capitals. I notable of the ancient capitals, namely, Anuradhapura and Polon- 
A naruwa, both situated in the North Central Province. The former 
city, a royal residence for more than a thousand years, was superseded in the 
eighth century by Polonnaruwa, the glory of which lasted, with interruptions, 
until the early years of the thirteenth century. The antiquities, therefore, 
belong to two widely separated series. Those at Anuradhapura go back to 
the time of Asoka, but mostly date from the earliest centuries of the Christian 
era; whereas the most important buildings at Polonnaruwa were constructed 
during the second half of the twelfth century (a.d. 1153-97) in the reigns of 
Parakrama Bahu the Great and Kirti Nissanka Malla. 1 
Anuradha- Anuradhapura, when in its prime, was a city of colossal proportions, 'une 
pura. veritable Rome bouddhique', at least 8 miles in diameter, and crowded with 
magnificent buildings. After the removal of the court everything went to 
ruin, but many edifices were repaired and restored by Parakrama Bahu, to 
whose energy the splendours of Polonnaruwa also are largely due. After his 
death the ancient capital again became desolate, and remained buried in 
dense forest until recent times. During the last forty years the ruins have 
been systematically and efficiently explored, with the result that the principal 
remains have been exposed, mapped, and more or less completely described. 
Dagabas. The most conspicuous structures are the great Buddhist dagabas (stupas), 
far exceeding in dimensions anything of the kind now standing in India. That 
commonly called the Jetawanarama, still 251 feet high, stands on a stone plat- 
form nearly 8 acres in extent, while the space included within the walled 
enclosure measures nearly 14 acres. The Abhayagiri dagabas, almost equal 
in mass, is said to have been originally erected in the first century B.C. The 
earliest, the Thuparama, built in the days of Asoka, has been covered up in 
recent times, like most of the others, by later additions. 2 

1 Polonnaruwa, alias Kalingapura, or Pulasti- Ruined Structures. Measured, drawn, and de- 
pura, the modern Topavewa or Topawa, repre- scribed by James G. Smither, F.R.I.B.A., late 
sents a much more ancient city, Wijitapura, of Architect to the Government of Ceylon. Sixty- 
which some remains seem to be traceable seven Plates. Published by order of the Ceylon 
(Parker, Ancient Ceylon, pp. 239-41). For the Government (Atlas folio, N.D.). The drawings, 
dates of the medieval kings see Epigraphia finished in 1877, were not published until 1894. 
Zeylanica, vol. i, p. 156. The traditional date Mr. Parker (Ancient Ceylon, p. 300) gives good 
for the foundation of Anuradhapura is 457 B.C. reasons for believing that the real Abhayagiri is 
Polonnaruwa was abandoned finally in a.d. 1240. now miscalled the Jetavana. The true Jetavana, 

2 See Architectural Remains, Anuradhapura, Cey- according to him, stands to the east of the Sela 
Ion: comprising the Dagabas and certain other Chaitya. 














A. Stone railing at Anuradhapura, as restored 

B. Siva temple Xo. i, at Polonnaruwa, west wall 

C. Stucco reliefs on porch of the Hetadage, Polonnaruwa 



The dagabas, huge masses of masonry, wonderful as stupendous monu- other sacred 
ments of laborious engineering, are not in themselves interesting as examples buildings. 
of architectural art. The work of the artist must be sought in the numerous 
and splendid associated buildings. The stone railing never attained in Ceylon 
the development which in India made it the vehicle for much of the highest 
art of the country. The only considerable example in the island, situated at 
Anuradhapura, and discovered and rebuilt by Mr. Bell, was a well-designed 
structure of uncertain date, perfectly plain, as at Sanchi, except for sculptured 
guard-posts at the entrance. It surrounded a rectangular pillared hall, not 
a dagaba (Plate 96 a). 1 The monasteries and temples connected with the 
dagabas included every variety of edifice needful for the accommodation of 
thousands of monks and for the ritual of a highly ceremonial religion. 

Mr. Bell's description of the Vijayarama at Anuradhapura, erected in or a typical 
about the eighth century for the use of a community of Tantric Mahayanist monastery. 
Buddhists, will serve to give a notion of the form and extent of an early 
monastic establishment of the more important kind in Ceylon. 

'Here existed', he writes, 'a typical sangharama, or Buddhist establishment, perfect 
in itself, with its shrines and meeting-hall, its priesdy residences, bath-house, store- 
rooms, ponds, &c. 

Broadly, the monastery consisted of a raised quadrangle, 288 feet north and south by 
268 feet east and west, walled, with entrances at the cardinal points, enclosing a dagaba 
and three vihares [temples], and having an open hall attached to the north. Outside 
this temenos was first a walk, then twelve annexes, evenly grouped, surrounded by a 
moat, with the chief pansala [monks' residences], a bathing-house, and a few other 
buildings on the south and west; the whole covering an area of 12$ acres, bounded by 
a quadrangular wall of stone, 200 yards by 300 yards, traces of which may still be 
seen. From the lodge (mura-ge) a broad street led straight to the inner quadrangle.' 2 

It would be difficult to point out the ruins of an Indian monastery equally 
extensive. The unlimited field for the exercise of the painter's and sculptor's 
arts presented by such a mass of buildings was sedulously cultivated. 

The Buddhist temples in Ceylon, differing widely from Indian models, Temples, 
ordinarily were rectangular buildings of either brick or stone, approached 
through a vestibule, and sometimes with only a single entrance, but often 
with four entrances facing the cardinal points. They were frequently arranged 
quincunx fashion in groups of five, four small shrines being placed symmetri- 
cally round a larger central one. 

Shrines of the Hindu gods find honoured places among the Buddhist 
buildings, Vishnu, for instance, being regarded as the protector of Ceylon, 
and worshipped in subordination to Buddha. Hindu temples intended for 

1 Full details in Ann. Rep. A. S., Ceylon, for 1892 * Am. Rep. A. S., Ceylon, for 1891 (xxxvi, 

(xxxvii, 1004), p. 1. As a decorative pattern the 1904), p. 4. The existence of Tantric Maha- 

railing was familiar in the island (see Anc. Ceylon, yanist Buddhism in Ceylon deserves special 

p. 278). notice - 


Brahmanical worship, as practised by the Tamil invaders, also exist. One 
illustration of such a temple at Polonnaruwa, dedicated to Siva, and dating 
probably from the eleventh or twelfth century, may be given to show how 
far the Ceylonese Hindu buildings resemble the South Indian Chola types. 1 
The Tivanka Vihare at Polonnaruwa, built by Parakrama Bahu, and generally 
miscalled the Thuparama, has a high pyramidal roof in Dravidian style, and, 
generally speaking, the Polonnaruwa buildings have a distinctly Dravidian 
character, but the huge Dravidian gateways (gopuram) are unknown in 
Stucco The basements are sometimes adorned with relief figures in stucco of some 

reliefs. mer i t (Plate 96 C). 

Circular Circular temples or shrines, of which three notable examples are known, 
shrines. are tfe most original and peculiar of Ceylonese buildings. That at Polon- 
naruwa, erected by King Nissanka Malla at the close of the twelfth century, 
is considered by Mr. Bell to be 'the most beautiful specimen of Buddhistic 
stone architecture existing in Ceylon'. He declares that 'no photographs or 
drawings can adequately reproduce, nor can words but faintly outline, the 
inexpressible charm' of the inner platform. The structure, about 80 feet in 
diameter, is circular, standing on a terrace, also circular, and 125 yards in 
diameter. It was intended for the reception of the tooth-relic. The centre was 
occupied by a small dagaba surrounded by sixteen statues, and two concentric 
circles of granite columns, twenty and sixteen in number respectively. The 
entrance was through a portico on the north-east. The elaborate decoration 
was lavished chiefly on the stylobate of the inner platform and on the staircase. 
A portion of the exterior is shown in Plate 97 a, and the western stairs in 
Plate 97 b. 2 
The Medi- A second and earlier building of the same class has been discovered at a 
rigiriya place called Medirigiriya in the Tamankaduwa District, North Central 
examp e. p rov j nce> hidden in the heart of the forest, six miles from the nearest village. 
It stands on the highest point of a mass of rock, and like its fellow at Polon- 
naruwa is surrounded by a slab wall, carved with surface ornament. There 

'three concentric rows of graceful columns (sixty-eight in all) of the type seen at 
Thuparama and Lankarama, Anuradhapura. The inner and second row[s] of pillars 
bear single lions and pilasters on their capitals, the outermost [bears] posturing ganas 
(dwarfs). In height this row of columns is but 9 ft. 9 in., while the two inner rows 

1 Fully described and illustrated in Ann. Rep. extension of the eastern wall of the enclosure of 

A. S., Ceylon, 1907, pp. 17-24, 36, Pis. XVI- this temple, to which they evidently belonged. 

XIX, and plan. This purely Brahmanical build- * The building is fully described and illustrated 

ing is locally miscalled the Dalada Maligawa, or by half-tone blocks in Mr. Bell's Ann. Rep. for 

'Shrine of the Tooth Relic'. The fine Hindu 1903, 1904, and 1907 (Sess. Papers LXV, LXVI 

bronzes described post, chap, vii, sec. 7 ft, were of 1908, and V of 191 1). It has been extensively 

found by digging a trench outside a southern restored by the replacing of fallen members. 

A. Circular shrine (zvata-da-ge) at Polonnaruwa ; part of north-eastern quadrant 

15. The same; western stairs 

C. Circular shrine (wata-da-ge) at Medirigiriya, N.C.P. 


A and B. Capitals at Abhayagiri Viharc, Anuradhapura 
C. Column in Ruwanveli area, Anuradhapura; 9 ft. 4 in. high 



reach 16 ft. All are octagonal, and all are unbroken, save four; but several have lost 
their spreading capitals. Within the circle of pillars, seated on an asanaya, is a Buddha 
in stone; probably one of four cardinally placed, with their backs to a small central 
dagaba. The design on the stone slab wall encircling the dagaba and columns is the 
"Buddhist railing" pattern, in this differing from the flowered ornamentation of the 
Polonnaruwa "Wala-da-ge".' 

An inscribed pillar close by was erected in the third year of King Kasyapa V 
(929-39), which may be taken as the date of the building. 1 Plate 97 c shows 
the best-preserved part of the enclosure. 

The third example, discovered in 1894, t0 the north of the g reat Toluvila The 
monastery at Anuradhapura, is of small size, with an enclosure 37 feet in Toluvila 
diameter, surrounding a miniature dagaba with a diameter of only 8 feet, and exam P le - 
two concentric rings of slender columns. 2 

Such concentric circles of detached, slender, monolithic columns are a Circles of 
characteristic feature of Ceylonese architecture. They occur, in addition to columns, 
the examples already cited, at the Thuparama and Lankarama dagabas of 
Anuradhapura, as well as at the Ambusthala dagaba of Mihintale, distant eight 
miles from the early capital. Their purpose has been much discussed. Mr. 
Smither has demonstrated that those at the Thuparama could not have 
carried a roof of any kind. 3 It is possible that in some cases they may have 
been used to support sacred Buddhist symbols, but ordinarily, as Mr. 
Parker argues, those round the large buildings appear to have been intended 
primarily as a barrier against evil spirits, and secondarily to support festoons 
of lamps suspended on great occasions. At the wata-da-ge shrines, according 
to Mr. Bell, the pillars were intended to 'hold up a roof to shelter the small 
stupa and worshippers at the shrine'. 4 The forms of shaft and capital, differ- shafts and 
ing widely from Indian types, are illustrated on a larger scale on Plate 98. capitals of 
But it is impossible to go into detail here, or to discuss the age and evolution columns - 
of the various types. Mr. Parker supposes the Thuparama columns to date 
from the period between 100 B.C. and a.d. ioo. 5 

This necessarily slight notice of architecture in Ceylon may be concluded Sat Mahal 
by mention of a unique building at Polonnaruwa known as the Sat Mahal Prasadaya. 
Prasadaya, a seven-storied square brick tower, built in diminishing stages, 
and rising from a low basement, which measures 39 feet 2 inches each way 
at ground level. The brickwork was covered with fine lime plaster, probably 

1 Bell, Ann. Rep. for 1897, p. 7; for 1907, Pis. Toluvila. These seem to be different from the 
XXVIII, XXIX (Sess. Papers XLII, 1904; V, wata-da-ge type described in the text. 

191 1). J General de Beylte maintains that the Thupa- 

2 Ann. Rep. A. S., Ceylon, 1904, p. 2 (Sess. rama columns supported 'un toit k Tindienne, a 
Paper LXVI of 1908). In Ann. Rep. A. S., etages superposes' (V Architecture Undone en 
Ceylon, 1907, p. 3, Mr. Bell notes the existence ExtrSme-Orient, Paris, 1907, p. 361). 

of six small circular brick shrines (wata-geval) 4 Ancient Ceylon 9 p. 289; Ann. Rep. A. S. 9 Ceylon, 

at the Vessagiriya Monastery, Anuradhapura, 1904,^.2. 

besides one at the Abhayagiriya and one at the s Ancient Ceylon, p. 268. 

3>9« u 


once coloured, and twenty niches contained as many stucco statues, eleven of 
which still exist. The edifice was erected by order of King Nissanka Malla 
a little before a.d. 1200, in imitation of Cambodian models, and probably for 
the use of the Cambodian mercenaries then in the service of the Ceylonese 
monarch, 1 
Ceylonese Ceylon is rich in sculpture of many kinds, beginning probably from the 
sculpture ear jy ce nturies of the era. Fergusson's belief that the 'almost total absence 
a un ant. ^ sculpture' was one of the most striking peculiarities of Ceylonese art has 
been disproved abundantly by the fruitful researches of the Archaeological 
Department. But it is extremely difficult to affix dates, even approximate, to 
the numerous specimens of the Ceylonese sculptors' skill. Dated dedicatory 
inscriptions, so common in India, are rare in the island, and the principal 
monuments have been subject to such extensive alterations at various times 
that it is almost impossible to distinguish the sculptures of different periods. 
It is possible that when systematic study shall be applied to the local styles of 
art closer discrimination will be feasible, but in the present state of knowledge 
anything like accurate chronological classification of the sculptures of Ceylon 
is unattainable. The brief discussion of the subject which limits of space 
permit will be arranged under two headings, Early and Medieval; the former 
comprising everything up to about a.d. 700, and the latter everything later. 
Ceylon has not produced any noticeable modern sculpture. Mr. Hocart, the 
present Commissioner, is engaged upon an analysis of the sculptures which 
bids fair to solve these problems. An epigraphist has also recently been 
Inferior in The general impression on my mind is that, with the exception of some of 
quality to th e colossal statues, the bronzes, which are very good, but may have been 
n ian * cast in India, and a few other works, the production of the island sculptors 
is by no means equal to that of the best artists on the mainland. The style 
is Indian, with a difference. We must remember that many of the Ceylonese 
images were originally plastered and coloured, and that the rough, weather- 
worn blocks now visible do not produce the effect designed by the artists. 
Stelae. The highly decorated stelae at the entrances to chapels connected with the 
great dagabas are characteristic of Ceylonese art. The examples chosen from 
the Abhayagiri dagaba at Anuradhapura may be assigned with considerable 
probability to the time of King Gajabahu I, in the second century of the 
Christian era, but it is possible that they may be later, or even earlier. The 
flora] patterns differ widely from those used in the medieval stelae of Polon- 
naruwa. The devices springing from vases (Plate 99 c) recall many examples 
of the same motive in Alexandrian and Indian art. 
The human figures in panels have a general resemblance to those at Sanchi, 

1 Ann. Rep. A. 5., Ceylon, 1903, pp. 14-16, Pis. XIII-XV (Sess. Paper LXV of 1908). See also 
Ann. Rep., 1906, p. 17. 




PLATE 99. Sculptured stelae at Ahhayagiri dagaba, Anuradhapura 

B. Napa door-keeper, Ruwanveli vihare 

A. North stele, east chapel, 

PLATE 100 


but are more advanced in style. 1 The dwarf in the Atlas pose may be noticed 
in Plate 100, Fig. A. The seven-headed Naga or cobra shown in Fig. B is 
a good example of an art form extremely common in Ceylon, and usually well 
sculptured; the number of heads varies, nine being the maximum. Door- 
keepers intended to ward off the attacks of evil spirits were deemed essential 
for most Ceylonese buildings. This Naga at Ruwanveli (Plate 100, Fig. B) is 
a good example. 2 Ugly dwarfs were regarded as very effective janitors. The 
specimens from the Ruwanveli and Jetawanarama dagabas (Plate 101, Figs. 
A, B) are typical. They may be compared with the somewhat similar figures 
on the capitals of the western gateway at Sanchi, but are much later. 3 Fig. C 
in the same plate is a characteristic example of the small grotesque figures 
used decoratively in Ceylonese art. Like Gothic gargoyles, they are cleverly 
done, though ugly, and very like the Badami dwarf -friezes. 

Portrait statues supposed to be those of ancient kings are said to be a Reputed 
speciality of Ceylonese art. Mr. Smither has described two battered examples g atues of 
which seem to be of high antiquity. One of these, traditionally believed to ngs * 
represent King Devanampiya Tissa, the contemporary and friend of Asoka, 
which was found near the Ambusthala dagaba at Mihintale, eight miles from 
Anuradhapura, may be correctly attributed by the popular voice. It is 
described as follows: 

'The stone was in four pieces, but these have been put together and the statue 
placed erect on its circular base. The figure, which is 6 feet 5 inches in height, originally 
stood facing the dagaba, and doubtless in a devotional attitude; the arms, however, 
are broken off close to the shoulders and cannot be found. The king is clothed in the 
"dhoti", or waist-cloth wrapped round the loins and falling to the ankles, the upper 
part of the body being uncovered. The head-dress consists of a plain and slightly 
elevated pear-shaped cap, encircled by a jewelled band, or diadem; the ears are adorned 
with pendant ear-rings, and the neck with a jewelled neck-piece. The base is carved to 
represent an expanded lotus-flower, and is precisely similar in design to that found at 
the Thuparama dagaba. Both statue and base are much weather-worn, although origin- 
ally sheltered beneath a covered structure of which three stone octagonal pillars, for- 
merly surmounted by capitals, are the only remains.' 4 

The second example is the reputed portrait of King Bhatika Abhaya 
(Batiya Tissa), who reigned during the first century of the Christian era. It 
was found near the Ruwanveli dagaba, and has been set up, after undergoing 
repair. The material is hard dolomite, much weather-worn, and the height 
is about 8 feet. The dress of the figure resembles that of another statue com- 
monly believed to represent King Dutthagamini, which stands on the terrace 
of the Ruwanveli dagaba, and has been published by Mr. Havell. It seems 

1 Third century (?). In India these figures can and also the pose, 
only be compared to Kushan art. 3 Seventh century (?). 

2 Seventh century (?). The costume is medieval 4 Smither, Anuradhapura, p. 11. 



probable that these works represent saints or religious teachers rather than 
kings. 1 

A curious collection of eight life-size images on the embankment of a tank 
at Minneriya, N. C. P., is popularly believed to represent King Mahasena 
(c. a.d. 300) with his wives and courtiers. The images obviously are ancient, 
but too much injured for appraisement as works of art. 2 
Buddhas. Large and often colossal images of Buddha, seated, standing, or recumbent, 
are numerous in the island, some of which undoubtedly must be very ancient. 
One of the oldest, probably, is a battered seated figure at Tantrimalai, which 
wears a conical cap, and is believed by Mr. Parker to date from about the 
beginning of the Christian era. 3 

One of the best Buddhas of early age is the now well-known image from 
the Toluvila ruins, Anuradhapura, represented in situ in Plate 102 B, with 
a native seated beside it in exactly the same attitude. The photograph helps 
the European reader to realize the facts on which the forms of the canonical 
images are based. 
The Kapila I think that I am right in including among the early works a fine sculpture 
of uncertain date, proved by Dr. Coomaraswamy to represent Kapila, a 
legendary sage (Plate 102 a). It is cut in rather high relief on the face of the 
rock on the right-hand side of the Isurumuniya Vihara at Anuradhapura, 
where many other notable works of sculpture exist. They appear to be of 
various ages and to deserve more attention than they have received. 

The subject is a man curled up in the attitude technically described as 
c kingly ease* (maharaja lila)> with his left hand resting on the seat, and his 
right hand extended over the raised knee, holding the halter of a horse, the 
head of which appears on the rock, but is not included in the photograph. 
The man's head, covered with thick hair, is partly turned towards the proper 
left and averted from the horse, which he seems to ignore. The expression 
is that of calm and abstracted but not unconscious dignity, while the difficult 
pose is modelled with consummate skill and yet with perfect simplicity. 4 

The legend, as told in the Ramayana, may be briefly summarized as fol- 
lows: Sagara, King of Ajodhya, had by his queen Sumati 60,000 sons, whose 
impiety was such that the gods complained to Vishnu and the sage Kapila. 
King Sagara, having undertaken to perform the rite of the horse sacrifice 

1 Smither, op. cit., p. 11 ; Havell, Indian Sculp- 4 Bell, A. S. Rep., 1906, p. 8. Neither Mr. Bell 
ture and Painting, PL XII; Parker, Ancient nor Mr. Cave mentions the Kapila relief, the 
Ceylon, Fig. 72. The dates of early kings of merit of which was first recognized by Dr. 
Ceylon are uncertain ; Bhatika Abhaya is assigned Coomaraswamy. The critical opinion expressed 
to a.d. 42-70, and Dutthagamini to 106-84 B.C. in the text is confirmed by Mr. Lawrence 

2 Ann. Rep. A. S., Ceylon, 1893, P« I0 "> photo- Binyon, who holds that 'the rock-carved "Ka- 
graphs A. 344-7, C. 806. pila" in Ceylon is a tremendous work, impossible 

3 Parker, Ancient Ceylon, pp. 219, 244; Bell, to forget when once seen' (Sat. Rev., 18 Feb. 
A. S. Rep., 1896, p. 8. x 9")» 

A. Dwarf door-keeper, Ruwanveli 

B. Dwarf right-handed door-keeper 

of south porch of west chapel 

of Jctawanarama 

fr***^**^ $9 &F T %^ ^™*^± 



C. Part of dado, Rirwanveli dagaba, vihare 
PLA'IE 1 01 

A. Kapila relief, Isurumuniya, Anuradhapura 

B. Seated Buddha [5 ft. 9 in.] in situ at Toluvila, Anuradhapura; 
now in Colombo Museum 

PLATE 102 

A. Seated Buddha [6 ft. 9 in.]; Pankuliya 
Yi hare, Anuradhapura 

B. Seated Buddha; from Yiharc, No. 2, Polon- 
naruwa ; now in Colombo Museum 

C. Colossal Buddha at Awkana, N.C.P. 
PLATE 103 

PLATE 104. Colossal statue of 'Ananda*, 


(asvamedha) in token of his universal sovereignty, deputed the duty of guard- 
ing the intended victim to his 60,000 sons, who failed in the trust committed 
to their charge, and allowed the animal to be carried off to the nether regions 
(Patala). Their father having directed them to recover the horse, they dug 
down and down until they found him grazing in Hades, with the sage Kapila 
seated close by, and engaged in deep meditation. The princes menaced him 
with their weapons, but were reduced to ashes by the flames which darted 
from his person, when he turned his glance upon them. 

This relief seems to me to be one of the most remarkable productions of 
Indian art, whether on the mainland or in the island of Ceylon. 1 

The 'moonstone', a semicircular slab placed at the foot of a staircase and 'Moon- 
carved elaborately in low relief, is specially characteristic of, although not stones -' 
absolutely peculiar to, Ceylonese art. The design is always based on the open 
lotus flower, the pattern being arranged in concentric circles. At Anuradha- 
pura, where some specimens may be very ancient, the standard arrangement 
is that of an outermost circle with the 'cobra pattern', resembling acanthus 
leaves in effect ; then a procession of quadrupeds in a fixed order moving from 
left to right— horse, elephant, humped bull, and lion; next, a belt of graceful 
foliage, which is followed by a row of sacred geese, while the central circles 
represent the lotus in bud, leaf, and flower. The animal symbolism is perhaps 
the same as that of the Asoka pillars. 2 

The notable statue of an aged bearded man cut in the face of a boulder to The 
the east of the Topaveva embankment at Polonnaruwa, popularly known as so-called 
the image of King Parakrama Bahu the Great, who reigned from a.d. 1153 Bahu 'statue, 
to 1 1 86, certainly is not what it is supposed to be. The figure, cut in gneiss 
(granite), and 1 1£ feet high, stands full face, fronting nearly south, in an easy 
attitude, with the right leg slightly bent. The costume is confined to a tall 
cap and simple loin-cloth held up by a band knotted in front. The hands 
support a model of a palm-leaf book {old) held across the body. The ex- 
pression of the face is grave, and the half-closed eyes look down upon the 
manuscript. A long rounded beard and drooping moustache add to the 
gravity of the countenance. These details are inconsistent with the popular 
attribution. Mr. Bell is of opinion that the book and the whole appearance 
and pose of the figure stamp it unmistakably as the portrait of a reverend 
religious teacher from the Indian continent. He suggests that the statue may 
represent an ascetic named Kapila, for whom Parakrama Bahu built a richly 
adorned dwelling. 3 

' For the identification see Coomaraswamy in * A good early example from the Dalada Mali- 

SpoUa Zeylanica, vol. vi (1909), p. 132. The gawa in Smither, PI. LVII. See also Tennent, 

legend is given in Dowson, Classical Diet, of Ceylon, 2nd ed., p. 619. 

Hindu Mythology, s.v. Sagara. Some description 3 A. S. Rep., 1906, p. n, PI. XI; Guide to 

of the shrines of Isummuniya is given in Cave, Colombo Museum (1905), p. 21 ; Ruined Cities of 

The Ruined Cities of Ceylon, pp. 47"9- c 9*« ( l8 97). P- "9- 


Seated Two seated Buddhas strike me as being excellent works and out of the 

Buddhas. comm0 n — namely, the colossal image at the Pankuliya Vihare, Anuradhapura, 
and the smaller image from Vihare, No. 2, Polonnaruwa (Plate 103 AandB). 
The characteristic points of each appear sufficiently from the photographs 
without detailed comment. Mr. Bell conjectures that the Pankuliya statue may 
date from the tenth century ; the Polonnaruwa image maybe two centuries later. 
Standing The largest statue in the island, and perhaps the most impressive, is the 

Buddhas. colossal standing image of Buddha at Awkana, N. C. P., 46 feet in height, 
including the pedestal. It is cut from the face of an enormous boulder,. prac- 
tically in the round, being joined to the rock only by slight support. Local 
tradition attributes the work to the reign of Parakrama Bahu. The expression 
of calm majesty is given successfully (Plate 103 c). A similar, and nearly as 
large, but less effective colossus, carved merely in high relief, and inferior 
in execution, stands at Saseruwa, N. W. P., and may be assigned to the same 
period. 1 

'Ananda.* The stately colossal standing image at the Gal-vihare, Polonnaruwa, popu- 
larly known, and apparently rightly, as that of Ananda, the disciple of Buddha, 
is one of the most imposing and interesting statues in Ceylon (Plate 104). 
The faithful attendant stands watching a colossal recumbent figure of his 
dying Master. 
The h t0 ? e ^° monument * n the island is more extraordinary than the gigantic 'stone 
bo ° • book' (gal-pota) at Polonnaruwa, a monolith brought from Mihintale, eighty 
miles distant, at the close of the twelfth century by Nissanka's 'mighty men', 
as recorded in a long inscription on its surface. It is nearly 27 feet long, 4 feet 
7 inches broad, and varies in depth from 1 foot 4 inches, to 2 feet 2 inches. 
The relief sculpture treats of the common Indian subject, elephants pour- 
ing water over Sri or Lakshmi — the goddess of good fortune. 

Bas-relief More artistic bas-reliefs of uncertain date occur elsewhere. Perhaps the 
scenes. most rem arkable is that at Pokuna (masonry tank) A, Anuradhapura, which 
vividly depicts elephants bathing, and then charging away when scared. The 
relief is so low that the photographs are not sufficiently distinct for successful 
reproduction. Mr. Bell describes this work, which is in two sections, as an 
absolutely unique piece of carving, and without exception the most spirited 
and life-like to be seen anywhere among the ruins of Anuradhapura. 2 It is 
supposed to date from the time of Parakrama Bahu. 

Ceylonese The remarkable richness of Ceylon in art- works of metal, chiefly copper, 
cratin^ was not rea ^ ze d unt ^ recent discoveries compelled attention to the fact. 
^' Before 1905 a few objects of interest had been collected by the casual exertions 

1 Bell, A. S. Rep., 1895, pp. 6, 12; Tennent, 2 Ann. Rep. Arch. S. t igoi, p.6; photographs A. 
Ceylon, 2nd ed., vol. ii, p. 604, with woodcut. 405, 406, C. 1304. 


of individuals, but since that date the numerous additions to the public col- 
lections have been acquired by the systematic operations of the Archaeo- 
logical Survey. Few, if any, of the castings are earlier than the tenth century, 
and most of them are a great deal later. As a whole there is little that is 
distinctive about them and it is better to consider them as one with the 
Southern Indian castings. 

Perhaps the most notable of the Ceylon bronzes is an image of the goddess Pattini Devi. 
Pattini Devi, found near the north-eastern coast somewhere between Trinco- 
malee and Batticalwa, and presented to the British Museum in 1830. It 
stands 4 feet 9 \ inches in height and is composed of a metal which looks like 
brass, but may be a pale bronze (Plate 105). It seems to have been originally 
gilt. The age of the work is doubtful. The cleverness with which the trans- 
parency of the skirt is shown recalls similar skill exhibited in the Gupta sculp- 
ture of the fifth century in Northern India, but it would be rash to attribute 
such an early date to the Ceylonese image for that reason only, and it is 
difficult to find any other test of its age. 

The nudity above the waist, which may offend the European eye, is in 
accordance with the ancient custom of Southern India and Ceylon, not wholly 
disused even in these days. The waist is rather too much attenuated, in con- 
formity with common Indian practice, examples of which may be found even 
in the Bharhut sculptures; but, except for that defect, the modelling is good, 
and the hands especially are admirable. 

Pattini is one of the most popular deities in Ceylon, and her worship is 
still kept up on the mainland also, whence it was introduced into the island, 
most probably in the reign of Gajabahu I at some time in the second century 
of the Christian era. The cult seems to have originated in the Chera territory 
(Coimbatore and Salem), but some of the legends connect its beginnings with 
a Pandya King of Madura. The goddess is considered to be the guardian 
of female chastity, and is also credited with power over epidemics, whether 
of man or beast. Two wooden images of her and her husband found in a 
cave at the Nikawaewa monastery are supposed to date from the eleventh 
century. The British Museum casting may be quite as early. 1 

Some good castings, believed to date from about the tenth century, have Bronzes 
been obtained from various localities within the area of the ancient capital, *"> m 
Anuradhapura. They include a pair of miniature feet apparently belonging to pu ^ 
a lost statuette, and only three inches in length, which are described as 'ex- 
cellently modelled'. 2 Like the great Buddha in the Birmingham Museum, 
they were cast on a core, in this case of iron. The best piece, from the aesthe- 

' The legend of the goddess is too long to quote, wooden images are figured in Parker, Ancient 

See The Tamilian Antiquary, No. 3 (1909). P- vii Ceylon, Fig. 272. 

note ; ibid., No. 5, p. 47 : and Dr. Coomaraswamy 2 Bell, Anuradhapura and the North-Central Pro- 

injf. R. A. S., 1909, p. 293, with references. The vince, 7th Progress Report (xiii, 1896), PI. XVII. 


tic point of view, is a statuette supposed to be that of a Bodhisattva (Plate 
107 a). The statuette, 20J inches high, was found to the south of the Thupa- 
rama. The person represented stands in the pose with a double bend, known 
technically as tivanka. The drapery is gracefully treated, the modelling, 
especially of the hands, is truthful, and the serene expression of the face is 
pleasing. 1 The style closely resembles that of some of the Polonnaruwa 
castings, which are ascribed to the twelfth century, and the Anuradhapura 
statuette may be as late. 
Bronzes The few figures collected at Polonnaruwa in 1906, forming the first series 
from Polon- j n ^ e Colombo Museum (Nos. 40-52), are not of much importance; but the 
naruwa. seconc j anc j third series, excavated in 1907 and 1908 from the Siva Dewale 
and neighbouring sites, may be fairly said to add a new chapter to the history 
of art in Ceylon. Nothing like them was known before, except the Anuradha- 
pura Bodhisattva, if that be the correct designation for it. A few of the best 
have been selected from a set of good photographs taken by Dr. Andreas Nell 
and kindly supplied by the Government of Ceylon. The identification of the 
images has been effected by the Honourable Mr. P. Arunachalam. These 
figures, all massive and very heavy, are ascribed to the twelfth and thirteenth 
centuries. Their large size proves that the artists of those days knew how 
to overcome the difficulties of casting copper on a considerable scale, and 
gives the images an importance and dignity which cannot be claimed by 
miniature works a few inches high. In the opinion of Dr. A. Willey, F.R.S., 
late Director of the Colombo Museum, they 'are Polonnaruwa bronzes for 
better or for worse', and certainly were not imported from the mainland. 2 
But I am disposed to agree with Mr. Bell that they were executed in India. 
Siva The place of honour may be given to the spirited images of Siva as 
Nataraja. Nataraja, 'Lord of the Dance', the first of their kind to be found in Ceylon 
(Plates 106, 107 b); which compare favourably with the best examples of 
similar compositions in Southern India. A specimen in the Madras Museum 
arouses enthusiasm, which few can share fully, in the breast of Dr. A. K. 
Coomaraswamy, who first published photographs of the work. In order to 
make the Ceylonese bronzes intelligible, the explanation of the legend of 
Siva's manifestation as 'Lord of the Dance', given in the Koyil Puranam, and 
said to be familiar to all southern worshippers of the god, is quoted from the 
eloquent pages of the author referred to: 

The legend. 'Siva appeared in disguise amongst a congregation of the thousand sages, and in the 
course of disputation, confuted them and so angered them thereby, that they en- 
deavoured by incantations to destroy Him. A fierce tiger was created in sacrificial 
flames, and rushed upon Him, but smiling gently, He seized it with His sacred hands, 
and with the nail of His little finger stripped off its skin, which He wrapped about 
Himself as if it had been a silken cloth. Undiscouraged by failure, the sages renewed 

1 See Burlington Magazine, 1910, p. 87, PI. I, 3. 2 Spolia Zeylanica, Sept. 1909, p. 67 note. 

PLATE 105. Pattini Devi. Cast brass. Sinhalese, 
now in the British Museum 

PLATE I0(>. Siva Nataraja (3 ft.)- Copper casting. No. 1, from Polonnaruwa 

(Colombo Museum) 


their offerings, and there was produced a monstrous serpent, which He seized and 
wreathed about His neck. Then He began to dance; but there rushed upon Him a last 
monster in the shape of a hideous malignant dwarf. Upon him the God pressed the tip 
of His foot, and broke the creature's back, so that it writhed upon the ground; and so, 
His last foe prostrate, Siva resumed the dance of which the gods were witnesses. 

One interpretation of this legend explains that He wraps about Him as a garment, 
the tiger fury of human passion; the guile and malice of mankind He wears as a neck- 
lace, and beneath His feet is for ever crushed the embodiment of evil. More character- 
istic of Indian thought is the symbolism, in terms of the marvellous grace and rhythm 
of Indian dancing, the effortless ease with which the God in his grace supports the 
cosmos ; it is his sport. The five acts of creation, preservation, destruction, embodiment 
and gracious release are his ceaseless mystic dance. In sacred Tillai, the "New Jeru- 
salem", the dance shall be revealed; and Tillai is the very centre of the Universe, that 
is, His dance is within the cosmos and the soul/ 1 

The more prosaic description of the group by Mr, Arunachalam, slightly The 
condensed, will enable the student to appreciate the intention of the formal symbolism, 
symbolism. The god's hair is braided, forming a crown at the top and a 
circular coil at the back, the lower braids whirling in the dance, which is 
named Tandava. The mermaid on the right braid (indistinct in the photo- 
graph) symbolizes the Ganges; a crescent moon and serpent decorate the left 
braid. Other serpents coiled round his body are regarded as symbols of 
Siva's energy. His three eyes, one in the forehead, represent the sun, moon, 
and fire; the skull at the base of the crown is a symbol of destruction, and 
the necklace, composed of skulls of Brahmas, Vishnus, and Rudras, sym- 
bolizes the evolution and involution of the universe throughout the aeons. 
The bisexual nature of the deity is indicated by the long man's ear-ring in 
the right, and the woman's circular ear-ring in the left ear. Fire, a symbol of 
both destruction and divine purifying grace, is held in the left upper hand, 
and also surrounds the group. The small drum in the right upper hand is 
supposed to suggest vibration, the first stage in evolution. The right lower 
hand is raised in assurance of protection to the worshipper, while the left 
lower hand points to the uplifted foot, the refuge of the suppliant. The mon- 
ster trampled on personifies the powers of evil and illusion from which the 
deity delivers the soul. The composition as a whole is understood to represent 
the control of the operations of the universe by Siva. 

The greater part of the foregoing commentaries has nothing to do with the Limits of 
merits of the compositions as works of art. Any competent coppersmith can *^' s 
make to order rings symbolizing fire and other formal attributes in accor- 
dance with written rules, and such accessories, whether well or ill made, will 
be equally significant to the devout Hindu versed in the legends and meta- 
physics of his faith. The general lines of the principal image, too, are deter- 
mined by pattern sketches, of which Dr. Coomaraswamy has published a 

1 The Aims of Indian Art (pamphlet, Essex House Press, 1908). 

339a x 


specimen. Consequently, a perfectly correct group, with all the needful ap- 
paratus for edification, can be made passably well by any skilled bronze 
founder, whose work need not be anything higher than mere manufacture. 
The scope for the display of aesthetic feeling and creative skill, which dis- 
tinguish an artist from a skilful mechanic, is restricted almost exclusively to 
the manner of rendering the action of dancing with passion, including, of 
course, the modelling of the principal figure. When various examples of the 
treatment of the prescribed theme are examined and compared they will be 
found to differ widely according to the degree of artistic power possessed by 
the maker of each. Among good examples may be classed Dr. Coomara- 
swamy's favourite in the Madras Museum, the Tanjore specimen, and No. i 
from Polonnaruwa (Plate 106). The No. 15 Polonnaruwa image (Plate 107 b), 
without the ring of fire, is the most artistic of all. It is described as being 
'the best finished of all the bronzes', and is deserving of the care spent on its 
production. A third Polonnaruwa specimen (No. 24) is coarsely executed 
and of inferior quality. The same criticism applies to a second example in 
the Madras Museum, to another in the British Museum, and to the South 
Kensington image from Malabar. The 'belle statue de bronze ancien' in the 
Musee Guimet may be placed in the higher class. 1 

The standing image of Siva (No. 12), striking an attitude in another of his 
dances (sandhyanirtta), is gracefully posed, and well modelled, save for the 
excessive thickness of the arms. The figure of the Sun-god (No. 18), with 
a halo, holding a lotus bud in each hand, is dignified, and the type is unusual 
(Plate 108 b). One ideal of the goddess Parvati, consort of Siva, is expressed 
in No. 7, with the characteristic Indian bend (Plate 109 a). The image closely 
resembles that labelled as Lakshmi in the Musee Guimet {Petit Guide III., 
Plate p. 62). Another conception of Parvati (No. 20) is shown in Plate 109 b. 
The figure and pose are natural and pleasing. 
Images An interesting group of images deals with popular Tamil saints, whose 
of J*™} effigies have been identified by Mr. Arunachalam. Probably the best of this 
group is No. 16, representing Sundara-murti Swami, an apostle and psalmist 
of Siva about a.d. 700. He was a native of Tiruvarur, near Negapatam in the 
Madras Presidency; called to be an apostle on his wedding-day, hence dressed 
in the clothes and ornaments of a bridegroom. The clothes are somewhat 
scanty. The artist has rendered with remarkable success the attitude and 
facial expression of religious ecstasy powerful enough to tear away a bride- 

1 The S. Kensington bronze, two feet in height, of Higginbotham's edition, Madras, 1864), but 

found long ago at Chaoghat in Malabar, was pre- the engraving is not faithful, having been 'im- 

sented by Jonathan Duncan, Governor of Bom- proved* by the artist. For the 'belle statue* see 

bay, to the India Office Museum, and thence figure on p. 94 of the Petit Guide Illustri du 

has passed into the Indian Section of the Victoria Musie Guimet. The subject is often treated in 

and Albert Museum. It is engraved as PI. XIV stone sculpture, 
of Moor's Hindu Pantheon (i8io;=frontispiece 



groom from the side of his bride (Plate 108 c). The image has strong claims 
to be considered the finest of the Polonnaruwa bronzes, or, at least, to be 
placed second only to the Nataraja, No. 15. 

Certain small miscellaneous bronze images from Ceylon, of which the exact Statuette of 
find-spots are not recorded, are of sufficient interest to deserve special notice. Avalokites- 
A little figure, presumably that of the Mahay anist deity (if the expression be vara * 
allowed) named Avalokitesvara or Padmapani, only 3I inches in height, in 
Dr. Coomaraswamy's collection now at Boston, and ascribed to the sixth or 
seventh century, is regarded by him, and not without reason, as the best of 
all the Ceylonese images (Plate no a). He praises the 'perfection and ab- 
straction of the style', claiming that 'the divine ideal is fully realized both in 
expression and in physical form'. 

Another excellent little image, 3^ inches high, from the same collection, Jambhala, 
represents the minor deity Jambhala, or Kuvera, the well-contented god of or Kuvera - 
riches (Plate hob), whose effigy in various forms is frequently found in the 
ruins of Buddhist monasteries in India and Java. His right hand grasps a 
fruit; the left rests upon the mongoose, or ichneumon, sacred to him. Dr. 
Coomaraswamy's criticism is as follows: 

'The artistic interest of this figure lies in its frank realism, contrasting with the 
idealistic treatment of the figures so far referred to. The God of Wealth, far less 
remote and hard to reach than so exalted a being as a Bodhisattva, is worshipped for 
material rather than spiritual benefits ; he is represented as the very image of a fat 
trader seated in his booth awaiting customers. The patron saint of prosperity and 
trade is a comfortable, worldly person ! The realistic treatment of the firm flesh is as 
masterly in its own way as the generalization of the more ideal types, such as the 
Avalokitesvara.' l 

The Colombo Museum possesses many other bronze objects, including The Badulh 
several Buddhas. One of these (Plate hoc), a Buddha 'of unique design' Buddha, 
and uncertain date, found below Badulla, possesses considerable merit. 2 It 
belongs to the same early period as the two Boston figures described 

1 Coomaraswamy, 'Mahayana Buddhist Images No. xii, St. P&ersbourg, 1910). 'Jambhala of 
from Ceylon and Java',J.-R. -4. £., 1909, p. 288. Ceylon' was known even in distant Nepal 
That valuable article, and another entitled (Foucher, Iconographie bouddhique, PI. IX. 2). 
'Indian Bronzes' in Burlington Magazine, May 2 A rough list of bronzes and other objects is 
1910, discuss in detail many images which can- printed in the Catalogue of Finds, Archaeological 
not be noticed here. Kuvera (= Jambhala = Survey of Ceylon; deposited in the Colombo 
Vaisravana) was chief of the Yakshas. He was Museum, 1906-7, p. 27, supplied by the Govern- 
specially honoured in Khotan and Chinese ment of Ceylon. The Guide to the Museum 
Turkestan generally. A manuscript from Turfan (1905) is published in Spolia Zeylanica, Part IX. 
calls him 'the highest of the gods' (von Holstein , Badulla is in the hill country ; the image was found 
'Tisastvustik,' pp. 97, 122 note, Bibl. Buddhica, in the plain below. 


Indian The extensive and long-continued emigration from India to the Far East — 
colonies including Pegu, Siam, and Cambodia on the mainland, and Java, Sumatra, 
F *V he Bali, and Borneo among the islands of the Malay Archipelago — and the con- 
sequent establishment of Indian institutions and art in the countries named, 
constitute one of the darkest mysteries of history. 1 The reality of the debt 
due to India by those distant lands is attested abundantly by material remains, 
by the existence to this day of both the Buddhist and Brahmanical religions 
in the island of Bali to the east of Java, by Chinese history, and by numerous 
traditions preserved in India, Pegu, Siam, and the Archipelago. But when 
the attempt is made to transmute vague, conflicting traditions and imperfectly 
known archaeological facts into orderly history the difficulties in the way of 
success appear to be largely insurmountable. But, in order to render at all 
intelligible the fact of the existence of magnificent achievements of Indian 
art in Java, to which island the summary observations in this work will be 
confined, some attempt, however imperfect, at historical explanation is indis- 
pensable. In Java the forms of art are thoroughly Indian in subject and style, 
of high aesthetic quality, and sufficiently dated to permit of their correlation 
with the art of India. The less purely Indian and less meritorious ramifica- 
tions of Hindu art in the other countries of the Far East must be left un- 
Some It is certain that during the early centuries of the Christian era India 
traditional possessed an active and enterprising seafaring population on both coasts — 
ates ' that of the Bay of Bengal on the east, and that of the Arabian Sea on the west ; 
and it is highly probable that from the first to the eighth century emigration 
to the Malay Archipelago continued to proceed from both sides of India. If 
Javanese tradition may be believed, a large body of Indian emigrants, led by 
Aji Saka, landed in the island from the east of India in the year i of the local 
era, equivalent to a.d. 75 or A.D. 78 according to various computations, but 
the details of the story are obviously open to sceptical criticism. 

The observation of Fa-hien, the Chinese Buddhist pilgrim who visited 
Java in a.d. 414, and found 'plenty of heretics and Brahmans, but not enough 
Buddhism to be worth mentioning', 2 is excellent evidence that a strong Indian 
colony professing the Brahmanical religion must have been then already 
settled in the island for a long time. The statement made in a late Chinese 
work that an Indian colony arrived in Java during the reign of the Han 
emperor, Kwang Wu-ti (a.d. 25-57), is credible, although the authority on 

1 The Hinduized Javanese founded considerable tially with Giles. Fa-hien's statement is cor- 

colonies in Madagascar during the early cen- roborated by certain nearly contemporary in- 

turies of the Christian era (Journal Asiaiique, scriptions in Java and at Koetei in Borneo (Kern, 

1910, p. 330). 'Gedenkteekenen der oude indische beschaving 

1 Travels y chap, xl, in Giles's version. The other in Kambodja,' Onze Eeuw, 4 Jan. 1904, p. 46). 
versions (Laidlay, Beal, Legge) agree substan- 

1N °-")'""»'"<'*C,.l<.mb„ M 

H-ATE 107 

— o 


i O 
<tf Qh 




J '^ 

6/ ? 


s § 


£ 1 




<u c 


JC ~z 





















A. Parvati. [lit. i ft. Sin.] 
Polonnarwwa [No. 2}] 

B. Parvati. [lit. i ft. 4.I in.] 
Polonnaruwa [No. 20] 

PLATli TOC). Copper castings in the Colombo Museum 

A. Avalokitesvara, or Padmapani. 
Sinhalese. Boston Museum 

H. Jambhala,or Kuvera. Sinhalese. 
Boston Museum 

C. Seated Buddha. Copper casting. Colombo Museum 
PLATE 110 

Part III JAVA 157 

which it is based has not been found, 1 From the testimony of Fa-hien and 
other indications there is no doubt that Brahmanical Hinduism reached Java 
long before Buddhism . According to the Chinese History of the Sung Dynasty, 
the conversion of the island to Buddhism was effected by Gunavarman, 
Crown Prince of Kashmir, who had renounced his rank in order to become 
a monk. He then joined a monastery in China and died at Nanking in A.D.431 . 
This statement dates the conversion immediately after Fa-hien's visit. 2 

Javanese writers, supported to some extent by local traditions of Gujarat 
and Southern Marwar in Rajputana, affirm that in the year A.D. 603 a 
numerous body of colonists sailed from Western India to Java. 3 The Siamese 
annals record that in the year a.d. 685 (Saka 607) 

|great political disturbances occurred all over India, and the inhabitants, finding it 
impossible to make a living, were forced in large numbers to leave their home and 
country and settle among other nations. ... At that time four tribes of Brahmans, 
consisting of a considerable number of persons, made their way eastward from 
"Wanilara" to Burma, Pegu, then independent, the Laos States, Siam, and Cambodia/ 4 

Traditional dates like those cited notoriously require to be treated with 
caution, but in this case both the dates in the seventh century happen to be 
credible, as marking times of ascertained political disturbance in India. The 
earlier date, a.d. 603, which falls within the period of anarchy and strife due 
to the Hun invasions, precedes by a few years the consolidation of the empires 
founded by Harsha in the north and by Pulakesin II Chalukya in the Deccan. 
The later date, a.d. 685, approximately coincides with the fall of Valabhi, 
which is believed to have been destroyed about that time by the Arabs then 
settled in Sind. 5 The Chinese statement in the History of the Sung Dynasty 
dating the conversion of the island between a.d. 414 and 431 is the most 
trustworthy of all, though of course the assertion that the whole population 
was converted cannot be accepted. As in India, Brahmanical Hinduism con- 
tinued to exist side by side with Buddhism. The earliest known dated Indo- 
Javanese inscription is said to be one of the year a.d. 732.° We are, therefore, 
justified in believing that the ancient Indian Brahmanical colonies in Java 
received strong reinforcements from the mother-country during the fifth, 
seventh, and eighth centuries. Considering that all, or nearly all, the Bud- 

1 The Pilgrimage of Fa Hian (Calcutta, 1848), Mr. Sewell {J. R. A. S. y 1906, p. 421). Earlier 
p. 363, Laidlay's translation of Klaproth's note. Indian inscriptions not bearing precise dates 

2 de Beylie, U Architecture hindoue en ExtrSme- exist from the fifth century. In Cambodia the 
Orient, p. 335 ; Pelliot, Bull. E. F. E. O., iv. 274. earliest recorded Indian ruler, Srutavarman or 

3 A. M. Jackson in Bombay Gazetteer (1896), Kaundinya, lived in the middle of the fifth 
vol. i, Part I, App. century. In the following age Bhavavarman 

4 A. Steffen, art. No. 125, Man, 1902. 'Wani- founded many temples in honour of Indian 
lara' has not been identified. Quaere does lara= deities, especially Siva, at which daily readings 
Lata = Gujarat? of the epics and Pur anas were held. Indian 

5 A. S. W. /., vi. 3 ; ix. 4. influence was at its height in Cambodia in the 

6 According to the late Dr. Brandes quoted by sixth century (Kern, op. cit., p. 47). 




dhist remains in the island are later than the middle of the eighth century, 
we may further infer that the new-comers were largely Buddhist in religion, 
and included many skilled craftsmen. The most ancient objects in the island 
possessing value as works of art are Buddhist. The late Dr. Brandes, who had 
a good right to express an authoritative opinion, held that the buildings at 
Borobudur, with their incomparable sculptures, should be dated between 
a.d. 778 and 928 ( =700-850 Saka). According to M. Tissandier the Kali 
Bening and Sari temples at Prambanam (Brambanam) were begun in A.D.779. 1 
Other Indo- Javanese works, however, are much later, the Chandi Sewa 
temple, for example, being assigned to a.d. 1098. The Hindu kingdom of 
Majapahit in Eastern Java was overthrown by the Muhammadans in a.d. 1478, 
when the persecuted Hindus fled to Bali, where their descendants still practise 
Brahmanical rites, including sati (suttee) in its most appalling form, while 
another section of the population is Buddhist. 

From these facts it follows that the whole history of Indo- Javanese Bud- 
dhist art must lie between a.d. 420 and 1478, a period of more than a thousand 
years. The finest works may be assigned to the ninth century. 

In Java, as elsewhere, the late Mahay anist Buddhism so closely approxi- 
mated to Hinduism that sculptures which at first sight appear to be purely 
Brahmanical may be really Buddhist. 'Brambanam and Chandi Sewa\ Mr. 
Sewell observes, 'are to all external appearances purely Brahmanical, though 
we learn on examination that Brahma, Vishnu, and Siva were there held to be 
Bodhisattvas and not gods. And this is the case everywhere in Eastern Java, 
the temples being mostly Hindu in type (though always with a difference), 
and having statues adapted generally from Brahmanical originals/ 

The best known monument in the island is the vast pyramidal pile of Boro- 
budur, 'a hill in nine stages', combining the character of a stupa or dagaba 
with that of a temple. As an architectural composition the building, more 
than 400 feet square at the base, is of small account. Its importance in the 
history of art depends upon the immense series of about 2,000 bas-reliefs 
adorning the galleries, which, if laid end to end, would extend more than two 
miles. The best reliefs are the panels of the so-called 'second gallery', ex- 
ceeding two hundred in number, which are arranged in two series. The upper 
series presents in easily recognizable stone pictures the life of Buddha, as 
told in the ancient Sanskrit work the Lalita Vistara. The scenes of the lower 
series, artistically of equal merit, resisted interpretation until lately, but have 
now been proved to be illustrations of the Divyavadana and other Buddhist 
romances, including some of the Jatakas, or stories of the former lives of 
Buddha. About two-thirds of the 120 panels in that series have now been 
identified, and in time the balance probably will yield their secrets. The 
intention of the designer of the monument was that the worshipper, while 

1 Cambodge etjava, Paris, 1896, p. 126. 

PLATE III. Offerings to a Bodhisattva 

A. The Hodhisattva's descent to earth. H. The story of Prince Sudhana. 
Relief-panels in the first gallery, Borobudur, Java 

A. Sarasvati enthroned; from 
Jogyokaita, Java 

H. Prajna Faramita. Javanese 

PLATE 1 1 2 

Part III JAVA 159 

making his ritual perambulation (pradakshina) of the building, should be 
instructed ocularly in the whole doctrine of Buddhism, according to the 
system of the Mahayana, or 'Great Vehicle'. 1 

It is difficult to choose among the numerous beautiful reliefs of the 'second Example of 
gallery* of Borobudur. Several of the best have been reproduced by Mr. bas-reliefs. 
Havell and in the new edition of the History of Indian and Eastern Architec- 
ture. I select one from the lower series (Plate m). 

All critics can go so far as to concur in M. Tissandier's rather faint praise Criticism, 
that the bas-reliefs are 'motifs ciseles dans la pierre avec une puissance rare'; 
or M. Foucher's more liberal criticism that they are justly celebrated for 
their good proportions, naturalness of gesture, and the variety of attitude in 
the figures. But not everybody can agree with Mr. Havell that the reliefs 
exhibit 'supremely devout and spontaneous art', far excelling by their sim- 
plicity, unaffected naivete, artistic feeling, imagination, and magnificent con- 
ventionalism of the accessories the work of Ghiberti on the bronze doors of 
the Baptistery at Florence, which Michael Angelo declared 'to be worthy to 
be the gates of Paradise \ The same critic holds that the simple life led by 
the artists of Borobudur left them in peace to concentrate their whole soul 
on this work, and kept their minds free and able to listen to the voices of 
Nature and of their own inspiration — 'the soul of Nature speaking to the soul 
of man\ In reality, as M. Foucher truly observes, the immense processions 
of scenes at Borobudur have a 'caractere livresque' in virtue of their being 
illustrations of sacred story-books, which deprives them of the spontaneity 
and emotional (vibrant) expression that can spring only from contact with 
living oral tradition. The compositions were prompted, not by the 'voices of 
Nature', but by a business-like, systematic endeavour to give visual expression 
to set passages in favourite authors; and we have not the slightest reason for 
believing that the artists led particularly simple lives. We know, in fact, 
nothing whatever about them or their lives. A certain uniformity of effemi- 
nacy (mollesse) characterizes the forms, as it does some of the much earlier 
compositions of Gandhara. But, although it is true that the reliefs are care- 
fully planned and must be criticized as selected book illustrations rather than 
as the spontaneous utterance of simple souls in direct contact with nature, 
they are extremely good and charming. Fergusson thought that the art of 
the later cave-temples was 'nearly identical' with Borobudur. 2 As a whole 

1 The name Boro-Budur means 'the many Bud- encased in a structure of later masonry. The 

dhas* (cf. Sanskrit, Brihad-devata). The older literary works illustrated by the reliefs all belong 

books give erroneous interpretations. The build- to the Mula-Saroastivadin school of Buddhism, 

ing, although apparently a staged pyramid, is to which the seventh-century pilgrim, I-tsing, 

really constructed on the plan of a circular stupa, adhered (Foucher, 'Notes d'archiologie boud- 

all the angles being inscribed in circles. The so- dhique*, B. E. F. E. O., Janv.-Mars, 1909, 

called 'second gallery', designed to be the first, pp. 1-50). 

became the second when the original plinth was 2 Hist. Ind. and E. Arch. (1910), ii, p. 426. 


the Pallava and Chola sculpture of the South is nearer to the Javanese work. 

They deserve the most careful critical study by professional sculptors, who 

alone would be in a position to realize how much praise is due to artists 

capable of executing more than two miles of stone pictures, almost uniform 

in beauty and the display of technical skill of a high order. 

Numerous Notable sites, crowded with ancient buildings, are far too numerous in 

ancient j ava to be even named. The most important, perhaps, after Borobudur is 

cities. p ram banam (Brambanam), an early capital, where the temples are said to 

include six large and 150 small ones, supposed to date from about the tenth 


Detached The Javanese sculptures, in addition to reliefs, comprise multitudes of 

images and j ar g e detached stone images and small bronzes, of which only a small number 

ronzes. o ^ S p ec j mens can b e illustrated here. From van Kinsbergen's plates I select 

a very pleasing image of Sarasvati, consort of Brahma and goddess of speech 

and learning, who is represented enthroned. The mongoose or ichneumon 

is her special attribute (Plate 112 a). 

The other illustrations are from photographs kindly supplied by Dr. 
Coomaraswamy and already published by him, and also, in part, by 
Mr. Havell. 

The stone Buddha (Plate 113 a) is one of several similar images, nearly 
equal in quality, which exhibit the Indian yogi ideal in an exceptionally 
dignified and agreeable manner. The expressive modelling of the right hand 
deserves special commendation. 

Plate 1 12 B gives a side view of the beautiful image of Prajna-Paramita now 
at Leyden, of which Mr. Havell has published a front view. The name is 
that of the most sacred book of the Mahay anist scriptures, ascribed to Nagar- 
juna, and thence transferred to a personification of Supreme Wisdom in 
female form. Mr. Havell, who regards the image as being 'one of the most 
spiritual creations of any art, Eastern or Western', compares it with the 
Madonnas of Giovanni Bellini. 

The little bronze (Plate 113 b), supposed to represent Manjusri, is one of 
the most attractive of the Raffles collection in the British Museum. 

A. Seated Buddha. Javanese 

Manjusri. Copper casting. Javanese. 
British Museum 


PLATE 1 14. Persian Bodhisattva ; rev. of wooden panel, D. vii, 6, from Dandan-Uiliq 

Chapter Twelve 


The explorations carried on since 1896 by Sir Aurel Stein, Professor Recent 
Griinwedel, Dr. v. Le Coq, and other savants, in the vast regions of discoveries. 
Chinese Turkistan, lying north of Tibet, to the west of China, and 
both north and south of the Taklamakan Desert ('Gobi' of the older maps), 
have revealed 'sand-buried* and other ruins full of the remains of ancient 
civilizations. 1 Those remains, which include thousands of manuscripts 
written in many scripts and languages, known and unknown, also comprise 
multitudes of works of art, pictorial and plastic, which, by their character- 
istics, mark Chinese Turkistan as the meeting-ground of Hellenistic, Indian, 
Persian, and Chinese forms of civilization. 

The wide extension of Indian languages, literature, and art from the second Immense 
century of the Christian era thus demonstrated has been a surprise to the amount of 
learned world, but the hugt mass of material collected is so unmanageable 
that many years must elapse before 'the most interesting subject*, as Dr. v. 
Le Coq calls it, of the relations between the early civilizations of India, Persia, 
China, and the Far East can be worked out so as to admit of firmly established 
conclusions. At present it is not possible to present in a few pages a satis- 
factory abstract of the new knowledge concerning the diffusion of Indian 
art and learning in the Chinese Turkistan countries. The paintings seem to 
be assignable mostly to the seventh or eighth centuries, and so help to fill 
up the gap in the story of Indian painting between Ajanta and Akbar. In 
this section no more can be attempted than a slight indication of the extent 
to which Indian schools of painting, modified by external influences, pene- 
trated Turkistan, and, through it, the Far East. 

The discoveries made by Sir Aurel Stein during his various expeditions Remains of 
into the Desert having been published in considerable detail, a fair idea ei e hth 
can be formed of the achievements of painters following Indian models more SSL** 
or less closely during the seventh and eighth centuries in Turkistan. Numer- uiliq. 
ous fresco or distemper paintings on wood and plaster were found at a place 
called Dandan-Uiliq, which was abandoned soon after a.d. 791. All these 
works may be referred with confidence to the eighth century, and thus afford 
evidence of a sufficiently dated stage in the evolution of Indian painting when 
exposed to the influence of the Persian and Chinese schools. A few of the 
more striking examples are reproduced by permission. 

1 The word 'Gobi' simply means 'desert' (Stein). 

3*9» v 


Mounted One of the best preserved paintings is that on a panel (D. vii, 5), 15 inches 

personages high an( j near iy y inches broad, which represents two sacred or princely per- 

on panel. sona g eSj m0 unted, one on a piebald Yarkandi pony and the other on a camel 

(Plate 116 a). The nimbus behind the head of each rider indicates either his 

high rank or his sacred character. The artists of the Mughal court in India 

were accustomed to give this emblem of sanctity to the emperors and even 

to members of their families, and in Khotan during the eighth century the 

same practice seems to have prevailed. The picture speaks for itself so clearly 

that detailed description is unnecessary, but the blending of Indian and 

Chinese features in the face of the horseman may be noted, and the free 

drawing of the camel deserves commendation. The horseman is repeated 

on D. x, 5 (Stein, Plate LXII), but the identity of either figure has not yet 

been determined. 

Painting in The ugly picture on the obverse of panel D. vii, 6, measuring i2§ x8 

Tibetan j^kes, representing a three-faced, four-armed deity, supposed to be a Tan- 

m ' trie form of Avalokitesvara, squatting on a chequered cushion supported by 

two white bulls, is purely Indian, and is so closely related to the modern 

Lamaist compositions that it might be described as the oldest extant Tibetan 

painting. The body and front face of the deity are dark blue, the face on the 

proper right, with a feminine expression, is white, and the demoniac face on 

the proper left is yellow. The outline is drawn in thick black lines, and the 

work has little aesthetic merit (Stein, Plate LX). 

A Persian The reverse of the same panel offers a surprise by presenting a picture of 

Bodhisattva. a four-armed Buddhist saint or Bodhisattva in the guise of a Persian with 

black beard and whiskers, holding a thunderbolt (vajra) in his left hand. 

The combination on one panel of this almost purely Persian figure with 

the Indian image on the other side suggests questions, at present insoluble, 

concerning the forms which Buddhism may have assumed in Iranian 

lands. The art, seemingly of higher quality than that of the obverse 

picture, is certainly more pleasing. The four arms are a distinctly Indian 

feature (Plate 114). The existence of this queer figure may help us in some 

measure to understand the introduction of Persian figures into the Buddhist 

pictures of Ajanta, which may yet be proved to be an Indian development 

of Central Asiatic Buddhist art. But that hypothesis at present lacks historical 


The water- The most interesting of the Dandan-Uiliq paintings is the fresco depicting 

sprite fresco. gome j e g en( j connec ted with a female water-sprite, probably the tale told by 

Hiuen Tsiang of the minister who married the widow of the Naga king in 

order to secure the flow of water over the lands of Khotan. However that 

may be, the design and execution of the composition are of considerable 

merit, and well illustrate the variety of elements combined in the medieval 

art of Khotan. The pose of the lady, whose figure in the original projects 

*.■ >-**?{ 



? : *Wi'« 

PLATE 1 1 v Water-sprite, &c. ; fresco at Dandan-Uilip 













.5 ^ 

£ c 
.2 W 

c E 


■ »- 

. ^_ 






r -l 



f m 
























about 18 inches above the water, is plainly a reminiscence of some Hellenistic 
Venus, such as the de' Medici or the Capitoline, and the vine-leaf guarding 
her modesty equally recalls the conventional fig-leaf. Her ornaments are 
Indian, her face Chinese. Thus in this one figure we can trace the meeting 
of the three civilizations, Greek, Indian, and Chinese. The seated figures 
are more Chinese in type than anything else. I do not perceive any Persian 
factors in this work (Plate 115). 

Another painting (D. x, 4), more primitive in style, illustrates the story of Chinese 
the Queen of Khotan, a Chinese princess, who secretly introduced silk princess 
cocoons into her adopted country by concealing them in the folds of her fresco - 
head-dress. The central effigy of the princess, boldly sketched with a few 
etching-like strokes, will suffice as an example of the style (Fig. 116 b). It 
will be observed that the head of the princess, like the heads of the three 
other persons in the picture, is surrounded by a nimbus or halo, apparently 
affording clear evidence that in Khotan art of the eighth century, as in 
Mughal art of the seventeenth, the nimbus was given to persons of royal 
birth as well as to divinities and saints. The lady's features are Indian rather 
than Chinese. 

Further east, at Endere, between Niya and Cherchen, in ruins of somewhat Sketch of 
earlier date than those at Dandan-Uiliq, Sir Aurel Stein found a scrap of can } el and 
faded fresco on stucco with 'delicate and harmonious colouring', and an ca ' 
Indian-ink sketch on paper depicting a Bactrian she-camel suckling her calf, 
drawn in the fewest possible bold strokes with considerable spirit and vigour. 
A slight attempt to indicate the solidity or roundness of the body has been 
made by adding a wash of faint colour round the contours (Plate 1 16 c). The 
drawing seems to have been executed with a brush, not a pen, and is free 
from conventionality. 

The countries to the north of the great desert have proved to be equally Art to the 
fertile in finds of astonishing richness. At the ruined city of Idiqut-i-Shahri J orth of the 
the German explorers found the remains of Buddhist, Manichean, and Nes- e 
torian buildings and art associated in such a way as to show that for centuries 
the adherents of the rival creeds managed to live together. Ultimately, in 
or about the ninth century, the Buddhists were massacred by the Chinese, 
a fact of which Dr. v. Le Coq discovered terrible proof when he opened a 
chamber filled with the skeletons of monks and other signs of ruthless slaugh- 
ter. At this site curious votice flags, both Manichean and Buddhist, were 
found, with designs painted on plaster applied to long strips of cotton, in 
the manner still practised by Tibetan Lamas. 

At Yar-Khoto, to the west of Turfan, paintings on silk, described as being 
exquisitely wrought and harmoniously coloured, were obtained. Stein also 
brought home from his second expedition a large quantity of similar silk 
designs obtained in a walled-up temple near the Kan-su border, many of 


which are in the finest condition. 1 The Yar-Khoto pictures are both Mani- 
chean and Buddhist, the latter never failing to retain a distinctly Indian 
character. At the Basaklik monastery Dr. v. Le Coq discovered wall-paintings 
executed on a surface of plaster composed of loam and chopped straw. 

The pictures at Chiqqan Kol and Toyoq are in the archaic Indian style, 
whereas in other places the Indian features have been much modified by 
Chinese and Persian influence. It is evident that the Turkistan paintings 
range over a long time, and that, when their sequence shall have been worked 
out, much light will be thrown upon the development of the pictorial art of 
Asia, including India. 
Indian Students of Chinese and Japanese painting have been aware for some years 
° n S in of past that the specially Buddhist forms of art in China were derived from India 
an of the through Khotan, and passed on through Korea to Japan, the principal agent 
Far East, in the transmission to Korea, and so to Japan, having been Wei-chi I-song, 
son of Bajna of Khotan. Bajna was one of the numerous foreign artists in 
the service of the Chinese Emperor, Yang-ti (a.d. 605-17), and had been 
preceded at the imperial court by two Indian monks, with names something 
like Kabodha and Dharmakuksha. Both Bajna and his son, according to 
Chinese critics, worked in a foreign, that is to say, Indian manner, and 
enjoyed high repute as Buddhist artists. The marked Indian element in early 
Japanese art is thus amply accounted for. 2 
Indian Although the descent of the specially Buddhist varieties of the art of the 
con&TdT ^ ar ^ ast fr" om I n dia, an d more particularly from Indo-Greek prototypes in 
Buddhist Gandhara on the north-western frontier, is abundantly proved, the evidence 
art. does not warrant the larger inference drawn by Mr. Anderson that 'a pre- 
viously undeveloped art* in China was dependent upon importations from 
India for its growth and development. 3 The earliest extant Chinese painting, 
the fourth-century picture by Ku K'ai-chih in the British Museum, does not 
show the slightest trace of either Indian or Greek influence. Buddhist pic- 
tures form but a single subdivision of Chinese painting, the subjects of which, 
according to Professor Giles, may be classified under seven heads, namely — 
(1) history, (2) religion (including Buddhism and Taoism), (3) landscape, 
(4) flowers, (5) birds, (6) beasts, and (7) portraiture. 4 Excepting the Buddhist 
designs under the second head, China learned nothing, and had nothing to 
learn from 'the land of the Brahmans\ I am disposed to agree with Mr. 
Binyon, who finds in China 'if not the parent art of Asia, its earliest mature 
flower in painting \ 5 

1 A series was exhibited in the Indian Court of 3 Anderson, Descriptive and Hist. Catal. of a 
the Festival of Empire, 1911, and described by Collection of Japanese and Chinese Pictures, in 
Stein in the Catalogue of the Court, pp. 14- the B. M. (1886), p. 482. 

26. * An Introduction to the History of Chinese Pic- 

2 Hirth, Ueber fremde Einflusse in der chines, torial Art, 1905, p. 7. 

Kunst, pp. 34, 38, 39, 43-60. 5 Painting in the Far East (1908), p. 48. 


Mr. Griffiths thought that he could discern marks o£ Chinese influence in Wall- 
the paintings at Ajanta; and he may be right, although such marks are not painting of 
very distinct, and may, perhaps, be explained as derived from the common J^jjjj! 
stock of Asiatic art. However that may be, the art of Ajanta certainly pro- japan.' 
duced no effect upon the general development of painting in China; and in 
Japan the only conspicuous instance of imitation of the Ajanta style is the 
wall-painting in the temple of Horiuji at Nara, supposed by some critics to 
date from ajd. 607, but according to others about a century later. 1 


The art of Tibet is so closely related to that of Nepal that the paintings Tibetan anc 
of both countries may be grouped together. The style is a combination of Nepalese 
Indian and Chinese characteristics, traceable back to the earlier style of sc oos ' 
Turkistan, specimens of which have been cited above. Nepal probably 
imitated Indian painting before Tibet was sufficiently civilized to do so. 
According to Taranath, the earliest Nepalese school followed the model of 
the school of the c Ancient West' founded by Sringadhara of Marwar in the 
seventh century, while subsequent Nepalese artists inclined rather to favour 
the methods of the Bengal 'Eastern' school of the ninth century. The latest 
Nepalese artists before Taranath's time in a.d. 1600 are said to have had 'no 
special character'. All the existing specimens of Nepalese painting, with the 
exception of the miniatures in MSS., apparently are later than the seven- 
teenth century. Most of the extant Tibetan pictures are believed to be not 
older, but it is not possible to determine exact dates. 

Painting is still extensively practised by Tibetan Lamas for the purposes Modem 
of their ritualistic worship and as a source of income. Usually the composi- Ti f> et . an 
tions are depicted on long narrow banners of either silk or cotton. They may P amtm S- 
be painted either directly on the fabric or on a coat of plaster applied to it. 
Pictures on paper also exist. The silken banners obtained by Stein from a 
walled-up temple near the Chinese frontier, and dating from the seventh or 
eighth century, closely resemble those now made by the Lamas, who follow 
strictly prescribed ritual rules. The Lamas also execute frescoes on the 
temple walls, some of which, according to travellers, are remarkable com- 

Tibetan painting is generally more a matter of skilled craftsmanship than Mechanical 
of fine art. The canonical process of manufacture has been fully described by methods. 
Godwin Austen, who explains in detail the way in which a figure of Buddha 
is built up. The draughtsman starts by drawing a long vertical rectangle, 
within which are inscribed a medial perpendicular line and sundry horizontal 
parallels at prescribed distances. The different organs of the body are then 
plotted out for insertion at certain intersections of the lines. For example, 
1 A tracing (No. 148, Anderson's Catal.) is in the B. M. Mr. Okakura favours the later date. 


the face is plotted from the starting-point determined by the intersection 
of the medial perpendicular with the transverse line No. 17. The remaining 
parts of the body are worked out in a similar way, and other sacred objects, 
such as a stupa (chorten, or dagaba), are imaged on like principles. 1 Travellers 
tell us that the monks of the Greek communities at Mount Athos manufac- 
ture the sacred ikons in an equally mechanical fashion. 
Good The examination of specimens of Tibetan ritualistic paintings confirms the 
colouring expectation formed from knowledge of the mechanical process enjoined. The 
and details, j.^ j^^y^ j s tru jy drawn and full of subtlety. The colouring is often 
rich and harmonious, shades of indigo blue in particular being combined 
with black in a very effective manner. The execution of details, too, is often 
finished with characteristic Indian minuteness. 2 
Banners in The British Museum possesses a considerable collection of Tibetan banner- 
British paintings on silk, mounted on rollers. Most of the pictures are distinctly 
useum. c^jjjggg j n style, with little trace of Indian influence on the art, as distin- 
guished from the subjects. But a few are more Indian than Chinese. One 
such is No. 5^0, 63 (measuring 2 feet 3 inches X20 inches), with an em- 
broidered border. The central figure is a seated Buddha of Indian style in 
the 'earth-touching' pose. An unpleasant Tantric Bodhisattva of little artistic 
value is depicted on No. ^2, 57. The most characteristically Tibetan 
specimen, combining Indian with Chinese peculiarities, is No. ^, 62, 
which possesses considerable beauty as a scheme of colour, dark indigo 
blue predominating. The central figure is a horrible and repulsive Yama, 
or Death, wearing a garland of skulls. The field is mostly occupied by a 
series of scrolls in dark tints, of distinctively Tibetan form. In the upper 
section three small figures, a seated Buddha in Chinese costume, with on 
each side a Bodhisattva, or Tibetan Lama, wearing a tall, conical head-dress, 
are tolerably well executed. The painting does not look old. 
Cotton The Tibetan pictures on cotton exhibited in the Indian Section of the 
P X g Ur ^ Victoria and Albert Museum, South Kensington, include many of great 
Kensington, aesthetic interest, especially No. 2451 from the Schlagintweit collection. 

Th e The considerable collection of Tibetan drawings and paintings, presented 

Hodgson by Mr. Brian Hodgson to the Institut de France and still preserved there, has 

collection, had the good fortune to have been described by two eminent scholars, M. 

Barthelemy St. Hilaire and M. A. Foucher. M. St. Hilaire, who formed a 

higher estimate of the artistic quality of a set of ten numbered paintings in 

1 'On the System of outlining the Figures of doivent etre purifies au moyen de pri&res et de 

Deities and other Religious Drawings, as prac- formules d'exorcisme de peur que quelque 

tised at Ladak' (J.A.S. B. f Part I, vol. xxxiii demon ne viennei en prendre possession; c'est 

(1864), p. 151, with plates). 'II est de r&gle, ce qui explique que tous les peintres sont des 

quand il s'agit de personnages, qu'on commence Lamas' (de Millou6, Bod- Youl ou Tibet, p. 295). 

toujours par les yeux, qui, aussitot terminus, 2 Indian Sculpture and Painting. 

PLATE II 7. Bi:ddha and worshippers. Tibet 
(Tibet, N\ 5', Institut de France) 

PLATE 1 1 8. Portrait of young Lama. Tibet 
(Hodgson Collection, Institut de France) 


the collection than M. Foucher can accept, criticized them generally in the 
following terms: — 

'Mais les monuments de sculpture et de peinture que nous venons de passer en revue 
sont tr&s loin d'etre denu^s de m6rite ; le dessin en est quelquefois tres pur, les attitudes 
des personnages sont 616gantes et naturelles. II y a meme, quoique plus rarement, 
une onction profonde dans la physionomie du Buddha et des principaux Bhikshous. 
... La composition est ordinairement r^guliere, quelquefois vaste et tres bien ordonn6e, 
comme Tatteste la description que j'ai donn^e plus haut du troisieme tableau tib&ain.' 

No. 3, alluded to in the passage quoted, depicts various Buddhas and a Nos. 3,9, i< 
crowd of worshippers. The uppermost scene represents a Dhyani-Buddha of Hodgson 
holding his Sakti, or female counterpart, in close embrace, a common subject colIectlon - 
in these paintings. M. St. Hilaire considered No. 9, a large work measuring 
i\ metres in length by 70cm. in breadth,to be 'd'un travail presqu'aussi delicat 
que celui du numero 3*. No. 10, which includes representations of devil- 
dances performed by Lamas wearing horrible masks, is also commended. 

M. Foucher has kindly selected No. 5 as being one of the best and most No. 5 of 
suitable for reproduction (Plate 117). It depicts Buddha in the 'earth- Hodgson 
touching* pose surrounded by a host of worshippers on earth and in the coIlectlon - 
clouds, and is framed in a oretty border. The figures of the adoring Lamas 
are numbered. Similar numbers are inserted in other pictures. 

The valuable collection of objects illustrative of Buddhism formed by Tibetan anc 
Prince E. Ukhtomskij, and once preserved in the Museum of the Emperor Mongolian 
Alexander III, St. Petersburg, includes many Tibetan and Mongolian pictures, P* ct £ res 
of which select specimens, including portraits, have been engraved in outline ukhtomskij 
as illustrations of the Catalogue in Russian prepared by Prof. A. Griinwedel. collection. 

Very little can be recorded concerning the pictorial art of Nepal, which, as Nepalese ar 
known to us, is only a modern variety of the Tibetan school. The extant a variety of 
specimens are all Buddhist, and seem to possess little aesthetic value. The Tlbetan - 
Hodgson collection in Paris includes ten pictures, two of which have been 
reproduced by M. Sylvain Levi in his learned work, 'Nepal*. The first of his 
plates is a reduced copy of No. 6, a large pen-and-ink drawing, 2 m. 85 cm. 
long and 1 m. high, believed to have been prepared to the order of Mr. 
Brian Hodgson. The subject has been identified as a procession in honour 
of Padmapani or Avalokitesvara, marching round the walls of a town in the 
valley. The drawing is carefully executed and shows a knowledge of linear 
perspective presumably due to European teaching, but as a specimen of 
Indian art it is of no interest. M. Levi's second and larger folding plate 
reproduces in six sections a photograph of the illustrated manuscript giving 
the sacred legend of Nepal. This work, too, possesses little merit as art. 

The Buddhists of the Northern School are fond of constructing magic Magic 
circles (mandala) crowded with figures of Buddhas, worshippers, monks, jjj^jj 
lotus-plants, and other sacred persons or things, believing that the maker C0 Ji e £ ? n. 


or user of such a picture will have a claim on the protection of all the influen- 
tial beings and lucky objects depicted. No. 10 of the Hodgson collection is 
such a magic circle, filled with more than 200 figures. Another magic circle 
is No. 7 of the Hodgson collection. The silken magic circle in the British 
Museum (MS. Add. 8898) is accompanied by a description from the com- 
petent pen of Col. Waddell. The composition, as a whole, is ugly and bar- 
barous, not worth copying, though the floral border is pretty. 
Ancient The only relics of an ancient school of Nepalese painting are the miniature 
m * ni MSS S til ustrat i° ns °f two manuscripts, Add. 1643, Cambridge, and A. 15, Asiatic 
Society of Bengal, Calcutta, which have been minutely studied and in large 
part reproduced by M. Foucher. 1 Unfortunately, the age of the manuscripts 
and miniatures is not quite certain, but probably both date from the eleventh 
century. The older document, that at Cambridge, cannot be later than 
a.d. 1015; the Calcutta manuscript may be some fifty years posterior in date. 
The miniatures, numbering 85 in Add. 1643, and 37 in A. 15, equal in height 
(o m, 055) the narrow strip of palm-leaf on which the text is written, and each 
depicts a holy place, a sacred personage, or an incident in Buddhist legend. 
Most of them being plainly labelled, they are of high archaeological and 
historical value, but from the purely aesthetic point of view are not of much 
Technique The technique is simple. The outlines were drawn in red ink and filled 
and quality. i n ^fo co l ur washes, only the five canonical colours being used — white, 
blue, red, yellow, and green. The designs evidently transmit an ancient 
tradition, and are the production of an art long stereotyped; but, notwith- 
standing the mechanical monotony of treatment, M. Foucher holds that 
these little paintings, although not masterpieces, cannot be regarded as merely 
vulgar daubs. They have been drawn and coloured by illuminators 'tres 
suffisamment maitres de leurs moyens'. If they date from the eleventh cen- 
tury, they may represent the 'Eastern* school of Dhiman, which according 
to Taranath was favoured in Nepal at about that time. 
Close rela- The plastic art of both Tibet and Nepal is Indian in origin and essentially 
*!£ n of one. The art of Nepal, apart from wood-carving, is represented by images 
and Ne- most ly ca st in copper or cut in slate or coarse marbles, 2 all being comparatively 
palese art. recent in date, none, perhaps, being more than three or four centuries old. 
The castings are made specially for Lamaist use in Tibet. They include large 
and small figures of gilt copper, and many ritualistic instruments, such as 
candlesticks, thunder-bolts, and daggers. The latter are usually in brass. 

1 Similar miniatures are executed in Tibet (de ale, Paris (Blochet, 'Inventaire,' Revue des 

Millou£, Bod-Youl ou Tibet, p. 237), but I am Biblioth&ques, 1899, p. 265). 

not in a position to cite examples. Good speci- a E. B. Havell, Stone Carving in Bengal, thin 

mens, dating from the seventeenth century, or quarto, 16 pp., 5 plates (Bengal Secretariat 

earlier, are inserted in a manuscript of the Kah- Depot, Calcutta, 1906). 

gyur (Tib&ain io') in the Bibliothfeque Nation- 

PLATE I ig. Portrait of a Lama evoking a demon. Tibet 
(Hodgson Collection, Institut de France) 



" * ^ ^ *> -, '• >^ 


A. A teacher. Pitt-Rivers Museum, 

B. The Bodhisattva Manjusri. Pitt-Rivers 



C. The goddess Sarasvati. Ukhtomskij 

]). Tsong-kapa. Pitt-Rivers Museum 
PLATE 120. Bronze images of Tibetan deities and saints. Copper castings. Tibet 


M. de Millou£ gives a summary account of Tibetan fine-art work in 

'Copper is found both native and in the form of pyrites in Tibet, where it is wrought Tibetan art 
with uncommon perfection. Several localities are well known for their famous foundries, industry in 
which supply the whole of the Buddhist East with statuettes of divinities. Lhasa has copper. 
a special reputation for small figures in gilt copper, which are esteemed the more the 
smaller they are. Its productions are easily recognized by their graceful and somewhat 
arch (mi&vre) style. The statuettes made by the monks and craftsmen of Tashilumpo 
are equally esteemed. Most of the bronze l statuettes come from the workshops of the 
Tsang and Khams provinces. The bronzes from the region last named are famous 
for the perfection of their execution in details and their wonderful patina, qualities 
especially noticeable in the examples which go back to the sixteenth or seventeenth 
century, notwithstanding the impurity of the metal. Tsiamdo, Jaya, Bathang, and 
Lithang seem to be the principal centres of this art industry, which possesses an 
eminently religious character/ 2 

A special characteristic of Tibetan art is the abundance of realistic, highly Portrait 
individualized portrait statuettes of holy Lamas and other Buddhist saints, statuettes. 
How far such reputed portraits are actual likenesses and how far merely typical 
forms it is impossible to say. They may be authentic portraits transmitted 
by tradition through contemporary paintings. A good example of such a 
traditional portrait is the seated image of the 'Dalai Lama of the Third Re- 
birth', also known as the 'Apostle of the Mongols', whom he converted to 
Buddhism in the sixteenth century (1543-89). The original is in the large 
collection formed by Prince E. Ukhtomskij, until lately in the Museum of 
H.I.M. Alexander III, St. Petersburg, which has been carefully catalogued by 
Professor Griinwedel. 3 The presentment is thoroughly realistic, and possibly 
may be from the life. No criterion seems to exist by which the age of such 
images can be determined. The Prince's collection contains many equally Other 
good portrait statuettes. One notable portrait is that of the Lama reproduced portraits, 
in Guide, Abb. 72. An ancient image in Chinese crackled porcelain vividly 
represents in Indian pose a follower of the teacher known as Bhaisajya-guru, 
or Man-la, the 'Buddha of Medicine' {Guide, Abb. 94). 

Other artistic examples of the same portrait class are in the Musee Guimet, 
among which may be specially noted the bronze images of Padmasarribhava 
and Tsong-kapa, the founder of 'Yellow Lamaism' (Petit Guide Illustrd, pp. 
143, 144). A reproduction of a statuette of Tsong-kapa from an original in 

1 The Indian castings as a whole are copper, 3 See Prof. Griinweders Mythologie des Bud- 
with brass coming into use in comparatively dhismus in Tibet und der Mongolei, Fiihrer durch 
modern times. In Burma and Indo-China the die Sammlung des Fursten E. Ukhtomskij (Leip- 
castings are of true bronze as are the oldest in zig, 1900), cited as Guide; and his illustrated 
Tibet which show strong Chinese influence. Catalogue of the collection in Russian (Biblio- 
Most of these Tibetan 'bronzes' are copper cast- theca Buddhica, No. vi, 2 fasc, St. Petersburg, 
ings, the term being used in a very wide sense. 1905). The Tibetan name of the Apostle is 

2 Bod-Youlou Tibet (Paris, 1906), p. 130. Tsang mK'as-grub-bSod-nams-rgya-mtso. 
and Khams lie to the east. 

3»* Z 


the Pitt-Rivers Museum, Oxford, is here given (Plate 120). Portraits of this 
kind do not come from Nepal, so far as I know. 
Images of The effigies of Buddhas and deities, although similar in style to the human 
deities, portrait statuettes, are necessarily more conventional. They are often gilt 
and decorated with turquoises. The goddess Tara in her various forms is, 
perhaps, the favourite, but many deities are represented. 1 In Plate 120 
illustrations are given of three other figures — an unnamed teacher; the Bod- 
hisattva Manjusri; and his consort, Sarasvati, goddess of music and poetry. 
The last-named object, which is gracefully and freely modelled, closely 
resembles the best Nepalese work. 

1 The well-executed nearly black bronze Tara in the Pitt-Rivers Museum holds in her left hand 
a phallus of amethyst-colour glass. 

Chapter Thirteen 


Within about eighty years after the death of Muhammad in a.d. 632 Origin of 
the followers of his religion reigned supreme over Arabia, Persia, Musalman 
Syria, Western Turkistan, Sind, Egypt, North Africa, and Southern art 
Spain, the marvellously rapid extension of Muhammadan power having been 
rendered possible by the barbarism and weakness of the subjugated kingdoms 
in Asia, Africa, and Europe, The first contact of Islam, as MM- Le Bon and 
Saladin observe, was stimulating to what remained alive of the older forms 
of civilization. Muslim armies, recruited in Persia, Syria, and Egypt, carried 
with them crowds of Asiatic skilled craftsmen, who introduced everywhere 
the arts of Asia, and modified the various local forms of art so as to suit the 
needs of the new faith and satisfy the luxurious tastes of magnificent courts. 
The Arabs, although possessing little art of their own, succeeded in impress- 
ing upon the local styles which they utilized for Muslim purposes a general 
character of uniformity, which we now recognize as that of Musalman art. 

The Muhammadan conquest in a. d. 712 of Sind, which at that time was Indo-Mu- 
regarded as distinct from India, did not seriously affect India proper, and hammadan 
the occupation of Kabul in a.d. 870 was equally without appreciable influence **£ lf^ 
on Hindu polity, which continued its isolated course unchanged by external 
forces, developing on the political side the Rajput kingdoms, and on the 
aesthetic side the Brahmanical art already described. India did not feel the 
impact of Muslim ideas until the beginning of the eleventh century, when 
the repeated fierce raids of Mahmud of Ghazni compelled her to take notice 
of the new force which had arisen. Before his death in a.d. 1030 the Panjab 
had become a province of the Muhammadan Sultanate of Ghazni. But, until 
the closing years of the twelfth century, Islam made no further progress in 
India. The early Arab conquerors of Sind seem to have left nothing but 
ruined Hindu temples behind them, nor are there tangible traces of the rule 
of the Ghaznivide rulers of the Panjab. 

The history of Indo- Muhammadan art begins with the year a.d. 1200 in 
round numbers. Between 1193 and 1236 Muhammad of Ghor, Kutb-ud-din 
Ibak, and Sultan Altamsh had compelled all Northern India, including 
Bengal, to submit, more or less completely, to the Muslim government 
established at Delhi. The earliest Muhammadan monuments in India date 
from the reigns of the three princes named; the principal works of that 
time being the mosque at Ajmer, the Kutb mosque and minor at Delhi, the 
gateway of the chief mosque at Budaun (a.d. 1223), 1 and the tomb of the 
Sultan Altamsh at Delhi. 

1 Cunningham, A. S. Rep., vol. xi, p. 5, PI. III. 


Essentials of The simple, world-wide worship of Muslims, who adore the One God and 
a mosque. j^te ever y y n d Q f idol, can be performed satisfactorily without any building. 
But it is convenient to have a spacious edifice in which the faithful can 
assemble on Friday, the Musalman Sabbath, to join in public prayer, and 
occasionally hear a sermon. During prayer the worshippers should turn 
towards Mecca, the direction of which is indicated by a niche or niches in 
the appropriate wall. The Muhammadan mosque, or church, therefore, 
consists essentially of an enclosure, with a niche in one wall to indicate the 
direction of Mecca. There should be also a pulpit, and a tank for ablution. 
All other things, such as cloisters, chambers, and lofty portals are unessential, 
being needed only for purposes of convenience and dignity. The mosque 
may be wholly open to the air, or wholly or partially roofed. Examples of 
wholly roofed mosques are very rare in India, the only one on a large scale 
being that at Gulbarga in the Deccan. Ordinarily a large open quadrangle is 
the principal feature of an Indian mosque. The covered portions of the more 
considerable buildings usually consist of an aisle or aisles (liwari), at the 
western side, with cloisters round the enclosing walls, and often include huge 
gateways with many chambers, and sundry minor structures. The roofs are 
invariably domed in some fashion or other, and pointed arches are a promi- 
nent feature. 1 
Origin of The almost universal presence of domes and arches, usually of the pointed 
arche? kind, * n Muhammadan buildings is due to the fact that Muslim architecture 
' is based on the style practised at Baghdad in the time of the great Abbasid 
Khalifs (Caliphs), of whom Harun-ar-rashid (786-809) is the best known. 
The Baghdad style was derived from the ancient vaulted architecture of 
Mesopotamia, as transmitted through the modified developments of Sassa- 
nian times (a.d. 226-641). The beginnings of the familiar forms of Muham- 
madan architecture have been recently traced by General de Beylie in the 
buildings of Samara in Mesopotamia, erected in the early part of the ninth 
century, and abandoned in 875, when Baghdad became the capital of the 
Khalifate. From Baghdad the style spread rapidly throughout the Muham- 
madan world, and became to such a degree universal that it is hardly possible 
to imagine a mosque of brick or stone without domes and arches. 2 

At the beginning of the thirteenth century, when Kutb-ud-din undertook 
to build mosques and tombs at Delhi and Ajmer, domes and pointed arches 

1 The growth of the mosque was radically ture in Kashmir', Ann. Rep. A. S., India, 1906- 
affected by the extension of its purely reli- 7, pp. 161-70, with plates; a valuable treatise), 
gious function to include education. The great My text is based on the article 'Les Origines de 
mosques of Asia were universities as well as rArtmusulman\byM.LouisBrehier(La/?CTW 
places of worship. des Idies, Paris, No. 75, Mars 1910, pp. 189-99), 

2 The exceptional wooden mosques of Kashmir and on Saladin, Manuel d'Art musulman, tome i, 
have tall spires, probably derived from Buddhist chap. I (Paris, 1907). 

architecture (Nicholls, 'Muhammadan Architec- 








PLATE 122. The Kutb Minar, Delhi 


were recognized to be essential. But the conquerors were obliged to employ Hindu con- 
Hindu masons, unaccustomed to turning true radiating arches and domes, stmction 
and ordinarily used only to make the semblance of such by means of the jjj^ 681 
horizontal corbelled construction familiar to them, with which the Muslim mosques, 
architects had to be content. The cloisters were easily made up from the 
materials of overthrown Hindu temples, and retained a manifest Hindu 
character without objection. 

At the Kutb mosque of Delhi the glory of the building is the screen of 
eleven pointed arches, eight smaller and three larger, Muslim in form, but 
Hindu in construction (Plate 123 a). The faces of these structures are 
decorated with 'a lace-work of intricate and delicate carving', considered by 
Fergusson to be 'the most exquisite specimen of its class known to exist any- 
where'. It bears some resemblance to the decorations of the Sassanian 
palace of Mashita and those of certain parts of Santa Sophia at Constanti- 
nople. The similar screen at Ajmer, built between a.d. 1200 and 1235, con- 
sists of seven arches, the central one being 22 feet 2 inches wide. 'Each arch 
is surrounded by three lines of writing, the outer in the Kufic and the other 
two in Arabic characters, and divided from each other by bands of Arabesque 
ornament boldly and clearly cut and still as sharp as when first chiselled. 1 
In the centre the screen rises to a height of 56 feet' (Plate 121). The illustra- 
tion shows clearly the Hindu mode of construction, and the peculiar low 
conical dome appearing within. 

The mosque colloquially known as 'the Kutb' is commonly believed to be Origin of 
named after the Sultan Kutb-ud-din Ibak (1205-10), and it is true that it * e I j ame 
was completed in its original form in the year a.d. i 198 by him while he was 
still Viceroy of Delhi and the Indian territory under the Sultan of Ghazni. 
But the building is really named after a famous saint, Kutb-ud-din of Ush 
near Baghdad, who lies buried near, and is popularly remembered as Kutb 

Muslim usage requires that the faithful should be summoned to prayer at The Kutb 
the stated times by a loud call uttered by an official known as muazzin. In Minar. 
order to facilitate his duty many mosques, although by no means all, were 
furnished with a minaret, or two minarets, from which the summons could 
be proclaimed. Sometimes the minarets were attached to the mosque, some- 
times they were detached. The Kutb Minar at Delhi, originally about 250 
feet high, and even now not much less, is the most remarkable example of 
the detached minaret in existence. Like the adjoining mosque, it derives its 
familiar name from the saint, not the prince. It is, however, some thirty 
years or more later in date than the mosque, having been erected about 
a.d. 1232 by the Sultan Altamsh when he made large additions to the mosque. 
The details of the building are due to its Hindu sculptors. The structure 
1 This foliated scroll-work is unmistakably Indian in character. 


has been so often described at length, that it will be sufficient to give a 
photograph (Plate 122), and to cite Fergusson's authority for the statement 
that the Minar is 'the most beautiful example of its class known to exist 
anywhere'. Fine specimens of minors of later date will be illustrated presently. 
The form, a specially Muslim one, offers much scope for variety of treatment. 
'The minaret', as Sir George Birdwood observes, 'is the one original feature 
the Saracens contributed to architecture.' 1 
Gateways of The magnificent gateway erected in a.d. 13 10 by the Sultan Ala-ud-din 

Ala Khan Kk^Ji on the south side of &* enlarged Kutb Mosque marks an advance in 
ajI " Indo-Muhammadan architecture. Here the true arches with keystones were 
no longer constrained to execute the designs of their foreign masters by the 
structurally inferior Hindu methods. The building consists of a rectangular 
chamber surmounted by a low-spreading dome. The ornament is composed 
mainly of geometrical designs and artistic Arabic inscriptions, but sundry 
details show influence of Hindu tradition (Plate 124 a). 2 
Tughlak The Kings or Sultans of the Tughlak dynasty of Delhi in the fourteenth 
DeM centurv introduced a new style of architecture marked by massiveness and 
' extreme simplicity, qualities which have suggested a comparison with the 
early Norman work in England. The most characteristic example of this 
severe style is the tomb of Ghiyas-ud-din Tughlak, who was killed by a care- 
fully devised 'accident' in 1324 (Plate 124 b). The plan is a square measuring 
38! feet inside and 61 £ feet outside, and the height to the top of the dome is 
70 feet. The enormously thick walls slope inwards. The exterior decoration 
is effected in an austere manner by the free use of bands and borders of white 
marble, varied with a few panels of black marble, showing against the large 
surfaces of red sandstone. 3 No trace of Hindu tradition is evident. The style 
is more or less unique. 
The At the close of the fourteenth century many provinces broke away from the 
style* suzeraint y of the Sultans of Delhi, and set up as independent kingdoms. 
* Among such mushroom states one of the most notable was that known as 
the Sharki, or Eastern Sultanate, with its capital at Jaunpur, forty miles from 
Benares. Its independence lasted until 1476. During its short period of 
glory the local sovereigns occupied themselves by destroying Hindu temples 
and replacing them by mosques designed on a grand scale, and in a dis- 
tinctive style. The handsomest of the Jaunpur mosques is the Atala, com- 
pleted in 1408, of which the main portal is shown in Plate 125. The gateways 
and great halls are thoroughly Muslim, with radiating arches and true domes, 
but in the cloisters and interior galleries, where there was no need to roof 
large spaces, square pillars, often borrowed from Hindu temples, are used, 

1 3- Xty' Soc. Arts, Jan. 1911, p. 179. aj>. 1494 at Khairpur near Delhi (Cunningham, 

* This building was copied for the gateway of A. S. Rep., vol. xx, p. 156). 

the tomb of Sultan Sikandar Lodi, built in J Cunningham, A. S. Rep., vol. i, p. 123. 

- \ 

f Jl*fl3Sa^*- 




O^* 7 












A. Gateway of Ala-ud-din Khalji, Delhi 

15. Tomb of Tughlak Shah, Tughlakahad, Old Delhi 
PLATE 124 


and the construction is Hindu. The style, while it has much of the massive- 
ness of the Tughlak buildings at Delhi, is less severe and more attractive, a 
curious hybrid of Muslim and Hindu. 

Under the patronage of its independent kings Bengal developed a Muham- The Bengal 
madan style of its own. 8tvle « 

'It is' (Fergusson observes) 'neither like that of Delhi, nor Jaunpur, nor any other 
style, but one purely local, and not without considerable merit in itself; its principal 
characteristic being heavy short pillars of stone supporting pointed arches and vaults 
in brick— whereas, at Jaunpur, for instance, light pillars carried horizontal architraves 
and flat ceilings.' 

The second characteristic of the style is the curvilinear cornice copied 
from bamboo structures. The best examples are to be seen among the ex- 
tensive ruins of the cities Gaur and Pandua in the Malda District. The 
buildings are mostly in brick and possess little beauty. But one mosque, 
known as the Small Golden, or Eunuch's Mosque at Gaur, is built wholly of 
basalt with massive solidity. This elegant building, which has been called 
'the gem of Gaur', was erected about a.d. 1500, and is covered inside and 
out with beautifully chiselled designs, including the Indian lotus. The gate- 
way is shown in Plate 127 A. There are fifteen domes. 

The buildings at Mandu, the capital of the kingdom of Malwa, which was The Mandu 
independent from a.d. 1401 to 1531, are purely Muslim in style, closely 8t y le - 
related to those of the Sultans of Delhi. They are distinguished from the 
later Mughal buildings by the absence of groining and by the spreading domes. 

Unquestionably, the most beautiful of the provincial styles of Muham- The Gujarat 
madan architecture in Northern and Western India is that of Gujarat. By 8tvlc - 
good fortune it has been studied more carefully than any other Indian style, 
all the chief examples having been elaborately described and illustrated by 
Dr. Burgess and his staff in three quarto volumes, fully furnished with plans, 
sections, elevations, and photographs. The style is that of the late medieval 
Hindu and Jain temples with such modifications as were necessary for the 
purposes of Muslim worship, and is characterized by all the richness of orna- 
ment distinctive of the temples of Gujarat and Southern Rajputana— a strange 
contrast to the stern simplicity of the Tughlak buildings contemporary with 
the earlier examples. Hindu construction, too, is freely used, but the indis- 
pensable domes and pointed arches are introduced. The entrance to the chief 
mosque at Cambay, for instance, erected early in the fourteenth century, is 
simply a Hindu temple porch, with a low dome, plain on the exterior, put on 
top of it. 1 The exquisite roofed pulpit of Hilal Khan Kazi's mosque at 
Dholka, built in a.d. 1333, has a purely Hindu pyramidal roof, and much of 
the panelled ornament with which the whole surface has been covered is 
equally Hindu. 2 

' Burgess, A. S. R., Western India, vol. vi, PI. XIX, p. 28. * Ibid., PI. XXX, p. 31. 


Ahmadabad. The finest examples of the style, which, of course, gradually discarded 
some of its Hindu features, are to be seen at and near Ahmadabad, the ancient 
provincial capital, to the architecture of which two of Dr. Burgess's volumes 
are devoted. The name of the city is derived from Ahmad Shah, Sultan of 
Gujarat from 141 1 to 1443, and the earliest Muhammadan buildings date 
from his time. The domes of his cathedral (Jatn'i) mosque are constructed 
in the Hindu fashion. The elaborate traceries and other decorative accessories 
of the Ahmadabad buildings will be noticed in the next chapter. 
Mosque of The best preserved mosque in Ahmadabad, and one of the prettiest build- 

M £h afiz * n S s * n ^ e c ^ y 1S ***** k^t by Mahafiz Khan at the close of the fifteenth 
century. The minarets are adorned with panels of rich floral tracery under- 
cut to such an extent that it is almost detached from the masonry. The archi- 
tecture still largely retains a Hindu character (Plate 126). 
Tomb of The tomb of Abu Turab, about a century later than Mahafiz Khan's 
Abu Turab. m0 sque, although still preserving the Ahmadabad character, is constructed 
with arches throughout, and is completely free from Hindu pillars (Plate 
127 b). The perforated screens which formerly connected the internal 
columns have disappeared. 
Buildings in The buildings designed in the distinctive Ahmadabad style have no 
Persian specially Persian features, and are thus sharply distinguished from the styles 
s y e# which we are about to notice. But two exceptional edifices at Ahmadabad, 
the mosque and tomb of Nawab Sardar Khan, built about 1680, are quite 
Persian in style. The mosque is very elegant. 1 
Styles of The Bahmani Sultanate of the Deccan, established in 1347 by a successful 
th 5,^ cc ^ n: revolt against the authority of Sultan Muhammad bin Tughlak of Delhi, 
o on a. b ro k e U p j nt0 g ve s t a t es a t the close of the fifteenth century. The rulers of 
all those kingdoms encouraged architecture, and, consequently, ancient build- 
ings of greater or less importance exist at all the local capitals. The covered 
mosque of Gulbarga has been already mentioned (ante, p. 172), and other 
notable edifices exist in the same town. Bidar possesses an imposing mosque, 
several remarkable tombs, and the ruins of a great college. The royal tombs 
at Golkonda, near the Nizam's capital, Hyderabad, are more or less familiar 
to tourists. The special peculiarities of the Golkonda style, high clerestories, 
stucco work in minarets, and domes of peculiar shape with narrow bases, 
may be illustrated from a tomb built in that style at Bijapur in the seventeenth 
century (Plate 128 a). 

The Deccan buildings, except a few of the earliest, are free from Hindu 

forms and constructions, and are related to the Mughal Indo-Persian style. 

But each kingdom had fashions of its own. 

The Bijapur By far the most important of the Deccan styles is that of Bijapur. The 

8t y le - buildings in it date between the years 1557 and 1686. The most ornate is 

1 Burgess, A. S. /?., Western India, vol. viii, p. 55, Pis. LX, LXI. 

PLATE T2v Main entrance of Ataladcvi Mosque, Jaunpur 

PLATE 126. Mosque of Mahafiz Khan, Ahmadabad 


the comparatively small tomb of Ibrahim Adil Shah II (1579-1626), the 
character of which may be judged from Plate 128. 

The stately tomb of Muhammad Adil Shah (1636-60) is covered with a Tomb of 
dome the second largest in the world, c a wonder of constructive skill', Muhammad 
balanced internally by an ingenious arrangement of pendentives, fully ex- Shah * 
plained by Fergusson, and with an internal height of 178 feet. 

'The external ordonnance of this building is as beautiful as that of the interior. 
At each angle stands an octagonal tower eight storeys high, simple and bold in its 
proportions, and crowned by a dome of great elegance. The lower part of the building 
is plain and solid, pierced only with such openings as are requisite to admit light and 
air; at the height of 83 feet a cornice projects to the extent of 12 feet from the wall, 
or nearly twice as much as the boldest European architect ever attempted. Above 
this an open gallery gives lightness and finish to the whole, each face being further 
relieved by two minarets/ 

The name of the architect of this wonderful structure, commonly known 
as the Gol Gumbaz, or Circular Dome (Plate 129 a), does not seem to be 
recorded. Foreigners, Asiatic or European, were frequently employed by the 
Indo-Muhammadan sovereigns, and the Bijapur style is thought to show the 
influence of Ottoman architects. An expert critic truly observes that 'under 
Mohammedan influence the dome-builders of India attained a mastery over 
this form unknown to and seemingly unappreciated by the builders of the 
Western world V 

At Tatta in Scinde is a fine group of tombs. The earliest of these is the Tatta 
tomb of Jam Nizam-ud-din which was built in 1508. Most of the tombs, Scmde - 
however, were built within twenty-five years before or after the year 1600, 
and in fact arc monuments to Moghul officials. The tomb of Sharfa Khan 
(c. 1638) stands on a stone base, but is of brick and tile- work. It is rather 
Persian in style than Indian. 2 

We now pass on to the Indo-Persian styles of the North, the only forms of Sur style: 
Muhammadan architecture in India familiar to the world in general. The *?"* °* 
short-lived and unstable Sur dynasty (1540-55), of which Sher Shah was the 
most distinguished member, had such a hard fight for existence that it could 
not have been expected to pay much attention to architecture. Nevertheless, 
several meritorious buildings are due to the Sur Sultans, and the mausoleum 
of Sher Shah at Sasaram (Sahasram), built on a lofty plinth in the midst 
of a lake, is one of the best designed and most beautiful buildings in India, 
unequalled among the earlier buildings in the northern provinces for grandeur 
and dignity. Cunningham was half inclined to prefer it even to the Taj. The 
dome, although not equal in size to the Gol Gumbaz of Bijapur, is 13 feet 
wider than that of the Agra monument. 3 Externally, the architecture is 

1 R. F. Chisholm, F.R.I.B.A., F.S.A., in J. Roy. Scinde Tiles. 

Soc. of Arts y Jan. 1911, p. 173. 3 Cunningham, A. S. Rep., vol. xi, pp. 133, 

2 For examples of Tatta tiles see Cousins, 137. 

3*9* A a 


wholly Muhammadan, but Hindu corbelling and horizontal architraves are 
used in all the inner doorways, as at Jaunpur. The style may be described 
as intermediate between the austerity of the Tughlak buildings and the femi- 
nine grace of Shahjahan's masterpiece. Plate 130 may suffice to give a good 
notion of the merits of this admirable style. The plan is octagonal, and 
coloured glazed tiles were used for decoration. Both the octagonal form and 
the use of glazed tiles were importations from Persia. 1 
Babar's Babar, the versatile founder of the Mughal dynasty, was an active builder 
buildings. d u ri n g his brief and stormy Indian reign of five years (1526-31). Holding a 
poor opinion of all Indian products, he summoned from Constantinople pupils 
of the celebrated architect Sinan, an Albanian officer on the staff of the Janis- 
saries, who had planned hundreds of important buildings in the Ottoman 
empire. 2 Out of the numerous edifices erected by these foreigners to Babar's 
order at Agra, Delhi, Kabul, and other places, only two are now visible, 
namely, the large mosque in the Kabul Bagh, Panipat, built after the great 
victory of 1526, and the J ami Masjid at Sambhal in Rohilkhand, bearing the 
same date (a.h. 933). The Panipat building is said to be still in fair condition. 
The Sambhal mosque has a remarkable ovoid dome. 3 Although the Indian 
buildings are much more Persian than Ottoman in style, there is some reason 
for thinking that the grandeur of the proportions of the existing monuments 
in Northern India and Bijapur may be partly due to the teaching of the school 
of Sinan. 
Humayun's Babar's accomplished son and successor, Humayun, the rival and opponent 
buildings. £ gj^j. ghaj^ found time in the midst of his unceasing wars to do some 
building. But most of his works have been destroyed, and, as in his father's 
case, only two are known to have survived, namely, a ruinous mosque near 
Agra, and one much better preserved at Fathabad in the Hissar District, 
Panjab, which is decorated in the Persian manner with enamelled tiles. 4 The 
buildings of Babar and Humayun are purely foreign in decoration. 
Akbar's Akbar's strong liking for Hindu ways induced him to revert to Hindu styles 
Hindu f decoration, and many of the buildings erected during his long reign (1556— 
nmg ' 1605) are more Hindu than Muslim. A conspicuous instance of such rever- 
sion is afforded by the well-known palace in the Agra Fort, commonly called 
the Jahangiri Mahal, which really dates from Akbar's time and might have 

1 Octagonal memorial mosque of the fourteenth A. S. Rep., vol. iv, p. 100; and by Moin-ud-din, 
century at Sultaniyah in Persia (Saladin, Manuel History of the Taj, p. 1 1 1 . It is dated a.h. 937= 
d'Art musulman, tome I, Fig. 267). A.D. 1530-1. The Fathabad mosque is a massive, 

2 Saladin, op. ciU, pp. 509, 561, with reference well-proportioned building with domes rather 
to Montani, Architecture ottomane. more than hemispherical, built to the order of 

3 Cunningham, A. S. Rep., vol. xii, p. 26 ; E. W. Humayun about 1540 or 1541, when he was on 
Smith, Akbar's Tomb, p. 4, editor's note. his way to Sind (Garrick, in Cunningham, A. S. 

4 The ruinous mosque at Kachpura opposite Rep., vol. xxiii, p. 12, Pis. Ill, IV). 
Agra is described by Carlleyle, in Cunningham, 

A. Gateway of Small Golden (Eunuch's) Mosque, Gaur 

A. Tomb in Golkonda style at Bijapur 

B. Ibrahim Rauza, Bijapur; front view 
PLATE 128 


been built for a Hindu Raja. 1 The other buildings of Akbar in the Fort were 
demolished by Shahjahan. 

The splendid mausoleum of Humayun, near Delhi, erected early in Akbar's Tomb of 
reign, while distinctly Persian in style, is differentiated by the free use of Humayun. 
white marble, a material little employed in Persia, and by the abstinence from 
coloured tile decoration so much favoured by the architects of that country. 
The building (Plate 129 b) is of special interest as being to some extent the 
model of the inimitable <r Taj\ The dome is built entirely of white marble, 
the rest of the masonry being in red sandstone, with inlaid ornaments of 
white marble. The four corner cupolas and the narrow-necked dome now 
make their first appearance in India. 2 

Space fails to enumerate even in the most summary fashion the architec- The 
tural marvels of Akbar's palace-city of Fatehpur Sikri, begun in 1569, ^ uIand , 
finished fifteen years later, and practically abandoned after its founder's death arwaza • 
in 1605. 3 That wondrous city bears in every part the impress of Akbar's tact 
and genius, and justifies the courtly phrase of his biographer, who declares 
that 'His Majesty plans splendid edifices, and dresses the work of his mind 
and heart in the garments of stone and clay\ The fullest possible details will 
be found in the four well-illustrated quarto volumes devoted to the subject 
by the late Mr. E. W. Smith, a work not easily to be matched. But a few 
words must be devoted to the southern gateway of the great mosque, known 
as the Buland Darwaza, or Lofty Portal, a name justified by the fact that it 
is the highest of Indian gateways, and among the largest in the world. The 
height to the summit of the finials from the pavement at the top of the stairs 
is 134 feet, and reckoned from the road at the foot of the stairs is 176 feet. 
The structure is a magnificent example of the Persian form of gateway, 
deriving its dignity from the great semi-dome in which the actual doors are 
inset — an arrangement extolled by Fergusson. The mosque, purporting to 
be copied from one at Mecca, was built in 1571. The Buland Darwaza was 
added in 1601-2 as a triumphal arch to commemorate Akbar's conquest of 
Khandesh, and probably replaced a more ordinary edifice consonant with 
the other entrances. It may be taken as typical of the innumerable similar 
gateways on a smaller scale which characterize the Mughal style (Plate 131). 

1 Ann. Rep. A. S., India, 1902-3, p. 62; and moved to Agra in the latter year, but continued 
1903-4, p. 170. to prefer Fatehpur Sikri as a residence until his 

2 Cunningham, A. S. Rep., vol. i, p. 224. Mr. death. The regular issue of coins from the 
Chisholm points out that in both the Tomb of Fatehpur mint {Dams-sultanat) continued only 
Humayun and the Taj the small corner domes until a.d. 1581 (a.h. 989). No more is heard of 
are much earlier in style than the main dome the mint until a.d. 1628-9 (a.h. 1038), the first 
and facade. year of Shahjahan, when one coin is known to 

3 From 1569 to 1584 Fatehpur Sikri was the have been struck there (Wright, Cat. Coins in 
principal residence of the Court. From 1585 to /. AT., vol. iii, p. xlvii). 

1598 Lahore was the capital of Akbar, who 


It is the most beautiful specimen of the second type of Indo-Persian archi- 
tecture, that in which marble is freely intermixed with sandstone, which was 
used alone in the earlier style exemplified by the Jahangir Mahal. 
Jahangir: The extant contributions of the Emperor Jahangir (1605-27) to 1 Indo- 
Mausoleum Persian architecture, although important, are not very numerous. The design 
of Akbar. Q £ ^ ma g n jfi cen t mausoleum of Akbar at Sikandra near Agra, in which 
Jahangir personally had an undefined share, is exceptional. The building, 
completed in 1612 (a.h. 1021), is said by one Muslim writer to have been 
under construction for twenty years, having been begun, according to custom, 
by the sovereign whose remains were to find their resting-place within it. 
But the inscriptions and the Memoirs of Jahangir seem to prove that it was 
wholly erected under his orders between 1605 and 1612. 2 It is composed of 
five square terraces, diminishing as they ascend, and the only edifice of the 
period at all resembling it is Akbar's five-storied pavilion, or Panch-Mahal, 
at Fatehpur Sikri. In all oriental houses and palaces the roof plays a part of 
great importance in daily life. From the earliest times it was used as an addi- 
tional room, being covered by awnings and screened in. These hangings, 
which were beautifully dyed and embroidered, are indispensible in Mughal 
architectural planning, being hung from pillar to pillar or supported on finely 
worked staffs. The design of Akbar's and Itimad-ud-daula's tombs is a trans- 
lation into stone of a tent-pavilion on the open roof, altogether in keeping 
with the Mughal conception of garden-tombs. 
Tomb of Another famous building of Jahangir's reign, the tomb of Itimad-ud- 
Itimad-ud- daula, near Agra, finished in or about 1628 by that nobleman's daughter, the 
Empress Nurjahan, is almost equally exceptional in other ways. The material 
is wholly white marble, enriched with pietra dura patterns in semi-precious 
stones, and equal to or surpassing in splendour the finest work of the kind 
executed in Shahjahan's reign. Although the architectural design does not 
wholly satisfy expert critics, there can be no question that the structure 
possesses rare beauty (Plate 133 a). 
The build- Passing by other notable buildings of Jahangir's reign at Lahore and else- 
ShaSSa? w ^ ere > we come to the reign of his son Shahjahan( 1627-58), during which the 
Ja ' Indo-Persian style, by universal consent, attained supreme beauty in the Taj 
Mahal (1632-53) (Plate 132), the Moti Masjid, or Pearl Mosque at Agra 
(1646-53) (Plate 133 b), and the palace at Delhi (Plate 134 a), begun in 1638. 3 
It is possible only to indicate briefly the general character of Shahjahan's 
modification of Indo-Persian architecture, give a few typical illustrations, and 
note certain points of special interest. 
isticBoTthe ^he style is essentially Persian, but with an undefinable difference of ex- 


1 Jahangir died in Oct. 1627, but Shahjahan was 2 Ann. Rep. A. S., India, 1903-4, p. 19. 

not able to ascend the throne formally until Feb. 3 Ibid., p. 184. 


A. (lol Gumba/,, or tomb of Muhammad Adil Shah, Bijapur, front view 

B. Tomb of Humayun. Cir. A.n. 1500. Photograph from the 
C. P. Bureau, Indian State Railways 

PLATE 1 21) 

















i?r "~* 


pression, and sharply distinguished from the fashions of Isfahan as well as 
those of Constantinople by the lavish use of white marble, carved and fretted, 
and supplemented by sumptuous decoration in pietra dura inlay and other 
enrichments. Coloured tiles were rarely used. Open-work tracery of incom- 
parable beauty is a marked feature, and spacious grandeur of design is success- 
fully combined with feminine elegance. It is, indeed, impossible to exag- 
gerate descriptions of the magnificence of the Delhi palace, nor is there any 
need to insist on the unearthly loveliness of the Taj, the noblest monument 
ever erected to man or woman: — 

'Not architecture! as all others are, 
But the proud passion of an Emperor's love, 
Wrought into living stone, which gleams and soars 
With body of beauty shrining soul and thought/ 1 

The chaste simplicity of the Moti Masjid commands admiration equally 
ungrudging. c Verily ,' says the inscription on its walls, 'it is an exalted palace 
of Paradise made of a single resplendent pearl, because, since the beginning of the 
population of this world, no mosque pure and entirely of marble has appeared as 
its equal, nor since the creation of the universe, any place of worship, wholly 
bright and polished has come to view to rival it. 9 That testimony is true. After 
many years there is nothing which I remember more distinctly or with greater 
pleasure than the pearly colonnades of this unequalled mosque. 

The immense enclosed complex of buildings and gardens familiarly desig- The Taj 
nated as 'the Taj*, comprises the central mausoleum, the mosque on the west, g r °|*P of 
a corresponding (jawab) edifice on the east, intended as a place of assembly U1 mgs ' 
for the congregation of the mosque and the persons invited to the annual 
commemoration services; huge gateways with many chambers, massive en- 
closing walls, and various minor structures, some of which have been ruined. 
The purpose of all was to honour the memory of Shahjahan's well-beloved 
wife, the Empress Arjumand Banu Begam, whose title Mumtaz Mahal ('The 
Chosen One of the Palace') has been corrupted into Taj. Outside the 
enclosure a considerable town grew up, named Mumtazabad, now repre- 
sented by Tajganj. The villas and tombs of the great nobles and many other 
buildings, few of which remain, once crowded the approaches and surround- 
ing space. 

The Empress died in childbirth, on 17 June 1631 N.S. (17 Zu'l Q'adah, Time 
a.h. 1040), while in camp at Burhanpur in the Deccan, where her remains 0CCU P ied . m 

a construction 

rested for six months. They were then conveyed to Agra, and the wondrous 
tomb destined to give her immortal fame was begun early in a.d. 1632, corre- 
sponding to the fifth year of Shahjahan's reign. When the plans had been 
settled to the Emperor's satisfaction work was pushed on with eagerness, 

1 Sir Edwin Arnold. 


some 20,000 men being employed daily. 1 On 6 February 1643 N.S. (17 Zu'l 
Q'adah, a.h. 1052), the anniversary of the death of the Empress, the annual 
funeral ceremony was celebrated by the bereaved husband at the new mauso- 
leum which was then regarded as complete. But the construction of the sub- 
sidiary buildings continued for many years longer. The latest inscription, 
one on the entrance gateway, was set up in a.d. 1647 (a.h. 1057). We know, 
however, from Tavernier, who witnessed both the commencement and com- 
pletion of the buildings, that operations did not cease finally until 1653, 
nearly twenty-two years after they had begun. The general superintendence 
was entrusted to Mukramat Khan and Mir Abdul Karim. 
Cost. The statements of cost recorded by writers in Persian vary enormously. 
The Badshah namah gives Rs. 50,00,000 (50 lakhs) as the cost of the mauso- 
leum itself. The highest estimate of the cost of the whole amounts to the 
huge sum of Rs. 411,48,826: 7: 6 (411 lakhs, 48 thousand, 826 rupees, seven 
annas, six pies), as stated with curious minuteness, equivalent, at the rate 
of 25. 3</. to the rupee, in round numbers to four and a half million pounds 
sterling. Intermediate estimates put the expense at three millions sterling, 
said to have been about the sum which Shahjahan resolved to spend. If the 
full value of materials be included, the highest figure is not excessive, and 
may be considered as approximately correct. Tavernier notes that the expense 
was increased enormously by the necessity of using brick scaffolding and 
centring. Such lavish expenditure on a single monument and its adjuncts is 
not likely to be repeated anywhere in the world. Shahjahan planned for him- 
self a mausoleum of equal magnificence to be erected on the opposite side 
of the river and united with the Taj by a marble bridge, but his family 
troubles prevented the realization of this gigantic conception, and so he sleeps 
beside the 'Lady of the Taj\ 'They were lovely and pleasant in their lives and 
in their death they were not divided.* 2 
Question as The foregoing details, rarely to be found stated with accuracy, help us to 
t0 ld of the rea ^ ze t * ie grandiose scale on which the whole composition known collectively 


1 The number rests on Tavernier's excellent due to S. Muhammad Latif, Agra, Historical 
authority. According to Manrique the staff of and Descriptive, p. 113. From the artistic point 
'maestros, officiales, y obreros (workman)' num- of view, the structure was essential to the perfect 
bered only about 1,000 in 1640 (Itinerario, ed. symmetry aimed at by the architect. The high- 
1649, chap, lx, p. 352). No doubt the numbers est estimate of the cost is given in Anderson's 
varied much from time to time. translation of one of the Persian MSS., now No. 

2 Correct dates have been kindly supplied by II in Or. 2030, B. M. Much of the more costly 
Wm. Irvine, Esq., I. C. S. Ret. Those in the material was presented by tributary princes, and 
books are usually wrong. For the value of the its value probably was excluded from the lower 
rupee see Tavernier, Travels in India, transl. V. estimates. Mr. Chisholm is mistaken in believing 
Ball, vol. i, 413, and Manrique (chap, lx) 'una that 'these domes seem to have been built with- 
rupia medio peso Espanor ; and for time of com- out centres' {J. Roy. Soc. Arts, Jan. 191 1 , p. 173). 
pletion, Tavernier, p. no. The explanation of He overlooked Tavernier's statement. 

the practical purpose of the jawab building is 

A. Tomb of Itimad-ud-daula, near Agra 


K. ThcMoti Masjid, Agra 
FI.ATF 133 

A. Diwan-i-khas of Delhi Palace 

B. Shrine of Sayyid Salar, Bahraieh 
PLATE 134 


as 'the Taj* was designed, and the absolute disregard of cost in realizing the 
design. Much of the credit for the vastness of the scale must be given to 
Shahjahan himself, who, of course, is solely responsible for sanctioning the 
unparalleled expense. But nobody supposes that the Emperor was his own 
architect, and much interest attaches to the question, 'Who was the architect 
by whom this noblest of monuments was designed, and to what nation did 
he belong?' The controversy on the subject, lately revived, excites some heat 
in the disputants. I approach it simply as a case in which evidence should 
be weighed and appraised impartially. Sleeman's notion that Austin de 
Bordeaux, a skilled French engineer and craftsman employed by Shahjahan, 
was the architect, and identical with the Master (Ustad) Isa (Jesus), also 
called, more correctly, Muhammad Isa Effendi, certainly is erroneous, 1 and 
his statement, first published in 1844, seems to be the sole foundation for the 
current assertions about the connexion of Austin with the Taj. Balfour's 
Cyclopaedia of India (3rd ed., 1885) boldly asserts that * Austin de Bourdeaux 
[was] an artist who erected the Taj at Agra\ For that assertion I believe that 
Sleeman 's loose guessing is the only authority. The note recently printed 
by Dr. Burgess stating that the Taj was 'most probably designed by Ali 
Mardan Khan, a Persian refugee', is opposed to the evidence of the Persian 
History of the Taj, and I do not know on what grounds it is based. 2 

The Persian MSS purporting to give the history of the Taj, the names of Ustad Isa 
the chief artists and artificers, and the cost of the buildings, appear to exhibit according 601 
many discrepancies in details, but to agree in stating that the chief designer to Persian 
and draughtsman was * Ustad (or Master) Isa\ otherwise called Muhammad authorities. 
Isa EfFcndi, who drew a salary of Rs. 1,000 a month, and was assisted by his 
son, Muhammad Sharif. The Agra copy, in the possession of the hereditary 
custodians of the monument, says that he came from 'Rum', interpreted to 
mean Turkey or Constantinople, and that his son came from Samarkand. 
Other copies are alleged to assert that the Ustad came from Shiraz in Persia. 
The title 'Effendi* sometimes given to him is an Ottoman one. No details 
of his life seem to be on record. 

The rival statement is the categorical assertion made by Father Sebastian Geronimo 
Manrique, a Spanish Augustinian friar, Visitor of his order in the East, that ^Wtect* 6 

according to 
1 Sleeman, Rambles and Recollections, ed. V. A. the Ustad was a Christian (Isahi). If 'Ustad Isa* Manrique. 
Smith, vol. i, p. 385. The MS. used by Anderson and Muhammad Sharif really were Christians, 
(Calcutta Rev.> 1873, p. 237) alleges that artist one of the objections urged against Father 
No. 1, unnamed, 'a rare plan-drawer and artist', Manrique's statement is removed, 
was a Christian, as was also Muhammad Sharif. 2 Fergusson, Hist, of Indian and E. Archit., 2nd 
Sleeman appears to have used a similar docu- ed., p. 306 note. Ali Mardan Khan, a famous 
ment, and agrees with the custodian's MS. in engineer in Shahjahan's service, constructed or 
stating that Muhammad Sharif was the son of repaired canals, and laid out the Shalamar gar- 
No. 1, whom Sleeman names as Ustad Isa. The dens at Lahore (see Index, s.v. 'Ali Mardan 
name Isa may have suggested the notion that Khan', Imp. Gaz. 1908). 


the architect was a Venetian named Geronimo Veroneo, who drew a large 
salary from Shahjahan. Manrique's words, as translated by Father Hosten, 
S.J., are: 

'The architect of those structures was a Venetian, named Jerome Veroneo, who 
went to those parts in the ships from Portugal, and died in the city of Laor [Lahore] 
shortly before my arrival. Emperor Corrombo [=Khurram= Shahjahan] gave him 
large salaries ; but it is thought that he profited so badly by them that when he died, 
they say Father Joseph de Castro, of the [Jesuit] Society, a Lombard by birth, found 
on him much less than was imagined.' 

The author then proceeds to give merely as current gossip (Jama velocissima) 
the story of Geronimo Veroneo 's supposed interview with Shahjahan. The 
positive assertion quoted above seems to be made of his own knowledge, and 
not as hearsay. 1 I attach little importance to the hearsay gossip, but much 
to the categorical allegation of fact. 

Thecredi- Father Manrique spent about a month at Agra in December 1640 and 
father J anuai 7 1 641, and thence travelled to Lahore where he met Father de Castro. 

Manrique^ He thus had ample opportunities of learning facts as well as gossip, and, 
moreover, he was on friendly terms with the greatest of the Muhammadan 
nobles, Asaf Khan, 'the ancient and only protector of the priests', and father 
of the Lady of the Taj, who gave him 'a goodly alms'. Geronimo Veroneo 
died at Lahore as stated, but was buried at Agra, some four hundred miles 
distant, where his tomb, dated a.d. 1640, still exists. Before his death he had 
spent money, presumably a considerable sum, to ransom Christians from 
prison. Father Manrique 's accuracy is thus confirmed on several points, and 
the fact that Veroneo 's body was removed to Agra for burial indicates that 
he must have been a person of considerable importance and specially con- 
nected with Agra. 

I have no doubt that the good Father's positive assertion that Veroneo was 
the architect of the Taj was made in perfect good faith, and, indeed, nobody 
impugns his personal veracity. But it is argued that he must have been mis- 
informed. The most weighty objection raised is that the Taj unquestionably 
is Asiatic in style, a development from the tomb of Humayun, and that even 
in the decoration, except perhaps the technique, as distinguished from the 
designs of the pietra dura inlay, there is little trace of European influence. 
The objection, although deserving of attention, is not conclusive, because, 
so far as I can see, there is no reason why a seventeenth-century Venetian of 

1 The Spanish text is: 'El Architeto destas Joseph de Castro de la Compania, y de nacion 

fabricas fue un Veneciano por nombre Gero- Lombardo, muy menos de lo que si imaginava' 

nimo Veroneo que pass6 a aquellas partes en las (Itinerario de las Missiones que Hzo el Padre F. 

naves de Portugal, y murio en la Ciudad de Sebastian Manrique, p. 352; ed. Roma, 1649). 

La&r poco tiempo antes de mia llegada. A este A reprint with an altered title-page appeared at 

dava el Corrombo Emperador grandes salarios : Rome in 1653. Both impressions are in the 

mas supose aprovechar tan mal dellos, que British Museum. The Bodleian possesses only 

quando murio, dizan que le hallara el Padre the earlier one, which alone I have consulted. 



Fig i. Muhammad b. Tughlak 
Figs. 2-5. Akbar 
Figs, ft- 10. Jahangir 

PLATE 135. Indo-Muhammadan Coins 

A. From Jam'i Masjid, Fatehpur Sikri 

B. South end {jalla jalalahu) of Akbar's cenotaph 


fi 8 


if; ; - -^^i'i**'- 



,,V './r*»5Rt/ /^ f 

./■:?^;^y ■•■: 


TY)}ry >■ Vv' 

C. Panel in dado of 'false mosque ' (jawab) at the Taj, Agra 
PLATE 136. Mughal Reliefs 


genius, aided by skilled Asiatic technical advisers, should have been unwilling 
or unable to design a group of buildings on Asiatic lines, in accordance with 
a general idea prescribed by Shahjahan, who must have known and declared 
what he wanted. M. Saladin, writing without reference to or apparent know- 
ledge of the documentary evidence, simply in his capacity of architectural 
expert, expresses the opinion that 'the hand of a European architect has 
traced the exact symmetries and the outlines, perhaps too regular, of the 
monument' ; adding that the decoration combines Florentine elegance with 
Oriental richness, while the breadth and symmetry of the composition give 
the design the appearance almost of a classical conception. 1 

On the whole, after considering all the arguments, including that drawn Conclusion, 
from the silence of other authors, I do not see any reason sufficient to dis- 
credit the positive assertion of Father Manrique, published in 1649 before 
the work on the Taj buildings was completed. It is not inconsistent with the 
Persian authorities. I accept their evidence as proving that Ustad Isa* 
whether he was a Turk or a Persian, was the chief architect during the later 
stages of the construction; and it is easy to understand that when the history 
of the monument was being put on record no Muhammadan writer would 
have cared to recall the leading part taken by a long-deceased Christian 
European in framing the original design. Thus the matter stands. I abide 
by the opinion expressed by me in 1893 that 'the incomparable Taj is the 
product of a combination of European and Asiatic genius'. It should be 
observed that no authority ascribes the design to an Indian architect. The 
credit for it belongs to either or both of two foreigners, one a Venetian, the 
other most probably a Turk. The lively interest felt in the question of the 
authorship of the building, which may fairly claim to be the most beautiful 
in the world, will, I trust, be considered justification sufficient for this long, 
although much condensed, disquisition on the subject. 3 

1 'Le Tadj-Mahal a Agra. — II semble que la 3 The principal Persian authorities, as enu- 
main d'un architecte europien a trac£ les symi- merated by Mr. Wm. Irvine, are (1) the con- 
tries exactes et les profils peut-etre trop rlguliers temporary Badshah namah by Abd-ul-hamid, 
de ce monument . . . cet art qui, en effet, allie Lahori (BibL Ind., 2 vols., Calcutta, 1868; vol. 

1 "elegance florentine k la richesse orientale' ii, pp. 322-31); (2) Aml-i-Salih, by Muhammad 
{Manuel d'Art Musulman, tome i, p. 571). 'Le Salih, Lahori (a.h. 1052), perhaps copied from 
Tadj n'est que le centre de la composition . . . No. 1 ; (3) a group of MSS., mostly anonymous, 
On voit done que, par l'ampleur de la composi- purporting to give the history of the Taj. One 
tion et par la sym&rie, ce plan est presque de copy at least is in the hands of the custodians 
conception classique' (Ibid., p. 575). M. Sala- at Agra, and another is in the Imperial Library, 
din's wide experience of Muslim art in countries Calcutta. The MSS. in the B. M. with similar 
where it comes in contact with that of Europe contents are Addl. 8910 (62 foil.), Or. 194 (94 
entitles his critical opinions to respectful con- foil.), Or. 195 (55 foil.)— all in Rieu, CataL, p. 
sideration. 430: Or. 2030, containing two MSS., viz. (1) 

2 The name Ustad Isa commonly used is in- by Manik Chand, foil. 1-30, and (2) notice of 
correct. The fuller form, Muhammad Isa, really the Taj Mahall, foil. 32-81 , nearly identical with 
means 'Muhammad the son of Isa*, as Professor Addl. 8910. This is the version partly translated 
Margoliouth points out. by Capt. Anderson in C ale Mevietv,volAvu (1873), 

3>P Bb 


Archi- The long and unhappy reign of Aurangzeb Alamgir (1659-1707) 1 was 
te AurauiB- mar ^ ec ^ by a ra P^ decline in art, including architecture. The emperor was 
zeb's reign; Inore ea g er to throw down Hindu temples than to construct great edifices of 
and after- his own. Some few buildings of his time, however, are not without merit; 
wards, for instance, the tall minarets of the mosque which he caused to be erected 
at Benares on the site of the holiest temple are graceful objects well known 
to all travellers in India. The principal mosque at Lahore (1674), almost a 
copy of the great mosque at Delhi, but inferior to that noble building, is 
described by Fergusson as being 'the latest specimen of the Mughal architec- 
tural style'. The emperor's own tomb at Khuldabad near Aurangabad in the 
Deccan is insignificant. The buildings in Persian style of Aurangzeb 's age, 
being merely examples of growing deterioration, are not worth detailed study 
or illustration. The tomb of Nawab Safdar Jang of Oudh near Delhi (1756), 
a passable copy of the mausoleum of Humayun, is marred by wretched plaster 
decoration in the interior. The shoddy buildings of the Nawab Vazirs at 
Lucknow are pretentious abominations. 
Composite In many places modern architects have effected a graceful compromise 
style, between the Hindu and Muhammadan styles by combining Persian domes 
with Bengali bent cornices and Hindu or half-Hindu columns. Excellent 
examples of this pretty though feeble style, as used for both civil and religious 
buildings, are to be seen at Mathura and in hundreds of other localities. It 
is quite impossible to tell merely from inspection of the architecture whether 
a building is intended for Muslim or Hindu use. The modern part of the 
ancient shrine of Sayyid Salar in Northern Oudh (Plate 134 b) is a good 
example of the style in its more Muhammadan form. 

Thus the story of Indo- Muhammadan architecture ends, as it began, with 
the subjection of foreign innovations to the irresistible pressure of native 
taste and methods. 

PP- 2 33-7- The abov e are on p. 958 b of Rieu, vol. vi (1910), pp. 281-8). Mr. Keene also ae- 
tata/. Or.2031 (Rieu,p.i044«),No.IV,foll.i48- cepted his statement (Turks in India, pp. 251-5). 
226 is another copy of Manik Chand's account. Mr. Havell erroneously denounces the positive, 
The Agra version, used and partly translated contemporary evidence of the Spaniard Father 
by Sleeman, S. Muhammad Latif (Agra, His- Manrique as 'the old Anglo-Indian legend' (J. 
torical and Descriptive, Calcutta, 1896) and Mu- Roy. Soc. Arts, Jan. 191 1, p. 180). 
hammad Moin-ud-din (History of the Taj, Agra, The question has not been throughly threshed 
1905), seems to exist in more than one form, out yet, the Persian MSS. especially requiring 
F. Manrique's account is discredited by the two careful examination and comparison. Mr. Irvine 
Muhammadan writers named, as well as by Mr. has made a beginning at my request by examin- 
E. B. Havell, 'The Taj and its Designers* (Nine- ing the Manik Chand MSS., Or. 2030 and 203 1 . 
teenth Century and After, June 1903 ; reprinted They are of no independent value as authorities, 
in Essays on Indian Art, &c, Madras, n. d.), and the text of Manik Chand's late compilation 
and by Sir J. Marshall in Ann. Rep. Arch, in Or.2031 is merely a copy of that in Or. 2030, 
S., India, 1904-5, pp. 1-3. Father Hosten, S.J., made for the use of Sir H. Elliot, 
stoutly defends Manrique in his article, 'Who « Shahjahan was deposed in 1658. Aurangzeb's 
Planned the Taj ?' (J. and Proc. A. S. B., N. S. formal accession took place in 1659. 

Chapter Fourteen 



uhammadan architecture, excluding the styles most deeply affected Limitations 
by Hindu influence, and in spite of infinite variety in detail, presents, ^ a ^^ al " 
as we have seen in the last chapter, a character of general uniformity 
throughout the Muslim world, partly due to the practically universal use of 
pointed arches and domes, and partly to the free interchange of architects 
between different countries, resulting in the frequent imitation of foreign 
models. Muhammadan decorative art presents a similarly uniform character 
by reason chiefly of the Koranic prohibition of images, which, although not 
universally respected, was observed in all ages and countries sufficiently to 
impose narrow limits on the field open to the creative artist. The orthodox Elements of 
Muslim decorator has found himself in practice constrained to restrict his Musalman 
invention to the dexterous use of calligraphy, geometrical patterns, and floral decoratlon - 
devices. However varied in detail the application of those elements may be, 
the effect is necessarily flat and somewhat monotonous. 

In this chapter a few pages will be devoted to the art of calligraphy as dis- Scope of 
played in coinage, to the rare figure types on coins and gems, and to the ex- this chapter, 
ceptional attempts at stone sculpture in the round or in high relief. They will 
be followed by a condensed account of the leading forms of Musalman archi- 
tectural decoration arranged under the heads of Calligraphy and Decorative 
Reliefs, Lattices, Inlay and Mosaic, and Enamelled Tiles. No attempt will be 
made to follow the Muhammadan decorator in his treatment of minor objects 
of luxury, which is essentially the same as that of architectural ornament. 
Even in his floral designs the tendency of the Muslim artist is in favour of 
a formal, over-symmetrical conventionalism, calculated to harmonize with 
his favourite geometrical patterns. Akbar's taste inclined to a more interest- 
ing naturalism, as displayed in the exquisite ornament on his cenotaph 
executed a few years after his death, and designed in his spirit. The art 
of painting, in the exercise of which greater liberty was assumed, will be 
discussed at considerable length in the concluding chapter. 

It is a common error to suppose that the ancient Semitic prohibition of Transitional 
images, repeated in the Koran, invariably prevented Muhammadan artists Musalman 

r ** * i r r i • • \. 1 A COinS With 

from representing the forms of living creatures, real or imaginary. As a figure types 
matter of fact, the prohibition, although respected as a rule, has been dis- 
regarded frequently in almost every Musalman country from the earliest ages 
of Islam to the present day, and especially in those countries, like Persia, 
where the Shia sect prevails. The introduction of figure types in many 


ancient Muhammadan coinages was due to the business necessity of main- 
taining for a time the forms of currency to which people had become accus- 
tomed. For example, when the Sassanian dynasty of Persia fell in the seventh 
century the newly appointed Arab governors continued to issue coins in the 
familiar national form with the king's head, distinguished from the native 
issues merely by the insertion of Arabic legends in minute characters. In 
India Muhammad of Ghor was obliged to accept a similar compromise and 
even to issue coins bearing the image of a Hindu goddess. 
Orthodox In most Muhammadan kingdoms such numismatic compromises with 
calligraphic idolatry were only temporary, and the die-cutters of the Muslim sovereigns 

coinage. were ordinarily obliged to content themselves with calligraphic devices, on 
which much skill was lavished. The coins issued by Sultan Muhammad bin 
Tughlak of Delhi (a.d. 1324-51), who has been called 'the prince of moneyers', 
are exceptionally brilliant examples of calligraphic art. A specimen is shown 
in Plate 135, Fig. 1, and may be taken as a typical illustration of well- 
executed Muhammadan orthodox coinage. 

Peculiar Akbar, notwithstanding his scant respect for orthodoxy, submitted as a 
iSS Akh ° f ru ^ e t0 Koranic restrictions in the types of his coinage, which exhibits many 
ar ' varieties of artistic ornamental writing. A highly elaborated specimen, a rupee 
struck at Agra, is shown in Plate 135, Fig. 5. On three occasions only did he 
permit himself the luxury of figure types, and the pieces struck on those 
three occasions are medals rather than ordinary current coins. A falcon 
(ibid., Fig. 2) commemorates the capture of Asirgarh, the strong fortress 
commanding the road to the Deccan. The Brahmini goose appears on an 
Agra coin (ibid., Fig. 4). Both birds are well designed and surrounded by 
pretty floral scrolls. A curious piece, exhibiting the figures of a crowned 
archer and a veiled lady (ibid., Fig. 3), is a memorial of the submission in 
a.h. 1013 (a.d. 1604-5) of the King of Bijapur, who gave his daughter in 
marriage to Prince Daniyal, Akbar's youngest son. 
The freaks Jahangir, although officially a better Musalman than his father, was less 
of Jahangir. orthodox in his coinage. He alone of all the Muhammadan sovereigns of India 
dared to put his own portrait on coins intended for circulation. He habitually 
disregarded the Prophet's prohibition of strong drink, and was not ashamed to 
show himself on the coinage holding a goblet of wine (ibid., Figs. 7, 8). He 
also indulged in the freak of issuing a coinage, both gold and silver, on which 
the months were indicated by pictorial symbols of the zodiacal signs, instead 
of by words or numbers (ibid., Figs. 9, 10). The figure of Virgo is a European- 
ized angel. 1 The great bulk, however, of Jahangir's coinage is perfectly 
orthodox in form. His five-mohur piece (ibid., Fig. 6) is an excellent example 

1 'Mais les beaux chefs-d'ceuvre numismatiques nages d'un dessin si parfait et d'un relief si 
sont les dilicieuses monnaies d'or de Djehangir prlcis et si vif (Migeon, Manuel (Tart musulman, 
(1605-1628) frappees d'animaux et de person- tome ii, p. 164). 


of first-class calligraphy. Many of the coins of the later Mughal emperors are 
well executed, but the specimens given are enough to illustrate the general 
character of calligraphic dies. 

Muhammadan gems and seals with artistic devices other than calligraphic Cameo with 
are necessarily extremely scarce. Mr. King, after referring to the rarity of g? p S^ f 
cameos in purely Oriental style, mentions one conspicuous Muhammadan a J an# 

4 The most remarkable example of all in the Oriental class/ he writes, "although of 
modern origin, came to my knowledge among the Webb gems (when sold by Christie 
and Manson in 1854), the subject being the teatperformed by Shahjahan in cleaving 
asunder a lion which was mauling a courtier. The inscription consists of two parts 
[namely], "The portrait of the Second Sahib-Qiran, Shahjahan the victorious emperor/, 
and the artist's signature "Made by Kan Atem" [sic, the reading is impossible]. The 
gem probably must be dated early in his reign, for it shows Shahjahan with a moustache 
but no beard. He wears a long double row of big pearls round his neck, and, as a 
pendant, a great convex gem, perhaps the Kohinoor.' l 

The actual feat commemorated here was performed by Shahjahan, as Prince 
Khurram, when he rescued Anup Rai from the jaws of a tiger. 2 

Another notable Indo- Muhammadan artistic gem which has come to my Elephant 
notice is the beautiful sardonyx cameo of the Mughal period, bought by Sir cameo - 
John Marshall some years ago and now in the Lahore Museum, which is 
36 inches broad and 3-3 high. It represents two elephants with riders, lock- 
ing their tusks and trunks together apparently in combat. 3 


Musalman representations of living forms in stone or stucco of various ages Musalman 
from Egypt, Mesopotamia, Asia Minor, and Spain have been published, and ?£ ul 2 ure 
although rare in any one country, amount in the aggregate to a considerable countr j eSt 
number. A few bronze figures of a ruder kind, mostly dating from the time 
of the Fatimite sovereigns of Egypt and Syria (a.d. 969-1 171), are also 
known. 4 

In India the examples of sculpture in the round or in high relief, executed Indian 
to the order of Muhammadan princes, but probably by the hands of Hindu ^^J 1 * 8 ' 
artists, are extremely few; the most notable of which any remains exist being con fined to 
the elephants, sometimes with riders, set up at the gateways of fortresses, in elephants 
continuance of Hindu custom. Nearly every stronghold of importance had and riders - 
its Elephant Gate (Hathipol). The portal of that name at Akbar's city of 
Fatehpur Sikri is still guarded by the mutilated figures of two colossal 

1 Ancient Gems and Rings, London, 1872, pp. 3 Ann. Rep. A. S., India, 1905-6, p. 40, Fig. 1. 
314-16. 4 Catalogued by Migeon, Manuel (Tart tnusul- 

2 Memoirs ofjahangir, transl. Rogers and Beve- man, tome ii, chaps, ii, vii. 
ridge (1909), pp. 185-8. 


elephants, perched on supports iz\ feet high, whose trunks originally were 

interlocked across the entrance. Aurangzeb caused the heads to be knocked 

off. The elephants, being clumsily made up of large blocks of hewn stone 

laid in mortar and joined by iron cramps, are of no account as works of art. 1 

Other statues, presumably of Hindu origin, which once guarded the Elephant 

Gates of Gwalior, Mandu, and other fortresses have been destroyed. 

Elephants William Finch, the English traveller, who visited Agra early in the reign of 

and riders Jahangir (1610), there saw 'a second gate, over which are two Rajaws in stone, 

at Fort* w ^° were s ^ a * n * n l ^ e K* n g s Derbar before the King's eyes, for being over- 

' bold in speech, they selling their lives bravely, in remembrance of which they 

are heere placed*. From a note appended by Purchas, it would seem that the 

two 'Rajaws* were mounted on elephants. The note states: — 

'It is said that they were two Brothers, Resboots, Tutors to a Prince, their nephew, 
whom the King demanded of them. They refused, and were committed; but drew on 
the Officers, slew twelve, and at last, by multitudes oppressing, were slain, and here 
have Elephants of stone and themselves figured.' 2 

Mr. Keene is of opinion that 'the allusion probably is to the three sons of 
Akhiraj, son of Akbar's brother-in-law, Raja Bhagwan Das of Jaipur, killed 
in a fight arising out of a tumult caused by themselves in the Palace'. 3 Who- 
ever the originals may have been, Finch's testimony is clear that two statues 
of men over one of the gates of the Agra Fort were erected by order of either 
Akbar or Jahangir, and Purchas's note indicates that they were mounted on 
Current The similar, but wholly distinct, statues of elephants with riders which 
errors about formerly stood at the Delhi Gate of the Delhi Fort, and of which fragments 
statues st ^ ex * st > h ave been l ^ e subject of so much discussion and misunderstanding 
' that it is desirable to state the facts as recently elucidated by the officers of 
the Archaeological Survey. The statements in all the ordinary books of 
reference are erroneous. The Delhi groups certainly possessed considerable 
merit as works of art, and the riders at least must be counted as examples 
of sculpture executed to Musalman order. 
Life-size But before going into the history of the much debated Delhi statues it is 
Agra^dace we ^ to note *^ at J a ^angir, in the eleventh year of his reign, had caused life- 
garden. s * ze figures of the Rana of Chitor (Amar Singh) and his son Karan to be 
carved in marble and set up in the palace garden at Agra, below the window 
(darshanjharokha) where the Emperor made his daily public appearance. 4 

1 E. W. Smith, Fathpur-Sikri, Part III, p. 33, for his note. 

PL LV. Small elephants, poorly modelled, occur 3 The incident, which occurred 28 Dec. 1605, is 

among the decorative sculptures of various described by Jahangir {Memoirs of Jahangir \ 

Mughal buildings. transl. Rogers and Beveridge (1909), p. 29). 

2 Hakluytus Posthumus or Purchas His Pilgrims, 4 Beveridge, J. R. A. S., 1909, p. 743; Memoirs 
in 20 volumes, MacLehose, Glasgow, mcmv, of Jahangir, transl. Rogers and Beveridge (1909), 
vol. iv, p. 72. Purchas does not cite any authority p. 332. 


This undoubted fact, recorded by Jahangir himself, is clear proof that the 
early Mughal emperors had no objection to life-size statues of men, and some- 
times had them made. No trace has been found of the garden effigies, which 
appear to have been carved at Ajmer and thence sent to Agra. 

The history of the Delhi groups may be summarized as follows: 

In 1663, early in Aurangzeb's reign, Bernier saw and warmly admired the History of 
effigies of two elephants with riders which then stood at the Delhi Gate of ^ ^ elhl 
the Delhi Fort. A few years later they were seen still in position by Thevenot. ^JLJJ* 
Subsequently they were broken up by order of Aurangzeb, and the fragments 
cast away. In 1863 the buried fragments were found, and after an interval 
some of them were pieced together and made into an absurd monster, which 
was set up in the Queen's Gardens, with a false inscription based on an erro- 
neous guess of Alexander Cunningham. Lord Curzon expressed a desire 
to reconstruct the groups from the broken pieces, but it proved impossible 
to carry out his wishes. A skilled European artist, Mr. R. D. Mackenzie, 
was commissioned to make a new model. He did so, and his work is 
preserved in the Delhi Museum. Native sculptors were then instructed to 
make two elephants without riders from that model. They carried out their 
orders as well as they could, and their productions have been erected on the 
old pedestals. 

The original elephants were made of black stone (? marble), and according Material, 
to Sir John Marshall, who has examined the fragments carefully, were 
'moulded with masterly skill and care' and 'true to nature'. They are believed 
to be of Hindu origin. The riders were carved in red sandstone, and 'their 
material, style, and technique establish beyond a doubt', according to the 
same authority, that they were 'carved by Mughal sculptors', a phrase pre- 
sumably to be interpreted as meaning 'sculptors of the Mughal period'. The 
actual artists are more likely to have been Hindus than Musalmans. Whoever 
wrought them, the statues of the riders also seem to have been good, well- 
finished work. 

Bernier was told that the riders represented the brothers Jaimall and Palta, Identity of 
the brave heroes of the defence of Chitor in 1568, who 'with their still braver the riders, 
mother, immortalized their names by the extraordinary resistance which they 

opposed to the celebrated Ecbar It is owing to this extraordinary devotion 

on their part that their enemies have thought them deserving of the statues 
here erected to their memory.' I see no reason to doubt the truth of this 
explanation, which is confirmed by the fact already noticed that Jahangir 
erected statues of two other chiefs of Chitor in the palace garden at Agra. 
But if the statues of the riders date from the time of either Akbar or Jahangir, 
they must have been placed originally somewhere else, and subsequently 
shifted by Shahjahan who built the Delhi (Shahjahanabad) Fort. There is, 
however, nothing in Bernier's statement to indicate that the statues were not 

Statue of 
a horse. 


ordered by Shahjahan, who may have been influenced by the precedent set 
at Agra by his father. 1 

A life-size statue of a horse in red sandstone standing on the left-hand side 
of the Sikandra road about two miles from Agra, near the garden of Suraj 
Bhan, and opposite a masonry Muhammadan tomb, may be a work of Mughal 
age, but nothing definite about it is known, and no photograph is available. 
Mr. Beglar's conjecture that it may date from the time of Sikandar Lodi, the 
idol-breaking Sultan in the fifteenth century, is extremely improbable. 2 

Calligraphic The Arabic alphabet in its various forms, as used for writing both the 
decoration; Arabic and Persian languages, is so well adapted for decorative purposes, 
Ajmer. ^^ a i most every Muhammadan building of importance is freely adorned 


From Kadam Rasul Mosque, Gaur, a.d. 1480. 

with texts from the Koran or other inscriptions arranged decoratively to form 
part of the architectural design, and often signed as the work of famous calli- 
graphists. A good early Indian example of such calligraphic decoration is 
afforded by the great arch of the Ajmer mosque, where the outer line of 
writing is in the angular Kufic script, while the other lines are in a more 
rounded Arabic character. Later examples from Indo- Muhammadan build- 
ings of all styles and ages might be multiplied indefinitely. 

Musalman figure sculpture in the round has, as we have seen, slight artistic 
value and is interesting chiefly as a curiosity. But Musalman decorative 
sculpture in bas-relief applied to architecture may fairly claim on its merits 
to take at least equal rank with first-rate Italian work of the kind. 

1 For fuller details see Ann. Rep. A. S., India, p. 183; Latif, Agra, Descriptive and Historical, 
1905-6, pp. 33-42. p. 183. 

2 Cunningham and Beglar, A. S. Rep., vol. iv, 

PLAT]* 137. Panels from Sarangpiir Mosque, Ahmadabad. Cir. \.n. 1500 

PLATE 138. Windows of Sidi Sayyid's Mosque, Ahmadabad. Cir. a.d. 1500 


'L'on ne saurait/ writes M. Migeon, 'trop recommander l'&ude des arts de l'lslam 
aux artistes d£corateurs et aux ouvriers d'art. Par la puissante beaut6 de ses formules, 
par sa fantaisie toujours r6gie par les lois les plus rigoureusement logiques, par le 
rayonnant 6clat de la couleur, il n'est pas d'art qui offre plus de richesse d6corative et 
plus de souveraine harmonie. II renferme des germes feconds qui transplants doivent 
fructifier k l'infini.' 1 

The validity of the concluding proposition may be doubted, and it seems 
to me by no means certain that the teaching of Musalman art to European 
craftsmen would produce satisfactory results. But, however that may be, 
M. Migeon's enthusiastic praise of the decorative quality of Muslim art 
generally may be accepted. The best Indian specimens, with which alone 
we are concerned at present, could not be surpassed as pure decoration. 
Among all the many varieties of Muhammadan decorative designs none are 
more agreeable than the best of those carved in relief on the Mughal build- 
ings, from the time of Akbar to that of Shahjahan. The work of Akbar's time 
being more naturalistic, is more interesting than that of the later period, 
which is formally conventional, with a tendency to monotony. 

The choicest Italian work does not surpass, if it equals, the superb carving Akbar's 
on the white marble cenotaph of Akbar, which occupies the centre of the top- cenotaph, 
most story of his mausoleum at Sikandra. 

'The two oblong sides and the top are adorned with the ninety-nine titles of the 
Creator in alto-relievo, set in delicate Arabic tracery (Plates XI and XV of Akbar* s 
Tomb). The words Allahu Akbar jalla jalalahu are inscribed on the head and foot, 
set in panels surrounded by most beautiful and delicate floral ornamentation (ibid., 
Plates XVI, XVII ; ante, Plate XCIX, Fig. C). The carving, which is most exquisitely 
done, is in very low relief, and savours of Chinese workmanship. Amongst other 
flowers and plants portrayed one recognizes the lily, the almond, and the dahlia, all of 
which are found carved or painted upon Akbar's palace at Fatehpur-Sikri. In the left- 
hand corner of each of the panels, cloud-forms carved after a most distinctive Chinese 
¥pe are noticeable. Similar cloud-forms are met with upon the dado panels in the 
urkish Sultanah's house at Fatehpur-Sikri, and it is generally supposed that they were 
executed by Chinese workmen/ 

But forms of a like kind so often appear in Persian art that it is unnecessary 
to assume the employment of Chinese craftsmen by Akbar. 

4 Small butterflies and insects flitting from flower to flower are carved upon the 
panels. Upon the top of the cenotaph a qalam-dan or pen-box is sculptured, signifying 
that the tomb is a man's, in distinction from a woman's, which is generally provided 
with the takhti or slate.' 2 

Shahjahan's architects relied on inlay rather than relief sculpture for Relief at 
decoration; but at the Taj dados are very effectively adorned by conventional Taj. 
flowers cut on red sandstone in low relief (Plate 136). 

1 Manuel d'art tnusulman, tome ii, p. 454. 2 Akbar's Tomb, Sikandarah, p. 15. 

3292 c c 

Inscribed principal Mihrab % Jam'i Mosque, Fatehpur-Sikri. 

Vase motive panel, east false gate of Akbar's tomb. 


Ahmadabad Plate 137 illustrates the totally different style adopted in the much earlier 

reliefs. Sarangpur mosque at Ahmadabad, erected about a.d. 1500. The tree motive 

is characteristic of Ahmadabad. The whole design is far more Hindu than 



Hindu Pierced stone screens or lattices used as windows were not unknown to 
lattices. Hindu architects, and were especially favoured by the builders of the highly 
decorated temples in the Mysore, Deccan, or Chalukyan style. At Pattadkal 
and in the Kailasa at Ellora beautiful lattices are to be found. At Belur there 
are twenty-eight such windows, all different. Some of these are pierced with 
merely conventional patterns, generally star-shaped, with bands of foliage 
between; others are interspersed with figures and mythological subjects. 1 
Musalman. But the Musalman architects, who were more restricted than the Hindus 
in their liberty of decoration, developed the art of designing and executing 
stone lattices to a degree of perfection unknown to other schools. Endless 
variations of geometrical patterns, generally pleasing, although wearisome 
when examined in large numbers, are the most characteristic forms of Muham- 
madan lattice-work, which is seen at its best in the Gujarat (Ahmadabad) and 
Mughal buildings. The designs both in Gujarat and the earlier Mughal work 
have been often influenced by Hindu tradition. The Muslim artists used the 
lattice, not only for windows, but also for the panels of doors and for screens 
or railings round tombs with excellent effect. 
Ahmadabad. The most beautiful traceries at Ahmadabad are to be seen in ten nearly 
semicircular windows of Sidi Sayyad's mosque built about a.d. 1500, which 
may be fairly described as the most artistic stone lattice-work to be found 
anywhere in the world. I give two examples — one with geometrical patterns, 
and the other with the tree motive of Hindu origin, which should be com- 
pared with the modern carving in the Mysore Palace. 

'It would be difficult', Fergusson observes, 'to excel the skill with which the vegetable 
forms are conventionalized just to the extent required for the purpose. The equal 
spacing also of the subject by the three ordinary trees and four palms takes it out of 
the category of direct imitation of nature, and renders it sufficiently structural for its 
situation; but perhaps the greatest skill is shown in the even manner in which the 
pattern is spread over the whole surface. There are some exquisite specimens of 
tracery in precious marbles at Agra and Delhi, but none quite equal to this/ 2 

The material of the Ahmadabad windows is Gujarat sandstone. (Plate 


1 Fergusson, Hist, of Ind. and E. Archit., ed. menon, not unfamiliar to the Indian traveller, of 

1910, vol. i, p. 440, with a bad illustration. a banyan-tree growing out of and around a palm, 

1 Hist. Ind. and E. Archit., ed. 1910, vol. ii, until in its snake-like entanglements of root and 

p. 236. The companion window (PI. IV of branch the banyan strangles its foster parent' 

Burgess) represents more distinctly 'the pheno- (Indian Art at Delhi, p. 122, PI. XXVII). 

'• ' ,[; ;\\v '■! I ;!. , 1 - , .!j|i , 'l"! , '! , !!: , ;; 1 |ji" , .! , r M'Hj:;"; ''I!:"""!";!" m- ij-^jmi t ■;i- , l'!!v'- 1 , !i|i: , . i ;: : - ij:||lj»|!.,||i: 

PLATE 139. Marb.c verandah screen, tomb of Salim Chisbti, Fatehpur Sikri. a.d. 1571 

PLATE 140. Marble screen round the cenotaph of the Taj 


PLATE 141 . l T pper part of a corner turret, Itimad-ud-daula's tomb, Agra, showing pietra dura inlay 

and marble mosaic 

PLATE 142. Pietra dura inlay on the cenotaph of the Taj 


The examples of well-designed and well-executed open-work tracery, Mughal, 
chiefly in marble, at Agra and Delhi are so numerous that it is difficult to 
select typical specimens. But it is impossible to do better than to illustrate 
the style of Akbar's time from the tomb of Salim Chishti at Fatehpur-Sikri, 
built a.d. 1571. Plate 139 reproduces some of the marble screen-work en- 
closing the verandah, exhibiting an elegant and effective combination of a 
geometrical pattern with a conventionalized plant design. 

The well-known railing round the cenotaph in the Taj may be taken as an 
unsurpassed example of the art in Shahjahan's time (Plate 140). The lines 
of the repeating pattern in this case are more like Italian renaissance than 
Asiatic work. According to Sir John Marshall this is the only case in which 
Italian influence can be discerned in the decorations of the Taj. However, 
it suggests a textile design translated into relief in stone, and considered as 
such is purely eastern. 


The device for breaking the monotony of a wide wall surface by inserting Marble inlay 
broad bands of white marble, as employed in the fourteenth century on the and mosaic - 
tomb of Tughlak Shah, and a few years earlier on Ala-ud-din's gateway, was 
commonly used in the Musalman art of Central Asia, Syria, and Egypt, and 
was freely adopted for Christian buildings in Italy. In Akbar's time this 
early severe form of decoration was supplemented by mosaics made up after 
the Roman and Byzantine fashion from small tesserae, which were combined 
in Persian geometrical patterns. The great mosque at Fatehpur-Sikri offers 
many examples. Sometimes the effect was enhanced by the insertion of little 
bits of blue or green enamel. 

A great innovation was effected by the introduction of the form of inlay Pietra dura. 
known technically by the Italian name of pietra dura, which is composed of 
hard precious or semi-precious stones, such as onyx, jasper, carnelian, &c, 
cut into thin slices and neatly bedded in sockets prepared in the marble. This 
process, of which the best comparatively small specimens are to be seen at 
Florence, is capable of producing charming decorative effects when executed 
by capable workmen. In India, where expense was disregarded, it was 
applied to buildings on an enormous scale. The bold floral mosaics made of 
marble or red sandstone which appear on the south gateway of Akbar's tomb 
(1605-12) are nearly equivalent in effect to pietra dura work, but are not 
identical with it. 1 The Mughal kings evidently loved flowers, which are 
admirably treated in all forms of art patronized by them. The motives, 
are borrowed from Persian art. Nowhere else are the assimilating, trans- 
forming powers of the Indian genius more evident, both in the colour and 
the perfect freedom of the lines. 

1 E. W. Smith, Akbar's Tomb, Pis. XLI, XLII. 


Earliest The earliest Indian example of true pietra dura, according to Major Cole, 
example. j s ^jj to b e ^^ j n t h e q \ Mandril, a domed pavilion in the small Jagmandir 
palace, at Udaipur in Rajputana, built in or about 1623 f° r Prince Khurram, 
afterwards the Emperor Shahjahan, while he was an exile from his father's 
court. The process is very extensively employed on the approximately con- 
temporary mausoleum of Itimad-ud-daula near Agra, erected by his daughter 
Nurjahan after her father's death in a.d. 1621. The general effect of the 
pietra dura decoration is well shown (so far as it can be without colour) in 
Plate 141, which represents one of the white marble turrets at the corners of 
the tomb. The older style of marble mosaic is seen in the lower panels. 

Shahjahan 's Shahjahan (1627-58) wholly abandoned mosaic in favour of pietra dura, 
buildings. w hich probably he learned to admire while residing in the Jagmandir palace 
at Udaipur before his accession. The decoration is applied so lavishly in the 
Taj and the palaces of Agra and Delhi that volumes might be filled with 
reproductions of the designs, which are familiar to most people from modern 
copies. One plate will be enough to show their character (Plate 142). They 
are remarkable for their restraint and good taste, and are superior to the 
similar work in the Delhi palace. 
Origin of The Florentine pietra dura inlay, a revival of the ancient Roman opus sectile, 

pietra dura fi rst appears, according to Major Cole, in the Fabbrica Ducale built by 
wor " Ferdinand I, Grand Duke of Tuscany, in 1558. The earliest certain Indian 
examples being considerably later in date and identical in technique, a strong 
presumption arises that the art must have been introduced into India from 
Italy. There is no doubt that the Mughal sovereigns freely entertained artists 
from Europe as well as from most parts of Asia. The presumption is not 
rebutted by the obvious fact that the designs of the Mughal work are essen- 
tially Asiatic, and in the main Persian, because the ordinary Indian practice 
is to transpose foreign importations, so to speak, into an Indian key. Persian 
designs were readily assimilated, but in the seventeenth century nobody in 
India cared much for outlandish European forms, or wanted to have them. 
Now, of course, things are different, and European forms are fashionable 
because the government is English. If Sir John Marshall was correctly in- 
formed when he wrote some years ago that 'pietra dura work in a rougher 
and earlier stage than was hitherto known' had been discovered in the ruins 
of the Khalji mausoleum at Mandu in Central India, the presumption of 
Italian origin would no longer hold good, because Mahmud Khalji, in whose 
honour the mausoleum seems to have been erected, died in 1475. 1 But the 
details given in an earlier report suggest that the remains found were those 
of marble mosaic, not of pietra dura inlay. 
Modern The decline and fall of the Mughal empire during the eighteenth century 

ptetradura necessarily involved the rapid decay of the arts which had ministered to the 

1 Ann. Rep. A. S., India, 1904-5, p. 5. 


splendour of the imperial court. Among other arts that of producing pietra 
dura inlay had been almost forgotten until about 1830, when Dr. Murray, 
Inspector-General of Hospitals, induced the craftsmen to revive it for com- 
mercial purposes. Since that time it has been practised sufficiently to provide 
a constant supply of pretty trifles for European tourists and visitors, but 
nobody dreams of decorating a building in the fashion which appealed to 
Shahjahan the Magnificent. The plaques and other inlaid objects now made 
at Agra are too familiar to need illustration. A selection of first-class speci- 
mens is figured in Indian Art at Delhi, Plate 17-A. 

Inlay with mother-of-pearl occurs at Salim Chishti's tomb, Fatehpur Sikri, Glass 
and elsewhere. Glass mosaics are to be seen in several Shish Mahals, or mosaics, 
'glass chambers', at Udaipur, Amber, Agra, Lahore, and other places. Those 
in the ceiling of the Shish Mahal, Lahore, are said to be particularly well done. 
But such meretricious bedizenment certainly is not fine art, and need not be 
further discussed. 


The practice of decorating wall surface with coloured enamelled bricks or Ancient 
tiles was of very ancient date in Persia, and derived ultimately from Baby- Persian tiles - 
Ionia. The Lion and Archer friezes from Susa now in the Louvre, and well 
reproduced by Perrot and Chipiez, are the best examples of the art as prac- 
tised in Achaemenian times. 1 But the style of those friezes is not imitated 
in any extant Indian work. The Indo-Muhammadan enamelled or glazed 
tiles were copied from a much later development of the art in Persia, where 
the ancient technique apparently was never wholly forgotten. This later 
Persian work shows traces of Chinese influence. 

M. Migeon believes that the Muhammadan use of enamelled tiles in Persian tiles 
numerous Persian buildings of the fourteenth and fifteenth centuries was of Timund 
derived from Khorasan. From that province it seems to have spread to 
Samarkand, where we find coloured tile facings on the tomb of Timur at 
the beginning of the fifteenth century. Coloured tiles had become known in 
India at an earlier date, certainly in the first quarter of the fourteenth and 
possibly in the thirteenth century, but the Timurid tradition of the Mughal 
emperors made them still more fashionable. The Indian work, although 
sometimes very good, is not admitted by experts to equal the best Persian 
in either the beauty of the colours or the brilliancy of the enamel. 2 

The tomb of Baha-ul-hakk at Multan, built between a.d. 1264 and 1286, Early tile- 
still retains, or retained in 1882 when Cunningham wrote, 'some fairly pre- JJjJjJ 
served specimens of diaper ornament in glazed tiles', which may or may not 

1 Perrot and Chipiez, Hist, of Art in Persia, pavements. 

London, 1892, p. 420 and plates. Persian and 2 Migeon, Manuel d'art musulman, tome ii, pp. 

Indian tiles are not strong enough for use in 295, 296. 



be contemporaneous with the building in its original form. The tomb was 
extensively rebuilt in the seventeenth century, and Sir John Marshall is of 
opinion that most of the tile-work belongs to that age. The tomb of Baha- 
ul-hakk's grandson, Rukn-ud-din (a.d. 1320), a well-designed octagonal 
domed building of brick, in the same city, has its whole exterior 

'elaborately ornamented with glazed tile panels and string courses and battlements. 
The only colours used are dark blue, azure, and white; but these are contrasted with 
the deep red of the finely polished bricks, and the result is both effective and pleasing. 
These mosaics are not, like those of later days, mere plain surfaces, but the patterns 
are raised from half an inch to two inches above the background. This mode of con- 
struction must have been very troublesome, but its increased effect is undeniable, as 
it unites all the beauty of variety of colour with the light and shade of a raised pattern.' 

The tile from Baha-ul-hakk's tomb figured by Cunningham exhibits the 'key 
pattern* in white on a dark blue ground; that from Rukn-ud-din's tomb has 
a white ground with interlacing circles in dark blue, the interspaces being 
partly filled by six-petalled stars and polygonal blocks in pale azure. 1 
Gaur tiles. Two of the mosques at Gaur in Bengal, the Tantipara and Lotan (Lattari), 
erected between a.d. 1475 and 1480, are decorated with true encaustic tiles. 
Those of the Lotan mosque are the best preserved. A collection of earlier 
glazed tiles from Gaur in the Victoria and Albert Museum, South Kensing- 
ton, is described as having 'a marked Hindu character, quite distinct from 
the blue, and diapered, and banded tiles which are distinctive of Mahom- 
madan manufacture elsewhere in India, before the florid designs of the Mogul 
period came into vogue.' 2 It is possible that the art, however introduced 
originally, may have been known to the Hindus of Bengal in an imperfect 
form before the Muhammadan conquest. 
Gwalior The palace of Raja Man Singh at Gwalior, built at the beginning of the 
tiles, sixteenth century, 'was once profusely decorated with glazed tiles of various 
colours', as noticed by Babar, who recorded in his Memoirs: 'The outside of 
the walls they have inlaid with green painted tiles. All around they have 
inlaid the walls with figures of plantain trees made of painted tiles/ Cunning- 
ham, writing in 1871, states that 

'the plantain [i.e. banana] trees mentioned by Babar still exist. They are of the natural 
size, but the leaves made of bright green glazed tiles are very regularly disposed on 

1 Cunningham, A. S. Rep., vol. v, pp. 13 1-3, into relief patterns, which are covered with a 
PL XXXIX. At Sitpur in the Muzaffargarh poor vitreous dip, forming a ground of opaque 
District, where similar tile decoration occurs, the dark blue, upon which patterns in opaque white 
colours include yellow. The Sitpur tombs date — either enamel or clay — have been laid. The 
from the fifteenth century. patterns include Muhammadan (Saracenic) and 

2 Birdwood, Industrial Arts of India, p. 322. Hindu forms, and may be referred to the 
These objects are rather enamelled bricks or eleventh or twelfth century (Furnival, headless 
terra-cotta than tiles. The body is similar to Decorative Tiles, p. 118, Figs. 72-5). 

that of red bricks, moulded on the edges or sides 

A# ; 

U *o 



9 9:A'J>S>J>f>J)J>AA 


PLATE T 43- Minaret of Wazir Khan's Mosque, Lahore 

PLATE 144. Chini-ka-Rauza, Ajjra 

Part VI TILES 201 

each side of the yellow stems, and the effect is consequently too stiff and formal. The 
diamond patterns in blue tile, and the long narrow lines of the same colour are, 
however, both effective and pleasing/ 1 

We now pass on to the more highly developed and artistic use of glazed Tile-work 
tiles after the Persian manner on the walls and domes of Mughal buildings. of Mughal 
Most of the Mughal tiling is of the kind called Kashi or Chad, composed of p 
pieces cut out from a painted sheet and laid as mosaic. The larger part dates 
from the seventeenth century, with a range of colours considerably more 
extensive than that employed on the early Panjab tiles already noticed. 2 Such 
Kashi tile casing, sparingly employed on the tombs of Sher Shah and Huma- 
yun, came largely into favour in the reigns of Jahangir and Shahjahan (1605- 
58), and continued to be used in Aurangzeb's time. The art is now extinct. 

The most remarkable series of tile pictures in the world is the huge band Tile pictures 
on the walls of the Lahore Fort, extending from the Elephant Gate (Hathi ™™ 1 * of 
Pol) to the north-eastern tower of Jahangir's quadrangle for a length of 497 Fort! * 
yards, with a height of 17 yards. Nearly the whole of this enormous surface 
is faced with painted tiles representing elephant fights, a game of polo, and 
other scenes. Dr. Vogel has obtained tracings of 116 panels, of which many 
select examples have been reproduced on a reduced scale in colour. 3 

The most beautiful example of Kashi tile- work on a large scale is universally Mosque of 
recognized to be the mosque built in 1634 at Lahore by the governor, Wazir ^J* 21 * 
Khan. The building is a well-designed domed structure with four handsome 
minarets , constructed of small thin bricks . The exterior is panelled , the panels 
and minarets being veneered with Kashi tile-work of great brilliancy, still 
in fairly good preservation. 

Passing by several interesting buildings exhibiting more or less decoration Chini-ka- 
in coloured tiles, we come next to the tomb near Agra known as the Chini-ka- Rauza - 
Rauza, which has had the advantage of being exhaustively described and 
illustrated by the late Mr. E. W. Smith in a volume mainly devoted to it. 
The building, a large octagonal domed tomb of uncertain date, supposed to 
have been built early in the reign of Aurangzeb, in memory of Afzal Khan, 
a poet who died in 1639, was originally covered on the outside from top to 
bottom with mosaic in Kashi tiling of various colours, worked up into 
numerous patterns so as to form one unbroken flat surface. It is now much 
dilapidated. The tiles, t of an inch thick, are bedded in a layer of fine plaster 
an inch thick, which was laid on a stratum of coarser plaster two inches in 
thickness. The principal colours include blues, greens, orange, vermilion, 
lake, &c, in a variety of delicate shades with a metallic lustre, the unavoidable 

1 Cunningham, A. S. Rep., vol. i, pp. 347-9- but Mr. F. H. Andrews, after making experi- 

2 Opinions differ as to the mode of manufacture, ments, believes that the shaped pieces were cut 
Mr. J. L. Kipling thought that the designs were after glazing and firing (Ibid., vol. x, pp. 27-30). 
painted on large sheets, which were cut up into 3 Progress Rep. A. S., Panjab Circle, 1901-2, 
tiles before firing (J.Ind.Art, vol. ii, pp. 17, 18); par. 13; J. I. A. /., 1911. 

3292 d d 


slight irregularities of the surface producing wonderful play of light. One 
illustration may be given to show the style (Plate 146). The tomb also 
exhibits some painted internal decoration in excellent taste. 

Eighteenth- Sir John Marshall describes as follows a third type of Indian tile decora- 
century tion . 

square tiles. 

'A third kind of tiles is found on buildings of the eighteenth century, such as the 
mosque of Muhammad Amin at Lahore (beginning eighteenth century) and the 
mosque of Zakariya Khan near Lahore. The founder of the latter was a viceroy of 
the Punjab from a.d. 1717 to 1738. It is strange to find the same type combined with 
Kashi work on the tomb of Asaf Khan at Shahdara as early as a.d. 1634. The tiles of 
this class are square. They form, consequently, not a tile-mosaic as the two earlier 
types, in which each separate piece has its own shape and colour, but are similar to 
the tiles known in Europe, from where presumably they were introduced into India. 
The colours are faint as compared to [sic] those of the Kashi tiles, pale green, blue and 
yellow being the most prominent. In one case, the tomb of Sharf-un-nissa, known as 
the cypress tomb (Sarwali maqbara), not far from Begampura near Lahore, we find, 
besides Kashi work on the lower part of the walls, square blue and white tiles of a type 
well known in the west of Europe. This building also would seem to belong to the 
eighteenth century/ 1 

In Plates 145 and 146 reproductions are given from photographs specially 
taken of six artistic square tiles in the Victoria and Albert Museum, Indian 
Section, all believed to date from either the sixteenth or seventeenth century. 
Plate 145, Fig. A, showing the complete figure of a young woman seated on 
her heels, with part of another woman offering her a fruit, appears to be 
unique, and is supposed to date from the seventeenth century. The drawing 
is good, and the general aspect suggests European influence, of which there 
was plenty in those times. The broken tile from Delhi (Fig. B) presents a 
humped bull and flowers in brilliant colours. The fragments of a hunting 
scene on two perfect tiles (Plate 145, Fig. C, and Plate 146, Fig. A) from 
Lahore are vividly designed and, I think, rightly referred to the seventeenth 
century, when such pictures of Persian origin were much in fashion. The 
floral devices on the Lahore tiles (Plate 146, Figs. B and C) are pretty and 
well coloured. 
Sind tiles. The modern tile-work of Sind and Multan is described in various books 
dealing with the industrial arts. The oldest Sind tiles on the Dabgir mosque 
and Mirza Jani Beg's mosque at Tatta, dating from about a.d. 1509, exhibit 
only two colours, a deep rich blue and a pale turquoise blue, on a white 
ground, and so resemble the early Multan tiles. 2 Multan used to be reckoned 
as in Sind, not in the Panjab, as it is now. 

1 For a long list of Indian buildings decorated 6 and Appendix C. 

with tiles, prepared by Mr. C. Stanley Clarke, 2 Cousens, H., Portfolio of Sind Tiles, Griggs, 

see Furnival, Leadless Decorative Tiles, pp. isi- 1906. 

A. (ilazt'il eirthenware tilt' from Punjab. 
17th century 

B. Enamelled earthenware tile from Delhi. 
1 6th century 

(\ Enamelled earthenware tile from Lahore. 
17th century 

PLATF 145. Mughal tiles in the Victoria and Albert Museum 

PLAT!* 146. Knamellcd earthenware tiles from Lahore, in the Victoria and Albert Museum, 

South Kensington. 17th century 

Chapter Fifteen 



THE study of Indian painting has of late been greatly advanced by the 
work of Goetz, Mehta, and Percy Brown. However, a great deal 
remains to be done, before an acceptable classification of the various 
schools is arrived at. Dr. Coomaraswamy was among the first investigators 
of Indian painting. It is therefore only right to outline his pioneer views at 
the head of this chapter. To him, above all else, Indian miniature-painting 
is divided into two, the foreign Muhammadan school which rose under Persian 
tuition during the reigns of the Mughal emperors, and in contradistinction, 
an ancient, indigenous, wholly Indian school, which he designates 'Rajput', 
and treats of as persisting 'in Rajputana and the Himalayas ... up to the end 
of the eighteenth century, comparatively little affected by the Persian and 
European influences which enter so largely into the art of the Mughal 
Court'. 1 c . . . Rajput painting', he writes, 'has none of the characteristics of 
a new art. It is, on the contrary, related to the classic art of Ajanta, as the 
Hindu language and literature are related to the older Prakrits and to Sanskrit. 
. . . The Rajput paintings, indeed, show a remarkable combination of folk 
idioms with ancient hieratic design.' 2 Mughal art, on the other hand, is a 
purely miniature art, unrelated to the ancient Indian frescoes. It is courtly 
not popular, secular not religious, material not spiritual. 

Several objections may be made to this radical division. For one thing, Modern 
it is perfectly evident that both schools share a common technique, seemingly trend °* 
derived from Persian painting. Furthermore, a closer study of 'Rajput Paint- 
ing', shows it also to be a 'courtly' art, associated with the capitals of various 
ruling dynasties. It is also evident that on the one hand, the Hindu Krishna 
and Ragmala subjects of the 'Rajput' schools are often embodied in purely 
'Mughal' renderings, although they are of course commoner in Hindu 
Jaipur and Garwhal than in Delhi; on the other hand, it is equally evident 
that magnificent examples of 'Rajput' portraiture exist, fulfilling the same 
demand as 'Mughal' portraiture. Lastly, Goetz' study of costume and, still 
more conclusively, various dated examples of 'Rajput' paintings, prove with- 
out a shadow of doubt that the bulk of 'Rajput' painting is posterior to, 
rather than contemporary with the great 'Mughal' work of the court artists 
of Akbar and Jahangir. 

This criticism of Coomaraswamy's primary classification is reinforced by Jain 

1 Museum of Fine Arts Bulletin, Boston, xvi. 49. z Burlington Mag., 1912, p. 315. 

2o 4 PAINTING Chap. XV 

a study of the few examples of pre- Mughal Indian miniature-paintings 
known to us. These are mostly illustrations to Jain palm-leaf manuscripts, 
and the school has therefore become generally known as 'Jain*. 1 This is 
unfortunate for the art was not confined to religious subjects. It appears 
likely that many purely secular examples exist, such as is the MS. of Vasanta 
Vilasa described and illustrated by Mehta. 2 This was written during the 
reign of Ahmad Shah Kutb-ud-din of Gujarat in a.d. 145 1; only two 
older examples of the school are known. The MS. in question is written 
on a long roll of prepared cotton, 35 feet 6 inches long and just over 9 inches 
wide. The colours are laid on flat and there is a preponderance of red 
and yellow, the body-ground being yellow. Features are usually rendered 
half-face, but occasionally side-face, the long almond-shaped eye of the 
Indian canon of beauty being greatly exaggerated. Trees are portrayed 
formally as lozenges containing branch and foliage; this treatment is usual in 
Indian art but is not found in the frescoes at Ajanta and Ellora, where foliage, 
blossom, and fruit are luxuriously reproduced. Here, except in the case of 
banana trees and mangoes, the treatment is strictly formal in a rather slovenly 
way; only here and there does any attempt at design lighten the arid con- 
vention. The figure-drawing is weak, but fortunately the costume with its 
detail of jewellery and floating scarf and waist-cloth is faithfully and delight- 
fully set down. On the whole one is impressed by the candour of this naive 
art, the purpose of which is frankly book-illustration, as indeed was the 
primary purpose of the masters whose work still glows on the dark walls of 
Ajanta and Ellora. 
Indian Indian costume as shown in these paintings is proven conservative. The 
costume. men wear th e W aist-cloth (dhoti), long or short, with a scarf for the shoulders. 
Jewelled head-dresses of various kinds are worn, but more commonly the 
hair was dressed with flowers. The pyjama and the women's veil do not 
appear. It is evident that the costume of fifteenth-century Gujarat must 
be treated of as being akin to that of Ajanta, not of Delhi and Agra. The 
subsequent change speaks clearly of a far deeper penetration of Mughal 
influence than has hitherto been allowed for. 
Examples As has been said the bulk of the known illustrated MSS. of this school are 
of Jain j a i n of these the earliest appears to be the Kalpa Sutra in the Patan library 
painting, ^ated in the year a.d. 1237. Two representative MSS. are the Kalpa Sutras 
respectively in the India Office Library (a.d. 1427), and the British Museum 
(a.d. 1464). The MSS. in the Boston Museum, illustrated in volume iv of the 
Catalogue of the Indian Collections, form perhaps the best group for com- 
parative study. There are also several excellent examples in the Ghose 

1 See Coomaraswamy's pioneer work on Jain also his 'Jaimi Art', Jour. Ind. Art., vol. xvi, 

Painting, Part IV of the catalogue of the In- p. 82. 

dian Collection, Museum of Fine Arts, Boston: 2 be. cit. f p. 15. 

PLATE 147. Kalpasutra (Jain) MS. Guj 

Or. 51J.0. British Museum 

,, M uu. V j—/ — -Aijarati School, a.d. 1464. 
Or. 5149. British Muse iim 

PLATE 148. Painting on cotton cloth from a copy of the Amir Hamzah. [a.d. 1550 75] 
Victoria and Albert Museum, Indian Section, South Kensington 

PLATK 149. Illustration from the Akbarnama, painted by Basawanand Chatar; 

subject-' Akbar in an elephant fight on a boat bridge across the Jhelum ; 

Victoria and Albert Museum, Indian Section, South Kensington 


ATF I qo. Illustration of the Durabnamah; by Whzad and Abdus 
J (B.M.Or.4615) 



collection in Calcutta, including one dated in a.d. 1480. The two illustrations 
are from a MS. in the British Museum. 

Since all these fifteenth-century paintings seem to belong to Gujarat, Origin of 
'Gujarati' would be a preferable title to 'Jain'. As a local school they are the School, 
closely comparable with the few examples of medieval Nepalese paintings in 


It is undeniable that Indian miniature-painting is largely derived from Islam 
Persia, although the essential 'Indianess' of the work is also undeniable. and Art - 
Persian painting is divided chronologically into three periods, the Mongol, 
the Timurid, and the Safavid. As a branch of Islamic art, it exists as the 
result of a direct breach of the Law, for it is written that whosoever makes a 
representation of a figure, human or animal, shall give it his soul at the Day 
of Judgement and so come to perdition. Until the collapse of the Caliphate 
of Baghdad it seems that the Law was upheld, for no illuminated Arab 
manuscripts are known to exist before the end of the thirteenth century. It 
seems that Islamic painting came into existence under the somewhat hetero- 
dox Aiyubite sultans, whose coins bear on the reverse the head of the Byzan- 
tine Christ. The most important early MS. is the Schafer Makamat of 
ifonW (Bibliotheque Nationale: Arabe, No. 5847), made in the year A.D. 1237. 
In these pictures one sees a vital pictorial sense struggling to embody itself 
in foreign and decadent forms. Byzantine influence is obvious, the nimbus, 
vestments, drapery, and the architectural setting being borrowed en bloc. One 
is also reminded of the art of the Persian potters and of the older tradition 
that lies at the root of Sassanian art. . . . Islamic art was created in a land that 
had witnessed the rise, modification, and decay of many schools of art. 
Assyrian and Greek influences are blended in the colonnades of the Apadana 
of Xerxes, and East and West, Bactria and India, Sassanian and Chinese met 
in the markets of the Taklamakan trade-routes, long before the coming of 
the Mongols released Persian art from its religious bonds. As has been said, 
foreign influence is paramount in these early pictures, but the drawing is full 
of interest, and when displaying familiar things, horses and horsemen, and 
their furniture and arms, has a native vigour of its own. 

Under the Abbasids Arab painting flourished in the great cities of the The Mongol 
Tigris valley, Baghdad, Wasit, and Basrah. Its development and very exis- 
tence was cut short with everything else that represented the Caliphate at its 
greatest, in the year a.d. 1258, the date of the Mongol invasion. The flood 
of destruction passed away and good arose out of evil. Under Mongol rule 
China was in direct contact with Persia. Byzantine influence and whatever 
remnants of decadent classicism that still lingered on, died away before a 
steady current of influence from the East. It has been pointed out that 


206 PAINTING Chap. XV 

Hulagu had Christian wives and that the Mongols favoured Christianity in 
the face of Islam. It has also been stressed that Central Asia and especially 
the Tarim Basin was a polyglot meeting-place of foreign cultures, western, 
Chinese, and Indian. However, from the point of view of Persian painting, 
Chinese art was the dominant art of the period. China was the source of 
Persian technique and inspiration, not only indirectly by the importation of 
Chinese wares, but, it is said, directly by the introduction of Chinese crafts- 
men, potters and embroiderers, as well as painters. So arose in the Mongol 
cities of Maraghan, Sultania, and Tabriz, a well established art destined to 
a long and illustrious history. 1 
The house The Mongol period (1258-1335) drew to its turbulent close, and out of 
of Timur. c h aos emerjged the house of Timur under whom civilization and art awoke 
to new life in the cities of the Oxus, Bokhara, and Samarqand. Of this house 
came Babar, the founder of the Mughal Empire of India. In Timurid Persia 
architecture flourished. Shah Rukh of Herat, son of Timur, himself a poet, 
maintained court-painters, one of whom journeyed to China with an embassy. 
At the end of the fifteenth century Sultan Husain Mirza gathered at his 
court the most famous artists of his time, among them Bihzad, the painter. 
TheSafavids. After the death of Husain Mirza in 1506 Bihzad was employed at the 
court of the Safavid Shah Ismail. Under the new dynasty Persian painting 
entered upon its period of romanticism. Timurid clarity and restraint were 
cast aside and design and colour are lavishly conceived. At its best under 
Shah Abbas at the end of the sixteenth century, this period inevitably led to 

Mughal The history of Mughal painting begins with the name of Mir Sayyid Ali. 
painting. j n the year IS ^ fiabar get out upon the conquest of i n ^[ Zy a \ m ^ however, 

of which he did not conceive highly. Five years later he was dead. In 1546 

Humayun, his son, was deprived of his empire by the Afghan, Sher Shah, 

and until his final victory in 1555 existed as a landless refugee. One year of 

this period was spent at the Safavid court at Tabriz, where Shah Tahmasp 

now ruled. Bihzad was dead, but the work of a young painter, Mir Sayyid 

Ali, was already attracting attention. His father, Mir Mansur of Badakshan, 

who was also a painter, was a contemporary of Bihzad's. Another painter 

of growing reputation also attracted the notice of the exiled emperor; this 

was Abdus Samad. 

Amir In 1550 both these artists joined Humayun's court at Kabul. It was here 

paSSs! that Mir Savvid Ali was commissioned to supervise the illustration of the 

' romance of Amir Hamzah (Dastan-i-Amir Hamzah) in twelve volumes of 

a hundred folios each. Sixty of these illustrations painted in tempera colours 

1 For minor influences in Persian painting, a few motives does not postulate the survival of 
especially Manichean, see Arnold, Some Sur- stylistic influence. 
vivak in Persian Painting. But the survival of 


on prepared cotton cloth are in Vienna, and twenty-five of them in the 
Indian Museum, South Kensington. They must probably be attributed to 
the artists of the imperial court working under Mir Sayyid Ali, rather than 
to that painter himself. After Humayun's death Mir Sayyid Ali continued 
to work at the court of Akbar, and also performed the pilgrimage to Mecca. 
The style of these early Mughal paintings is, of course, largely Safavid, but 
it is evident that modification and developments have already taken place. 
It is said that Bihzad added skill in portraiture to the art of painting; por- 
traiture is further developed in Mughal painting. Also a greater use is made 
of relief and the range of colours is larger and more striking. There is some- 
thing, too, about the use of flower and foliage that is un-Persian and wholly 
Indian. A certain simplicity and breadth of design dominates the wealth of 
detail; the microscopic rendering of costume and accoutrements, textile 
hangings and architectural details is doubly delightful in so much as it is 
never obtrusive. 

Such paintings on prepared fabric are common in India. It appears that 
paper itself was rare, or at any rate that large sheets were hard to obtain. 

Summing up the technique and quality of early Mughal painting, it may Early 
be said that it was an offshoot of the Safavid school, the handiwork of artists Mughal 
trained in the school of Bihzad. 1 However, as has been said, the local P amtm &- 
character of the detail as shown in the portrayal of the Indian countryside 
and of its flowers and foliage is proof of complete acclimatization, promising 
vigorous development. 

Akbar succeeded to the insecure throne of his father when still a boy with Painting 
this distinction: that whereas Babar and Humayun were rulers in a foreign under Akbar. 
land, he was native born. The culture of his court did not merely reflect at 
a distance the splendour of Bukhara and Samarqand. The building of Fateh- 
pur Sikri in 1569 heralded a new era of Indian rule. And after the architects, 
masons, and sculptors had done their work, painters were called in to 
decorate the walls of the public halls and private apartments. The art of 
these paintings, as far as may be judged from what remains of them, was 
closely allied to that of the Mughal miniatures, the colours being applied 
upon a ground of white pigment laid directly upon the sandstone. Some 
of the paintings are purely Persian in style: others are Indian. It is evi- 
dent that many artists were employed, each working in his own style. As 
the result of this co-operation under royal patronage, a school of court- 
painters was set up under the Emperor's direct control, the Persian artists, 
Mir Sayyid Ali and Abdus Samad being, of course, prominent. The latter 
rose to be master of the mint at Fatehpur, and afterwards Divan or Controller 
of the Revenues of Multan.* 

1 Percy Brown, Indian Fainting, p. 56. 

2 His son attained the dignity of Amk-ul-Umara under Jahangir. 

208 PAINTING Chap . XV 

Wall- The wall-paintings of Fatehpur are fully described in E. W. Smith's book, 
paintings, pfo Mxighal Architecture of Fatehpur-Sikri, from which Plate 151 is taken. 
The treatment is flat, no shading being made use of. One scene in which 
winged figures play a part is called 'The Annunciation* by the local guides. 
Chinese motives such as dragon-clouds appear as an inheritance from the 
parent Persian school. That the arts of painting and sculpture were closely 
united is proved by many of the bas-relief panels where flowering trees and 
animal forms are represented in a very naturalistic manner. 1 It has been said 
that Mughal miniature-painting are wall-paintings in little, a statement 
which tends to be confusing, since neither branch of Mughal painting has 
anything in common with the ancient Indian schools of painting of Ajanta 
and Gujarat, except certain inclinations to bright colouring and fine line- 
drawing which seem temperamentally inherent in Indian artists. 
Calligraphy. In Persia and India, as in China, calligraphy was regarded as a fine art 
worthy of the most serious study, and masters of it enjoyed fame throughout 
Asia like that of great painters in Europe. They were careful to sign and date 
their works, which were eagerly collected by connoisseurs. Abul Fazl gives 
a list of calligraphic experts, among whom in Akbar's time the most eminent 
was Muhammad Husain of Kashmir, who survived the emperor for six years. 
Many of the albums in the London collections containing 'miniatures' include 
hundreds of specimens of beautiful writing in various styles and of different 
periods, which often seem to have been more valued than the drawings and 
paintings associated with them. Abul Fazl enumerates eight calligraphical 
systems as current during the sixteenth century in Iran (Persia), Turan 
(Turkistan), India, and Turkey, distinguished one from the other by differ- 
ences in the relative proportion of straight and curved lines, ranging from the 
Kufic with five-sixths of straight lines to the Nastalik, Akbar's favourite 
script, with nothing but curved strokes. The forms of the Arabic alphabet 
used for writing Persian, although not distinctly reminiscent of pictorial 
hieroglyphs, as the Chinese characters are, lend themselves readily to artistic 
treatment, and even Europeans may understand to some extent the high 
technical skill of the masters of the calligraphic art, and admire the beauty 
of their productions. But full enjoyment and appreciation are possible only 
to persons familiar with the character from infancy and sensitive to all the 
associated ideas. 3 

Close con- 'Among the general characteristics of Chinese painting the most striking, and the 
nexion of one which has prevailed most strongly throughout its long historical evolution, is the 


- l See casts of the panels from the Turkish 

Sultana's house in the Indian Museum, South 


2 The technicalities of the art are explained by 

Huart in Les Calligraphes et les Miniaturistes de 

VOrient Musulman, Paris, 1908. He gives (p. 


256) a list of Indian calligraphists in the eigh- 
teenth century, and also mentions Jawahir 
Raqam, Aurangzeb's librarian, who died in 1683. 
The Department of Design, &c, at South 
Kensington, possesses specimens of the work 
of Roshan Raqam, one of the artists named. 

, lM \Van-pa' ,nlinp 

;T . c ^htmcn 

in a boat; in. 

s bedroom, 



PLATE 152. Picture of a Plane Tree. [dr. a.d. 1610]. 
Johnson Collection, vol. i, fol. 30 


graphic quality of the painting; Chinese painters are, first of all, draughtsmen and 
calhgraphists . . . The different legends all carry out the leading idea of the common 
origin and essential unity of writing and painting, and this unity is constantly insisted 
upon by Chinese critics of the two arts.' 1 

The same idea dominated the Persian artists and their Indian imitators at 
Akbar's court. Abul Fazl, accordingly, devotes Ain 34 of his Institutes of 
Akbar to the discussion of the * Arts of Writing and Painting', passing naturally 
from the account of calligraphic systems summarized above to the invaluable 
notice of the early history of Indo-Persian painting, which forms our only 
source of knowledge of the subject other than the information to be gleaned 
laboriously by minute study in detail of individual works. M. Huart sums up 
the close relations between calligraphy and Asiatic painting in the phrase: 
'En Orient la miniature n'est que la servante de la calligraphic ' The phrase, 
however, is not applicable to the ancient Hindu schools of painting, which, 
except in so far as they may have been influenced by Chinese and Persian 
ideas, were independent of the scribe's art. None of the many varieties of 
the square Brahmi or Sanskrit script ever tempted the calligraphist to regard 
his manuscript as a picture, nor did anybody dream of collecting specimens 
of writing in that script merely for the sake of their beauty. 

The rapidity with which the teaching of Abdus Samad and his Musalman Daswanth 
colleagues was assimilated and then modified by scores of Hindu artists of Kahar - 
various castes is in itself sufficient proof that the foreign teachers must have 
found trained indigenous scholars with whom to work. Men accustomed to 
draw and paint could easily learn new methods and a foreign style, but not 
even the despotic power of Akbar would have been able to create a numerous 
school of Hindu artists out of nothing. 

This inference, inevitable from a general survey of the facts, is established 
with certainty by the positive testimony of Abul Fazl that Daswanth, who 
disputed with Basawan the first place among the Hindu painters of Akbar's 
court, had 'devoted his whole life to the art, and used, from love to his pro- 
fession, to draw and paint figures even on walls'. He was the son of a poor 
man, a member of the Kahar or palanquin-bearer caste; and when such a 
man, in spite of all social disadvantages, could become a professional artist, 
many others more favourably situated must have done the same. Daswanth's 
genius was rescued from obscurity by the royal favour. 'One day,' writes 
the courtly historian, 'the eye of His Majesty fell on him; his talent was dis- 
covered, and he himself handed over to the Khajah (scil. Abdus Samad). In 
a short time he surpassed all painters and became the first master of the age. 
Unfortunately the light of his talents was dimmed by the shadow of madness; 
he committed suicide. He has left many masterpieces/ Abul Fazl goes on to 

1 Bushcll, Chinese Art, ii. 207. 

3292 £ e 

210 PAINTING Chap. XV 

say that the work of Basawan is so excellent that many connoisseurs preferred 
him to Daswanth. 
Disregard of As has been said the Koran, following the Semitic principle formulated 
P of ^ma^es 1 * n t ^ ie Mosaic Second Commandment, absolutely forbids Muslims to make 
images. ^ e likeness of anything in heaven or on earth; and the prohibition has been 
and is strictly obeyed, with rare exceptions, in all countries and at all times, 
so far as the decoration of mosques and other buildings devoted to religious 
purposes is concerned, 1 In book illustrations, however, such liberty is com- 
monly assumed. The Persians, adherents of the Shia sect of Islam, always 
have been especially lax in their open disregard of the Koranic prohibition. 
The Mughal emperors of India looked to Iran for the graces of civilization, 
and it was natural that Akbar should desire to add the charms of Persian 
pictorial art to the amenities of his court. Regarding himself as Head of the 
Church and pontiff of a new religion, he cared little about the Prophet, and 
at a private party was heard by his Boswell to observe: 

'There are many that hate painting, but such men I dislike. It appears to me as 
if a painter had quite peculiar means of recognizing God ; for a painter in sketching 
anything that has life, and in devising its limbs one after the other, must come to feel 
that he cannot bestow individuality upon his work, and is thus forced to think of God, 
the Giver of life, and will thus increase in knowledge.* 

Akbar's He found no difficulty in gratifying his taste. Liberal pay and abundant 
atrona^ 1 ^ onour drew crowds of artists, both foreigners and Indians, Muslims, and 
Hindus, to his magnificent court, where the more distinguished were enrolled 
as mansabdars, or members of the official nobility, and assigned ample salaries. 
His system of government making no distinction between civil and military 
employ, or rather giving military titles to all official rank, the successful 
artists ranked as army officers of good standing, while their assistants and 
allies, gilders, binders, and the like, were enrolled either as members of the 
imperial bodyguard {ahadi), or as private soldiers, with pay ranging from 
fifteen to thirty rupees a month, sufficient for comfortable subsistence. The 
industry of all grades was stimulated by weekly inspections, at which His 
Majesty generously rewarded merit. 
Imperial Imperial libraries of large extent were formed at Agra, Delhi, and other 
libraries. pi aces ^ stored with all that was best in Asiatic literature, both originals and 
Persian translations, the volumes being enshrined in the richest bindings, 
and adorned with miniatures regardless of expense. 

For example, the Razmnamah, or Persian abridged translation of the Maha- 
bharata, with preface dated A.D. 1588, now at Jaipur, is said to have cost 

1 Two exceptional cases are cited by Migeon. walls two paintings, one of Ali, son-in-law of 

The Khalif Abd-ul-Malik (a.d. 685-705) erected the Prophet, and another, perhaps representing 

a mosque at Jerusalem decorated with images of Fatima veiled (Manuel d'art musulman, tome ii, 

the Prophet and paintings of heaven and hell. pp. 1, 56). 
The Jumai Mosque at Isfahan exhibits on the 


£40,000 sterling; and Colonel Hanna estimates that his copy of the Ramayana, 
now at Washington, must have cost quite half that sum. 1 The Akbarnamah, 
from which 117 large paintings are preserved at South Kensington, was a 
similar work, and Abul Fazl mentions many others. 2 According to the 
Spanish priest, Father Sebastian Manrique, who was at Agra in 1641, the 
imperial library at that city contained 24,000 volumes, valued by him at the 
astounding figure of 6,463,731 rupees, or £720,000 sterling, an average per 
volume of almost 270 rupees, equivalent then to about £30. 3 

The libraries thus formed were maintained and increased by Jahangir, Destruction 
Shahjahan, and Aurangzeb (1 605-1 707); and even the weak successors of the °/ th ^ 
last Great Mogul were not indifferent to the delights of choice books and llbranes - 
dainty pictures. 4 But the political convulsions of the eighteenth and nine- 
teenth centuries destroyed the imperial libraries, with most of the similar 
collections formed by subordinate potentates like the Rohilla chief and the 
Nawab-Vazir of Oudh. 5 Fragments of these wonderful accumulations are 
now scattered over the world in private and public collections, and although 
constituting but a small fraction of the great mass once in existence, supply 
ample material for the history of Indo-Persian calligraphy and the sister art 
of the miniaturist. Many of these paintings have had adventurous histories. 

When Shahjahan began to grow old, his four sons, each eager to secure for 
himself the succession to the throne, engaged in bitter, internecine strife. 
Aurangzeb, the third son, a master of craft and guile, won the prize, im- 
prisoned his father, and assumed power in 1658. Dara Shikoh, his eldest 
brother, doubly hateful as a rival and a heretic, was pursued to the death with 
unrelenting rigour. Driven into the deserts of Sind, he was foully betrayed, 
and, to augment his affliction, before reaching the house of his betrayer, 

'received by a foot messenger the sad intelligence of the death of that one of his wives 

1 Vol. iv of Hendley, Memorials of the Jeypore present. Mir Muhammad, the artist from whom 
Exhibition, 1883, 4to, is solely devoted to repro- Manucci obtained the portraits of the imperial 
ductions from the Razmnamah, of which two family which he brought to Venice before 1712, 
are in colour. was in the service of Shah Alam (Irvine, Storia 

2 The Ain-i-Akbari, usually regarded as a separate do Mogor, vol. i, pp. liii-lvi). 

work, was really part of the Akbarnamah, or 5 B.M. MS. Add. 22470 belonged to Hafiz 

'History of Akbar\ Rahmat of Rohilkhand, and came into the pos- 

3 Manrique, Itinerario de las missiones que hizo el session of an English officer after Hastings's 
padre F. Sebastian Manrique, Roma, 1649, p. 417. Rohilla war, in the course of which the Bareiliy 
See ante, chap, xii, p. 417. Some of Manrique's library was plundered. Asaf-ud-daulah, Nawab- 
observations are summarized in English in Mur- Vazir of Oudh, secured most of the books for 
ray, Historical Account of Discoveries and Travels Lucknow, where they were again plundered and 
in Asia, 1820, vol. ii, pp. 96-1 19* and again, more scattered in 1858. B.M. MS. Add. 18579 was 
briefly, in Oaten, European Travellers in India, illustrated for the last king of Bijapur in the 
1909, pp. 97-102. Deccan, whose capital was sacked by Aurangzeb 

4 For instance, the splendid B.M. MS. Add. in 1686. Most princes probably owned libraries 
20734 (Pers. Catal., p. 259) was given to an of considerable value. See paragraphs on Rajput 
English officer by Akbar II in 1815 as an official painting. 

212 PAINTING Chap. XV 

whom he loved most, and who had accompanied him always during his misfortunes. 
He learnt that she had died of heat and thirst, not being able to find a drop of water 
in the country to assuage her thirst. The Prince was so affected by the news that 
he fell as though he were dead/ ! 

The memory of this sad tale is recalled by a beautiful little album now 
preserved in the India Office Library, which bears the unhappy prince's 
autograph inscription written across a splash of gold smeared over the deli- 
cately decorated fly-leaf: 'This album was presented to his nearest and 
dearest friend, the Lady Nadirah Begam, by Prince Muhammad Dara 
Shikoh, son of the Emperor Shahjahan, in the year 1051 ( =a.d. 1641-2).' 
Albums. The illustration of manuscripts was only one form of Indo-Persian art, 
and that, as M. Blochet truly observes, was not always the most successful. 
The highest achievements of the Indian draughtsmen and colourists were 
often attained in separate pictures of varying sizes, which were frequently 
bound in albums, like that given by Dara Shikoh to his beloved wife. The 
British Museum collection includes many such albums, some of which, such 
as Hafiz Rahmat's volume, constitute historical portrait galleries of the deep- 
est interest. The fashion set by the court of Delhi and followed by all the 
feudatory courts and many individual nobles, was passed on to the wealthy 
English 'Nabobs' in the latter part of the eighteenth century, who gladly 
seized opportunities of procuring specimens and bringing them home. Cer- 
tain pictures in B.M. MS. Add. 18801 were much admired by Sir Joshua 
Reynolds in July 1777. 
Prices. Occasional memoranda of prices give some notion of the pecuniary value 
of such pictures. One of those specially noticed by Sir Joshua — a large sketch 
of Shahjahan holding court — is marked Rupees 200, equivalent in those days 
to at least £25 sterling. In the Johnson Collection at the India Office formed 
by Warren Hastings's banker, Richard Johnson, a drawing of Nawab 
Shayista Khan, a great noble of Aurangzeb's time (vol. xxii, fol. 5), is priced 
Rupees 170, and in another volume a number of more ordinary small portraits 
are priced at 25 rupees each. During the nineteenth century the taste for the 
work of the school was lost by both Europeans and Indians, and very few 
persons seemed to care what happened to the pictures, which were then pro- 
curable for nominal sums. Interest in them has now been revived, chiefly 
by reason of Mr. Havell's efforts and the publications of French scholars. 
According to Badaoni, Akbar's hostile critic, the courtiers' taste for illu- 
minated books had been stimulated in his time by a certain amount of com- 
pulsion, and it was natural that, during the 'great anarchy' of the Maratha 
period, when the influence of the Delhi court sank to nothing, the amount 
of liberal patronage by the minor native courts should diminish. Neverthe- 
less, even during those stormy times much meritorious portrait work was 

1 Tavernier, Travels, transl. V. Ball, i. 350. 

PLATE 153. Faqirs resting under trees, [cir. a.i>. 1650]. 
Collection of Huron Maurice Rothschild, Paris 

PLATE 154. Sketch portrait of the Emperor Jahangir. \cir. a.d. 1625I. 
Collection of M. Carrier, Paris 

PLATE 15 v Portraits of Shcr Muhammad Nawal, Jahangir, 
and Shahjahan, by Muhammad Nadir of Samarkand. 
(B.M. Add. 1 8801, No. 40) 


produced, and some good portraiture was executed as late as the nineteenth 

When Bernier was writing to Colbert in 1669, early in the reign of Aurang- Position 
zeb, who had the Puritan dislike for art, the position of artists had become ? f a **ists 
much less favourable than that enjoyed by them in the days of Akbar, in 1 9 " 
Jahangir, and Shahjahan. The observant French physician, a thoroughly 
trustworthy witness, described as follows the relations between artists and 
their patrons, or rather taskmasters, as seen by him: 

'Can it excite wonder that under these circumstances [of general misery] the arts do 
not flourish here as they would do under better government, or as they nourish in our 
happier France ? No artist can be expected to give his mind to his calling in the midst 
of a people who are either wretchedly poor, or who, if rich, assume an appearance of 
poverty, and who regard not the beauty and excellence, but the cheapness of an article ; 
a people whose grandees pay for a work of art considerably under its value, and accord- 
ing to their own caprice, and who do not hesitate to punish an importunate artist or 
tradesman with the korrah, that long and terrible whip hanging at every Omrah's 
[nobleman's] gate. Is it not enough to damp the ardour of any artist when he feels that 
he can never hope to attain to any distinction? . . . The arts in the Indies would long 
ago have lost their beauty and delicacy if the monarch and principal Omrahs did not 
keep in their pay a number of artists who work in their houses, teach the children, and 
are stimulated to exertion by the hope of reward and the fear of the korrah. The pro- 
tection afforded by powerful patrons to rich merchants and tradesmen who pay the 
workmen rather higher wages tends also to preserve the arts. I say "rather higher 
wages", for it should not be inferred from the goodness of the manufactures that the 
workman is held in esteem, or arrives at a state of independence. Nothing but sheer 
necessity of blows from a cudgel keeps him employed/ 

In a subsequent passage the author describes the workshops attached to great 

'In one hall embroiderers are busily employed, superintended by a master. In an- 
other you see the goldsmiths, in a third, painters, &C.' 1 

Bernier 's description of the servile position of artists, while applicable 
specially to the experts in the industrial arts, must have been generally true 
also for that of the professors of the fine art of painting, A tyrannical 'Omrah' 
and his henchmen would not have drawn nice distinctions between the artist 
who painted the miniatures and the embroiderers or carvers who executed 
the binding of a sumptuous manuscript. Indeed the binding is sometimes 
as much a work of art as the pictures are. 2 

Excepting the modern Delhi miniatures on ivory, the frescoes, the early Technique, 
paintings on cotton, and a few pictures on vellum, the Indo-Persian paintings 

1 Bernier, Travels, transl. Constable, pp. 228, examples of rich early bindings in the Museum 
258. collection. See Indian Art at Delhi, p. 203, 

2 B.M. MS. Add. 18579, a copy of the Anwar- several papers in J. L A. /., and Migeon, Manuel 
i-Suhailim a beautiful minute script, has a hand- d'art musulman, tome ii, p. 59. 

some stamped gilt binding, and there are other 

2i 4 PAINTING Chap. XV 

are all executed on paper. 1 I do not know any Indian examples of painting 
on silk in the Chinese manner. The Indo-Persian, like other Asiatic artists, 
conceived every object as being bounded by firm lines, and consequently, 
his first step was the drawing of an outline. For the illustration of ordinary 
Persian books, according to M. Blochet, the outline drawn directly on the 
page in red or black chalk was filled in with colours at once. For more costly 
and elaborate volumes the process was more complicated, the illustrations 
being executed upon a separate sheet subsequently applied to the blank 
space left in the manuscript. That sheet was first covered with a layer of 
very fine plaster, mixed in a solution of gum arabic. The outline was then 
drawn upon the perfectly smooth surface thus obtained, and opaque body- 
colours, mixed with water, were laid on in successive layers, just as in oil- 
painting, but with the difference that mistakes could not be rectified. Jewels 
and ornaments were indicated by needle prickings in sheets of gold-leaf, or 
even by the insertion of pearls or diamond chips. 2 The work was all done 
by the Indian artists with fine squirrel-hair brushes, the most delicate strokes 
being executed with a brush of a single hair, an instrument requiring the 
utmost correctness of eye and steadiness of hand. The collections in London 
contain many examples of unfinished drawings and paintings, which, if 
examined critically by experts, would reveal fully the Indian methods of 
work, and show how far they agreed with or differed from the Persian methods 
described by M. Blochet. It must be pointed out that portraits often exist 
in duplicate and triplicate. 

Pigments. The blue was ordinarily obtained from powdered lapis lazuli, imported 
from Badakshan, but indigo blues appear in early book illustrations of Hindu 
subjects. The reds used were cinnabar, vermilion, or cochineal. 3 The yellow 
was chrome, and other colours were made up by mixing these. Gold was 
freely used in the form of gold-leaf, and also as a wash of which the Indians 
had the secret. 4 The Persians applied an admirably transparent varnish made 
of sandarac and linseed-oil, mixed as a paste and dissolved in either petroleum 
or highly rectified spirits of wine. 5 Probably the Indians used all the Persian 
appliances with some additions and modifications, but the ascertainment of 
full details would require special expert study and hardly repay the trouble. 6 

Collabora- The practice of beginning a picture by laying down a firmly drawn outline 

x Col. Hanna's Collection, now at Washington, 4 Before the discovery of cochineal in 15 18, 

U. S. A., included three examples on vellum, kermes, a pigment obtained from Coccus Indicus, 

namely, No. 28, Jahangir standing on globe ; an insect found in Persia, must have been used 

No. 52, a Sultan of Turkey; and No. 86, Babar. (Burlington Magazine, vol. iv, p. 144). Other 

2 Blochet, 'Musalman MSS. and Miniatures as authorities call the species Coccus ilicis. 
illustrated in the Recent Exhibition at Paris/ 5 Recipe in Ozias Humphrey MSS. in B.M., 
Burlington Magazine, vol. ii, June to Aug. 1903. No. 15962, first leaf. See also Moor, Hindu 

3 C.Stanley Clarke's introduction to the Wantage Pantheon (ed. 1810), p. 63 n. 

Paintings (Victoria and Albert Museum). 6 Blochet, Burlington Magazine, vol. ii, ut supra. 

PLATE 156. Jahangir as Prince Salim, anonymous. (Fol. 18 of Dara Shikoh's album) 


led to a curious division of labour, the outline often being drawn by one 
man and the painting done by another. For example, in the Clarke MS. of 
the Akbarnamah at South Kensington the picture (No. ^) of the execution 
of Adham Khan was drawn by Miskin and painted by Shankar. Sometimes 
three artists collaborated in one work, and I have noticed one instance in 
which the collaborators numbered four, namely, the audience scene (n?) in 
the Clarke MS. The outlines in that picture were drawn by Miskin, the 
painting was done by Sarwan, the faces (chihra-nami) by an artist whose 
name is indistinct, and the figures (surat) by Madho. It is not clear how such 
a complicated arrangement was worked. The method, whether only two 
artists or four collaborated, necessarily tended to reduce their art to the level 
of a skilled mechanical craft; and, as a matter of fact, the mechanical nature 
of much of the fine Indo-Persian work is its greatest defect. 1 

The early Indo-Persian book illustrations, such as those in the Clarke MS. The . carf y 
of the Akbarnamah, are wrought in excessively brilliant colours, chiefly red, ^dian book 
yellow, and blue. As has been said they are avowed imitations or, rather, 
developments of Persian work. 

In Persia, at the close of the fifteenth century, the character of Timurid art Change of 
began to change, passing into the more delicate and sentimental style of the character in 
Safavid period in the sixteenth century. During the seventeenth century the ersian art ' 
refined Safavid style, with its lowered scale of colour, became familiar in 
India, where further local modifications were effected under the influence of 
Hindu tradition. The Indian artists 'had a truer feeling for colour and more 
sober tonality' than their Persian teachers, according to M. Blochet, who is 
disposed to think that the Indians sometimes carried the policy of softening 
colour to an undue extreme. They were wonderfully successful in their 
grisaille drawings of a single colour, frequently a pale sepia, with delicate 
gradations of tint, very pleasing to my eye. At the same time they developed 
a mastery over individual characteristic portraiture never equalled, I think, 
by the Persians. The best Indian work dates from the first half of the seven- 
teenth century, but good portraits are to be found executed as late as the 
early years of the nineteenth century. 

During Akbar's reign (1556-1605) and a portion of Jahangir's (1605-27) Indian 
the standing portrait figures are usually represented in profile in a formal, ™°* [ " 

1 The word -Jo y tarh> or tarrah, primarily means tures unintelligible. Tainting' or 'colouring', 

'foundation' ; e.g. tarh afgandan, 'to lay a founda- as distinguished from 'outline', is expressed by 

tion,' tarh-kash, a 'plan-drawer' (Steingass, Pers. either the Arabic word J^, ofmal, 'execution', 

Diet.). The transition to the meaning 'outline' or the Persian term rang-amezi, 'colouring', 

was easy, and the word always has that meaning When dmal stands alone, it implies execution 

in the signatures to the Indo-Persian drawings, of the picture by a single artist. The term rang- 

as M. Blochet rightly perceived. Blochmann's amezi, to signify 'colouring', is preferred in the 

erroneous rendering 'back-grounding' in his Jaipur Razmnamah. 
translation of the Ain-i-Akbari made the signa- 

216 PAINTING Chap. XV 

conventional manner, with the right hand holding up a flower or jewel, and 
the feet placed one in front of the other. Gradually this stiff formalism was 
dropped, and men and women were drawn in natural attitudes. The more 
ancient Indo-Persian works, like their Persian models, follow unreservedly 
a style marked by the total lack of roundness, depth of tone, and aerial per- 
spective, every object being represented as absolutely flat. During the later 
years of Jahangir's reign and subsequently, this flat style was modified by the 
Indian artists, who frequently introduced slight line shading with admirable 
effect, so contriving to give their figures a sufficient degree of roundness with 
wonderfully few strokes. The change adds much to the attractiveness of 
seventeenth-century Indian work in European eyes, and was due to foreign 
influence. But chiaroscuro was imported to the detriment of colouring and 
line drawing. Delicacy and subtlety are bought at the cost of strength and 
vitality. Highly developed skill in portraiture seems to have swamped the 
sense of design and decoration. Foreign influence is also particularly notice- 
able in the treatment of clouds and foliage: such influence is often of a late 
eighteenth-century kind. 
Christian This improvement, if it may be so called, was the result of European 
subjects. i n fl uence> w hich certainly became a potent factor in Persian and Indian art 
at that time. Most of the albums show it plainly. For instance, Dara Shukoh's 
album includes two wood engravings (fol. 42ft, 43), one of S. Caterina di Siena, 
dated 1585, and the other of S. Margarita of about the same period, while 
the picture on folio 74 exhibits a lady and gentleman in European costume. 
Biblical subjects were frequently treated by the artists, and were specially 
favoured by the royal family, who used them for palace decorations at both 
Fatehpur-Sikri and Lahore. The treatment at times seems very quaint, as 
when we see the Good Shepherd depicted in the form of a stout middle-aged 
man with a black beard, wearing a Muslim's robe and a twisted turban of 
gold brocade. A Good Shepherd in vol. xvi, fol. 1 of the Johnson Collection 
is signed by Ustad Miskin, probably to be identified with Muhammad Miskin, 
the author of a lady's portrait in vol. xxi, fol. 1 of the same collection, and with 
Miskin, Akbar's artist, who signed some of the pictures in the Clarke MS. of 
the Akbarnamah. Many other biblical subjects will be found in the collec- 
tions, and it must be confessed that the pictures are not usually equal to 
those devoted to topics more congenial to the artists. 
The so- One subject, frequently treated with variations, has been mistakenly identi- 
called g ec j as Christian, and dubbed Angels ministering to Christ, although all the 
ministering compositions dealing with it are purely Muslim. The main motive is the 
to Christ, miraculous supply of food to a hermit saint dwelling in the wilderness by 
angels, who vary in number in different replicas, and are generally, if not 
always, provided with wings in the conventional fashion borrowed by Chris- 
tian art from the Greek figure of Victory. Most of the pictures show a second 

PLATE 157. Wild duck, anonymous. (Fol. 10, Dara Shikoh's album) 


figure, a discontented darvish sitting sulking in a corner or at the mouth of 
a cave. As is proved in several instances by the labels, the principal figure 
undoubtedly is that of Ibrahim, son of Adham, who resigned the kingdom 
of Balkh, and withdrew as a hermit into the wilderness already haunted by 
a darvish, whose food had been provided regularly by the angels. When the 
ex-king appeared on the scene, the angels, while continuing to supply their 
old client the darvish with a single daily dish as a bare subsistence, liberally 
brought ten dishes to the retired monarch, in recognition of the sacrifice made 
by him. The darvish naturally was annoyed, and whenever he is introduced 
into the picture his feelings are indicated by the artist. 1 

Many of the attempts to combine the methods of the West with those of 
the East are decided failures, as similar attempts in China have failed, but 
some few attain a high level of executive excellence. 

The origin of such influence is not far to seek. The Persian kings admired Persian art 
European art, and deliberately sought to introduce its methods into their ^^ ents m 
country. During the residence of Sir Robert Sherley at the Persian court, 
sometime about a.d. 1606, Shah Abbas I (1587-1629) sent to Rome a party 
of students, one of whom became a Christian and published a book under 
the name of Don John of Persia. Shah Abbas II (1642-67) repeated the ex- 
periment and dispatched a second party. One of these, by name Muhammad 
Zaman, also was converted, and returned to Persia as a Christian under the 
name of Paolo Zaman. Having been obliged to quit his native land, he ob- 
tained in India the protection of Shahjahan, who granted him, with other 
exiled Persians, allowances as a mansabdar in Kashmir. Early in Aurangzeb's 
reign all the Persian refugee mansabdars were summoned to court for the 
verification of their grants, and on that occasion, about a.d. 1660, Manucci 
made the acquaintance of Muhammad or Paolo Zaman, who avowed his 
Christian profession, while continuing to live in the ordinary Musalman 
manner. The three Europeanized pictures in B.M. Or. 2265 evidently are 
from his brush. 

To this day the painters and illuminators of Isfahan, the earlier, and 
Teheran, the later, capital of Persia, cherish as their ideal the ambition to 'paint 
like Raphael', and pride themselves on their descent from certain of the students 
sent long ago to Rome who survived to return to the home of their fathers. 2 

The attempt to weld Asiatic ideals and methods with those of Europe, Modern 
although responsible for some pretty pictures, was not a permanent success ^Eastern 011 

1 J. R. A. S.> 1909, p. 751 ; 1910, p. 167. artists when he was visiting Persia in 1874-6. It m & Western 

2 The strange story of the Persian missions to is said that Muhammad Zaman, who was con- art * 
Rome is pieced together from Irvine, Storia do verted, had been sent abroad to learn how to 
Mogor, ii. 17; and Sir C. Purdon Clarke's article confute the Christian missionaries. The ad- 

'A Tradition of Raphael in Persia', J. Ind. Art, venturous lives of the Sherley (Shirley) brothers 
vol. vii (1897), pp. 25, 26, Pis. XLII-XLVI. The may be read in the Diet, of Nat. Biography. 
author of that article heard the tradition of the 
3292 f f 

218 PAINTING Chap. XV 

in either Persia or India. It is now being renewed by the clever Bengali 
artists of Abanindro Nath Tagore's school in a different form, and with con- 
siderable ability, but I fear, without much prospect of producing any really 
important results. 
Enormous The Indo-Persian or Mughal school of drawing and painting having lived 
output. j n considerable vigour from about a.d. 1570 to 1820 or 1830 — a period, 
roughly speaking, of two centuries and a half — and not being quite dead even 
now, naturally produced an enormous output. The extant works, notwith- 
standing all the mishaps to which Indian art has been exposed, still can be 
numbered by thousands. Almost at the very beginning of the operations of 
the school, about the year 1590, when Abul Fazl, the minister of Akbar, 
wrote his memorable description of his sovereign's administration, a hundred 
artists were reckoned to be masters of their craft, while tolerable practitioners 
were past counting. During the reigns of Akbar's son and grandson, in the 
first half of the seventeenth century, when the new form of art grafted upon 
the stock of ancient Indian tradition attained its highest development, the 
number of proficients must have increased. Although the long-continued 
political and social agony which accompanied the decline and fall of the 
Mughal empire necessarily limited the opportunities for the practice of art 
and diminished its rewards, art did not die; a synthesis between Hindu tradi- 
tion and Persian technique produced a new variety of Indian pictorial art 
possessing high merits. It is plain, therefore, that even when the eighteenth- 
century mythological painting is placed on one side for separate treatment, 
the mass of material to be dealt with by the historian is enormous, and that 
it is not possible within reasonable limits to do more than select a small 
number of typical examples. 
100-200 Many, perhaps most, of the extant Indo-Persian compositions are anony- 
artists nious, but hundreds are signed, and it would not be difficult to compile a list 
recor e . Q £ ^ names f f rom one h un cired to two hundred artists. Abul Fazl's list 
of those considered by him to be the most eminent numbers seventeen 
persons, all of whom, with possibly one exception, are represented by extant 
works. In one manuscript, the Waqiat-i-Babari y or history of Babar, written 
and illustrated about a.d. 1600, towards the close of Akbar's reign (B.M. 
Or. 3714), I noted the names of twenty-two artists, and probably overlooked 
several. Unfortunately, a great many of the names thus freely recorded are 
mere names, nothing being known concerning the men who bore them, so 
that the perusal of nominal lists offers little of interest. 
Hindu Perhaps the most fruitful general observation arising from such perusal is 
Xmmate" t ^ iat °^ ^ e predominance of Hindu names. For instance, in the Waqiat-i- 
" Babari above mentioned, out of twenty-two names, nineteen are Hindu, and 
only three Muslim. Similarly, in Abul Fazl's catalogue of seventeen artists, 
only four are Muhammadan, while thirteen are Hindu. 


The four Muhammadans named are: — (i) Mir Sayyid Ali, the illustrator Muham- 
of the story of Amir Hamzah, whose work probably is represented by the two madan . 
large pictures in B.M. Or. 3600 (ante, p. 468); (2) Khwajah Abdus Samad SJJJ 8 ^., 
(ante, p. 452); (3) Farrukh the Qalmak (Calmuck); and (4) Miskin (ante, list. 
p. 464). Farrukh certainly deserves high praise. He contributed good work 
to the Clarke MS. of the Akbarnamah, and was the author of a remarkable 
painting in three scenes occupying a full page on the reverse of folio 13 of 
B.M. Or. 3714. Miskin, who drew the outlines of two pictures (ir 7 , £7) in 
the Clarke MS., seems to be identical with the Ustad (scil. 'Master') Miskin 
who painted the Good Shepherd in the Johnson Collection and the Muham- 
mad Miskin, author of a lady's portrait in the same collection (LVIII, 15). 
Both those works are early in style. 

The thirteen Hindu names in Abul Fazl's list are:— (5) Daswanth; (6) Hindu 
Basawan; (7) Kesu (Kesava); (8) Lai; (9) Mukund; (10) Madho; (11) artists in 
Jagan[nath]; (12) Mahesh; (13) Khemkaran; (14) Tara; (15) Sanwlan; (16) 8ame ' 
Haribans; and (17) Ram. The signatures of all the seventeen artists named 
by Abul Fazl appear in the Clarke MS., except Haribans, No. 16; and re- 
appear in the Jaipur Razmnamah, excepting Nos. 1, 2, and 16. I do not 
remember seeing any picture signed by Haribans. There were two Madhos, 
the Elder (Kalan) and the Younger (Khurd). Kesu (Kesava) and some other 
artists are similarly duplicated in the signatures. Abul Fazl probably referred 
to the elder persons bearing the names. In the Razmnamah I have noted 
twenty-eight names, of whom twenty or twenty-one are Hindu. 

The sad story of Daswanth has been told already. Good specimens of Daswanth. 
his work as draughtsman are to be seen in Plates XII and XV of Col. 
Hendley's reproduction of the Jaipur Razmnamah, both of which were drawn 
in outline by him, and coloured respectively by Madho the Elder and Kanha. 
The subjects are Hindu legends, treated in the Persian manner, but with 
differences. The principal figures are distinctively Indian in feature and 
form, and even in the minor figures, where the chubby cheeks characteristic 
of the Persian style are preserved, the bodies are much less elongated than 
in Persian pictures. The scheme of colour too is lowered in brilliancy, and 
indigo blue is introduced for the bodies of deities. 

Basawan, whom some critics preferred to Daswanth, is represented by Basawan. 
Plate XXI of the Razmnamah, illustrating the story of the Raja who married 
the daughter of the King of Frogs. The lady, divesting herself of her fine 
clothes, returned to the water and resumed her froggy form, whereupon the 
angry husband proceeded to kill all the frogs he could find, until the lady 
was restored to him. The prevailing colour is green in various shades. The 
birds, frogs, trees, and flowers are drawn and painted with the utmost 
delicacy, but the general effect is marred by the intrusion of blocks of manu- 
script. The perspective convention is the same as that of the ancient bas- 

220 PAINTING Chap . XV 

reliefs. If the spectator imagines that all the persons, trees, &c. are on hinges 
and can be raised to their feet, they will then all fall into their proper relative 
positions. The artist saw with his mind's eye all the figures standing up, 
but in order to paint them, conceived them all to be laid down on one side. 
The subject seems to be regarded and viewed from above, all the parts being 
equally bathed in light, which is not represented as coming from any particular 
direction. Consequently, there are no shadows, and there is hardly any 
shading. Strong sunlight is indicated by a wash of gold behind the big tree. 
The drawing is by Basawan, the colouring by Bhawani. I am inclined to 
prefer Basawan to Daswanth. 
Kesu, &c. The two Kesus, or Kesavas, like Daswanth, were members of the lowly 
Kahar or palanquin-bearer caste. The elder (Kesava-dasa) dedicated a col- 
lection of pictures, including copies and imitations of Christian works, to 
Akbar in a.d. 1588 (Sam. 1646). 1 
Animals. The Indo-Persian artists excelled in the delineation of animals, both quad- 
rupeds and birds, and a delightful album might be composed of their pictures 
of animal life. The celebrated artist Mansur, who enjoyed the special favour 
of Jahangir, and was honoured by him with a title of nobility, began his career 
in Akbar's reign. Two hunting scenes (■£ and f, a 7 ) in the Clarke MS. of the 
Akbarnamah are his work. The Waqiat-i-Babari, B.M. Or. 3714, contains a 
series of eight exquisite little miniatures from his brush (Persian, Nos. 1 10-17, 
on folios 387-9). Mansur, however, excelled as an animal and bird painter. 
His work is further represented in the India section of the Victoria and 
Albert Museum by Nos. 21, 22, and 23 of the Wantage Bequest, paintings of 
a pheasant, a turkey-cock, and a blue-throated barbet. Mr. Havell has repro- 
duced successfully a beautiful white crane by Mansur in the Calcutta Art 
Gallery {Indian Sculpture and Painting, Plate LXI). 
Date of In Dara Shikoh's album (ante> p. 457) only three pictures (folios 25, 27, 
Sh'k^h* and 21 b) are dated— the dates being a.h. ioi4=a.d. 1605-6; a.h. 1018-= 
album! A ' D ' * 609-10; and a.h. 1043 =-a.d. 1633-4. The first of those years was that 
in which the sceptre passed from the hands of Akbar to those of Jahangir ; 
the third falls in the reign of Shahjahan. Six of the paintings (folios 17 b> 18, 
J 9 b> 33 bj 35 b y and 45 b) seem to include portraits of Jahangir (Prince 
Salim) in his youth and early manhood. The collection, as a whole, therefore, 
may be ascribed to the time of Jahangir and the earlier part of Shahjahan's 
reign, or in other words, to the first forty years of the seventeenth century. 
Muhammad The only signed composition is that on folio 21 6, dated 1633-4, which 
Khan - bears the name of Muhammad Khan. The picture is characteristic of 
Jahangir's bibulous court. It represents a young man clad in a bright 
yellow robe and large green turban, kneeling before a vase of flowers and 
a golden dish containing four earthenware jars, and engaged in pouring red 
1 'Assess. 9278, 9360' in Royal Library, Berlin; cited by Weber, Jnd. Ant., vi. 353. 











'tfliJUcAjtJ. u~>.*.;.~d.. 

















"»T A T'T-' 

rA, T)„1 : /L* -.,\ Clw.,.l T «*« »G*U ....nt.iinr V ..rw1 A VTi.c^lll 


wine from a jewelled goglet into a cup held in his left hand. No shading is 

The birds in this album, exquisitely drawn and coloured, are worthy of Beautiful 
Mansur and may possibly be from his brush. I admire particularly the birds - 
picture on folio 8 of a long-legged, brown bird standing by the side of a pool 
fringed with grass, flowers, and bamboos in tolerably good perspective. The 
blue sky, unfortunately, is rather crude. Another remarkable bird study is 
that on folio 10 representing admirably a wild duck standing by the side of 
a pool at the foot of a hillock. The sunlight on the face of the hillock is boldly 
indicated by a wash of gold, with surprisingly fine effect. No Chinese work 
could surpass the picture of the turkey-cock, ordered specially by Jahangir, 
and now in the Calcutta Art Gallery, reproduced by Mr. Havell in Plate LXII 
of Indian Sculpture and Painting. 1 

The works of the Indo-Persian draughtsmen and painters furnish a gallery Historical 
of historical portraits, lifelike and perfectly authentic, which enable the portraits, 
historian to realize the personal appearance of all the Mughal emperors and 
of almost every public man of note in India for more than two centuries. 2 It 
may be doubted if any other country in the world possesses a better series 
of portraits of the men who made history. Pictures of this class are so 
numerous, and so many ot such excellence, that it is difficult to make a repre- 
sentative selection. 

Portraits of Akbar are too many for specification in detail. One (B.M. ??£ raits ° f 
Add. 1 8801, fol. 10) shows him standing with Prince Salim (Jahangir) as a ^is friends, 
child beside him; and another (B.M. Add. 22470, fol. 4) exhibits him as the 
enthroned monarch hearing a woman's petition. The principal courtiers in 
this latter scene are all represented by careful likenesses with the names 
attached in minute script. Volume lvii of the Johnson Collection in the India 
Office Library, presented in 1816 by Dr. Buchanan (Hamilton), contains 
fifty-three rather rough sketches of princes and nobles, including Akbar's 
friends, Abul Fazl, Birbal, and Raja Man Singh. Volume lviii of the same 
collection is mostly filled with similar sketches of better quality and some 
unfinished portraits. No. 18, a finished work by Miskin, may be the earliest 
in the set. 

All critics, presumably, would admit that Indo-Persian art attained its high- ^{JV^F 1 
est achievements during the reign of the magnificent Shahjahan (a.d. 1627- ° a ja an# 
58), when the land enjoyed comparative peace, and a luxurious court offered 
liberal encouragement to all artists capable of ministering to its pleasure. 
The fierce scenes of bloodshed in which the earlier artists delighted were 

1 See Memoirs of Jahangir, transl. Rogers and nesses of Nurjahan and other ladies. The rigid 
Beveridge (1909), p. 215. seclusion of females prescribed by Muslim usage 

2 They are, as stated, perfectly authentic for the seems to preclude the possibility of real portraits 
men, but I share Manuccfs doubts about the of ladies of rank. 

authenticity of the numerous supposed like- 


replaced by pageants of peaceful courtly splendour, the old aggressive colour- 
ing was toned down or dispensed with, and a general refinement of style and 
execution was cultivated. In the portraits of men and favourite animals a 
little shading executed by a few delicate strokes was dexterously introduced, 
sufficient to suggest solidity and roundness, and yet managed with such 
reserve that the Asiatic reliance on the power of line was not interfered with. 
Some of The compositions of this period comprise a variety of subjects and are the 
the artists. wor j c f ma ny artists. The names of a few whose productions have attracted 
my special attention may be mentioned:— Chitarman, alias Kalyan Das; 
Anupchhatar; Rai Anup (possibly the same person), court painter to Prince 
Dara Shikoh; Manohar; Muhammad Nadir of Samarkand; Mir Hashim; 
and Muhammad Fakirullah Khan. 
Pictures One of the richest albums in the British Museum is the manuscript Add. 
selected by 18801, inscribed with a note stating that the volume was dedicated as a pious 
Sir Joshua donation in a.h. 1072 =a.d. 1661-2. Sir Joshua Reynolds examined the col- 
Keynoi lection in July 1777, and expressed his particular admiration for the following 
six works: 

No. 20. Pencil sketch of an officer of Shahjahan, by Chitarman, who was 
also called Kalyan Das; 

No. 21. Similar sketch of Azam Khan Koka, by Muhammad Nadir of 
No. 27. Similar sketch of Asaf Khan, anonymous; 
No. 28. Large anonymous sketch of Shahjahan holding court, surrounded 
by nobles whose portraits are named. The price is marked as 200 rupees, 
equivalent at that time to £25 or more; 

No. 30. Sketch of head of Hakim Masih-uz-zaman, a noble who had lived 
in Akbar's time, by Mir Hashim, very small and very good; and 

No. 40. Three portraits. The principal one is a sketch of Sher Muham- 
mad Nawal, by Muhammad Nadir of Samarkand. The minor ones are small 
coloured miniatures of Jahangir and Shahjahan by the same artist. 

No. 41. A delicate little head of Mirza Nauzar, a noble of Shahjahan's 
court, by Mir Hashim, is worthy to rank with Sir Joshua's selections. 
Charger by Turning to animals, we find in the Johnson Collection (vol. iii, fol. 1) a 
Manohar. life^e portrait of Dilpasand, or 'Heart's Delight', a favourite charger of 
Dara Shikoh, by an artist named Manohar. An equestrian portrait of the 
same prince mounted on another charger is also notable (ibid., vol. iv, fol. 9) 
and of unusually large size, about 1 1 inches by 9. 
Cats. The tiny cat sitting up, in vol. liii, Fol. 5 , of the same collection, is excellent. 
This is not the only example of pictures of cats. One appears at the feet of the 
Emperor Farrukhsiyar in volume No. 5 of Exhibition Case B in the King's 
Library, British Museum, and a few others occur in other compositions. 
Elephants. Perfectly drawn elephants are numerous. Indian artists, whether sculptors 




~ C 
ca *^ 








■ ■v-v; 






O 3 

■5 s 






or painters, rarely failed to produce good representations of the huge quad- 
ruped, the nature of which they understood thoroughly. Volume lxvii in the 
Johnson Collection is specially devoted to elephants, several of which are 
admirable. One of the best is that on folio 7, by Nadir-uz-zaman (Abul 
Hasan). 1 Another fine picture is that on folio 15. The main subject is a 
magnificent elephant standing in a palace courtyard, with other elephants, 
a bullock, &c, as accessories. The drawing is grisaille in a brownish sepia 
tint, no other colour being used, except that the golden ornaments of the 
elephant are yellow. 

The many charming pictures treating of miscellaneous subjects including Miscel- 
illustrations of popular stories, offer a wide field for description and selection, lai J e . 0US 
far too large to be treated exhaustively. subjects. 

A favourite subject was the story of Baz Bahadur, king of Malwa, and his Baz Bahadur 
lady-love, Princess Rupmati, who are represented in several pictures as riding and Rupmati. 
together by torchlight. A good example in the Calcutta Art Gallery has been 
reproduced in Plate LXIV of Mr. Havell's Indian Sculpture and Painting] 
another, from the Hercules Read's rich collection, is of special value because 
of the label indicating the subject; and a third is on folio 22 of B.M. Add. 
21928. Other romances frequently illustrated are the tales of Laila and 
Majnun, Khusru and Shirin, and Kamrup and Kamta. 

Mr. Havell has rightly drawn attention to the skill with which the Indian Contrast 
artists treated the contrast between the pitchy darkness of night and the flare between 
of artificial light. Several pictures are extant which exhibit this contrast in J}f^ n and 
scenes of hunting by night, flaming torches being used to dazzle and hypno- 
tize the deer. Colonel Hanna's Collection, now in Washington, includes two 
such scenes, Nos. 42 and 102, of which the latter excited the warm admiration 
of the late Sir Frederick Burton. A more modern specimen in the Calcutta 
Art Gallery is reproduced in Mr. Havell's Plate LXV. 

The same motive, which also attracted Rembrandt, inspires the pictures 
representing a lady standing on a balcony watching the effect of fireworks over 
the dark waters of the Jumna. Sometimes she is shown in the act of dis- 
charging a squib herself. 2 In folio 4 of vol. xv of the Johnson Collection, the 
lady, clad in bright scarlet and standing against a background of inky dark- 
ness, produces a very impressive effect. A picture by Muhammad Fakirullah 
Khan (folio 7 of the same volume) depicting the nocturnal pursuit of a warrior 
is equally successful in bringing out the opposition of light and darkness. 
Other compositions exhibiting people grouped round a camp-fire aim at like 

1 Nadir-uz-zaman was the official title of Abul hammad Afzal, with a Persian verse on the back, 
Hasan, a favourite artist of Jahangir. He seems dated a.h. 1069 = a.d. 1658-9, commendatory 
to have continued to work in the following reign, of the painter. Another of his works is in volume 

2 Dr. Coomaraswamy possesses a good picture xi of the Johnson Collection, 
of girls discharging fireworks, signed by Mu- 

224 PAINTING Chap. XV 

Pictures of Many artists took great delight in depicting holy men and ascetics of all 
holy men. sor ts, Musulman and Hindu, singly or in groups. Two of the most exquisite 
works dealing with this class of subject, and no doubt executed in the reign 
of Shahjahan, are the companion pictures, folios n b and 12, Dara Shikoh's 
album, representing an old fakir in two positions, holding a book in the one 
case, and a rosary in the other. The outline of the figure is drawn with less 
than the usual sharpness, and shading with fine lines is employed sufficiently 
to give an impression of roundness. In the old man's beard the delicacy and 
accuracy with which individual hairs are drawn displays a wonderful mastery 
over that most difficult instrument, the single-hair brush. The colouring is 
subdued, and the perspective fairly correct. 
The reader Another drawing in the same volume, in similar style, and probably by the 
of the same ar tist, is that on folio 60. The subject is the reading by a young mullah 
oran. (M u hammadan teacher) from a Koran resting on a stand. Two of his com- 
panions are listening attentively, while the third, in the foreground, is en- 
gaged in pouring water over the toes of his left foot held up in his hand. The 
drawing of the difficult position of this figure is extremely clever. 
Court scenes. Most of the albums contain examples of gorgeous court scenes elaborated 
with infinite patience and minuteness of detail, harmoniously coloured, and 
often enriched with gold. It would be next to impossible to reproduce the 
most splendid of these pictures in colours with success, and I think it better 
not to make the attempt. The composition being the weak point in these 
works, photographs do them an injustice. Colonel Hanna possessed two of 
the richest specimens in existence, Nos. 1 and 2 in his volume marked Persian 
Drawings. No. 2 is the largest Indo-Persian picture known to me, excepting 
the early illustrations of the Story of Amir Hamzah, the measurements being 
23 inches by 17 \ inches. The subject is a review of cavalry on the bank of the 
Jumna by Shahjahan mounted on an elephant. The portraits of the principal 
chiefs and officers in the crowd have their names attached. 
Court of The manuscript B.M. Add. 20734, an official present given by the titular 
Shahjahan. Emperor of Delhi in 1815, contains nine pictures in the most highly finished 
style, of which two may be specified. One representing the infant Shahjahan 
(Prince Khurram) lying in his mother's lap, surrounded by admiring atten- 
dants, is wrought with colouring so rich and decorative details so elaborate 
that an attempt to copy it would certainly fail. Another picture, extending 
across two pages (fol. 689, 690), and depicting Shahjahan seated on the pea- 
cock throne in all his glory, while Asaf Khan offers a present of costly pearls, 
gives a vivid notion of the extravagant magnificence of the Mughal court in 
its prime. 
The ladies. Volumes ix, x, xi of the Johnson Collection may be noticed as being 
specially devoted to the ladies, some of whom are represented half nude in 
the bath or at their toilet. The pictures in volume xi are particularly good, 

PLATE 163. Lady and sunlight effect, by Rao Gobind Singh 
(Johnson Collection, vol. xxi, fol. 8) 


the most noticeable being a charming portrait of a lady wearing a high conical 
head-dress, and admirably shaded. From an inferior replica (B.M. Add. 
1 1747, f°l- 5 2 ) we l earn ^at the lady's name or title was Malkah Zamaniya. 

Passing on to the reigns of Aurangzeb (1 658-1 707) and his decadent sue- Later art. 
cessors during the eighteenth century, we find the artists still numerous and 
specimens of their work abundant. Although Aurangzeb was too zealous a 
puritan to care for art himself, the fashion set by his predecessors had not 
died out, and princes and nobles still kept court painters. Portraiture con- 
tinued to be practised with great success, although the execution rarely 
attains the perfection of the first half of the seventeenth century. The art 
of this period and subsequent periods can only be justly treated of as the 
product of artists who gained a living at minor courts, Hindu or Muham- 
madan, and whose style and choice of subjects are modified by the local 
demand. Certain of these local styles, spoken of collectively as 'Rajput', are 
distinct, but much of the later work remains true to the decadent Mughal 


To Dr. Coomaraswamy must be given the credit of the primary study and 'Rajput* 
classification of non- Mughal Indian paintings. He begins his survey 1 with P amtm 6- 
a quotation from Abul Fazl who says of the Hindu painters at the Mughal 
court that 'their pictures surpass our conception of things. Few, indeed, in 
the world are equal to them.' As has been said, his classification is based 
upon a dual conception of two schools of Indian painting, Mughal and 
Rajput, which are 'utterly diverse' in temper, the Rajput school dating from 
the fifteenth and sixteenth centuries and therefore preceding the Mughal 
school. He acknowledges that in latter days the schools tended to converge 
and blend, but at all times the subject-matter was different. With regard to 
his early dating of Rajput painting, he supports his views by a direct com- 
parison with Ajanta and Sigiriya. The well-known 'Death of Bishma' is there- 
fore 'unmistakably . . . reminiscent of the great Buddhist Parinirvanas\ It is 
evident that much of his argument is based on the subject-matter. Rajput 
painting is Hindu, popular, spiritual. . . . Mughal painting is Muhammadan^ 
courtly, material. The primary fact that is overlooked is that the technique 
of the two schools is identical, and Persian in origin. 

Dr. Coomaraswamy's classification of Rajput painting is a geographical Classification 
one, which invites chronological inexactitude. There is a Rajasthant (low- 
land) school and a Pahari (Himalayan) school. Though these subdivisions are 
absolutely acceptable in themselves, it must be acknowledged that there are 
numerous local schools and certain period differences to be distinguished. 
Roughly speaking, the Kangra paintings with their flowing line and western- 
ized drawing of foliage and landscape are typical of the Pahari schools, while 
1 Burlington Magazine, March 1912, quoting Blochmann's Ain-i-Akbari, i. 107. 

3292 g g 

226 PAINTING Chap. XV 

the Jaipur paintings with their concentration on jewellery treated in relief 
and formal drapery are typical of the Rajasthani schools. Both of these lesser 
schools show Mughal, if not foreign influence, especially with regard to their 
architectural settings. Certain Rajasthani paintings, however, exist which are 
clearly earlier than the eighteenth and nineteenth century Kangra and Jaipur 
work. Most of these are Ragini subjects, but their technique and the details of 
costume and architecture will not allow of them being dated pre-seventeenth 

Origin of These Rajput paintings seem to have been the work of the court painters 
Rajput f the petty Rajput courts of the seventeenth and eighteenth centuries. As 

pamtmg. ^^ ^ p arent Mughal school, portraits are plentiful, especially of the Jaipur, 
Bijapur, and Hyderabad schools, and a survey of them would provide ac- 
curate chronological data. Nineteenth century work is plentiful, being chiefly 
of the copyist order. The colouring tends to be crude and the drawing 
slovenly. Moreover, certain painters are still at work, turning out the old 
subjects usually on old paper — to the great confusion of students. 
Bazaar Apart from the work of the court painters, much work exists which is the 

pamtmg. p ro duct of 'bazaar schools'. Of these the Calcutta brush drawings in colour 
of the Patua caste are especially notable for their vigorous line. At Trichino- 
poly also there flourished a bazaar-school during the last century, working in 
tempera colours on paper or talc. It may be taken for granted that most of 
the large cities have produced * bazaar- work' of a kind, very little of which 
has been preserved. The subjects depicted in this type of work are usually 
purely iconographical. 

The Modern Schools 

Pictures at At the Delhi Exhibition of 1902-3 many examples were shown of the oil- 

*? ,P. elhl paintings and water-colours produced in considerable quantities of late years 

ex 1 ltion. ^ students trained in European methods, chiefly at the Government Schools 

of Art in Calcutta, Bombay, Madras, and Lahore. In Sir George Watt's book 

Mr. Percy Brown, late Principal of the Calcutta School of Art, criticizes 

the Delhi exhibits as follows:— 

'Until its introduction from Europe, there was no oil painting of any kind practised 
throughout the country, but the number of pictures executed in the medium shown 
in the Exhibition reveals the fact that oil picture painting as a branch of study, as well 
as a means of livelihood, is being taken up seriously by a rapidly increasing class. 
Some of the work displayed in the Eastern Hall of the Exhibition was remarkably 
good ; in the life studies the modelling and feeling of living flesh being well reproduced, 
and one or two landscapes showed an atmosphere and a consideration for composition 
which is [sic] worthy of remark. Much, however, of the work shown was of a very 
ordinary character, the drawing being decidedly defective, and the technique and 
colouring in most cases crude.' * 

1 Indian Art at Delhi, p. 457. 

PLATE 164. Marble building, &c, by Muhammad Fakirullah Khan 
(Johnson Collection, vol. xvii, fol. 3) 


The most prominent representative of the Europeanized school of Indian Raja Ravi- 
artists was the late Raja Ravi-varma of Travancore, a connexion of the varma. 
Maharaja of that State. His works, which are extremely numerous, achieved 
wide popularity, and have been freely vulgarized by oleographs and other 
cheap modes of reproduction. The Raja practised both portrait and land- 
scape painting, and four of the portraits in the Banqueting Hall, Madras, 
are from his brush. 1 He was assisted by his relative, Raja Raja-varma, and 
other members of his family. He had received instruction from Theodore 
Jensen and other European artists who visited Southern India, as well as 
from Alagri Naidu, a native of Madura, in the Madras Presidency, who was 
patronized by Swati Tirumal, Maharaja of Travancore from 1829 t0 *847> 
and was considered in his day to be the best painter in India after the 
European fashion. Ravi-varma had a formidable rival in Ramaswamy Naidu, 
a member of the clan of Naiks at Madura, who was considered to excel in 
portrait painting. 

Stimulated by the active encouragement of the royal family of Travancore, 
the Gaikwar of Baroda, and other wealthy patrons, Ravi-varma turned his 
attention to the illustration of the Hindu legends and epics. 

In his own country his works in that kind are regarded as masterpieces Criticism of 
and adequate expressions of Indian feeling. At the hands of recent critics Ravi-varma's 
in Europe they have met with a different reception. works. 

'The art\ writes Mr. Havell, c which truly reflects the fictitious culture of Indian 
universities and the teaching of Anglo-Indian art schools, is exhibited in the paintings 
of Ravi-varma, who is the fashionable painter of modern India for those Indians who 

do not ignore Indian art altogether Certain it is that his pictures invariably manifest 

a most painful lack of the poetic faculty in illustrating the most imaginative Indian 
poetry and allegory ; and this cardinal sin is not to be atoned for by any kind of technical 
skill in the execution.' 2 

Dr. Coomaraswamy, a fellow mystic, is still more severe, and declares that 

'theatrical conceptions, want of imagination, and lack of Indian feeling in the treat- 
ment of sacred and epic Indian subjects are Ravi-varma's fatal faults. . . . His pictures 
are such as any European student could paint, after perusal of the necessary literature 
and a superficial study of Indian life/ 3 

In a more recent publication the same author gives his opinion with 
greater brevity and somewhat less severity to the effect that 

'the late Raja Ravi-varma was the best known of these painters in a purely European 
style, but neither he nor any other workers of the pseudo-European school attained 
to excellence. His work at the best reached a second-rate standard.' 

1 The four portraits are those of the Duke of Hall, Madras (Government Press, Madras, 

Buckingham and Chandos, Sir Arthur Havelock, 1903), p. 132). 

and the Ladies Mary and Caroline Grenville z Indian Sculpture and Painting, p. 251. 

(Col. H. D. Love, R.E., Descriptive List of 3 Modern Review (Allahabad), vol. ii, p. 107. 
Pictures in Government House and the Banqueting 

228 PAINTING Chap . XV 

Probably this last quoted judgement is not far wrong. 1 

The Bengali 'The work of the modern school of Indian painters in Calcutta', Dr. Coomaraswamy 
Nationalist writes, 'is a phase of the National reawakening. Whereas the ambition of the nine- 
School, teenth-century reformers had been to make India like England, that of the later 
workers has been to bring back or create a state of society in which the ideals expressed 
and implied in Indian culture shall be more nearly realized.' 

This new movement on the art side has been enthusiastically supported 
by Mr. E. B. Havell, late Principal of the Calcutta School of Art, who felt 
keenly the futility of training Bengali students on purely foreign methods, 
alien to their nature, and sought to turn their attention to the productions 
of the Indo-Persian and eighteenth-century Hindu schools as being more 
expressive of Indian ideals. With some difficulty Mr. Havell persuaded the 
authorities to let him have his way, and replace a collection of poor European 
works by a choice selection of Indian paintings. He found in Mr. Abanindro 
Nath Tagore, now Vice-Principal of the School of Art, a willing coadjutor, 
and a painter of considerable power. Mr. Havell recognized in his colleague 
a real artist 'who has come to pick up the broken threads of Indian pictorial 
tradition', and credited him with 'giving us a true interpretation of Indian 
spirituality, and an insight into that higher world, the fairy land of Eastern 
poetry and romance, which Eastern thought has suggested'. 2 

The critic proceeds to say that 

'if neither Mr. Tagore nor his pupils have yet altogether attained to the splendid 
technique of the old Indian painters, they have certainly revived the spirit of Indian 
art, and besides, as every true artist will, invested their work with a charm distinc- 
tively their own. For their work is an indication of that happy blending of Eastern 
and Western thought, from the full realization of which humanity has so much to gain.' 

These rather large claims are founded on a series of small works described 
in the Studio as 'water-colour drawings', and very far indeed from having 
'attained to the splendid technique of the old Indian painters', which they 
do not attempt to rival. The more sober criticism of Dr. Coomaraswamy is 
more closely in accordance with the facts. 

'The subjects chosen by the Calcutta painters', he observes, 'are taken from Indian 
history, romance, and epic, and from the mythology and religious literature and 
legends, as well as from the life of the people around them. Their significance lies 
in their distinctive "Indianness". They are, however, by no means free from European 

1 Thirty-five of Ravi-varma's pictures are repro- The prints are too rough for reproduction, 
duced in an illustrated collection of Hindi courteously permitted by the editor. A list of 
poems, entitled Kavita Kalap (Allahabad, 1909), Ravi-varma's works and an enthusiastic appre- 
edited by Mr. Mahavira Prasada Dirvedi, and ciation of his art will be found in V. Nazam 
shown to me by Dr. Grierson. That book also Aiya, Travancore Manual, vol. iii, p. 263, a corn- 
contains prints of pictures in a similar style by pilation which is a rich mine of information. 
Braj Bhushan Rai Chaudhri, Babu Vamapad 2 Indian Sculpture and Painting, pp. 256, 257. 
Bandhopadhvaya, and Sriyut M. V. Durandhar. 


and Japanese influence. The work is full of refinement and subtlety in colour, and of 
a deep love of all things Indian ; but, contrasted with the Ajanta and Mughal and Rajput 
paintings which have in part inspired it, it is frequently lacking in strength. The work 
should be considered as a promise rather than a fulfilment. So regarded, it has very 
great significance for the future of Indian Art.' l 

Mr. Roger Fry holds a poor opinion of the work of the modern artists. 'Such pictures 
as that of "The Siddhas of the Upper Air",' he observes, 'show that, however anxiously 
these artists strive to adopt the formulae of their ancestors, the spirit that comes to 
expression is that of the American magazine illustrator. Nothing, indeed, could 
provide a stronger proof of the profound corruption which contact with European 
ideas has created in Oriental taste than these well-intentioned but regrettable drawings/ 2 

The leader of the school, Mr. Abanindro Nath Tagore, began as a painter 
in oils, after the European fashion, but soon abandoned the oils medium, and 
devoted himself to the 'water-colour drawings'. Many works by him and his 
pupils have been exhibited at meetings of Societies in London, and repro- 
duced in Mr. Havell's book or in periodicals. One specimen, therefore, may 
suffice — the picture of the ' Exiled Yaksha', or demigod, an illustration of 
a passage in Kalidasa's poem, the Meghaduta, or ' Cloud Messenger', by Mr. 
Tagore. Another good picture is 'The Flight of Lakshman Sen' by the late 
Mr. Surendra Nath Gangooly (Mr. Havell's PI. LXXVIII). Other pupils 
of Mr. Tagore deserving mention are Nanda Lai Bose, Ishwari Parshad, 
a descendant of hereditary painters at Patna, Gogonendra Nath Tagore, 
brother of Abanindro Nath, Asit Kumar Haldar, and Hakim Muhammad 

All well-wishers to India will join in the hope that the promise shown by The future, 
this new Bengali school may lead to something more important than the 
works hitherto produced. Probably all critics will agree that nothing of high 
worth can be created by men who merely seek to imitate foreign models. 
If modern India is to evolve a new art of her own it must have its roots in 
the Indian past and appeal to Indian sentiment. 'L'art dans Tlnde sera 
indien, ou il ne sera pas' 4 : but 'to be, or not to be, that is the question' 
which at present no man can answer. 

1 Catalogue of the Indian Court, Festival of Review (Allahabad), May, 1907; Havell, Indian 
Empire, 19x1, p. 106. Sculpture and Painting, Pis. LXXIII-LXXVIII. 

2 Quart. Rev., 1910, p. 237. 4 M. le comte Goblet d'Alviella, Ce que YInde 

3 See Studio for 1902, 1905, 1908; Modern doit d la Grice (Paris, 1897), p. 94. 


vL. D. Barnett, The Antiquities of India. Medici Society, 1913 : 'Handbooks to Ancient Civilizations*. 
vSir G. Birdwood, The Industrial Arts of India. 1880. 
\ Cambridge History of India: Vol. i, Ancient India. Edited by E. J. Rapson. Cambridge University 

Press, 1922. 
v A. K. Coomaraswamy, A History of India and Indonesian Art. Goldston, 1927. 
v Vincent Smith, Oxford History of India. Edited by S. M. Edwardes. Clarendon Press, 1923. 

Sir G. Watt, Indian Art at Delhi. 1904. 


J. Burgess, Archaeological Survey of India Reports, Vols. i-v. For Cave Temples, &c. 

K. de B. Codrington, Ancient India (Benn, 1926); Medieval India. 

Sir A. Cunningham, Archaeological Survey of India Reports, Vols, i-xxiii. 

, The Stupa at Bharhut. London, 1879. 

, Bhilsa Topes. London, 1854. 

*J. Fergusson and J. Burgess, A History of Indian and Eastern Architecture. Murray, 1910. 

A Foucher, VArt Grko-Bouddhique du Ghandhdra. Paris, 1905 and 1923. 

, Etude sur VIconographie bouddhique de VInde. Paris, 1905. 

O. C. Gangoly, South Indian Bronzes. Calcutta, 1914. 
xG. Jouveau-Dubreuil, Iconographie du Sud de VInde. Paris, 1914. 

Archaeological Survey of India Reports and Memoirs. 
v Sir J. H. Marshall, Guide to Sanchi. Calcutta, 1918. 
x , Guide to Taxila. Calcutta, 191 8. 

Sir A. Stein, Serindia. Clarendon Press, 1921. 

\P. Brown, Indian Painting. Oxford University Press, 191 8. 
A. K. Coomaraswamy, Indian Drawings. London, 1910 and 191 2. 

, Rajput Painting. Oxford University Press, 1916. 

V. Goloubew, Peintures bouddhiques aux Indes; in Annales du Musie Guinet, xl. Paris. 

J. Griffiths, The Paintings . . . of Ajanta. London, 1896-7. 

E. Kuhnel and H. Goetz, Indische Buchmalereien. 

N. C. Mehta, Studies in Indian Painting. Bombay, 1927. 

PLATE 165. The exiled Yaksha, by Abanindro Xath Tagore 


Aachen 'Bacchus', 72, 73. 

Abanindro Nath Tagore, 228, 229. 

Abhayagiri, 142, 146. 

Abul Fazl, 208, 209, 211, 218, 219, 221, 225. 

Abul Hasan (Nadir-uz-zaman), painter, 223. 

Abu Turab, 176. 

Adham Khan, painter, 215. 

Aesthetics, 13, 14. 

Afzal Khan, tomb, 201. 

Agama y 14. 

Agra, 177, 178, 180, 181, 183, 184, 188, 190, 191, 

192, 196, 197, 198, 199, 201, 204, 210, 211. 
Ahinposh, 55. 

Ahmadabad, 176, 196. 

Ahmad Shah, sultan, 176. 

Ahmad Shah Kutb-ud-din, 204. 

Ajanta cave temple, 13, 23, 25, 27, 36, 46, 49, 51, 
53» 5 6 > 6 7> 78, 79» 8 5> 86, 87, 90, 91, 92-109, in, 
113, 121, 161, 162, 165, 203, 204, 208, 225, 229. 

Aji Saka, 156. 

Ajmer mosque, 171, 172, 191, 192. 

Akbar, 64,92, 100, 122, 161, 178-80, 187, 189, 191, 

193, 195, 197, 203, 207, 208, 210, 212, 213, 215, 
218, 220, 221. 

Akbarnamah MS.> 211, 215, 216, 219, 220. 

Ala-ud-din Khalji, sultan, 174. 

Alexander, 15, 70-2. 

Allahabad, 18, 78, 80, 83. 

Altamsh, sultan, 69, 171, 173. 

Amaravati, 22, 24, 27, 32, 36, 39, 46-50, 53, 71, 73, 

74i 75- 
Amar Singh, 190. 

Amber, 199. 

Ambusthala, 147. 

Amir Hamzah, 206, 224. 

Ananda, 23, 150. 

Ananta, 80. 

Anathapindika, king ?, 23, 31. 

Anderson, Dr., 5, 74, 124, 164. 

Andhra dynasty, 47, 77, 95. 

Anupchhatar, painter, 222. 

Anuradhapura, 99, in, 112, 142-52. 

Archaeological Survey, 35, 138, 151, 190. 

Architecture, 13, 15,21-8, 32, 54, 55, 74, 79, 113- 

19, 129-32, 142-50, 171-86, 187, 206, 208. 
Arcot, N. and S., 133, 137. 
Arjumand Banu Begam, 181. 
Arjuna, 133. 

Arunachalam, the Hon. Mr. P., 152, 153, 154. 
Asaf Khan tomb, 202, 224. 
Asanaya, 145. 
Asirgarh, 188. 
Asit Kumar Haldar, painter, 229. 

Asoka, 15, 17, 18, 20, 21, 24, 27, 29, 32-5, 40, 68, 

69, 70, 72, 78, 83, 84, 142, 147, 149. 
Assatha, 8. 
Asura, 8. 
Asvattha, 8. 
Atakur stone, 134. 
Atala, 174. 

Aurangabad, 85, 109. 
Aurangzeb, 126, 186, 190, 191, 201, 211, 212, 213, 

217, 225. 
AvalokiUsvara> 123, 155, 162, 167. 
Avantipura, 118. 
Avantivarman, king, 119. 
Avatar a, 13. 
Awkana, 150. 

Babar, founder of Mughal dynasty, 178, 206, 207, 218. 

Badakshan, 214. 

Badami, 13, 86, 87, 91, 121 , 130, 134, 147. 

Badaoni, critic of Akbar, 212. 

Badulla, 155. 

Bagh, 78, 85, 92, 95, 97, 108, 109. 

Bagur stone, 134. 

Baha-ul-hakk tomb, 199, 200. 

Bahmani, sultanate, 176. 

Baital Dewal, 124. 

Bajna, 67, 164. 

Bakhira, 17, 19. 

Bala Krishna, 125. 

Bandarkar, Mr. D. R., 117, 128. 

Bangalore, 134. 

Bargaon, 121. 

Basaklik monastery, 164. 

Basawan, 209, 210, 219, 220. 

Bathang, 169. 

Batiya Tissa (Bhatika Abhaya), king, 147. 

Bedsa cave temple, 26, 27, 38. 

Bell, Mr. H. C. P., no, in, 112, 143, 144, 145, 149, 

150, 152. 
Bellary, 131,132, 135, 136. 
Belur, 130, 131, 196. 
Benares, 40, 126, 186. 
Berar, 95. 
Bernier, 191, 213. 
Besnagar, 12, 22, 29, 30, 32. 
Bettiga, king, 131. 
Beylie, General de, 172. 
Bhairava, 88. 

Bhaja cave temple, 26, 27, 38, 100. 
Bhaniyar (Buniar) temple, 118. 
Bharhut, 7, 8, 11, 13, 21, 22, 24, 27, 29, 30, 32, 34, 

37. 38, 39> 4°> 4i» 43» 46, 48, 49> 53> 61 > 66, 68, 70, 

7i > 72. 73> 75» 79> 80, 93> 120, 151- 



Bhatgaon temple, 119. 

Bhatika Abhaya (Batiya Tissa), king, 147. 

Bhawani, painter, 220. 

Bhikshu, 23. 

Bhilsa, 16, 29. 

Bhitargaon, temple, 117. 

Bhitari, 83. 

Bhringi, 59. 

Bhutan, 83. 

Bhuvanesvar, 6, 13, 38, "5» Il6 > JI 7» * l8 > I2 4- 

Bidar, 176. 

Bihar, 15, 17, 82, 121, 122, 123. 

Bihzad, 206, 207. 

Bijapur, 176, 177, 178, 188, 226. 

Bilhari temple, 118. 

Bimbisara, king, 37. 

Bindusara, king, 15. 

Birbal, painter, 221. 

Birdwood, Sir George, 4, 174. 

Bishanpur, 124. 

Bodh-Gaya, 23, 27, 32, 33, 53, 58, 59. 69, 74, 115. 

Bodhisattva, 40, 42, 43, 44, 46, 56, 58, 59, 66, 85, 

123. 1 5 2 « *55. *58, 162, 166. 
Bokhara, 206, 207. 

Borobodur, 65, 85, 106, 120, 140, 158-60. 
Brahma, 53, 56, 153, 158, 160. 
Brahmin, 11, 17, 26, 28, 37, 39, 40. 53. 59. 6 5. 66 > 

76, 78, 79, 86, 113, 121, 122, 127, 144, 156, 171. 
Brambanam, 158, 160. 
Brick (burnt), 28, 117. 
Brick (sun-dried), 15, 21. 
Brown, Mr. Percy, 203, 207, 226. 
Budaun, 171. 
Buddha image, 24, 32, 36, 37, 41, 42, 43, 48, 53, 57, 

58, 81, 82, 85, 102, 120, 122, 123, 145, 148, 150, 

i5*t x 55. l6o » l6 5. x 7°- 
Buddhism, 10, 11, 13, 15, 17, 21, 24, 26, 28, 30, 31, 

32, 33. 34. 36, 37. 39. 42. 43. 45. 46, 4 8 , 53. 54. 5^, 
57. 58, 65, 66, 67, 71, 73, 76, 78, 79. 8o, 85, 86, 
91, 92, 95, 99, 104, in, 113, 120, 121, 122, 142, 
144, 155, 156, 158, 163, 164. 

Buner, 52. 

Burgess, Dr., 3, 47, 50, 54, 56, 62, 71, 85, 86, 88, 95, 
101, 102, 104, 175, 183. 

Burhanpur, 181. 

Burma, 73. 

Calcutta, 65, 124, 168, 205, 220, 221, 223, 226, 228. 

Calligraphy, 187, 192-6, 208, 209. 

Cambay, mosque, 175. 

Cave temples, 13, 24-7, 32, 38, 46, 49, 59, 79, 85- 

91,94, 121, 159. 
Ceylon, 15, 17, 24, 48, 53, 66, 78, 92, 93, 96, 99, 104, 

109, no, 131, 134, 137, 140, 142-55- 
Chaitya, 21, 23, 24, 27, 33, 91, 93, 94, 104. 
Chalukya dynasty, 77, 87, 95, 130, 132, 133, 135, 

157, 196. 

Chanda, yakshi, 71. 

Chandaka, 62. 

Chandel kings, 116, 126. 

Chandi Sewa, 158. 

Chandragupta Cave, 13, 79. 

Chandragupta Maurya, 15. 

Chandragupta II, Vikramaditya, 77, 80, 82, 83. 

Chera dynasty, 133, 151. 

Chhargaon, 45. 

China, 66, 68, 206, 208. 

Chinese Turkistan, 65, 67, 161-5, 208. 

Chingleput, 133. 

Chini-ka-Rauza y 201. 

Chini (tile-work), 201. 

Chiqqan Kol, 164. 

Chitarman, Kalyan Das, 222. 

Chitor, 127, 190, 191. 

Chola dynasty, 129, 132, 133, 134, 140, 144, 160. 

Chorten, 166. 

Choultry, 138, 139. 

Chunatn (plaster), 25, 97. 

Chunar, 40. 

Cire perdue casting, 82, 140. 

Clarke MS., 215, 216, 219, 220. 

Coimbatore, 151. 

Coins, 78, 82, 187-9. 

Conjeeveram (Kanchi), 129, 133. 

Coomaraswamy, Dr. A. K., 5, 64, 65, 86, 140, 148, 

i5 2 . x 53» *54. 155. l6o « 2 °3> 225, 227, 228. 
Cunningham, Sir A., 3, 30, 31, 32, 35, 41, 43, 50, 

55. 63, 79. 80, 83, 84, 119, 127, 177, 191, 199, 

Cyrene, 15. 

Dabgir mosque, 202. 

Dagaba, 24, 92, 112, 142-7, 158, 166. 

Dames Collection, 60, 61. 

Dandan-Uiliq, 62, 161-3. 

Dara Shikoh, 211, 212, 216, 220, 222, 224. 

Dasaratha, king, 17, 27. 

Das Avatara, 87, 88, 90, 91. 

Dastan-i-Amir Hamzah, 206. 

Daswanth, 209, 210, 219, 220. 

Deccan, 130, 132, 196. 

Delhi, 171, 172, 173, 175, 178, 179, 181, 188, 191, 

Deogarh, 12, 79, 81. 
DevalokaSy 25. 

Devanampiya Tissa, king, 147. 
Devata, 7, 8, 30, 31,37. 
Dhanamjaya Dasarupa, 14. 
Dharanikota, 46. 
Dharma, the Law, 10, 57. 
Dharmaraja Ratha, 129. 
Dhatarattha, king, 8. 
Dholka, 175. 
Digambara sect, n. 
Dravidian, 129, 130, 132, 138, 144. 



Durga, 12, 133. 
Dutthagamini, king, 92, 147. 

Early Period, 13, 21-38, 69, 74. 

Egypt. 15.7*. 

Elephanta, 13, 59, 65, 66, 79, 86, 89. 

Ellora, 13, 23, 25, 65, 66, 79, 85, 86, 87, 88, 90, 91, 

Enderde, 163. 
Epirus, 15. 
Eran, 68, 84. 
Eravana, 8. 

Fa-hian, Chinese traveller, 23, 66, 156, 157. 

Farrukh the Qalmak, painter, 219. 

Fatehpur Sikri, 100, 179, 180, 189, 193, 194, 197, 

199, 207, 208, 216. 
Fathabad, 178. 
Fergusson, Mr., 3, 21 , 25, 35, 47, 49, 50, 95, 102, 105, 

107, 114, 115, 116, 117, 119, 130, 131, 132, 137, 

J38. i39> J 59, 173. *74. *75» *77» x 79. 186, 196. 
Finch, William, traveller, 190. 
Foreign influence, 29, 33, 48, 49, 50-76, 161-70, 

177, 203, 217. 
Foucher, Mons. M. A., 52, 53, 56, 62, 63, 69, 70, 71, 

72,76, 159, 166, 167, 168. 

Gajabahu I, king, 146, 151. 

Gal-vihare, 150. 

Ganas y 144. 

Gandhara, 26, 27, 33, 36, 39, 41, 42, 43, 44, 45, 46, 

49. 5°. 5 1 . 5 2 » 53> 5 6 > 57. 5 8 . &>. 6l > 62 » 6 3. 6 4. 6 5. 
66, 67, 68, 71, 72, 73, 74, 75, 76, 80, 81, 102, 128, 

138, 139. 159. 164. 
Gandharva, 7, 33, 58. 
Ganesa Ratha, 129. 
Ganga, 79. 

Gangaikonda-Cholapuram, 129, 134. 
Gardner, Prof. P., 16, 61. 
Garhwa, 80, 81, 100. 
Garuda, 7, 8, 12, 61. 
Garwhal, 203. 
Gaur, 175, 192, 200. 
Gautama Buddha (Gotama), 8, 1 1, 15, 19, 42, 53, 56, 

57. 59. 61. 62, 73, 81,99. 
Ghazni, 171. 

Ghiyas-ud-din Tughlak, 174. 
Ghosundi, 12. 
Gill, Major, 51,95. 
Gingee, 137. 
Girnar, 11. 

Gogonendra Nath Tagore, painter, 229. 
Gol Gumbos, 177. 
Gol Mandal y 198. 
Gop temple, 119. 
Gopuram, gateway, 129, 130, 144. 
Gosain cave, 85. 
Graeco-Buddhist, 42, 50, 52. 

3292 11 h 

Griffiths, Mr., 96, 97, 98, 99, 100, 101, 102, 104, 105, 

106, in, 165. 
Growse, Mr. F. S., 44, 45, 46, 55, 126. 
Grunwedel,Prof., 25,58,65,67,71, 73, 161, 167, 169. 
Gujarat, 116, 157, 175, 176, 196, 203, 204, 205, 208. 
Gulbarga, 172, 176. 
Gunas y 14. 

Gunavarman, prince, 157. 
Gupta Period, 6, 13, 15, 28, 40, 58, 66, 70, 77-84, 

86, 89, 109, 114, 120, 122, 151. 
Gwalior, 190, 200. 

Hakim Muhammad Khan, painter, 229. 

Halebid, 130, 131. 

Hanuman, monkey god, 135. 

Haribans, painter, 219. 

Hariharesvara temple, 136. 

Haritiy 59, 60. 

Harsha, king, 77, 157. 

Hathipol, elephant gate, 189, 201. 

Havell, Mr. E. B., 5, 32, 49, 64, 86, 87, 98, 106, no, 

in, 125, 126, 140, 147, 159, 160, 212, 220, 221, 

223, 227, 228. 
Hazara Ramaswami temple, 135. 
Heliodora, 12. 

Hellenistic sculpture, 50-67. 
Herringham, Lady, 98, 100, 102, 107, 108. 
Hidda, 53. 

Hilal Khan Kazi mosque, 175. 
Hinayana Buddhism, 95. 
Hiuen-Tsiang, Chinese traveller, 22, 66, 162. 
Hodgson, Mr. Brian, 166, 167, 168. 
Hoysala, 130, 131, 135. 
Hoysalesvara temple, 131. 

Humayun, prince, 178, 179, 184, 186, 201, 206, 207. 
Husain Mirza, sultan, 206. 
Huvishka, king, 39, 41, 45, 66. 
Hyderabad, school of painting, 226. 

Ibrahim Adil Shah II, 177. 

Iconography, 12-14, 19, 36, 56, 58, 59, 62, 73, 86, 87, 

99, 104, 120, 122, 127, 133, 153, 158, 226. 
Idiqut-i-Shahri, 163. 
Indra, 46, 56, 57, 58, 61. 
Indrasila, cave, 46. 
Ishwari Par shad, painter, 229. 
Isurumuniya Vihara, 148. 
Itimad-ud-daula, 180, 198. 

Jagannathy 125. 

}agan[nath], painter, 219. 

Jagmandir palace, 198. 

Jahangir, emperor, 180, 188, 190, 191, 201, 203, 

211, 213, 215, 216, 220, 221, 222. 
Jahangiri Mahal, 178, 180. 
Jainism, 10, 11, 13, 17, 21, 26, 39, 43, 71, 74, 78, 79, 

83. 86, 93, 113, 116, 119, 120, 131, 175, 204. 
Jaipur, 7, 39, 203, 210, 219, 225, 226. 



Jamalgarhi, 50, 52, 55, 74. 

Jamalpur, 41. 

Jambhala, 155. 

Jami Masjid, 178, 194, 202. 

Jam Nizam-ud-din, tomb, 177. 

Jatakas, 8, 9, 25, 26, 30, 31, 33, 37, 46, 58, 70, 86, 99, 

Jaunpur, 174, 175, 178. 
Java, 66, 85, 120, 142, 155-60. 
Jaya, 169. 
Jayavijaya cave, 38. 
Jetavana monastery, 31. 
Jetawanarama dagaba, Ceylon, 24, 142, 147. 
Jewellery, 38, 41, 43, 44, 86, 140, 147, 204, 225. 
Jogimara cave, 93. 

Kabul, 50, 52, 55, 63, 67, 171, 178, 206. 

Kabul Bagh, 178. 

Kadam Rasul mosque, 192. 

Kadphises I and II, 39. 

Kadwar temple, 119. 

Kahaon, 83. 

Kailasa, 87, 88, 91, 130, 196. 

Kailasanatha, 129. 

Kali, 12, 87, 88. 

Kali Bening temple, 158. 

Kalidasa, poet, 78, 92, 229. 

Kalpa Sutras, 204. 

Kalyan Das, painter, 222. 

Kanara temple, 119. 

Kanauj, 126. 

Kanchi (Conjeeveram), 129, 133. 

Kangra paintings, 225, 226. 

Kanha, 219. 

Kanheri, cave temple, 27, 38. 

Kanishka, king, 22, 39, 40, 52, 62, 63, 67. 

Kankali mound, 41. 

Kan-su, 163. 

Kanthaka, 62. 

Kantonagar, 117. 

Kapila relief, 148. 

Kapilavastu, 61, 73. 

Karan, son of Rama of Chitor, 190. 

Karli, cave temple, 27, 38. 

Kashi (tile-work), 201, 202. 

Kashmir, 55, 118, 119. 

Kasia, 85, 124. 

Kasyapa I, king, 110-11. 

Kasyapa V, king, 145. 

Kasyapa of Uruvilva, 24, 37. 

Katas temple, 119. 

Kausambi, 83. 

Kedaresvara temple 131. 

Kesava (Kesu), painter, 219, 220. 

Khajuraho temple, 6, 13, 113, 116, 117, 118, 121, 

126, 128. 
Khandagiri, 38. 
Khandesh, 179. 

Kharoshthiy 30. 

Khemkaran, 219. 

Khotan, 52, 57, 62, 67, 162-4. 

Khuldabad, 186. 

Kinnara, 33. 

Kirti Nissanka Malla, king, 142. 

Konarak, temple, 115, 124, 125, 126, 128. 

Konch, temple, 117. 

Kondane cave temple, 27, 38. 

Korea, 66, 164. 

Korwas, 9. 

Krishna Deva Raya, king, 135, 140. 

Kshattriya race, 11. 

Kukargrama, 45. 

Kumaragupta, king, 80, 83. 

Kurkihar, 122, 124. 

Kuruvatti, 136. 

Kushan, 11, 12, 24, 27, 38, 39~49» 55. 6 3> 6 4» 66 » 6 7> 

73» 74> 77> 78. 81. 
Kutagara-Vthara, 25. 
Kutb Minar, 69, 171, 173. 
Kutb Mosque, 171, 173, 174. 
Kutb-ud-din Ibak, 171, 172, 173. 
Kutb-ud-din of Ush, 173. 
Kuvera, 7, 8, 29, 31, 45, 59, 128, 155. 

Laccadive, 134. 

Lahore, 59, 60, 66, 180, 186, 189, 199, 201, 202, 216, 

Lakshmi, 150. 
Lai, painter, 219. 
Lalitaditya, king, 118. 
Lama, 162, 163, 165, 167, 168, 169. 
Lankarama, dagaba, Ceylon, 144, 145. 
Lankesvara, 88. 
Lattices, 196-7. 

Lauriya-Nandangarh, pillar, 17, 18, 19. 
Le Bon, Mons., 76, 126, 171. 
Le Coq, Dr. v., 161, 163, 164. 
Leitner, D., 50, 51, 62. 
LeVi, Mons. S., 167. 
Lhasa (Tibet), 169. 
Licchavis, 26. 

Lithang (Tibet), 169. 
Lomas Rishi cave, 27. 
Lonasobhika tablet, 40. 
Loriyan Tangai monastery, 58, 65. 
Lot an, 200. 
Lucknow, 186. 
Lumbini garden, 18, 61. 

Macedonia, 15. 

Mackenzie, Col., 47. 

Mackenzie, Mr. R. D., 191. 

Madho, painter, 215, 219. 

Madras, 47, 129, 135, 137, 152, 154, 226. 

Madura, 130, 138, 139, 151, 227. 


2 35 

Magadha, 15, 122. 

Magala, 132. 

Mahabharata, 12, 133. 

Mahadeva, 88. 

Mahafiz Khan, 176. 

Mahamalla, king Narasimha-varman I, 133. 

Maharatta, 41. 

Mahasena, king, 148. 

Mahavira, 11. 

Mahayana Buddhism, 56, 95, 120, 121, 123, 143, 

155, 158, 159, 160. 
Mahayogi, 80. 

Mahendra-varman I, king, 133. 
Mahesh, painter, 219. 
Mahishasura, 133. 
Mahmud of Ghazni, 171. 
Maiscy, Mr., 35. 
Majapahit kingdom, 158. 
Makamat of Hariri MS., 205. 
Makara, 33, 4i>79> x 3 6 - 
Maldive, 134. 
Malikarjuna temple, 136. 
Mallitamma, sculptor, 131. 
Malot, temple, 119. 
Malwa kingdom, 175, 223. 
Mamallapuram, 129, 133. 
Mandu, 175, 190, 198. 
Mangalesvara, king, 87. 
Manichean, 163, 164. 
Manjusri, 160, 170. 
Mankuwar, 82. 
Manohar, painter, 222. 
Manrique, Father Sebastian, 183-5, 2II# 
Mansur, painter, 220, 221. 
Mara, 57. 

Maratha Period, 212. 
Margasahayar, 137. 
Marichi, 120, 123. 
Marshall, Sir John, 19, 34, 69, 78, 84, 123, 125, 189, 

191, 197, 198, 200, 202. 
Martanda, 118. 
Mata Kunwar, 124. 
Mathura, 24, 27, 29, 33, 37, 39-46, 49, 50, 53, 55, 

57. 58. 7i» 72, 73. 74. 75. 8l » 82 > 86 » I0 9. "6, 128, 

138, 186. 
Maurya, Chandragupta, king, 15. 
Mauryan Period, 15-20, 27, 29, 34, 63, 69. 
Maya, mother of Buddha, 61, 71. 
Medirigiriya, 144. 
Mehta, 203. 

Migeon, Mons. G., 188, 193, 199, 210. 
Mihintale, 147, 150. 
Minneriya, 148. 

Mir Abdul Karim, Taj builder, 182. 
Mir Hashim, painter, 222. 
Mir Mansur of Badakshan, 206. 
Mirpur Khas, 117. 
Mir Sayyid AH, 206, 207, 219. 

Miskin, painter, 215, 216, 219, 221. 

Miza Jani Beg mosque, 202. 

Mokalji, king, 127, 128. 

Mongol, 66, 205, 206. 

Mosaics, 197-9. 

Mother goddesses, 7, 9. 

Moti Masjid, 180, 181. 

Mudra, 81. 

Mughal, 162, 163, 175-80, 186, 189, 192, 193, 196-8, 

200, 203-5, 2 °7> 2o8 » 2IO » 2l8 > 22 4> 22 5» 22 9- 
Muhammad Adil Shah, 177. 
Mudammad Amin mosque, 202. 
Muhammad bin Tughlak, sultan, 176, 188. 
Muhammad Fakirullah Khan, painter, 220, 222, 223. 
Muhammad of Ghor, 171, 188. 
Muhammad Husain of Kashmir, painter, 208. 
Muhammad Nadir of Samarkand, painter, 222. 
Mukramat Khan, Taj builder, 182. 
Muktesvara shrine, 115, 124, 129. 
Mukund, painter, 219. 
Multan, 199, 202, 207. 
Mumtaz Mahal, 181. 
Murray, Mr. A., no. 
Musical instruments, 128. 
Mysore, 130, 131, 132, 134, 135, 139, 140, 196. 

Nachna-Kuthara, 80. 

Nadir-uz-zaman (Abul Hasan), painter, 223. 

Naga, 7, 8, 31, 33, 45, 46, 61, 104, 119, 147, 162. 

Nalanda, 121. 

Nanda Lai Bose, painter, 229. 

Nandangarh, 23. 

Narasimha, king, 115. 

Narasimha-varman, king, 129, 133. 

Narasimha (Avatara), 135. 
j Narayana, 12. 

Narbada, 132. 

Nasik, cave temple, 27. 

Nataraja {Siva), 6, 140, 152, 155. 

Nathu, 75. 
I Nats, 73. 
j Nawab Safdar Jang, tomb, 186. 

Nawab Sardar Khan, tomb and mosque, 176. 

Nawab- Vazir of Oudh, 211. 

Nell, Dr. Andreas, 152. 

Nepal, 119, 161, 165-70 

Nestorian, tomb and mosque, 163. 

Nikawaewa, monastery, 151. 

Nilagiri, elephant, 38. 

Nissanka Malla, king, 144, 146, 150. 

Nuggehalli, 130. 

Nurjahan, empress, 180. 

Oldenberg, Prof., 39. 

Oraons, 9. 

Orissa, 92, 99, 121, 122, 124, 126. 

Osia, 45, 117, 128. 

Oxus, 206. 

236 INDEX 

Padmapani, 155, 167. 

Pahari, Himalayan school of painting, 225. 

Painting, 92-112, 162-8, 203-29. 

Pala dynasty, 121, 122, 123. 

Pali Khera, 44, 45. 

Pallava dynasty, 77, 95, 129, 130, 133, 134, 140, 

Panchasika, 58. 
Panch- Mahal, 180. 
Pandua, 175. 

Pandya dynasty, 133, 151. 
Panipat, 178. 
Pankuliya Vihare, 150. 
Parakrama Bahu, the Great, king, 142, 144, 149, 

Parasuramesvara, 124. 

Parkham, 29. 

Parvati, 12, 88, 89, 141. 

Pataliputra (Patna), 15, 25, 77. 

Pathari, 84. 

Patna, 15, 55, 77, 229. 

Pattadkal, 28, 130, 196. 

Pattini Devi, goddess, 151. 

Payer shrine, 119. 

Persepolitan, 18, 30, 41, 55, 84. 

Peshawar, 22, 50, 51, 52, 54, 59. 

Phear, Sir J. B., 54. 

Philosophy, 9, 14. 

Pidurangala, temple, 11 1. 

Pietra dura inlay, 180, 181, 184, 197-9. 

Pillar Edicts, 17, 19, 20. 

Piprahwa, 23. 

Pitalkhora, cave temple, 27, 38. 

Political organization, 26. 

Polonnaruwa, 141, 142, 144, 145, 149, 150, 152, 


Pottery, 17. 

Pradakshina, 159. 

Prambanam, 158, 160. 

Pulakesin II, king, 77, 95, 104, 157. 

Puliigoda galkanda (Tamankaduwa caves), 112. 

Puri, 139. 

Ragtnala, 203. 

Rai Anup, 222. 

Rajagriha, 37. 

Raja Man Singh, 200, 221. 

Rajaraja, king, 129, 134. 

Raja Raja-varma, 227. 

Rajarani, temple, 115, 124. 

Rajasimha, king, 129, 133. 

Rajasthani painting, 225, 226. 

Rajendra-Choladeva, king, 129, 134. 

Rajgir, 80, 123. 

Rajmahal, 123. 

Rajput, 121, 171, 175, 203, 225, 226, 229. 

Ram, 219. 

Ramaswamy Naidu, 227. 

Ramayana, 92, 131, 136, 148. 
Ramesvaram, 130, 139. 
Ramgarh Hill, 93, 99. 
Rampurva, 18, 20. 
Rana of Chitor, 190. 
Rani Gumpha, 38. 
Rasa, 14. 

Rashtrakuta dynasty, 130. 
Ratha, 129, 133. 
Ratnasura, 88. 
Ravana, 87, 120. 
Ravi-varma, 140, 227. 
Razmnatnah MS., 210, 219. 
Rock Edicts, 20. 
Rudra, 12, 153. 
Rukmini, 140. 
Rukn-ud-din, 200. 
Rummindei pillar, 17, 18, 61. 
Rupbas, 39. 
Ruwanweli, 92, 112, 147. 

Safavid, 205-7, 215. 

Sahasram (Sasaram), 20, 177. 

Sahet-Mahet, 54. 

Sahri-Bahlol, 52, 59. 

St. Hilairc, Mons. B., 166, 167. 

Saivism, 11, 79, 88. 

Sakra, 56, 57, 58, 61. 

Sakya clan, 11. 

Sakyamuni) 53. 

Salem, 151. 

Salim Chishti, tomb, 197, 199. 

Samarqand, 199, 206, 207. 

Sambhal, 178. 

Samudragupta, king, 78, 83. 

Sanchi, 7, 11, 13, 16, 19, 21, 22, 23, 24, 25, 29, 32, 

33» 34. 36, 37. 38, 39. 4°. 4*. 43. 4^, 48, 53. 6l . 

66, 68, 69, 70, 71, 73, 74, 75, 80, 84, 93, 95, 101, 

120, 143, 146, 147. 
Sandrokottos (Chandragupta), 15. 
Sangha, Buddhist, 26. 
Sangharama, Buddhist monastery, 143. 
Sankhya, 9, 10, 14. 
Sankisa, 18. 
Sanwlan, painter, 219. 
Sapta Kannigai, 9. 
Sarangpur, mosque, 196. 
Sarasvati, 160, 170. 
Sardula, conventional lion, 131. 
Sari, temples, 158. 
Sariputra, 23. 
Sarnath, 18, 19, 35, 39, 40. 4*. 55. 6 9» 73. 78, 81, 

82, 122, 123. 
Sarwan, painter, 215. 
Sasaram (Sahasram), 20, 177. 
Sastra, 5, 6, 28, 56. 
Sat Mahal Prasadaya, 145. 
Satrunjaya temple, 11. 



Sayyid Salar, 186. 

Sculpture, 16, 19, 22, 23, 26, 27, 28-38, 39, 40, 42, 

46, 48, 49, 50-67, 69, 70, 71, 73, 74, 75, 78, 80, 85, 

86,87,91, 104, 109, 115, 120-8, 130, 132-41,142- 

50, 189-92. 
Seleukos Nikator, 15. 
Seven Pagodas, 129. 
Shah Abbas, 206. 
Shahdara, 202. 
Shah Ismail, 206. 
Shahjahan, 179, 180-5, ^9, 191, 192, 193, 197, 

198, 199, 201, 211, 212, 213, 217, 220, 221, 222, 

Shah Tahmasp, 206. 
Shankar, painter, 215. 
Sharf-un-nissa tomb, 202. 
Sharki, 174. 

Sher Shah, 177, 201, 206. 
Siddhartha, 53, 62, 73. 
Sidi Sayyad, mosque, 196. 
Sigiriya, Ceylon, 78, 96, 99, 109, in, 225. 
Sikandra, 180, 193. 
Sikri, monastery, 59, 60. 
Silenus, 44, 50. 
Silpa Sastra, 28, 56. 
Simpson, Mr., 55. 
Simhala Avadana, 104. 
Sind, 202, 211. 
Sirima Devata, 7. 
Siva, 12, 59, 79, 80, 88, 89, 95, 120, 134, 139, 140, 

144, 152, 153, 154, 158. 
Sivali, queen, 23. 
Skandagupta, king, 80, 83. 
Soma, 8. 

Somnathpur, 131. 
Sondani (Songni), 84. 
Sramana, 53. 
Sravasti, 31, 54, 58. 
Sringadhara of Marwar, 165. 
Stacy, Col., 44, 50. 

Stein, Sir Aurel, 62, 67, 124, 161, 163, 165. 
Strzygowski, 72-5. 

Stupa, 8 t ii, 2i, 25, 27, 30, 34, 38, 40, 41, 46-48, 
5°. S5» 6 9. 74» 82, 90, 113, 117, 142, 158, 
Sudarsana, yakshi, 32. 
Sultanganj, 82, 83. 
Sultania, 206. 
Sundara-tnurti Stvami, 154. 
Sunga dynasty, 21, 30, 39, 63. 
Suparna, 46. 
Sur dynasty, 177. 
Surashtra, king, 82. 
Surendra Nath Gangooly, painter, 229. 
Suryanarayanaswami, temple, 132. 
Sutrapada, temple, 119. 
Svetambara sect, 11. 
Swat, 52, 58. 

Symbols, 18, 19, 36. 
Syria, 72. 

Tabriz, 206. 

Tadpatri (Tarpatri), 137. 

Tagara (Ter), 28. 

Tajganj, 181. 

Taj Mahal, 177, 180-5, *97> J 9^- 

Takht-i-Bhai, 52. 

Taklamakan desert, 161. 

Tamankaduwa, 112. 

Tandava, dance, 88, 153. 

Tanjore, temple, 113, 129, 134, 141, 154. 

Tantipara, mosque, 200. 

Tantrimalai, 148. 

Tara, 123, 170. 

Tara, painter, 219. 

Taranath, 92, 165, 168. 

Tarpatri (Tadpatri), 137. 

Tatta, 177, 202. 

Tavernier, 182. 

Taxila, 12, 54, 55, 68. 

Ter (Tagara), 28. 

Thuparama, 142, 145, 147. 

Tibet, 161, 165-70. 

Tigowa, 79. 

Tiles, 199-202. 

Timur, house of, 206, 215. 

Tinncvelly, 130. 

Tirumalai, 140. 

Tirumal Naik, king, 130, 138, 139. 

Tirupati, 140. 

Tiruvarur, 154. 

Tissandicr, 158, 159. 

Tivanka Vihare, 144. 

Toluvila, 145, 148. 

Topaveva, 149. 

Torana, beam of gateway, 22, 24, 33, 136. 

Toyoq, 164. 

Tree-spirits, 8. 

Trichinopoly, 129, 133, 134, 138, 226. 

Trimurti, 90. 

Tsiamdo, 169. 

Tughlak dynasty, 174, 175, 176, 178, 188. 

Tughlak, shah, 197. 

Tungabhadra river, 131. 

Turan, 208. 

Udaipur, 198, 199. 
Udaiyarpalaiyam, 138. 
Udayagiri, 13, 38, 79. 
Ukhtomskij, prince E, 167, 169. 
Utna, 12. 

Upanishads, 7, 8, 9, 11, 12, 65. 
Uruvilva, 24, 37. 
Ushnisha, 42. 
Ustad Isa, 183, 185. 


Vaikuntha-Perumaly 129. 
Vaisali, 25. 

Vaishnavism, 11, 12,87. 
Vaisravana, 59. 
Vajrapani, 123. 
Vakataka dynasty, 90, 95. 
Vardhamana, 10. 
Vasanta Vilasa MS., 204. 
Vasantgarh, 128. 
Vasishka, king, 39. 
Vasudeva, 12. 
Vasudeva I, king, 39, 66. 
Vedanta, 9, 65. 
Vedasy 7, 8, 9, 12, 68. 
Vedisa, 16. 

Venkata-ramanasvami, 137. 
Veroneo, Geronimo, 183, 184. 
Vihara, 25, 79, 87, 94, 143, 148, 150. 
Vijayanagar, 135, 137, 140. 
Vijayarama, 143. 
Vikramaditya, king, 77, 82, 83. 
Virinchipuram, temple, 137. 
Virulhaka, 8. 
Virupakkha, 8. 


Vishnu, 12, 13, 80, 83, 88, 118, 124, 125, 128, 130, 

131, 136, 139, 143, 148, 153, 158. 
Visvakarma cave, 90. 
Visvanatha Sahitya Darpana, 14. 
Vitthalaswami, temple, 136. 
Vogel, Dr., 41,45, 57, 73, 201. 

Waqiat-i-Babari MS., 218, 220. 
*Wata-da-ge* shrines, 145. 
Watt, Sir George, 75, 105. 
Wei-chi I-song, 67, 164. 

Yaksha, 8, 25, 29, 31, 37, 41, 42, 45, 57, 59, 73. 229. 

Yakshini(Yak$hi), 8, 31, 32, 71, 73, 79. 

Yah, 137, 138, 139. 

Yama, 8, 88. 

Yamuna, 8, 79. 

Yarkand, 162. 

Yar-Khoto, 163, 164. 

Yasodharman, 84. 

Yoga (Yogi), 12, 42, 48, 57, 58, 65, 80, 120, 160. 

Zakariya Khan, mosque, 202. 
Zaman, Muhammad, 217.