Temples of the Indus
Mar 18, 2023
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
untitledVOLUME 35
Ancient Pakistan
Cover Illustration: Amb, Punjab, Temple A, northwest. Photo by author.
This book is printed on acid-free paper.
Library of Congress Cataloging-in-Publication Data
Meister, Michael W. Temples of the Indus : studies in the Hindu architecture of ancient Pakistan / by Michael W. Meister. p. cm. — (Brill’s Indological library ; v. 35) Includes bibliographical references and index. ISBN 978-90-04-18617-0 (hardback : alk. paper) 1. Hindu architecture— Pakistan—Salt Range Region. 2. Hindu architecture—Indus River Valley. 3. Hindu temples—Pakistan—Salt Range Region. 4. Hindu temples—Indus River Valley. I. Title. II. Title: Studies in the Hindu architecture of ancient Pakistan. III. Series.
NA6010.72.S25M45 2010 726’.1450934—dc22
2010018109
Copyright 2010 by Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Hotei Publishing, IDC Publishers, Martinus Nijhoff Publishers and VSP.
All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher.
Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change.
printed in the netherlands
Chapter One Salt Range and Indus Temples .......................... 1
Chapter Two Sites, History, and Comparative Chronology ... 11
Chapter Three Archaeology at Kfirkot and Problems of Platform Extensions ............................................ 39
Chapter Four Original Variations in Tenth-Century Architecture .......................................................... 51
Chapter Five Archaeology and Ethnography .......................... 63
Bibliography ........................................................................................ 73
Index .................................................................................................... 83
Figures
ACKNOWLEDGMENTS
To the Universities of Pennsylvania and Peshawar, the Department of Archaeology and Museums, Government of Pakistan, Professors Abdur Rehman and Farid Khan and the Pakistan Heritage Society, the American Institute of Pakistan Studies, Islamabad, to all participants in the integrated hi temple project; to draftspersons, surveyors, and excavators who contributed their skills; F. Masih, representative of the Department of Archaeology and Museums; Alka Patel, who proffered her camera and a long lens at an important site; to Dr. Elizabeth Stone and two annonymous readers for helpful suggestions about revision; and to Kamal Khan Mumtaz, whose project with young architects and seminar for the Anjuman Mimaran in 1989 was an important inspira- tion, I offer my thanks.
Michael W. Meister Philadelphia 01 January 2010
LIST OF ILLUSTRATIONS
All figures are by the author unless otherwise credited. 1. Maps: a) ‘Map of the Panjab’ (revised from Cunningham 1875,
pl. 1); b) sites in Pakistan’s North-West Frontier and Punjab Provinces.
2. Sirkap (Taxila), Punjab, ‘shrine of the double-headed eagle’ façade, ca. second/third century, detail of architectural models.
3. Taxila, Dharmarjik stpa, outermost kañjr-stone encasement, ca. fifth century.
4. Pandrethan, Kashmir, iva Temple (courtesy: American Institute of Indian Studies, Gurgaon).
5. Kats, Punjab, site from southwest. 6. Kats, Temple B (southwest sub-shrine), southwest. 7. Kats, Temple B, south. 8. Kats, Temple B, reconstruction (from a drawing by P. George). 9. Kfirkot, North-West Frontier Province, site looking east toward
the Mri and Indus River. 10. Kfirkot, site looking west over Temples D, B, and A toward the
citadel. 11. Kfirkot, Temple B, south. 12. Kfirkot, Temple B from east (before clearance and excavation). 13. Kfirkot, Temple A, west. 14. Kfirkot, Temple A from east (after clearance and excavation). 15. Bilot, N.W.F.P., site views of: a) Temples D, E, F–G; b) Temples C
and B (D is behind); c) Temple A. 16. Bilot, Temple D, south. 17. Kfirkot, Temple C, west. 18. Kfirkot, Temple C, detail of north wall and ikhara’s jla pattern. 19. Kfirkot, Temple C, southwest. Note central offset in the wall and
superstructure. 20. Bilot, Temple A, south. Note corner niches. 21. Kfirkot, Kañjar Koth (from Stein 1905). This temple is no longer
standing. For its location, see Hargreaves’ site map (fig. 63). 22. Bilot, Temple D, detail of south wall and ikhara. Note shrine
model on the wall. 23. Bilot, Temple E (NE sub-shrine of Temple D), east.
x list of illustrations
24. Bilot, Temple C, south. Note chambers in foundation platform. 25. Bilot, Temple B, southwest. 26. Mri-Indus, Punjab, Temple A, west. 27. Mri-Indus, Temple B, west. 28. Klar (Sasu-da-Kalra), Punjab, brick temple, southwest. 29. Klar, brick temple, south wall. 30. Kfirkot, Temple D, southeast. 31. Kfirkot, Temple D, detail of ikhara ornament. Note malakas
embedded in the corner web pattern. 32. Kfirkot, Temple B, transition from square sanctum to dome. 33. Kfirkot, Temple A, squinches as transition from sanctum to
dome. 34. Kfirkot Temple A, squinches and dome. 35. Kfirot, Temple D, squinches and dome. 36. Kfirkot, Temple A, sanctum and dome. 37. Kfirkot, Temple A, dome. 38. Klar, brick temple, squinch with wooden beam above. 39. Kfirkot, Temple A, squinch made of kañjr stone. 40. Bilot, Temple D, southeast platform chamber, dome. 41. Nandana, Punjab, Temple A, two-storeyed interior with domed
chambers. Note the upper chamber’s ambulatory corridor. 42. Mri-Indus, Temple B, south. 43. Mri-Indus, Temple A, south. 44. Mri-Indus, Temple A, southeast. 45. Mri-Indus, Temple A, entry vault (partly corbelled, but note the
use of voussoirs and keystone). 46. Mri-Indus, Temple B, entry hall: a) doorway with trilobed arch;
b) dome and squinches seen through arch. 47. Amb, Temple A, sanctum doorway. The image over the door has
been removed since this photograph was taken. 48. Amb, Punjab, Temple A, west. 49. Amb, Temple B, south. 50. Amb, Temple B, south wall, central offset, niche with cinquefoil
arch. 51. Nandana, Temple A, detail of the lower level of the superstructure
with shrine models. 52. Nandana, Temple A, east. 53. Formative shrines in Ngara temple-development: a) Srnth,
Uttar Pradesh, India, shrine model on Gupta-period lintel (Sarnath Museum); b) Kfirkot, Temple B, superstructure, east; c) Deogarh,
list of illustrations xi
Uttar Pradesh, ‘Gupta’ Temple, doorframe, shrine model with malakas; d) Nland, Bihar, India, shrine model on Buddhist votive stpa (courtesy: AIIS, Gurgaon).
54. Kfirkot, Temples B and A, schematic elevations without surface ornament.
55. Bilot, Temple D, south wall, shrine model. 56. Deogarh, ‘Gupta’ Temple, author’s reconstruction of the first level
of the superstructure. 57. Kfirkot, Temple A, ikhara, east. 58. Dhnk, Saurashtra, India, Srya Temple, ikhara, south (courtesy:
AIIS, Gurgaon). 59. Kfirkot, Temple A, ikhara, south. Note small pillarets preserved
at the second level. 60. Bilot, Temple D, detail of southwest corner of ikhara. Note pil-
laret next to malaka. 61. Bhnasar, Saurashtra, India, Temple No. 1, west (courtesy: AIIS,
Gurgaon). 62. Evolution of jla patterns: Bilot Temples D (top), A (center), and
C (bottom). 63. Kfirkot, site map marked ‘M. Sharif Del.’ in the lower left and
hand signed by H. Hargreaves, who first published it, in the lower right (from ASIAR 1921–22, pl. 26).
64. Kfirkot and Bilot forts, site plans (revised from ASIAR 1921–22, pl. 26, and ASI(FC) AR 1920–21, appendix).
65. Kfirkot, Temple C, excavated compound, ground plan, two phases (modified from archaeological field drawing, courtesy: A. Rehman).
66. Kfirkot, Temples E, A, and B site plan following excavation (modified from archaeological field drawing, courtesy: A. Reh- man); sectional elevations of A and B.
67. Evolution of ground plans: Kfirkot Temples B, A, C; Bilot Temple A.
68. Bilot, Temples D and E, ground plans (modified from drawing by F. Iqbal, courtesy: Pakistan Heritage Society).
69. Bilot, Temple A, ground plan (modified from drawing by F. Iqbal, courtesy: Pakistan Heritage Society).
70. Mri-Indus, Temples A and B, ground plans (B modified from drawing by F. Iqbal, courtesy: Pakistan Heritage Society).
71. Kfirkot, Temple C, ground plan with second-phase additions (modified from archaeological field drawing, courtesy: A. Rehman).
xii list of illustrations
72. Kfirkot, Temple C, east, after first phase of excavation. 73. Kfirkot, Temple C, pilaster exposed on northeast face of kañjr
platform compared with an early ‘Asiatic-Corinthian’ capital from Ai-Khanum (drawing by P. Bernard and M. Le Berre, from Wheeler 1968, p. 77). For location, see fig. 71, ‘b’.
74. Kfirkot, Temple C, inserted limestone pilaster base, northeast juncture, prsda and antarla. For location, see fig. 71, ‘c’.
75. Kfirkot, two-storeyed Mri structure, east face (the Mri is built of kañjr tuffa stone, the stairs are limestone).
76. Kfirkot, Temple E, east, after excavation (courtesy: F. Khan). 77. Kfirkot, view of fort, east, from citadel, before excavation. 78. Kfirkot, Temples E, A, and B site after excavation (the century-
old wall and footpath were moved forward to make excavation possible).
79. Kfirkot, Temple E during excavation (courtesy: F. Khan). Insert shows limestone extension before clearance.
80. Kfirkot, Temple E, kañjr platform’s façade, northeast chamber (courtesy: F. Khan).
81. Kfirkot, Temple E, southeast corner after removal of second-phase construction (courtesy: F. Khan). Note the first-phase kañjr stair- way exposed under the limestone stairs to right).
82. Kfirkot, Temple E, kañjr platform’s façade (modified from archaeological field drawing, courtesy: A. Rehman).
83. Kfirkot, Temple E, kañjr platform, façade, northeast sub-shrine (courtesy: F. Khan).
84. Kfirkot, Temples E, A, and B, section of site. 85. Kfirkot, Temple E, compound, remains of east entry and colon-
naded cloister. 86. Kfirkot, Temple E, remains of cloister built against foundations
of Temple B. 87. Kfirkot, Temple A, east, extended stair and ground-level chamber
after excavation. 88. Kfirkot, Temples E, A, and B during excavation, south (courtesy:
F. Khan). 89. Kfirkot, Temple B, remains of platform stairway and side cham-
bers after excavation. 90. Kfirkot, sculpture of an ascetic, excaved in front of Temple B. 91. Kfirkot, sculpture of an ascetic, detail: three heads, ‘skamba’ (the
‘pillow’-like shaft behind).
list of illustrations xiii
92. Mathur, lintel (Government Museum, Mathura), 1st century B.C.E., with a phallic linga, tree, and square railing at the center and a ‘jambu-dvaja’ held to left (courtesy: D. M. Srinivasan 1984: pl. 20).
93. Kfirkot sculpture, showing position of skambha column. 94. Mathur, relief with four auspicious signs from an astamangala
set, ca. second c. C.E. (from Coomaraswamy 1931, pl. 31, fig. 2). 95. Bhrhut, Mathur, and Amarvati reliefs, first centuries B.C.E.
and C.E., differentiating lotus and jackfruit; and the astamangala sign I interpret as the auspicious, upward-growing jackfruit (images compiled from Coomaraswamy 1931; Agrawala 1965).
96. Kfirkot figure interpreted as ‘yoga-mandala’. 97. Compounds: Bilot and Kfirkot. 98. Platform extensions: Bilot and Kfirkot (scales adjusted). 99. Bilot, Temple H, south. 100. Bilot, Temple D, platform, SE corner extension and Temple E,
south, Temple F is in the background. 101. Bilot, Temple D, platform, SE corner with east-facing chamber. 102. Bilot, Temple D, platform, SE corner, vaulted entry to domed
chamber. 103. Bilot, SE corner of Temple D’s compound, conjoined Temples
F–G, south. 104. Bilot, Temples B–C, ground plans (modified from drawing by F.
Iqbal, courtesy: Pakistan Heritage Society). 105. Bilot, Temple B, southwest. 106. Bilot, Temple C, south wall. 107. Salt Range typologies: platform and wall mouldings. 108. Amb, Punjab, Temples A and B, site plan; remains of Temple B’s
platform. 109. Amb, site, Temples B and A, north. 110. Amb, Temples A and B, ground plans. 111. Nandana, Punjab, Temple A, axonometric section, elevation, and
ground- and first-storey ground plans (drawing by H. Choudhury). 112. Nandana, site for south. 113. Pattan Munra, Punjab, ‘minr’ from south. 114. Pattan Munra, brick temple, south. Note remnants of dome. 115. Pattan Munra, brick temple, sanctum, southwest. 116. Mri-Indus, Temple B, entry hall, trefoil arch over doorway,
east.
xiv list of illustrations
117. Pattan Munra, brick temple, south wall, vaulted niche with tre- foil pediment.
118. Salt Range vs. Western Indian moulding typologies. 119. Pattan Munra, brick temple, sanctum, plan reconstructed with
balconies and circumambulatory corridor. 120. Ambulatory plans in Western India: a) Osiñ, Rajasthan, Mahvra
Temple, ca. 800; b) Chittor, Rajasthan, Klikmt Temple, ca. 700. Note different configurations of projecting balconies.
121. Pattan Munra, brick temple (right), detail of damaged tower before conservation (after Vats 1929: pl. 34d); Gyaraspur, Mad- hya Pradesh, Mldev Temple, ca. 950 (left), detail of multi- spired tower (courtesy: AIIS, Gurgaon).
122. Malot, Punjab, temple, south. 123. Malot temple, entry, east. 124. Malot temple, entry gate. 125. Malot temple sanctum, stepped squinch. 126. Malot temple, Ngara shrine model over south-wall chamber. 127. Malot temple, entry gate, peak-roofed shrine model. 128. Malot, ground plans: sanctum and gateway. 129. Malot temple, south wall, monumental chamber. 130. Malot temple, north-wall chamber, images over entry. 131. Mri-Indus site, Temples A and B looking east. 132. Mri-Indus, Temple C, northwest. 133. Mri-Indus, Temple C, reconstruction of ground plan. 134. Mri-Indus, Temple C, north wall. 135. Mri-Indus, Temple C, north wall, bi-level pediment above tre-
foil niche. 136. Mri-Indus, Temple C, superstructure, northwest corner, pillaret
with latina (Ngara) model. 137. Mri-Indus, Temple C, north wall, pediment with Srya image. 138. Saddan (Muzaffargarh), Punjab, cut-brick tomb of Shaikh Sadan
Shahid. 139. Khatti Chor, Punjab, ribt of Khalid Walid, west corridor, cut-
brick mihrb. Note trefoil candral-patterned arch on lower wall.
140. Gumbat, Swt, ruined temple (left); Karwn-balasi, near Kuzai Gumbhaz, Little Pamir, Afghanistan (after Stein 1921: fig. 3).
141. ‘Bhmi-prsda’ storeyed shrines: (left) Kats, Temple B recon- struction; (right) Aihole, Karnataka, Galagnth Temple, 7th cen- tury (courtesy AIIS, Gurgaon).
list of illustrations xv
142. Bodhgay, Bihar, brick Mahbodhi Temple: (left) Kumrahr (Patna), terra-cotta plaque, ca. 3rd century (courtesy: AIIS, Gur- gaon); (right) detail of tower after conservation.
143. Bhtargon, brick temple, ca. 450: (left) ground plan; (right) 19th-century photograph of temple from the northeast (after Cunningham 1880: pl. XIV–XV).
144. Reconstructions of ‘proto-Ngara’ superstructures: (left) Deog- arh, ‘Gupta’ Temple, ca. 525 (ground storey after Vats 1952); (right) Bilot, Temple D, ca. 600.
145. Saurashtra, Gujarat, Maitraka-period superstructures: (left) Bilevara, Bilvantha Temple, ca. 600 (courtesy: AIIS, Gurgaon); Dhnk, Sun Temple, partial reconstruction of ikhara, ca. 700.
146. Amb, three sculptures found by the ASI during clearance of Tem- ple B’s plinth, ca, 1920: ‘Mahdeva’, animal plaque, Narasimha (after Sahni ASNC[H]PR 1921: pl. VII).
147. Kapa (Kabul), Afghanistan, Guldra stpa, plinth façade. 148. Kapa (Kabul), Topdra stpa, (1967 photo: Prof. S.-W. Breckle);
trefoil chamber facing dome (after Lézine 1964: fig. 53). 149. Sequence of arches in the Northwest (top to bottom): Sirkap
(Taxila), Punjab, shrine of the double-headed eagle; Kapa, Guldra stpa; Taxila, Dharmarjik stpa, outer casing; Kfirkot, N.W.F.P., Temple E; Mri-Indus, Punjab, Temple B; Pattan Munra, brick temple, Punjab.
CHAPTER ONE
SALT RANGE AND INDUS TEMPLES
Early medieval temples in the Khisor range of hills along the west bank of the Indus River in the North-West Frontier Province and those on the escarpments and plateau of the Salt Range, between the Indus and Jhelum rivers on the southern edge of the Pothohar table- land in the western Punjab, were visited by Alexander Cunningham in the mid second half of the nineteenth century and again by Aurel Stein at the beginning of the twentieth century; many were placed on a Government List of Protected Monuments in 1904 (ASN[H]PR 1918: Appendix F); and intermittent visits by officers of the Archaeological Survey, Northern (Hindu and Buddhist) and Frontier Circles, occurred through the nineteen twenties. Daya Ram Sahni ibid.: 5, for example, reported on his first visit to temples at Amb that “useful repairs have already been carried out by the Public Works Department . . . though it must be confessed the work has lacked skilled supervision.”
Early Field Reports
Cunningham 1875: 82–85 during his first field tour to the region reported “a number of old Hindu temples in the Salt Range, which all belong to the Kashmirian style of architecture, with its fluted pillars and peculiar trefoil arches”; these he associated with the “last dynasty of the Hindu kings of Gandhra.” Referring specifically to temples at the sites of Amb, Malot, and Kats, he drew the conclusion that these “temples of the Salt Range may be assigned with much probability to the latter half of the 9th century,” in part on the basis of their use of “spiral twisted pillars” not found on most of the temples under discus- sion. Six years later, at his first opportunity to visit the fort of northern Kfirkot, above the west bank of the Indus south of its intersection with the Kurram River, Cunningham 1882: 27 in part revised his ear- lier opinion. Noting that no trefoil arches were found there, he related the architecture of these temples to a period “which succeeded the semi-Greek architecture of the Indo-Sythians.”
2 chapter one
When he visited Kfirkot in 1903, Stein 1905: 10–17 was able to map and carry out an extended survey, discovering a further temple at the foot of the fort “at the point where the narrow boulder-filled ravine which skirts the north-west face of the site debouches into the bed of the Indus.” He observed that the location of this fortress guarded important trade routes from the Indus and the Punjab up the Kurram Valley to Bannu and south along the Indus to Sind. “These little tem- ples,” he wrote, possess “distinct architectural interest as represent- ing a type which meets us in the temples of several ancient sites of the Salt Range, but there with additions characteristic of Kashmirian style which are conspicuously absent at Kafirkot.” Of his newly located temple, Kañjar Koth, however, he added that the “appearance of a rudimentary trefoil arch evolved out of the ‘beehive’ ornament [the jla web of a Ngara tower] may be an indication of Kashmirian influ- ence which in the Salt Range tracts on the opposite side of the Indus seems to have obtained a hold since the seventh century.” Stein would change his mind at a later point, but had been studying Cunningham’s Reports.
In an imprecise repetition of the conflation of Salt Range temples with those of Kashmir Cunningham began, David Ross 1883: 153, in The Land of Five Rivers and Sindh, commented on the pilgrimage site of Kats: “The architecture is the same as in Kashmir, beautiful fluted pillars, trefoil arches, dentils, and pointed roofs.” This is much closer to an appropriate description of Malot than of temple remains at Kats, Kfirkot, and Bilot. In The Ancient Monuments, Temples and Sculptures of India, James Burgess 1897: 28 chose to illustrate, of all the temples in the Salt Range, only the one at Malot. He remarked that the “pyramidal sikhara or spire has long ago disappeared,” but made no mention of the curvilinear North Indian Ngara shrine models that decorate the Malot temple’s walls, although he gave good illustrations of them in his plates. He followed his illustrations of Malot with a set of Kashmir temples, then of an atypical sixth-century temple at Gop in Saurashtra that has a pyramidal roof, of which he felt the need to remark “the roof of the sikhara looks not unlike that of a Kashmiri temple” (for an analysis of the temple at Gop, see Meister/Dhaky/Deva 1988: 177–179).
Part of the problem early in the twentieth century was that few offi- cers had experience of all the monuments, and of the few they had vis- ited, it was often many years before they returned. Cunningham 1882: 34 wrote of Amb: “The temples are all of the Kashmiri style, but they
salt range and indus temples 3
are almost certainly of late date [ca. 800–950], as all the arches have cinquefoil instead of trefoil heads, which is the only form in Kashmir.” Sahni ASN(H)PR 1918: 5, when reporting on conservation of tem- ples at Amb, still followed Cunningham’s analysis: “Like the temple at Malot, these temples are constructed in a semi-Kashmiri style with cinquefoil arches instead of trefoil ones as is the case in the temples of Kashmir.” He then observed, however, that the “spires of the temple are in the usual North Indian style.”
If Cunningham in 1872–3 took the presence of a trefoil arch to be the marker of ‘Kashmiri style’ in Salt Range temples, when he…
Ancient Pakistan
Cover Illustration: Amb, Punjab, Temple A, northwest. Photo by author.
This book is printed on acid-free paper.
Library of Congress Cataloging-in-Publication Data
Meister, Michael W. Temples of the Indus : studies in the Hindu architecture of ancient Pakistan / by Michael W. Meister. p. cm. — (Brill’s Indological library ; v. 35) Includes bibliographical references and index. ISBN 978-90-04-18617-0 (hardback : alk. paper) 1. Hindu architecture— Pakistan—Salt Range Region. 2. Hindu architecture—Indus River Valley. 3. Hindu temples—Pakistan—Salt Range Region. 4. Hindu temples—Indus River Valley. I. Title. II. Title: Studies in the Hindu architecture of ancient Pakistan. III. Series.
NA6010.72.S25M45 2010 726’.1450934—dc22
2010018109
Copyright 2010 by Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Hotei Publishing, IDC Publishers, Martinus Nijhoff Publishers and VSP.
All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher.
Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change.
printed in the netherlands
Chapter One Salt Range and Indus Temples .......................... 1
Chapter Two Sites, History, and Comparative Chronology ... 11
Chapter Three Archaeology at Kfirkot and Problems of Platform Extensions ............................................ 39
Chapter Four Original Variations in Tenth-Century Architecture .......................................................... 51
Chapter Five Archaeology and Ethnography .......................... 63
Bibliography ........................................................................................ 73
Index .................................................................................................... 83
Figures
ACKNOWLEDGMENTS
To the Universities of Pennsylvania and Peshawar, the Department of Archaeology and Museums, Government of Pakistan, Professors Abdur Rehman and Farid Khan and the Pakistan Heritage Society, the American Institute of Pakistan Studies, Islamabad, to all participants in the integrated hi temple project; to draftspersons, surveyors, and excavators who contributed their skills; F. Masih, representative of the Department of Archaeology and Museums; Alka Patel, who proffered her camera and a long lens at an important site; to Dr. Elizabeth Stone and two annonymous readers for helpful suggestions about revision; and to Kamal Khan Mumtaz, whose project with young architects and seminar for the Anjuman Mimaran in 1989 was an important inspira- tion, I offer my thanks.
Michael W. Meister Philadelphia 01 January 2010
LIST OF ILLUSTRATIONS
All figures are by the author unless otherwise credited. 1. Maps: a) ‘Map of the Panjab’ (revised from Cunningham 1875,
pl. 1); b) sites in Pakistan’s North-West Frontier and Punjab Provinces.
2. Sirkap (Taxila), Punjab, ‘shrine of the double-headed eagle’ façade, ca. second/third century, detail of architectural models.
3. Taxila, Dharmarjik stpa, outermost kañjr-stone encasement, ca. fifth century.
4. Pandrethan, Kashmir, iva Temple (courtesy: American Institute of Indian Studies, Gurgaon).
5. Kats, Punjab, site from southwest. 6. Kats, Temple B (southwest sub-shrine), southwest. 7. Kats, Temple B, south. 8. Kats, Temple B, reconstruction (from a drawing by P. George). 9. Kfirkot, North-West Frontier Province, site looking east toward
the Mri and Indus River. 10. Kfirkot, site looking west over Temples D, B, and A toward the
citadel. 11. Kfirkot, Temple B, south. 12. Kfirkot, Temple B from east (before clearance and excavation). 13. Kfirkot, Temple A, west. 14. Kfirkot, Temple A from east (after clearance and excavation). 15. Bilot, N.W.F.P., site views of: a) Temples D, E, F–G; b) Temples C
and B (D is behind); c) Temple A. 16. Bilot, Temple D, south. 17. Kfirkot, Temple C, west. 18. Kfirkot, Temple C, detail of north wall and ikhara’s jla pattern. 19. Kfirkot, Temple C, southwest. Note central offset in the wall and
superstructure. 20. Bilot, Temple A, south. Note corner niches. 21. Kfirkot, Kañjar Koth (from Stein 1905). This temple is no longer
standing. For its location, see Hargreaves’ site map (fig. 63). 22. Bilot, Temple D, detail of south wall and ikhara. Note shrine
model on the wall. 23. Bilot, Temple E (NE sub-shrine of Temple D), east.
x list of illustrations
24. Bilot, Temple C, south. Note chambers in foundation platform. 25. Bilot, Temple B, southwest. 26. Mri-Indus, Punjab, Temple A, west. 27. Mri-Indus, Temple B, west. 28. Klar (Sasu-da-Kalra), Punjab, brick temple, southwest. 29. Klar, brick temple, south wall. 30. Kfirkot, Temple D, southeast. 31. Kfirkot, Temple D, detail of ikhara ornament. Note malakas
embedded in the corner web pattern. 32. Kfirkot, Temple B, transition from square sanctum to dome. 33. Kfirkot, Temple A, squinches as transition from sanctum to
dome. 34. Kfirkot Temple A, squinches and dome. 35. Kfirot, Temple D, squinches and dome. 36. Kfirkot, Temple A, sanctum and dome. 37. Kfirkot, Temple A, dome. 38. Klar, brick temple, squinch with wooden beam above. 39. Kfirkot, Temple A, squinch made of kañjr stone. 40. Bilot, Temple D, southeast platform chamber, dome. 41. Nandana, Punjab, Temple A, two-storeyed interior with domed
chambers. Note the upper chamber’s ambulatory corridor. 42. Mri-Indus, Temple B, south. 43. Mri-Indus, Temple A, south. 44. Mri-Indus, Temple A, southeast. 45. Mri-Indus, Temple A, entry vault (partly corbelled, but note the
use of voussoirs and keystone). 46. Mri-Indus, Temple B, entry hall: a) doorway with trilobed arch;
b) dome and squinches seen through arch. 47. Amb, Temple A, sanctum doorway. The image over the door has
been removed since this photograph was taken. 48. Amb, Punjab, Temple A, west. 49. Amb, Temple B, south. 50. Amb, Temple B, south wall, central offset, niche with cinquefoil
arch. 51. Nandana, Temple A, detail of the lower level of the superstructure
with shrine models. 52. Nandana, Temple A, east. 53. Formative shrines in Ngara temple-development: a) Srnth,
Uttar Pradesh, India, shrine model on Gupta-period lintel (Sarnath Museum); b) Kfirkot, Temple B, superstructure, east; c) Deogarh,
list of illustrations xi
Uttar Pradesh, ‘Gupta’ Temple, doorframe, shrine model with malakas; d) Nland, Bihar, India, shrine model on Buddhist votive stpa (courtesy: AIIS, Gurgaon).
54. Kfirkot, Temples B and A, schematic elevations without surface ornament.
55. Bilot, Temple D, south wall, shrine model. 56. Deogarh, ‘Gupta’ Temple, author’s reconstruction of the first level
of the superstructure. 57. Kfirkot, Temple A, ikhara, east. 58. Dhnk, Saurashtra, India, Srya Temple, ikhara, south (courtesy:
AIIS, Gurgaon). 59. Kfirkot, Temple A, ikhara, south. Note small pillarets preserved
at the second level. 60. Bilot, Temple D, detail of southwest corner of ikhara. Note pil-
laret next to malaka. 61. Bhnasar, Saurashtra, India, Temple No. 1, west (courtesy: AIIS,
Gurgaon). 62. Evolution of jla patterns: Bilot Temples D (top), A (center), and
C (bottom). 63. Kfirkot, site map marked ‘M. Sharif Del.’ in the lower left and
hand signed by H. Hargreaves, who first published it, in the lower right (from ASIAR 1921–22, pl. 26).
64. Kfirkot and Bilot forts, site plans (revised from ASIAR 1921–22, pl. 26, and ASI(FC) AR 1920–21, appendix).
65. Kfirkot, Temple C, excavated compound, ground plan, two phases (modified from archaeological field drawing, courtesy: A. Rehman).
66. Kfirkot, Temples E, A, and B site plan following excavation (modified from archaeological field drawing, courtesy: A. Reh- man); sectional elevations of A and B.
67. Evolution of ground plans: Kfirkot Temples B, A, C; Bilot Temple A.
68. Bilot, Temples D and E, ground plans (modified from drawing by F. Iqbal, courtesy: Pakistan Heritage Society).
69. Bilot, Temple A, ground plan (modified from drawing by F. Iqbal, courtesy: Pakistan Heritage Society).
70. Mri-Indus, Temples A and B, ground plans (B modified from drawing by F. Iqbal, courtesy: Pakistan Heritage Society).
71. Kfirkot, Temple C, ground plan with second-phase additions (modified from archaeological field drawing, courtesy: A. Rehman).
xii list of illustrations
72. Kfirkot, Temple C, east, after first phase of excavation. 73. Kfirkot, Temple C, pilaster exposed on northeast face of kañjr
platform compared with an early ‘Asiatic-Corinthian’ capital from Ai-Khanum (drawing by P. Bernard and M. Le Berre, from Wheeler 1968, p. 77). For location, see fig. 71, ‘b’.
74. Kfirkot, Temple C, inserted limestone pilaster base, northeast juncture, prsda and antarla. For location, see fig. 71, ‘c’.
75. Kfirkot, two-storeyed Mri structure, east face (the Mri is built of kañjr tuffa stone, the stairs are limestone).
76. Kfirkot, Temple E, east, after excavation (courtesy: F. Khan). 77. Kfirkot, view of fort, east, from citadel, before excavation. 78. Kfirkot, Temples E, A, and B site after excavation (the century-
old wall and footpath were moved forward to make excavation possible).
79. Kfirkot, Temple E during excavation (courtesy: F. Khan). Insert shows limestone extension before clearance.
80. Kfirkot, Temple E, kañjr platform’s façade, northeast chamber (courtesy: F. Khan).
81. Kfirkot, Temple E, southeast corner after removal of second-phase construction (courtesy: F. Khan). Note the first-phase kañjr stair- way exposed under the limestone stairs to right).
82. Kfirkot, Temple E, kañjr platform’s façade (modified from archaeological field drawing, courtesy: A. Rehman).
83. Kfirkot, Temple E, kañjr platform, façade, northeast sub-shrine (courtesy: F. Khan).
84. Kfirkot, Temples E, A, and B, section of site. 85. Kfirkot, Temple E, compound, remains of east entry and colon-
naded cloister. 86. Kfirkot, Temple E, remains of cloister built against foundations
of Temple B. 87. Kfirkot, Temple A, east, extended stair and ground-level chamber
after excavation. 88. Kfirkot, Temples E, A, and B during excavation, south (courtesy:
F. Khan). 89. Kfirkot, Temple B, remains of platform stairway and side cham-
bers after excavation. 90. Kfirkot, sculpture of an ascetic, excaved in front of Temple B. 91. Kfirkot, sculpture of an ascetic, detail: three heads, ‘skamba’ (the
‘pillow’-like shaft behind).
list of illustrations xiii
92. Mathur, lintel (Government Museum, Mathura), 1st century B.C.E., with a phallic linga, tree, and square railing at the center and a ‘jambu-dvaja’ held to left (courtesy: D. M. Srinivasan 1984: pl. 20).
93. Kfirkot sculpture, showing position of skambha column. 94. Mathur, relief with four auspicious signs from an astamangala
set, ca. second c. C.E. (from Coomaraswamy 1931, pl. 31, fig. 2). 95. Bhrhut, Mathur, and Amarvati reliefs, first centuries B.C.E.
and C.E., differentiating lotus and jackfruit; and the astamangala sign I interpret as the auspicious, upward-growing jackfruit (images compiled from Coomaraswamy 1931; Agrawala 1965).
96. Kfirkot figure interpreted as ‘yoga-mandala’. 97. Compounds: Bilot and Kfirkot. 98. Platform extensions: Bilot and Kfirkot (scales adjusted). 99. Bilot, Temple H, south. 100. Bilot, Temple D, platform, SE corner extension and Temple E,
south, Temple F is in the background. 101. Bilot, Temple D, platform, SE corner with east-facing chamber. 102. Bilot, Temple D, platform, SE corner, vaulted entry to domed
chamber. 103. Bilot, SE corner of Temple D’s compound, conjoined Temples
F–G, south. 104. Bilot, Temples B–C, ground plans (modified from drawing by F.
Iqbal, courtesy: Pakistan Heritage Society). 105. Bilot, Temple B, southwest. 106. Bilot, Temple C, south wall. 107. Salt Range typologies: platform and wall mouldings. 108. Amb, Punjab, Temples A and B, site plan; remains of Temple B’s
platform. 109. Amb, site, Temples B and A, north. 110. Amb, Temples A and B, ground plans. 111. Nandana, Punjab, Temple A, axonometric section, elevation, and
ground- and first-storey ground plans (drawing by H. Choudhury). 112. Nandana, site for south. 113. Pattan Munra, Punjab, ‘minr’ from south. 114. Pattan Munra, brick temple, south. Note remnants of dome. 115. Pattan Munra, brick temple, sanctum, southwest. 116. Mri-Indus, Temple B, entry hall, trefoil arch over doorway,
east.
xiv list of illustrations
117. Pattan Munra, brick temple, south wall, vaulted niche with tre- foil pediment.
118. Salt Range vs. Western Indian moulding typologies. 119. Pattan Munra, brick temple, sanctum, plan reconstructed with
balconies and circumambulatory corridor. 120. Ambulatory plans in Western India: a) Osiñ, Rajasthan, Mahvra
Temple, ca. 800; b) Chittor, Rajasthan, Klikmt Temple, ca. 700. Note different configurations of projecting balconies.
121. Pattan Munra, brick temple (right), detail of damaged tower before conservation (after Vats 1929: pl. 34d); Gyaraspur, Mad- hya Pradesh, Mldev Temple, ca. 950 (left), detail of multi- spired tower (courtesy: AIIS, Gurgaon).
122. Malot, Punjab, temple, south. 123. Malot temple, entry, east. 124. Malot temple, entry gate. 125. Malot temple sanctum, stepped squinch. 126. Malot temple, Ngara shrine model over south-wall chamber. 127. Malot temple, entry gate, peak-roofed shrine model. 128. Malot, ground plans: sanctum and gateway. 129. Malot temple, south wall, monumental chamber. 130. Malot temple, north-wall chamber, images over entry. 131. Mri-Indus site, Temples A and B looking east. 132. Mri-Indus, Temple C, northwest. 133. Mri-Indus, Temple C, reconstruction of ground plan. 134. Mri-Indus, Temple C, north wall. 135. Mri-Indus, Temple C, north wall, bi-level pediment above tre-
foil niche. 136. Mri-Indus, Temple C, superstructure, northwest corner, pillaret
with latina (Ngara) model. 137. Mri-Indus, Temple C, north wall, pediment with Srya image. 138. Saddan (Muzaffargarh), Punjab, cut-brick tomb of Shaikh Sadan
Shahid. 139. Khatti Chor, Punjab, ribt of Khalid Walid, west corridor, cut-
brick mihrb. Note trefoil candral-patterned arch on lower wall.
140. Gumbat, Swt, ruined temple (left); Karwn-balasi, near Kuzai Gumbhaz, Little Pamir, Afghanistan (after Stein 1921: fig. 3).
141. ‘Bhmi-prsda’ storeyed shrines: (left) Kats, Temple B recon- struction; (right) Aihole, Karnataka, Galagnth Temple, 7th cen- tury (courtesy AIIS, Gurgaon).
list of illustrations xv
142. Bodhgay, Bihar, brick Mahbodhi Temple: (left) Kumrahr (Patna), terra-cotta plaque, ca. 3rd century (courtesy: AIIS, Gur- gaon); (right) detail of tower after conservation.
143. Bhtargon, brick temple, ca. 450: (left) ground plan; (right) 19th-century photograph of temple from the northeast (after Cunningham 1880: pl. XIV–XV).
144. Reconstructions of ‘proto-Ngara’ superstructures: (left) Deog- arh, ‘Gupta’ Temple, ca. 525 (ground storey after Vats 1952); (right) Bilot, Temple D, ca. 600.
145. Saurashtra, Gujarat, Maitraka-period superstructures: (left) Bilevara, Bilvantha Temple, ca. 600 (courtesy: AIIS, Gurgaon); Dhnk, Sun Temple, partial reconstruction of ikhara, ca. 700.
146. Amb, three sculptures found by the ASI during clearance of Tem- ple B’s plinth, ca, 1920: ‘Mahdeva’, animal plaque, Narasimha (after Sahni ASNC[H]PR 1921: pl. VII).
147. Kapa (Kabul), Afghanistan, Guldra stpa, plinth façade. 148. Kapa (Kabul), Topdra stpa, (1967 photo: Prof. S.-W. Breckle);
trefoil chamber facing dome (after Lézine 1964: fig. 53). 149. Sequence of arches in the Northwest (top to bottom): Sirkap
(Taxila), Punjab, shrine of the double-headed eagle; Kapa, Guldra stpa; Taxila, Dharmarjik stpa, outer casing; Kfirkot, N.W.F.P., Temple E; Mri-Indus, Punjab, Temple B; Pattan Munra, brick temple, Punjab.
CHAPTER ONE
SALT RANGE AND INDUS TEMPLES
Early medieval temples in the Khisor range of hills along the west bank of the Indus River in the North-West Frontier Province and those on the escarpments and plateau of the Salt Range, between the Indus and Jhelum rivers on the southern edge of the Pothohar table- land in the western Punjab, were visited by Alexander Cunningham in the mid second half of the nineteenth century and again by Aurel Stein at the beginning of the twentieth century; many were placed on a Government List of Protected Monuments in 1904 (ASN[H]PR 1918: Appendix F); and intermittent visits by officers of the Archaeological Survey, Northern (Hindu and Buddhist) and Frontier Circles, occurred through the nineteen twenties. Daya Ram Sahni ibid.: 5, for example, reported on his first visit to temples at Amb that “useful repairs have already been carried out by the Public Works Department . . . though it must be confessed the work has lacked skilled supervision.”
Early Field Reports
Cunningham 1875: 82–85 during his first field tour to the region reported “a number of old Hindu temples in the Salt Range, which all belong to the Kashmirian style of architecture, with its fluted pillars and peculiar trefoil arches”; these he associated with the “last dynasty of the Hindu kings of Gandhra.” Referring specifically to temples at the sites of Amb, Malot, and Kats, he drew the conclusion that these “temples of the Salt Range may be assigned with much probability to the latter half of the 9th century,” in part on the basis of their use of “spiral twisted pillars” not found on most of the temples under discus- sion. Six years later, at his first opportunity to visit the fort of northern Kfirkot, above the west bank of the Indus south of its intersection with the Kurram River, Cunningham 1882: 27 in part revised his ear- lier opinion. Noting that no trefoil arches were found there, he related the architecture of these temples to a period “which succeeded the semi-Greek architecture of the Indo-Sythians.”
2 chapter one
When he visited Kfirkot in 1903, Stein 1905: 10–17 was able to map and carry out an extended survey, discovering a further temple at the foot of the fort “at the point where the narrow boulder-filled ravine which skirts the north-west face of the site debouches into the bed of the Indus.” He observed that the location of this fortress guarded important trade routes from the Indus and the Punjab up the Kurram Valley to Bannu and south along the Indus to Sind. “These little tem- ples,” he wrote, possess “distinct architectural interest as represent- ing a type which meets us in the temples of several ancient sites of the Salt Range, but there with additions characteristic of Kashmirian style which are conspicuously absent at Kafirkot.” Of his newly located temple, Kañjar Koth, however, he added that the “appearance of a rudimentary trefoil arch evolved out of the ‘beehive’ ornament [the jla web of a Ngara tower] may be an indication of Kashmirian influ- ence which in the Salt Range tracts on the opposite side of the Indus seems to have obtained a hold since the seventh century.” Stein would change his mind at a later point, but had been studying Cunningham’s Reports.
In an imprecise repetition of the conflation of Salt Range temples with those of Kashmir Cunningham began, David Ross 1883: 153, in The Land of Five Rivers and Sindh, commented on the pilgrimage site of Kats: “The architecture is the same as in Kashmir, beautiful fluted pillars, trefoil arches, dentils, and pointed roofs.” This is much closer to an appropriate description of Malot than of temple remains at Kats, Kfirkot, and Bilot. In The Ancient Monuments, Temples and Sculptures of India, James Burgess 1897: 28 chose to illustrate, of all the temples in the Salt Range, only the one at Malot. He remarked that the “pyramidal sikhara or spire has long ago disappeared,” but made no mention of the curvilinear North Indian Ngara shrine models that decorate the Malot temple’s walls, although he gave good illustrations of them in his plates. He followed his illustrations of Malot with a set of Kashmir temples, then of an atypical sixth-century temple at Gop in Saurashtra that has a pyramidal roof, of which he felt the need to remark “the roof of the sikhara looks not unlike that of a Kashmiri temple” (for an analysis of the temple at Gop, see Meister/Dhaky/Deva 1988: 177–179).
Part of the problem early in the twentieth century was that few offi- cers had experience of all the monuments, and of the few they had vis- ited, it was often many years before they returned. Cunningham 1882: 34 wrote of Amb: “The temples are all of the Kashmiri style, but they
salt range and indus temples 3
are almost certainly of late date [ca. 800–950], as all the arches have cinquefoil instead of trefoil heads, which is the only form in Kashmir.” Sahni ASN(H)PR 1918: 5, when reporting on conservation of tem- ples at Amb, still followed Cunningham’s analysis: “Like the temple at Malot, these temples are constructed in a semi-Kashmiri style with cinquefoil arches instead of trefoil ones as is the case in the temples of Kashmir.” He then observed, however, that the “spires of the temple are in the usual North Indian style.”
If Cunningham in 1872–3 took the presence of a trefoil arch to be the marker of ‘Kashmiri style’ in Salt Range temples, when he…
Related Documents
